The non-Hodgkin lymphomas (NHL) are a heterogeneous group of lymphoproliferative malignancies with differing patterns of behavior and responses to treatment.

Like Hodgkin lymphoma, NHL usually originates in lymphoid tissues and can spread to other organs. NHL, however, is much less predictable than Hodgkin lymphoma and has a far greater predilection to disseminate to extranodal sites. The prognosis depends on the histological type, stage, and treatment.

Incidence and Mortality

Estimated new cases and deaths from NHL in the United States in 2023:

• New cases: 80,550.
• Deaths: 20,180.

Anatomy

NHL usually originates in lymphoid tissues.

Prognosis and Survival

NHL can be divided into two prognostic groups: the indolent lymphomas and the aggressive lymphomas.

Indolent NHL types have a relatively good prognosis, with a median survival as long as 20 years, but they usually are not curable in advanced clinical stages. Early-stage (stage I and stage II) indolent NHL can be effectively treated with radiation therapy alone. Most of the indolent types are nodular (or follicular) in morphology.

The aggressive type of NHL has a shorter natural history, but a significant number of these patients can be cured with intensive combination chemotherapy regimens.

In general, with modern treatment of patients with NHL, the overall survival rate at 5 years is over 60%. More than 50% of patients with aggressive NHL can be cured. Most relapses occur in the first 2 years after therapy. The risk of late relapse is higher in patients who manifest both indolent and aggressive histologies.

While indolent NHL is responsive to immunotherapy, radiation therapy, and chemotherapy, a continuous rate of relapse is usually seen in advanced stages. However, patients can often be re-treated with considerable success if the disease histology remains low grade. Patients who present with or convert to aggressive forms of NHL may have sustained complete remissions with combination chemotherapy regimens or aggressive consolidation with marrow or stem cell support.

Late effects of treatment of non-Hodgkin lymphoma (NHL) have been observed. Impaired fertility may occur after exposure to alkylating agents. For as many as three decades after diagnosis, patients are at a significantly elevated risk of developing second primary cancers, especially the following:

• Lung cancer.
• Brain cancer.
• Kidney cancer.
• Bladder cancer.
• Melanoma.
• Hodgkin lymphoma.
• Acute nonlymphocytic leukemia.

Left ventricular dysfunction was a significant late effect in long-term survivors of high-grade NHL who received more than 200 mg/m² of doxorubicin.

Myelodysplastic syndrome and acute myelogenous leukemia are late complications of myeloablative therapy with autologous bone marrow or peripheral blood stem cell support, as well as conventional chemotherapy-containing alkylating agents. Most of these patients show clonal hematopoiesis even before the transplantation, suggesting that the hematologic injury usually occurs during induction or reinduction chemotherapy. With a median 10-year follow-up after autologous bone marrow transplantation (BMT) with conditioning using cyclophosphamide and total-body radiation therapy, the incidence of a second malignancy was 21% in a series of 605 patients, and 10% of those were solid tumors.

Successful pregnancies with children born free of congenital abnormalities have been reported in young women after autologous BMT. Late-occurring venous thromboembolism can occur after allogeneic or autologous BMT.

Long-term impaired immune health was evaluated in a retrospective cohort study of 21,690 survivors of diffuse large B-cell lymphoma from the California Cancer Registry. Elevated incidence rate ratios were found up to 10 years later for pneumonia (10.8-fold), meningitis (5.3-fold), immunoglobulin deficiency (17.6-fold), and autoimmune cytopenias (12-fold). Similarly, there are impaired humoral responses to COVID-19 virus vaccination in patients with lymphoma who receive B-cell–directed therapies.

Some patients have osteopenia or osteoporosis at the start of therapy; bone density may worsen after therapy for lymphoma.

A pathologist should be considered for consultation before a biopsy because some studies require special preparation of tissue (e.g., frozen tissue). Knowledge of cell surface markers and immunoglobulin and T-cell receptor gene rearrangements may help with diagnostic and therapeutic decisions. The clonal excess of light-chain immunoglobulin may differentiate malignant from reactive cells. Since the prognosis and the approach to treatment are influenced by histopathology, outside biopsy specimens should be carefully reviewed by a hematopathologist who is experienced in diagnosing lymphomas. Although lymph node biopsies are recommended whenever possible, sometimes immunophenotypic data are sufficient to allow diagnosis of lymphoma when fine-needle aspiration cytology is preferred.

Historical Classification Systems

Historically, uniform treatment of patients with non-Hodgkin lymphoma (NHL) has been hampered by the lack of a uniform classification system. In 1982, results of a consensus study were published as the Working Formulation. The Working Formulation combined results from six major classification systems into one classification. This allowed comparison of studies from different institutions and countries. The Rappaport classification, which also follows, is no longer in common use.

Table 1. Historical Classification Systems for Non-Hodgkin Lymphoma (NHL)

Current Classification Systems

As the understanding of NHL has improved and as the histopathological diagnosis of NHL has become more sophisticated with the use of immunologic and genetic techniques, a number of new pathological entities have been described. In addition, the understanding and treatment of many of the previously described pathological subtypes have changed. As a result, the Working Formulation has become outdated and less useful to clinicians and pathologists. Thus, European and American pathologists have proposed a new classification, the Revised European American Lymphoma (REAL) classification. Since 1995, members of the European and American Hematopathology societies have been collaborating on a new World Health Organization (WHO) classification, which represents an updated version of the REAL system.

The WHO modification of the REAL classification recognizes three major categories of lymphoid malignancies based on morphology and cell lineage: B-cell neoplasms, T-cell/natural killer (NK)-cell neoplasms, and Hodgkin lymphoma (HL). Both lymphomas and lymphoid leukemias are included in this classification because both solid and circulating phases are present in many lymphoid neoplasms and distinction between them is artificial. For example, B-cell chronic lymphocytic leukemia (CLL) and B-cell small lymphocytic lymphoma are simply different manifestations of the same neoplasm, as are lymphoblastic lymphomas and acute lymphocytic leukemias. Within the B-cell and T-cell categories, two subdivisions are recognized: precursor neoplasms, which correspond to the earliest stages of differentiation, and more mature differentiated neoplasms.

Updated REAL/WHO classification

B-cell neoplasms

1. Precursor B-cell neoplasm: precursor B-acute lymphoblastic leukemia/lymphoblastic lymphoma (LBL).
2. Peripheral B-cell neoplasms.
   1. B-cell CLL/small lymphocytic lymphoma.
   2. B-cell prolymphocytic leukemia.
   3. Lymphoplasmacytic lymphoma/immunocytoma.
   4. Mantle cell lymphoma.
   5. Follicular lymphoma.
   6. Extranodal marginal zone B-cell lymphoma of mucosa-associated lymphatic tissue (MALT) type.
   7. Nodal marginal zone B-cell lymphoma (± monocytoid B-cells).
   8. Splenic marginal zone lymphoma (± villous lymphocytes).
   9. Hairy cell leukemia.
   10. Plasmacytoma/plasma cell myeloma.
   11. Diffuse large B-cell lymphoma.
   12. Burkitt lymphoma.

T-cell and putative NK-cell neoplasms

1. Precursor T-cell neoplasm: precursor T-acute lymphoblastic leukemia/LBL.
2. Peripheral T-cell and NK-cell neoplasms.
   1. T-cell CLL/prolymphocytic leukemia.
   2. T-cell granular lymphocytic leukemia.
   3. Mycosis fungoides (including Sézary syndrome).
   4. Peripheral T-cell lymphoma, not otherwise characterized.
   5. Hepatosplenic gamma/delta T-cell lymphoma.
   6. Subcutaneous panniculitis-like T-cell lymphoma.
   7. Angioimmunoblastic T-cell lymphoma.
   8. Extranodal T-/NK-cell lymphoma, nasal type.
   9. Enteropathy-type intestinal T-cell lymphoma.
   10. Adult T-cell lymphoma/leukemia (human T-lymphotrophic virus 1+).
   11. Anaplastic large cell lymphoma, primary systemic type.
   12. Anaplastic large cell lymphoma, primary cutaneous type.
   13. Aggressive NK-cell leukemia.

HL

1. Nodular lymphocyte-predominant HL.
2. Classical HL.
   1. Nodular sclerosis HL.
   2. Lymphocyte-rich classical HL.
   3. Mixed-cellularity HL.
   4. Lymphocyte-depleted HL.

The REAL classification encompasses all the lymphoproliferative neoplasms.

Indolent non-Hodgkin lymphoma (NHL) includes the following subtypes:

• Follicular lymphoma.
• Lymphoplasmacytic lymphoma (Waldenström macroglobulinemia).
• Marginal zone lymphoma.
• Splenic marginal zone lymphoma.
• Primary cutaneous anaplastic large cell lymphoma.

Follicular Lymphoma

Follicular lymphoma comprises 20% of all NHL and as many as 70% of the indolent lymphomas reported in American and European clinical trials. Most patients with follicular lymphoma are age 50 years and older and present with widespread disease at diagnosis. Nodal involvement is most common and is often accompanied by splenic and bone marrow disease. Rearrangement of the BCL2 gene is present in more than 90% of patients with follicular lymphoma; overexpression of the BCL2 protein is associated with the inability to eradicate the lymphoma by inhibiting apoptosis.

Prognosis

Despite the advanced stage, the median survival ranges from 8 to 15 years, leading to the designation of being indolent. Patients with advanced-stage follicular lymphoma are not cured with current therapeutic options. The rate of relapse is fairly consistent over time, even in patients who have achieved complete responses to treatment. Watchful waiting, i.e., the deferring of treatment until the patient becomes symptomatic, is an option for patients with advanced-stage follicular lymphoma. An international index for follicular lymphoma (i.e., the Follicular Lymphoma International Prognostic Index ) identified five significant risk factors prognostic of overall survival (OS):

1. Age (≤60 years vs. >60 years).
2. Serum lactate dehydrogenase (LDH) (normal vs. elevated).
3. Stage (stage I or stage II vs. stage III or stage IV).
4. Hemoglobin level (≥120 g/L vs. 120 g/L).
5. Number of nodal areas (≤4 vs. >4).

Patients with one risk factor or none have an 85% 10-year survival rate, and three or more risk factors confer a 40% 10-year survival rate. In a revised FLIPI-2, an elevated beta-2-microglobulin and lymph node size of more than 6 cm are proposed prognostic factors instead of serum LDH and the number of nodal areas. Although the FLIPI and FLIPI-2 indices can predict progression-free survival (PFS) and OS, the scores cannot be used to establish the need for therapy, nor can they be used to predict response to therapy. The primary use of FLIPI or FLIPI-2 is to assure a balance of prognostic factors or to define entry requirements in randomized clinical trials. Individuals with an adverse FLIPI score may well benefit from watchful waiting or may respond well to initial therapy. An alternative prognostic index using only beta-2-microglobulin and initial bone marrow involvement (PRIMA-PI) has the disadvantage of requiring an invasive test not usually required outside the context of a clinical trial. An alternative prognostic index using only noninvasive clinical variables outperformed FLIPI, FLIPI-2, and PRIMA-PI, using data from immunochemotherapy trials.

Three retrospective analyses, including one pooled analysis of 5,225 patients in 13 randomized clinical trials, identified a high-risk group that had a 50% OS rate at 5 years when relapses occurred within 24 months of induction chemoimmunotherapy. A fourth retrospective analysis of 296 patients who received bendamustine-rituximab found a 2-year OS rate of 38% (95% confidence interval , 20%−55%) among those with progression of disease before 24 months (POD24). Most of these patients (76%) had transformed disease (histological progression to diffuse large B-cell lymphoma ). These higher-risk POD24 patients represent a target population for clinical trials.

Follicular, small-cleaved cell lymphoma and follicular mixed small-cleaved and large cell lymphoma do not have reproducibly different disease-free survival or OS.

Therapeutic approaches

Because of the often-indolent clinical course and the lack of symptoms in some patients with follicular lymphoma, watchful waiting remains a standard of care during the initial encounter and for patients with slow asymptomatic relapsing disease. When therapy is required, numerous therapeutic options may be employed in varying sequences with an OS equivalence at 5 to 10 years. Rituximab can be given alone or in combination with various chemotherapy options. Rituximab can also be combined with the immunomodulating-agent lenalidomide to avoid the short- and long-term toxicities of cytotoxic agents. Another anti–CD20 monoclonal antibody, obinutuzumab, can be administered with combination chemotherapy. Inhibitors of phosphatidylinositol 3-kinase (PI3K) are also effective in patients with relapsed or refractory disease. CD19-directed chimeric antigen receptor (CAR) T cells may be used In patients who have disease progression after two or more prior lines of therapy. Mosunetuzumab, a bispecific CD20-directed CD3 T-cell engager, may also be used in this setting. Consolidation therapy for relapsed disease after reinduction therapy using autologous stem cell transplant (SCT) or allogeneic SCT can be considered.

Follicular lymphoma in situ and primary follicular lymphoma of the duodenum are particularly indolent variants that rarely progress and rarely require therapy. A so-called pediatric-type nodal follicular lymphoma has indolent behavior and rarely recurs; adult patients with this histological variant are characterized by a lack of BCL2 rearrangement in conjunction with a Ki-67 proliferation index greater than 30% and a localized stage I presentation.

Patients with indolent lymphoma may experience a relapse with a more aggressive histology. If the clinical pattern of relapse suggests that the disease is behaving in a more aggressive manner, a biopsy can be performed, if feasible. Documentation of conversion to a more aggressive histology requires an appropriate change to a therapy applicable to that histological type. Rapid growth or discordant growth between various disease sites may indicate a histological conversion. The risk of histological transformation was 30% by 10 years in a retrospective review of 325 patients from diagnosis between 1972 and 1999. In this series, high-risk factors for subsequent histological transformation were advanced stage, high-risk FLIPI, and expectant management (as opposed to treatment being initiated at diagnosis). The 5-year OS rate was more than 50% for patients who had biopsy-proven, aggressive-histology transformation in several multicenter cohort studies employing rituximab plus anthracycline or platinum-based chemotherapy, or similar therapy followed by autologous or allogeneic SCT.

In a prospective nonrandomized study, at a median follow-up of 6.8 years, 379 (14%) of 2,652 patients subsequently transformed to a more aggressive histology after an initial diagnosis of follicular lymphoma. The median OS after subsequent transformation was 5 years; however, among 47 patients with evidence of transformation in conjunction with follicular lymphoma at the time of initial diagnosis, the OS was no worse than that of the other nontransformed patients (5-year OS rate, 88%; 95% CI, 74%–95%).

Lymphoplasmacytic Lymphoma (Waldenström Macroglobulinemia)

Lymphoplasmacytic lymphoma is usually associated with a monoclonal serum paraprotein of immunoglobulin M (IgM) type (Waldenström macroglobulinemia). Most patients have bone marrow, lymph node, and splenic involvement, and some patients may develop hyperviscosity syndrome. Most patients with Waldenström macroglobulinemia carry the MYD88 mutation, which some pathologists consider pathognomonic for the disease. Other lymphomas may also be associated with serum paraproteins. Patients with lymphoplasmacytic lymphoma should be checked for associated hepatitis C virus infection.

Asymptomatic patients can be monitored for evidence of disease progression without immediate need for chemotherapy.

Prognostic factors associated with symptoms requiring therapy include the following:

• Age 70 years or older.
• Beta-2-microglobulin of 3 mg/dL or more.
• Increased serum LDH.

Therapeutic approaches

The management of lymphoplasmacytic lymphoma is similar to that of other low-grade lymphomas, especially diffuse, small lymphocytic lymphoma/chronic lymphocytic leukemia. If the viscosity relative to water is greater than four, the patient may have manifestations of hyperviscosity. Plasmapheresis is useful for temporary, acute symptoms such as retinopathy, congestive heart failure, and central nervous system (CNS) dysfunction but can be combined with chemotherapy for prolonged disease control. Symptomatic patients with a serum viscosity of four or lower are usually started directly on chemoimmunotherapy or biologically directed therapies. Therapy may be required to correct hemolytic anemia in patients with chronic cold agglutinin disease; rituximab, bendamustine, and steroids are often used. Occasionally, a heated room is required for patients whose cold agglutinins become activated by even minor chilling. Sutimlimab, an immunoglobulin G4 monoclonal antibody that selectively inhibits the complement pathway at C15, can reduce hemolysis when therapies directed at the lymphoplasmacytic lymphoma prove ineffective.

First-line regimens include rituximab and ibrutinib (a Bruton tyrosine kinase inhibitor), rituximab alone, the nucleoside analogues, and alkylating agents, either as single agents or as part of combination chemotherapy. In a randomized prospective trial, 150 symptomatic patients (including previously untreated and relapsing patients) received either ibrutinib and rituximab or rituximab and a placebo. With a median follow-up of 50 months, the 4.5 year PFS rate favored the ibrutinib-and-rituximab arm (68%; 95% CI, 55%–78%) versus the rituximab-and-placebo arm (25%; 95% CI, 15%–37%) (hazard ratio, 0.25; 95% CI, 0.15–0.42; P .0001), and the OS rate at 30 months was no different in the two arms (92%–94%). Zanubrutinib, another BTK inhibitor, was compared with ibrutinib in a randomized prospective clinical trial of 164 patients with relapsed disease and 38 previously untreated patients. With a median follow-up of 18 months, the PFS rate was the same in both groups at 84%. The zanubrutinib group had fewer cases of atrial fibrillation (11 vs. 1) and 50% fewer cases of hypertension (statistics not provided). BTK inhibition with ibrutinib allowed all 13 patients with cold-antibody–mediated autoimmune hemolytic anemia and acrocyanosis to attain clinical remission regardless of underlying pathology or MYD88 status.

Previously untreated patients who received rituximab had response rates of 60% to 80%, but close monitoring of the serum IgM is required because of a sudden rise in this paraprotein at the start of therapy. The rise of IgM after rituximab can be avoided with the concomitant use of an alkylating agent, such as cyclophosphamide or the proteosome inhibitors bortezomib or ixazomib. A combination of bortezomib, dexamethasone, and rituximab has been used with avoidance of an IgM rebound. Previously untreated patients with lymphoplasmacytic lymphoma who received the nucleoside analogues cladribine and fludarabine have shown similar response rates. Patients who received single-agent alkylators, bendamustine, bortezomib, venetoclax, and combination chemotherapy with or without rituximab also show similar response rates. In the rare case of lymphoplasmacytic lymphoma involving the CNS (Bing-Neel syndrome), ibrutinib resulted in an 85% response rate in an anecdotal series of 28 patients.

Myeloablative therapy with autologous or allogeneic hematopoietic stem cell support is under clinical evaluation. Candidates for this approach should avoid long-term use of alkylating agents or purine nucleoside analogues, which can deplete hematopoietic stem cells or predispose patients to myelodysplasia or acute leukemia. After relapse from alkylating-agent therapy, 92 patients with lymphoplasmacytic lymphoma were randomly assigned to either fludarabine or cyclophosphamide, doxorubicin, and prednisone. Although relapse-free survival favored fludarabine (median duration of 19 months vs. 3 months, P .01), no difference was observed in OS.

Marginal Zone Lymphoma

When marginal zone lymphomas involve the nodes, they are called monocytoid B-cell lymphomas or nodal marginal zone B-cell lymphomas, and when they involve extranodal sites (e.g., gastrointestinal tract, thyroid, lung, breast, orbit, and skin), they are called mucosa-associated lymphatic tissue (MALT) lymphomas. Splenic marginal zone lymphoma is a distinct clinical entity that usually presents with massive splenomegaly. A variant form of MALT lymphoma is known as immunoproliferative small intestinal disease (IPSID). A prognostic index for all of the marginal zone lymphomas has three adverse prognostic factors: age 70 years or older, stage III or stage IV disease, and high LDH level. Fewer than 10% of patients transform to a higher-grade lymphoma; risk factors in one retrospective review included elevated LDH, more than four nodal sites at the time of initial diagnosis of marginal zone lymphoma, and failure to achieve complete response after initial treatment.

Gastric MALT

Many patients have a history of autoimmune disease, such as Hashimoto thyroiditis or Sjögren syndrome, or of Helicobacter gastritis. Most patients present with stage I or stage II extranodal disease, which is most often in the stomach. Treatment of Helicobacter pylori infection may resolve most cases of localized gastric involvement. After standard antibiotic regimens, 50% of patients show resolution of gastric MALT by endoscopy after 3 months. Other patients may show resolution after 12 to 18 months of observation. Of the patients who attain complete remission, 30% demonstrate monoclonality by immunoglobulin heavy chain rearrangement on stomach biopsies with a 5-year median follow-up. The clinical implication of this finding is unknown. Translocation t(11;18) in patients with gastric MALT predicts for poor response to antibiotic therapy, for H. pylori–negative testing, and for poor response to oral alkylator chemotherapy. Stable asymptomatic patients with persistently positive biopsies have been successfully followed on a watchful waiting approach until disease progression. Patients who progress are treated with radiation therapy, rituximab, surgery (total gastrectomy or partial gastrectomy plus radiation therapy), chemotherapy, or combined-modality therapy. The use of endoscopic ultrasonography may help clinicians to follow responses in these patients. Four case series encompassing more than 100 patients with stage IE or IIE DLBCL with or without associated MALT (but H. pylori-positive) reported durable complete remissions in more than 50% of the patients after treatment of H. pylori.

Extragastric MALT

Localized involvement of other sites can be treated with radiation or surgery. Patients with extragastric MALT lymphoma have a higher relapse rate than patients with gastric MALT lymphoma in some series, with relapses many years and even decades later. Many of these recurrences involve different MALT sites than the original location. When disseminated to lymph nodes, bone marrow, or blood, this entity behaves like other low-grade lymphomas. A prospective, randomized trial of 401 patients with nongastric, extranodal MALT compared chlorambucil alone versus rituximab plus chlorambucil versus rituximab alone. With a median follow-up of 7.4 years, the event-free survival was better for the rituximab-plus-chlorambucil arm (68%) than for the rituximab-alone arm (51%) and for the chlorambucil-alone arm (50%) (P = .0009). However, the 5-year OS rate was 90% in all arms. For patients with ocular adnexal MALT, antibiotic therapy using doxycycline that targeted Chlamydia psittaci resulted in durable remissions for almost half of the patients in a review of the literature that included 131 patients. These responses to doxycycline are mainly seen in Italian trials and less often in trials conducted in other geographic sites. Large B-cell lymphomas of MALT sites are classified and treated as diffuse large cell lymphomas. A large, retrospective review of primary ocular adnexal MALT found that after 10 years of follow-up, 4% of stage I patients treated with radiation therapy transformed to DLBCL, and 3% of them developed CNS involvement.

Nodal marginal zone lymphoma

Patients with nodal marginal zone lymphoma (monocytoid B-cell lymphoma) are treated with the same paradigm of watchful waiting or therapies as described for follicular lymphoma. Similar to follicular lymphoma, patients with POD24 who required initiation of therapy had a worse prognosis (53% 3-year OS rate) than did the patients without POD24 (95% 3-year OS rate). Among patients with concomitant hepatitis C virus (HCV) infection, 40% to 60% attained a complete or partial remission after loss of detectable HCV RNA with antiviral treatment.

Mediterranean abdominal lymphoma

The disease variously known as Mediterranean abdominal lymphoma, heavy–chain disease, or IPSID, which occurs in young adults in eastern Mediterranean countries, is another version of MALT lymphoma, which responds to antibiotics in its early stages. Campylobacter jejuni has been identified as one of the bacterial species associated with IPSID, and antibiotic therapy may result in remission of the disease.

Splenic marginal zone lymphoma

Splenic marginal zone lymphoma is an indolent lymphoma that is marked by massive splenomegaly and peripheral blood and bone marrow involvement, usually without adenopathy. This type of lymphoma is otherwise known as splenic lymphoma with villous lymphocytes. Splenectomy may result in prolonged remission.

Management is similar to that of other low-grade lymphomas and usually involves rituximab alone or rituximab in combination with purine analogues or alkylating agent chemotherapy. Splenic marginal zone lymphoma responds less well to chemotherapy, which would ordinarily be effective for chronic lymphocytic leukemia. Among small numbers of patients with splenic marginal zone lymphoma (splenic lymphoma with villous lymphocytes) and infection with HCV, most attained a complete or partial remission after loss of detectable HCV RNA with treatment using interferon-alpha with or without ribavirin.; In contrast, no responses to interferon were seen in six HCV-negative patients.

Primary Cutaneous Anaplastic Large Cell Lymphoma

Primary cutaneous anaplastic large cell lymphoma presents in the skin only with no preexisting lymphoproliferative disease and no extracutaneous sites of involvement. Patients with this type of lymphoma encompass a spectrum ranging from clinically benign lymphomatoid papulosis, marked by localized nodules that may regress spontaneously, to a progressive and systemic disease requiring aggressive doxorubicin-based combination chemotherapy. This spectrum has been called the primary cutaneous CD30-positive T-cell lymphoproliferative disorder.

Patients with localized disease usually undergo radiation therapy. With more disseminated involvement, watchful waiting or doxorubicin-based combination chemotherapy is applied.

Aggressive non-Hodgkin lymphoma (NHL) includes the following subtypes:

• Diffuse large B-cell lymphoma.
• Mediastinal large B-cell lymphoma (primary mediastinal large B-cell lymphoma).
• Follicular large cell lymphoma.
• Anaplastic large cell lymphoma.
• Extranodal NK–/T-cell lymphoma.
• Lymphomatoid granulomatosis.
• Angioimmunoblastic T-cell lymphoma.
• Peripheral T-cell lymphoma.
• Enteropathy-type intestinal T-cell lymphoma.
• Intravascular large B-cell lymphoma (intravascular lymphomatosis).
• Burkitt lymphoma/diffuse small noncleaved-cell lymphoma.
• Lymphoblastic lymphoma.
• Adult T-cell leukemia/lymphoma.
• Mantle cell lymphoma.
• Polymorphic posttransplantation lymphoproliferative disorder.
• True histiocytic lymphoma.
• Primary effusion lymphoma.
• Plasmablastic lymphoma.

Diffuse Large B-cell Lymphoma

Diffuse large B-cell lymphoma (DLBCL) is the most common type of NHL and comprises 30% of newly diagnosed cases. Most patients present with rapidly enlarging masses, often with both local and systemic symptoms (designated B symptoms with fever, recurrent night sweats, or weight loss). For more information about weight loss, see Nutrition in Cancer Care.

Some cases of large B-cell lymphoma have a prominent background of reactive T cells and often of histiocytes, so-called T-cell/histiocyte-rich large B-cell lymphoma. This subtype of large cell lymphoma has frequent liver, spleen, and bone marrow involvement; however, the outcome is equivalent to that of similarly staged patients with DLBCL. Some patients with DLBCL at diagnosis have a concomitant indolent small B-cell component; while overall survival (OS) appears similar after multidrug chemotherapy, there is a higher risk of indolent relapse.

Prognosis

Most patients with localized disease are curable with combined-modality therapy or combination chemotherapy alone. Among patients with advanced-stage disease, 50% are cured with doxorubicin-based combination chemotherapy and rituximab, typically R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone).

The National Comprehensive Cancer Network International Prognostic Index (IPI) for aggressive NHL (diffuse large cell lymphoma) identifies the following five significant risk factors prognostic of OS and their associated risk scores:

• Age.
  • 40 years: 0.
  • 41–60 years: 1.
  • 61–75 years: 2.
  • >75 years: 3.
• Stage III/IV: 1.
• Performance status 2/3/4: 1.
• Serum lactate dehydrogenase (LDH).
  • Normalized: 0.
  • >1x–3x: 1.
  • >3x: 2.
• Number of extranodal sites ≥2: 1.

Risk scores:

• Low (0 or 1): 5-year OS rate, 96%; progression-free survival (PFS) rate, 91%.
• Low intermediate (2 or 3): 5-year OS rate, 82%; PFS rate, 74%.
• High intermediate (4 or 5): 5-year OS rate, 64%; PFS rate, 51%.
• High (>6): 5-year OS rate, 33%; PFS rate, 30%.

Age-adjusted and stage-adjusted modifications of this IPI are used for younger patients with localized disease. Shorter intervals of time between diagnosis and treatment appear to be a surrogate for poor prognostic biological factors.

The BCL2 gene and rearrangement of the MYC gene or dual overexpression of the MYC gene, or both, confer a particularly poor prognosis. Dose-intensive therapies, infusional therapies, and stem cell transplantation (SCT) consolidation are being explored in this high-risk group. A retrospective review evaluated 159 patients with previously untreated DLBCL who underwent double-hit genetic testing by fluorescence in situ hybridization (FISH) and achieved complete response (CR). The induction therapy did not alter 3-year relapse-free survival or OS when autologous SCT was employed.

In a retrospective review of 117 patients with relapsed or refractory DLBCL who underwent autologous SCT, the 4-year OS rate was 25% for double-hit lymphomas (rearrangement of BCL2 and MYC), 61% for double-expressor lymphomas (no rearrangement, but increased expression of BCL2 and MYC), and 70% for patients without these features. Patients at high risk of relapse may be considered for clinical trials.

Molecular profiles of gene expression using DNA microarrays may help to stratify patients in the future for therapies directed at specific targets and to better predict survival after standard chemotherapy. For example, true anaplastic lymphoma kinase (ALK)-positive large B-cell lymphomas are extremely rare, and they do not respond well to conventional R-CHOP therapy. Anecdotal responses to ALK inhibitors like lorlatinib or alectinib have been reported. Patients who have DLBCL with coexpression of CD20 and CD30 may define a subgroup with a unique molecular signature, a more favorable prognosis, and possible therapeutic implication for the use of anti-CD30–specific therapy, such as brentuximab vedotin. Patients with DLBCL who are event-free after 2 years have a subsequent OS equivalent to that of the age- and sex-matched general population.

Central nervous system (CNS) prophylaxis

The CNS-IPI tool predicts which patients have a CNS relapse risk exceeding 10%. It was developed by the German Lymphoma Study Group and validated by the British Columbia Cancer Agency database. The presence of four to six of the CNS-IPI risk factors (age >60 years, performance status ≥2, elevated LDH, stage III or IV disease, >1 extranodal site, or involvement of the kidneys or adrenal glands) was used to define a high-risk group for CNS recurrence (a 12% risk of CNS involvement by 2 years).

CNS prophylaxis (usually with four to six doses of intrathecal methotrexate) is often recommended for patients with testicular involvement. A retrospective analysis of the German RICOVER studies compared intrathecal methotrexate with no prophylaxis in patients with DLBCL. This study was completed during the R-CHOP treatment era. With the possible exception of patients with testicular involvement, the analysis showed that intrathecal methotrexate did not reduce the risk of CNS disease. Some clinicians employ high-dose intravenous (IV) methotrexate (usually four doses) as an alternative to intrathecal therapy because drug delivery is improved and patient morbidity is decreased. A retrospective study evaluated 1,162 patients from 21 U.S. academic centers where 77% received intrathecal methotrexate, 20% received high-dose IV methotrexate, and 3% received both sequentially (because of toxicity). There was no difference in CNS relapse rates between patients who received intrathecal methotrexate or high-dose IV methotrexate (5.4% vs. 6.8%, P = .40). Testicular involvement, nongerminal center subtype, and high extranodal involvement predicted increased CNS relapse regardless of the route of prophylaxis. Two retrospective studies evaluating high-dose methotrexate in patients with high-risk DLBCL also showed no improvement in CNS relapse rate. Patients deemed at high risk for CNS relapse (e.g., patients with four to six CNS-IPI risk factors) often receive intrathecal methotrexate or high-dose IV methotrexate, but the lack of confirmatory randomized studies calls this standard into question and shows an urgent need for better therapeutics verified in clinical trials. Patients with testicular involvement are an exception, as they show benefit from intrathecal or high-dose IV methotrexate.

The addition of rituximab to cyclophosphamide, doxorubicin, vincristine, prednisone (CHOP)-based regimens has significantly reduced the risk of CNS relapse in retrospective analyses. Patients with CNS dissemination at diagnosis or at relapse usually receive rituximab and high doses of methotrexate and/or cytarabine followed by autologous SCT, but this approach has not been assessed in randomized trials.

Primary Mediastinal Large B-cell Lymphoma

Primary mediastinal (thymic) large B-cell lymphoma (PMBCL) is a subset of DLBCL with molecular characteristics that are most similar to nodular-sclerosing Hodgkin lymphoma (HL). Mediastinal lymphomas with features intermediate between primary mediastinal B-cell lymphoma and nodular-sclerosing HL are called mediastinal gray-zone lymphomas. Patients are usually female and young (median age, 30–40 years). Patients present with a locally invasive anterior mediastinal mass that may cause respiratory symptoms or superior vena cava syndrome.

Prognosis and therapy are the same as for other comparably staged patients with DLBCL. Uncontrolled, phase II studies employing dose-adjusted R-EPOCH (etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin plus rituximab) or R-CHOP show high cure rates while avoiding any mediastinal radiation. These results suggest that patients who receive R-CHOP–based regimens may avoid the serious long-term complications of radiation therapy when given with chemotherapy. Posttreatment fluorine F 18-fludeoxyglucose (18F-FDG) positron emission tomography–computed tomography (PET-CT) scans are controversial; it remains unclear if PET scans can reliably identify patients who can take or omit radiation therapy consolidation.

A retrospective review of 109 patients with PMBCL showed that 63% had a negative end-of-treatment PET-CT (EOT-PET-CT) (Deauville score 1–3). No radiation therapy was offered and the 5-year time-to-progression rate (similar to disease-free survival, but restricted to lymphoma relapse) was 90%, and the 5-year OS rate was 97%. Patients with a positive EOT-PET-CT received radiation therapy consolidation. It is unclear from this study whether those patients might have done just as well without radiation therapy. Clinicians may follow improvement over time for Deauville 4 scores on EOT-PET-CT scans as an alternative to radiation therapy, but this has not been studied in a clinical trial.

In situations where mediastinal radiation therapy would encompass the left side of the heart or would increase breast cancer risk in young female patients, proton therapy may be considered to reduce radiation dose to organs at risk. For more information, see the Superior Vena Cava Syndrome section in Cardiopulmonary Syndromes.

Because PMBCL is characterized by high expression of programmed death-ligand 1 (PD-L1) and variable expression of CD30, a phase II study evaluated nivolumab plus brentuximab vedotin in 30 patients with relapsed disease. With a median follow-up of 11.1 months, the objective response rate was 73% (95% CI, 54%−88%). Similarly, phase I and II trials of pembrolizumab in 74 patients with relapsed or refractory disease showed an objective response rate of 45% to 48%. The median duration of response was not reached for the 21 patients with a median follow-up of 29 months or for the 53 patients with a median follow-up of 12.5 months.

Among those who had received two prior lines of therapy, more than one-half of patients who received CAR T-cell therapy with lisocabtagene maraleucel had disease response.

Follicular Large Cell Lymphoma

Prognosis

The natural history of follicular large cell lymphoma remains controversial. While there is agreement about the significant number of long-term disease-free survivors with early-stage disease, the curability of patients with advanced disease (stage III or stage IV) remains uncertain. Some groups report a continuous relapse rate similar to the other follicular lymphomas (a pattern of indolent lymphoma). Other investigators report a plateau in freedom from progression at levels expected for an aggressive lymphoma (40% at 10 years). This discrepancy may be caused by variations in histological classification between institutions and the rarity of patients with follicular large cell lymphoma. A retrospective review of 252 patients, all treated with anthracycline-containing combination chemotherapy, showed that patients with more than 50% diffuse components on biopsy had a worse OS than other patients with follicular large cell lymphoma.

Therapeutic approaches

Treatment of follicular large cell lymphoma is more similar to treatment of aggressive NHL than it is to the treatment of indolent NHL. In support of this approach, treatment with high-dose chemotherapy and autologous hematopoietic peripheral SCT shows the same curative potential in patients with follicular large cell lymphoma who relapse as it does in patients with diffuse large cell lymphoma who relapse.

Among those who had received two prior lines of therapy, more than one-half of patients who received CAR T-cell therapy with lisocabtagene maraleucel had disease response.

Anaplastic Large Cell Lymphoma

Anaplastic large cell lymphoma (ALCL) is a T-cell lymphoma associated with the CD30 antigen. The translocation of chromosomes 2 and 5 creates a unique fusion protein with a nucleophosmin-ALK. Patients whose lymphomas express ALK by immunohistochemistry are usually younger and may have systemic symptoms, extranodal disease, and advanced-stage disease. However, they have a more favorable survival rate than ALK-negative patients.

A prospective randomized trial included 452 patients with CD30-positive T-cell lymphoma, 70% of whom had ALCL (22% ALK-positive and 48% ALK-negative patients). The trial compared the previously used standard regimen, CHOP, with brentuximab vedotin (an anti-CD30 monoclonal antibody conjugated to a cytotoxic agent) combined with cyclophosphamide, doxorubicin, and prednisone. With a median follow-up of 35 months, the 3-year OS rate was 77% in the brentuximab vedotin arm and 68% in the CHOP arm (hazard ratio , 0.66; 95% CI, 0.46–0.95; P = .02). This established brentuximab vedotin plus cyclophosphamide, doxorubicin, and prednisone as a new option for patients with ALCL and other CD30-positive T-cell lymphomas, such as angioimmunoblastic T-cell lymphoma and peripheral T-cell lymphoma, not otherwise specified. For patients with relapsed disease, anecdotal responses have been reported for brentuximab vedotin, romidepsin, and pralatrexate.

In a phase II study (NCT00866047), 66% of 58 patients attained a complete response with brentuximab vedotin. At a median follow-up of 58 months, the 5-year PFS rate was 57% (95% CI, 41%–74%), and the 5-year OS rate was 79% (95% CI, 65%–92%). Of the patients achieving a complete response, 42% underwent hematopoietic SCT. In a retrospective review, 39 patients with relapsed disease had a 3-year PFS rate of 50% after autologous or allogeneic SCT. A retrospective review of 84 patients with ALK-negative ALCL suggested a survival benefit with autologous SCT. This hypothesis requires confirmation in a randomized prospective trial.

ALCL in children is usually characterized by systemic and cutaneous disease and has high response rates and good OS with doxorubicin-based combination chemotherapy. Patients with breast implant–associated ALCL may do well without chemotherapy after capsulectomy and implant removal if the disease is confined to the fibrous capsule, and no associated mass or lymphadenopathy is present. Most patients with breast implant–associated ALCL have a characteristic deletion at 20Q13.13 that may help diagnostically to distinguish it from cutaneous or systemic ALCL.

Primary cutaneous ALCL is a distinct entity that is typically ALK-negative and has a very indolent/low-grade clinical course.

Extranodal Natural Killer (NK)-/T-cell Lymphoma

Extranodal natural killer (NK)-/T-cell lymphoma (nasal type) is an aggressive lymphoma marked by extensive necrosis and angioinvasion, most often presenting in extranodal sites, in particular the nasal or paranasal sinus region. Other extranodal sites include the palate, trachea, skin, and gastrointestinal tract. Hemophagocytic syndrome may occur; historically, these tumors were considered part of lethal midline granuloma. In most cases, Epstein-Barr virus (EBV) genomes are detectable in the tumor cells and immunophenotyping shows CD56 positivity. Cases with blood and marrow involvement are considered NK-cell leukemia.

The increased risk of CNS involvement and of local recurrence has led to recommendations for radiation therapy locally, concurrently, before the start of chemotherapy or between cycle two and three of chemotherapy, and for intrathecal prophylaxis and/or prophylactic cranial radiation therapy.

A retrospective review of 1,273 early-stage patients stratified them into a low-risk group and high-risk group using stage, age, LDH, performance status, and primary tumor invasion. Low-risk patients fared best with radiation therapy alone, while high-risk patients fared best with a strategy of radiation therapy combined with chemotherapy.

In a retrospective review of 303 previously untreated patients from an international consortium who received nonanthracycline chemotherapy, the OS rates were identical for early-stage patients (72%−74% at 5 years) who received either concurrent chemotherapy and radiation therapy or chemotherapy followed by radiation therapy.

Higher doses of radiation therapy administered at more than 50 Gy are associated with improved outcomes according to anecdotal reports. The highly aggressive course, with poor response and short survival with standard therapies, especially for patients with advanced-stage disease or extranasal presentation, has led some investigators to recommend autologous or allogeneic peripheral SCT consolidation. Asparaginase-containing regimens have shown anecdotal response rates greater than 50% for relapsing, refractory, or newly diagnosed patients. Because of the lack of randomized clinical trials with more than 100 patients for this rare type of T-cell lymphoma, regimens containing pegaspargase have become the standard for systemic therapy. Pegaspargase is a less toxic formulation of asparaginase with less hypersensitivity reactions and a longer half-life. NK-/T-cell lymphoma that presents only in the skin has a more favorable prognosis, especially in patients with coexpression of CD30 with CD56. A benign NK-cell enteropathy (EBV negative) on endoscopic biopsy can be distinguished from NK-/T-cell lymphoma. In a phase II trial, the anti-PD-L1 antibody avelumab was given to 21 patients with relapsed or refractory disease. The CR rate was 24%, the overall response rate was 38%, and responses correlated with tumor PD-L1 expression. Treatment with pembrolizumab, an anti-programmed cell death protein 1 (PD-1) antibody, resulted in similar responses in patients with relapsed or refractory disease.

Lymphomatoid Granulomatosis

Lymphomatoid granulomatosis is an EBV-positive large B-cell lymphoma with a predominant T-cell background. The histology shows association with angioinvasion and vasculitis, usually manifesting as pulmonary lesions or paranasal sinus involvement.

Patients are managed like others with diffuse large cell lymphoma and require doxorubicin-based combination chemotherapy.

Angioimmunoblastic T-cell Lymphoma

Angioimmunoblastic T-cell lymphoma (AITL or ATCL) was formerly called angioimmunoblastic lymphadenopathy with dysproteinemia. Characterized by clonal T-cell receptor gene rearrangement, this entity is managed like diffuse large cell lymphoma. Patients present with profound lymphadenopathy, fever, night sweats, weight loss, skin rash, a positive Coombs test, and polyclonal hypergammaglobulinemia. Opportunistic infections are frequent because of an underlying immune deficiency. B-cell EBV genomes are detected in most affected patients. For more information about weight loss, see Nutrition in Cancer Care and for more information about skin rash, see Pruritus.

Doxorubicin-based combination chemotherapy, such as the CHOP regimen, is recommended as it is for other aggressive lymphomas. For CD30-positive cases, brentuximab combined with cyclophosphamide, doxorubicin, and prednisone is the standard of care. For more information, see the Anaplastic Large Cell Lymphoma section. The International Peripheral T-Cell Lymphoma Project involving 22 international centers identified 243 patients with AITL or ATCL; the 5-year OS and failure-free survival rates were 33% and 18%, respectively. Myeloablative chemotherapy and radiation therapy with autologous or allogeneic peripheral stem cell support has been described in anecdotal reports. Anecdotal responses have been reported for cyclosporine, pralatrexate, bendamustine, the histone deacetylase inhibitor romidepsin, and brentuximab vedotin (even if there is little or no CD30 expression on the lymphoma). Occasional spontaneous remissions and protracted responses to steroids only have been reported.

Peripheral T-cell Lymphoma

Patients with peripheral T-cell lymphoma have diffuse large cell or diffuse mixed lymphoma that expresses a cell surface phenotype of a postthymic (or peripheral) T-cell expressing CD4 or CD8 but not both together. Peripheral T-cell lymphoma encompasses a group of heterogeneous nodal T-cell lymphomas that will require future delineation. This includes the so-called Lennert lymphoma, a T-cell lymphoma admixed with a preponderance of lymphoepithelioid cells.

Prognosis

Most investigators report worse response and survival rates for patients with peripheral T-cell lymphomas than for patients with comparably staged B-cell aggressive lymphomas. Most patients present with multiple adverse prognostic factors (i.e., older age, stage IV, multiple extranodal sites, and elevated LDH), and these patients have a low (20%) failure-free survival and OS at 5 years. As with other lymphomas (e.g., DLBCL or follicular lymphoma), event-free survival at 24 months predicts a 5-year OS of 78%.

Therapeutic approaches

Therapy involves doxorubicin-based combination chemotherapy (such as CHOP or CHOPE ), which is also used for DLBCL. For CD30-positive cases, brentuximab combined with cyclophosphamide, doxorubicin, and prednisone is the standard of care. For more information, see the Anaplastic Large Cell Lymphoma section. For patients with early-stage disease, anecdotal retrospective series disagree on the value of consolidative radiation therapy after combination chemotherapy. Consolidation therapy using high-dose chemotherapy with autologous or allogeneic hematopoietic stem cell support has been given to patients with advanced-stage peripheral T-cell lymphoma after induction therapy in multiple phase II or retrospective trials. Evidence for this approach is anecdotal.

A randomized prospective trial included 104 patients younger than 61 years with stage II, III, or IV peripheral T-cell lymphoma (excluding ALK-positive ALCL). Patients received either autologous SCT or allogeneic SCT as consolidation therapy after induction with CHOEP (cyclophosphamide, doxorubicin, vincristine, etoposide, and prednisone) followed by DHAP (dexamethasone, cytarabine, and cisplatin). With a median follow-up of 42 months, the 3-year EFS rate was 43% for patients who received allogeneic SCT and 38% for patients who received autologous SCT. The 3-year OS rate was 57% for patients who received allogeneic SCT and 70% for patients who received autologous SCT (P = nonsignificant). None of the 21 responding patients who proceeded to allogeneic SCT relapsed, and 36% of patients who proceeded to autologous SCT relapsed. Eight of 26 patients (31%) who received allogeneic SCT died of graft-versus-host disease, and none of the 41 patients who received autologous SCT died of toxicity. The benefit of graft-versus-lymphoma effect was negated by increased transplant-related mortality.

In a prospective trial of 109 evaluable patients with relapsing disease, treatment with pralatrexate resulted in a 30% response rate and a median 10-month duration of response. Similar response rates were seen in 130 evaluable patients with relapsing disease who received romidepsin in a prospective trial. Anecdotal responses have been seen with a combination of pralatrexate and romidepsin, single-agent bendamustine, belinostat, and brentuximab vedotin (even if there is little or no CD30 expression on the lymphoma). Incorporation of these new agents with CHOP chemotherapy is under clinical evaluation.

An unusual type of peripheral T-cell lymphoma occurring mostly in young men, hepatosplenic T-cell lymphoma, appears to be localized to the hepatic and splenic sinusoids, with cell surface expression of the T-cell receptor gamma/delta. Another variant, subcutaneous panniculitis-like T-cell lymphoma, is localized to subcutaneous tissue associated with hemophagocytic syndrome. These patients have cells that express alpha-beta phenotype. Those with gamma-delta phenotype have a more aggressive clinical course and are classified as cutaneous gamma-delta T-cell lymphoma. These patients may manifest involvement of the epidermis, dermis, subcutaneous region, or mucosa. These entities have extremely poor prognoses with an extremely aggressive clinical course and are treated within the same paradigm as the highest-risk groups with DLBCL. An indolent T-cell lymphoproliferative disease of the gastrointestinal tract must be distinguished from peripheral T-cell lymphoma because no therapy may be indicated.

Enteropathy-type Intestinal T-cell Lymphoma

Enteropathy-type intestinal T-cell lymphoma involves the small bowel of patients with gluten-sensitive enteropathy (celiac sprue). Because a gluten-free diet prevents the development of lymphoma, patients diagnosed with celiac sprue in childhood rarely develop lymphoma. The diagnosis of celiac disease is usually made by finding villous atrophy in the resected intestine. Surgery is often required for diagnosis and to avoid perforation during therapy.

Therapy is with doxorubicin-based combination chemotherapy, but relapse rates appear higher than for comparably staged diffuse large cell lymphoma. Complications of treatment include gastrointestinal bleeding, small bowel perforation, and enterocolic fistulae; patients often require parenteral nutrition. For more information on parenteral nutrition, see Nutrition in Cancer Care. Multifocal intestinal perforations and visceral abdominal involvement are seen at the time of relapse. High-dose therapy with hematopoietic stem cell rescue has been applied in first remission or at relapse. Evidence for this approach is anecdotal.

Intravascular Large B-cell Lymphoma (Intravascular Lymphomatosis)

Intravascular lymphomatosis is characterized by large cell lymphoma confined to the intravascular lumen. The brain, kidneys, lungs, and skin are the organs most likely affected by intravascular lymphomatosis.

With the use of aggressive R-CHOP–based combination chemotherapy, as is used in DLBCL, the prognosis is similar to that of conventional stage IV DLBCL.

Burkitt Lymphoma/Diffuse Small Noncleaved-cell Lymphoma

Burkitt lymphoma/diffuse small noncleaved-cell lymphoma typically involves younger patients and represents the most common type of pediatric NHL. These types of aggressive extranodal B-cell lymphomas are characterized by translocation and deregulation of the MYC gene on chromosome 8. A subgroup of patients with dual translocation of MYC and BCL2 appear to have an extremely poor outcome despite aggressive therapy (median OS, 5 months).

In some patients with larger B cells, there is morphological overlap with DLBCL. These Burkitt-like large cell lymphomas show MYC deregulation, extremely high proliferation rates, and a gene-expression profile as expected for classic Burkitt lymphoma. Endemic cases, usually from Africa, involve the facial bones or jaws of children, mostly containing EBV genomes. Sporadic cases usually involve the gastrointestinal system, ovaries, or kidneys. Patients present with rapidly growing masses and a very high LDH but are potentially curable with intensive doxorubicin-based combination chemotherapy.

Therapeutic approaches

Treatment of Burkitt lymphoma/diffuse small noncleaved-cell lymphoma involves aggressive multidrug regimens in combination with rituximab, similar to those used for the advanced-stage aggressive lymphomas (diffuse large cell). Aggressive combination chemotherapy, which is patterned after that used in childhood Burkitt lymphoma, has been very successful for adult patients with more than 60% of advanced-stage patients free of disease at 5 years. Adverse prognostic factors include bulky abdominal disease and high serum LDH. Patients with Burkitt lymphoma have a 20% to 30% lifetime risk of CNS involvement. Prophylaxis with intrathecal chemotherapy is required as part of induction therapy. Patients with HIV-associated Burkitt lymphoma also benefit from less-toxic modification of the aggressive multidrug regimens in combination with rituximab. For more information, see Primary CNS Lymphoma Treatment and AIDS-Related Lymphoma Treatment.

Lymphoblastic Lymphoma

Lymphoblastic lymphoma (precursor T-cell) is a very aggressive form of NHL. It often, but not exclusively, occurs in young patients. It is commonly associated with large mediastinal masses and has a high predilection for disseminating to bone marrow and the CNS.

Treatment is usually patterned after that for acute lymphoblastic leukemia. Intensive combination chemotherapy with or without bone marrow transplantation is the standard treatment for this aggressive histological type of NHL. Radiation therapy is sometimes given to areas of bulky tumor masses. Because these forms of NHL tend to progress quickly, combination chemotherapy is instituted rapidly once the diagnosis has been confirmed. Careful review of the pathological specimens, bone marrow aspirate, biopsy specimen, cerebrospinal fluid cytology, and lymphocyte marker constitute the most important aspects of the pretreatment staging workup. For more information, see Adult Acute Lymphoblastic Leukemia Treatment.

Adult T-cell Leukemia/Lymphoma

Adult T-cell leukemia/lymphoma (ATL) is caused by infection with the retrovirus human T-lymphotrophic virus 1 and is frequently associated with lymphadenopathy, hypercalcemia, circulating leukemic cells, bone and skin involvement, hepatosplenomegaly, a rapidly progressive course, and poor response to combination chemotherapy. ATL has been divided into four clinical subtypes:

• Acute (aggressive course with leukemia, with or without extranodal or nodal involvement).
• Lymphoma (aggressive course with lymphadenopathy and no leukemia).
• Chronic (indolent course with leukemia and lymphadenopathy).
• Smoldering (indolent course with only leukemia).

The acute and lymphoma types of ATL have done poorly with strategies of combination chemotherapy and allogeneic SCT with a median OS under 1 year. Using combination chemotherapy, less than 10% of 807 patients were alive after 4 years. Anecdotal durable remissions have been reported after allogeneic SCT and even after subsequent donor lymphocyte infusion for relapses after transplant. Among 815 patients who underwent allogeneic SCT in two retrospective reviews, the 3-year OS rates were 36% and 26%.

The combination of zidovudine and interferon-alpha has activity against ATL, even for patients who failed previous cytotoxic therapy. Durable remissions are seen in most patients who present with this combination, but are not seen in patients with the lymphoma subtype of ATL. In a multicenter phase II study of 26 relapsed patients, 42% responded to lenalidomide (including four CR). Symptomatic local progression of all subtypes responds well to palliative radiation therapy. In the relapsed setting, an overall response rate above 50% was seen using mogamulizumab, a humanized monoclonal antibody against the C-C chemokine receptor 4 (CCR4). For CD30-positive cases, brentuximab combined with cyclophosphamide, doxorubicin, and prednisone is the standard of care. For more information, see the Anaplastic Large Cell Lymphoma section.

Mantle Cell Lymphoma

Mantle cell lymphoma (MCL) is found in lymph nodes, the spleen, bone marrow, blood, and sometimes the gastrointestinal system (lymphomatous polyposis). MCL is characterized by CD5-positive follicular mantle B cells, a translocation of chromosomes 11 and 14, and an overexpression of the cyclin D1 protein. MCL may be divided into two clinical subtypes: a classical version with lymphadenopathy with high SOX-11 expression that manifests with an aggressive clinical course and a worse prognosis versus a leukemic, non-nodal version with low SOX-11 expression and a more indolent course and a better prognosis. A complex karyotype predicts poor response to induction therapy and inferior survival. There is frequent overlap on presentation with these subtypes, and the therapeutic implication remains unclear. However, both of these versions can converge later in their course into a blastoid phenotype or treatment-resistant phenotype due to genomic instability and selection.

Like the low-grade lymphomas, MCL appears incurable with anthracycline-based chemotherapy and occurs in older patients with generally asymptomatic advanced-stage disease. The median survival, however, is significantly shorter (5–7 years) than that of other lymphomas, and this histology is now considered to be an aggressive lymphoma. A diffuse pattern and the blastoid variant have an aggressive course with shorter survival, while the mantle zone type may have a more indolent course. A high cell-proliferation rate (increased Ki-67, mitotic index, beta-2-microglobulin) may be associated with a poorer prognosis.

Therapeutic approaches

Asymptomatic patients with low-risk scores on the IPI may do well when initial therapy is deferred. There is no standard approach to MCL. Several induction chemotherapy regimens may be employed for symptomatic progressing disease. These regimens range in intensity from rituximab alone to rituximab plus ibrutinib, rituximab plus bendamustine, R-CHOP, or high-dose intensive regimens such as R-hyper C-VAD (hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone alternating with methotrexate and cytarabine). Some physicians use autologous SCT or allogeneic SCT consolidation next, while others prefer rituximab maintenance, reserving high-dose consolidation for a later time. Ibrutinib, lenalidomide, and bortezomib have shown activity in relapsing patients, and these drugs are being incorporated up front.

It is unclear which therapeutic approach offers the best long-term survival in this clinicopathological entity.

In a phase II trial of previously untreated patients with MCL older than 64 years, 50 patients received the B-cell receptor-inhibitor ibrutinib plus rituximab. With a median follow-up of 45 months, the overall response rate was 96%, the CR rate was 76%, the 3-year PFS rate was 87%, and the 3-year OS rate was 94%. In a phase II trial of 131 previously untreated patients with MCL aged 65 years or younger, 1 year of ibrutinib plus 4 weeks of rituximab resulted in a CR rate of 89% prior to any chemotherapy consolidation. Another phase II trial using ibrutinib plus rituximab included asymptomatic patients with previously untreated MCL; the CR rate was 87%. Previously treated patients who received ibrutinib had a response rate of 86% (21% CR rate) and a median PFS of 14 months. In a prospective randomized trial, 280 patients with relapsed or refractory MCL received either ibrutinib or temsirolimus. With a median follow-up of 15 months, the median PFS favored ibrutinib (14.6 months vs. 6.2 months; HR, 0.43; 95% CI, 0.32–0.58, P .0001). Ibrutinib was combined with another active agent, venetoclax, in a phase II study of 23 patients with relapsed or refractory MCL. An unprecedented 71% of patients had a CR and 78% of responding patients maintained response at 15 months.

A prospective randomized trial included 523 patients aged 65 years and older with MCL. Patients were randomly assigned to receive either ibrutinib, bendamustine, and rituximab or bendamustine and rituximab alone. With a median follow-up of 84.7 months, the median PFS was 80.6 months for patients who received ibrutinib, and 52.9 months for patients who received bendamustine and rituximab alone (HR, 0.75; 95% CI, 0.59–0.96; P = .01). There was no difference in the 7-year OS rate (55.0% vs. 56.8%; HR, 1.07; 95% CI, 0.81–1.40). It is unclear if patients who received ibrutinib alone could have achieved these same results without receiving conventional chemotherapy. The magnitude of benefit demonstrated by the PFS results contrasted with the insufficient OS benefit after 7 years may cast doubt on the long-term safety of this combination.

In a prospective randomized trial, 560 patients older than 60 years and not eligible for SCT were given either R-CHOP or R-FC (rituximab, fludarabine, cyclophosphamide) for six to eight cycles, followed by maintenance therapy in responders randomly assigned to rituximab or interferon-alpha maintenance therapy. With a median follow-up of 7.6 years, the median OS was significantly shorter after R-FC than after R-CHOP (3.9 years versus 6.4 years; P = .0054). In the same trial, with a median follow-up of 8 years for the 316 responding patients, rituximab maintenance resulted in improved OS over interferon maintenance (median OS, 9.8 years vs. 7.1 years; P = .009). Patients responsive to R-CHOP benefitted most from rituximab in OS (median 9.8 years vs. 6.4 years; P = .0026). A randomized trial compared bendamustine plus rituximab (BR) with R-CHOP and showed improved PFS (35 vs. 22 months; HR, 0.49; 95% CI, 0.28–0.79; P = .004) but no difference in OS. However, this trial failed to show any benefit for rituximab maintenance after BR. A prospective randomized trial of 487 patients compared VR-CAP (bortezomib, rituximab, cyclophosphamide, doxorubicin, prednisone) with R-CHOP. With a median follow-up of 82 months, the median OS was longer for VR-CAP (90.7 months) compared with R-CHOP (55.7 months) (HR, 0.66; 95% CI, 0.51−0.85; P = .001).

A prospective randomized trial of 497 patients younger than 65 years compared six cycles of R-CHOP with six cycles of alternating R-CHOP and R-DHAP (rituximab, dexamethasone, cytarabine, and cisplatin), with both groups then receiving autologous SCT. With a median follow-up of 10.6 years, the 10-year PFS rate was 73% for patients who received R-DHAP and 57% for patients who received R-CHOP (HR, 0.56; P = .038), but there was no difference in the 10-year OS rates (60% vs. 55% ; HR, 0.80; 0.61–1.06; P = .12). This is the randomized trial referenced by all subsequent articles establishing a role for cytarabine in induction therapy; the ultimate lack of survival advantage casts doubt on this assertion.

Randomized trials have not confirmed an OS benefit in patients who receive consolidation therapy with autologous or allogeneic SCT.

In a prospective trial (NCT00921414) of 299 patients who were previously untreated for MCL, 257 responders received four courses of R-DHAP and autologous SCT. The patients were randomly assigned to receive rituximab maintenance therapy for 3 years versus no maintenance therapy. After randomization, a median follow-up at 50.2 months showed the rate of PFS at 4-years favored the rituximab-maintenance arm at 83% (95% CI, 73%–88%) versus the no-maintenance arm at 64% (95% CI, 55%–73%; P .001). The 4-year OS rate also favored the rituximab-maintenance arm at 89% (95% CI, 81%–94%) versus the no-maintenance arm at 80% (95% CI, 72%–88%; P = .04).

Lenalidomide with or without rituximab also shows response rates of around 50% in relapsed patients, with even higher response rates for previously untreated patients.

Acalabrutinib (another B-cell receptor inhibitor via the Bruton tyrosine kinase pathway) was studied in 124 patients with relapsed or refractory MCL. In a phase II study, there was an 81% overall response rate, 40% CR rate, and 67% 1-year PFS rate. Similarly, the BTK inhibitor zanubrutinib was evaluated in a phase II study of 86 patients with relapsed or refractory MCL. After a median follow-up of 35.3 months, the overall response rate was 84%, the CR rate was 78%, and the median PFS was 33.0 months.

Patients with relapsed or refractory MCL whose disease did not respond to ibrutinib or acalabrutinib were enrolled in a phase II trial using brexucabtagene autoleucel, an anti-CD19 CAR T-cell therapy. With a median follow-up of 36 months, 74 patients had an objective response rate of 91% (95% CI, 82%−97%) and a CR rate of 68% (95% CI, 55%−78%). Grade 3 or higher cytokine release syndrome occurred in 15% of patients, and neurological events occurred in 31% of patients.

In summary, the optimal sequencing of these various therapies is unclear and is the subject of an ongoing Intergroup clinical trial. Rituximab, lenalidomide, ibrutinib, acalabrutinib, venetoclax, and zanubrutinib represent directed biological agents that may lead to chemotherapy-free treatment strategies for patients with MCL.

Routine administration of CNS prophylaxis in high-risk MCL has never been studied in a prospective randomized trial. The use of intrathecal or high-dose methotrexate or the use of systemic therapies with CNS penetration like ibrutinib, high-dose cytarabine, or venetoclax, have not been studied and proven efficacious in this situation.

Posttransplantation Lymphoproliferative Disorder

Patients who undergo transplantation of the heart, lung, liver, kidney, or pancreas usually require lifelong immunosuppression. This may result in posttransplantation lymphoproliferative disorder (PTLD) in 1% to 3% of recipients, which appears as an aggressive lymphoma. Pathologists can distinguish a polyclonal B-cell hyperplasia from a monoclonal B-cell lymphoma; both are almost always associated with EBV.

Prognosis

Poor performance status, grafted organ involvement, high IPI, elevated LDH, and multiple sites of disease are poor prognostic factors for PTLD.

Therapeutic options

In some cases, withdrawal of immunosuppression results in eradication of the lymphoma. When this is unsuccessful or not feasible, a course of rituximab may be considered, because it has shown durable remissions in approximately 60% of patients and a favorable toxicity profile. If these measures fail, doxorubicin-based combination chemotherapy (R-CHOP) is recommended, although some patients can avoid cytotoxic therapy. Localized presentations can be controlled with surgery or radiation therapy alone. These localized mass lesions, which may grow over a period of months, are often phenotypically polyclonal and tend to occur within weeks or a few months after transplantation. Multifocal, rapidly progressive disease occurs late after transplantation (>1 year) and is usually phenotypically monoclonal and associated with EBV. These patients may have durable remissions using standard chemotherapy regimens for aggressive lymphoma. Instances of EBV-negative PTLD occur even later (median, 5 years posttransplant) and have a worse prognosis; R-CHOP chemotherapy can be applied directly in this circumstance. A sustained clinical response after failure from chemotherapy was attained using an immunotoxin (anti-CD22 B-cell surface antigen antibody linked with ricin, a plant toxin). An anti-interleukin-6 monoclonal antibody is also under clinical evaluation.

True Histiocytic Lymphoma

True histiocytic lymphomas are very rare tumors that show histiocytic differentiation and express histiocytic markers in the absence of B-cell or T-cell lineage-specific immunologic markers. Care must be taken with immunophenotypic tests to exclude ALCL or hemophagocytic syndromes caused by viral infections, especially EBV.

Therapeutic options

Therapy is modeled after the treatment of comparably staged diffuse large cell lymphomas, but the optimal approach remains to be defined.

Primary Effusion Lymphoma

Primary effusion lymphoma presents exclusively or mainly in the pleural, pericardial, or abdominal cavities in the absence of an identifiable tumor mass. Patients are usually HIV seropositive, and the tumor usually contains Kaposi sarcoma–associated herpes virus/human herpes virus 8.

Prognosis

The prognosis of primary effusion lymphoma is extremely poor.

Therapeutic approaches

Therapy is usually modeled after the treatment of comparably staged diffuse large cell lymphomas.

Plasmablastic Lymphoma

Plasmablastic lymphoma is most often seen in patients with HIV infection and is characterized by CD20-negative large B cells with plasmacytic features. This type of lymphoma has a very aggressive clinical course, including poor responses and short remissions with standard chemotherapy. Anecdotal reports suggest using aggressive chemotherapy for Burkitt or lymphoblastic lymphoma, followed by SCT consolidation in responding patients, when feasible.

Stage is important in selecting a treatment for patients with non-Hodgkin lymphoma (NHL). Chest and abdominal computed tomography (CT) scans are usually part of the staging evaluation for all lymphoma patients. The staging system is similar to the staging system used for Hodgkin lymphoma (HL).

Common among patients with NHL is involvement of the following:

• Noncontiguous lymph nodes.
• Waldeyer ring.
• Epitrochlear nodes.
• Gastrointestinal tract.
• Extranodal presentations. (A single extranodal site is occasionally the only site of involvement in patients with diffuse lymphoma.)
• Bone marrow.
• Liver (especially common in patients with low-grade lymphomas).

Cytological examination of cerebrospinal fluid may be positive in patients with aggressive NHL. Involvement of hilar and mediastinal lymph nodes is less common than in HL. Mediastinal adenopathy, however, is a prominent feature of lymphoblastic lymphoma and primary mediastinal B-cell lymphoma, entities primarily found in young adults.

Most patients with NHL present with advanced (stage III or stage IV) disease that can often be identified with limited staging procedures such as CT scanning and biopsies of the bone marrow and other accessible sites of involvement. In a retrospective review of over 32,000 cases of lymphoma in France, up to 40% of diagnoses were made by core needle biopsy, and 60% were made by excisional biopsy. After expert review, core needle biopsy provided a definite diagnosis in 92.3% of cases, but it was less conclusive than excisional biopsy, which provided a definite diagnosis in 98.1% of cases (P .0001). Laparoscopic biopsy or laparotomy is not required for staging but rarely may be necessary to establish a diagnosis or histological type. Positron emission tomography (PET) with fluorine F 18-fludeoxyglucose can be used for initial staging and for follow-up after therapy as a supplement to CT scanning. Interim PET scans after two to four cycles of therapy did not provide reliable prognostic information because of problems of interobserver reproducibility in a large cooperative group trial (ECOG-E344 ) and lack of difference in outcome between PET-negative and PET-positive/biopsy-negative patients in two prospective trials and in a meta-analysis. For patients with follicular lymphoma, a positive PET result after therapy has a worse prognosis; however, it is unclear whether a positive PET result is predictive when further or different therapy is implemented.

In a retrospective study of 130 patients with diffuse large B-cell lymphoma, PET scanning identified all clinically important marrow involvement from lymphoma, and bone marrow biopsy did not upstage any patient's lymphoma. A retrospective study of 580 patients with follicular lymphoma from seven National Cancer Institute-sponsored trials showed no improvement in assessing response to therapy when bone marrow biopsy was added to radiological imaging. The workup of NHL should include bone marrow biopsy when management would change (e.g., determining limited stage vs. advanced stage) or when evaluating cytopenias.

Staging Subclassification System

Lugano Classification

The American Joint Committee on Cancer (AJCC) has adopted the Lugano classification to evaluate and stage lymphoma. The Lugano classification system replaces the Ann Arbor classification system, which was adopted in 1971 at the Ann Arbor Conference, with some modifications 18 years later from the Cotswolds meeting.

Table 2. Lugano Classification for Hodgkin and Non-Hodgkin Lymphomaa

Occasionally, specialized staging systems are used. The physician should be aware of the system used in a specific report.

The E designation is used when extranodal lymphoid malignancies arise in tissues separate from, but near, the major lymphatic aggregates. Stage IV refers to disease that is diffusely spread throughout an extranodal site, such as the liver. If pathological proof of involvement of one or more extralymphatic sites has been documented, the symbol for the site of involvement, followed by a plus sign (+), is listed.

Table 3. Notation to Identify Specific Sites

Current practice assigns a clinical stage based on the findings of the clinical evaluation and a pathological stage based on the findings made as a result of invasive procedures beyond the initial biopsy.

For example, on percutaneous biopsy, a patient with inguinal adenopathy and a positive lymphangiogram without systemic symptoms might have involvement of the liver and bone marrow. The precise stage of such a patient would be clinical stage IIA, pathological stage IVA(H+)(M+).

Several other factors that are not included in the above staging system are important for the staging and prognosis of patients with NHL. These factors include the following:

• Age.
• Performance status (PS).
• Tumor size.
• Lactate dehydrogenase (LDH) values.
• The number of extranodal sites.

The National Comprehensive Cancer Network International Prognostic Index (IPI) for aggressive NHL (diffuse large cell lymphoma) identifies the following five significant risk factors prognostic of overall survival (OS) and their associated risk scores:

• Age.
  • 40 years: 0.
  • 41–60 years: 1.
  • 61–75 years: 2.
  • >75 years: 3.
• Stage III/IV: 1.
• PS 2/3/4: 1.
• Serum LDH.
  • Normalized: 0.
  • >1x–3x: 1.
  • >3x: 2.
• Number of extranodal sites ≥2: 1.

Risk scores:

• Low (0 or 1): 5-year OS rate, 96%; progression-free survival (PFS) rate, 91%.
• Low intermediate (2 or 3): 5-year OS rate, 82%; PFS rate, 74%.
• High intermediate (4 or 5): 5-year OS rate, 64%; PFS rate, 51%.
• High (>6): 5-year OS rate, 33%; PFS rate, 30%.

Age-adjusted and stage-adjusted modifications of this IPI are used for younger patients with localized disease. Shorter intervals of time between diagnosis and treatment appear to be a surrogate for poor prognostic biological factors.

The BCL2 gene and rearrangement of the MYC gene or dual overexpression of the MYC gene, or both, confer a particularly poor prognosis. Patients at high risk of relapse may benefit from consolidation therapy or other approaches under clinical evaluation. Molecular profiles of gene expression using DNA microarrays may help to stratify patients in the future for therapies directed at specific targets and to better predict survival after standard chemotherapy.

Treatment of non-Hodgkin lymphoma (NHL) depends on the histological type and stage. Many of the improvements in survival have been made using clinical trials (experimental therapy) that have attempted to improve on the best available accepted therapy (conventional or standard therapy).

In asymptomatic patients with indolent forms of advanced NHL, treatment may be deferred until the patient becomes symptomatic as the disease progresses. When treatment is deferred, the clinical course of patients with indolent NHL varies; frequent and careful observation is required so that effective treatment can be initiated when the clinical course of the disease accelerates. Some patients have a prolonged indolent course, but others have disease that rapidly evolves into more aggressive types of NHL that require immediate treatment.

Radiation techniques differ somewhat from those used in the treatment of Hodgkin lymphoma. The dose of radiation therapy usually varies from 25 Gy to 50 Gy and is dependent on factors that include the histological type of lymphoma, the patient’s stage and overall condition, the goal of treatment (curative or palliative), the proximity of sensitive surrounding organs, and whether the patient is being treated with radiation therapy alone or in combination with chemotherapy. Given the patterns of disease presentations and relapse, treatment may need to include unusual sites such as Waldeyer ring, epitrochlear nodes, or mesenteric nodes. The associated morbidity of the treatment must be considered carefully. Most patients who receive radiation are treated on only one side of the diaphragm. Localized presentations of extranodal NHL may be treated with involved-field techniques with significant (>50%) success.

Table 4. Treatment Options for Non-Hodgkin Lymphoma (NHL)

Even though existing treatments cure a significant fraction of patients with lymphoma, numerous clinical trials that explore treatment improvements are in progress. If possible, patients can be included in these studies. Standardized guidelines for response assessment have been suggested for use in clinical trials.

Several retrospective reviews suggest that routine surveillance scans offer little to no value in patients with diffuse-large B-cell lymphoma (DLBCL) who have attained a clinical complete remission after induction therapy. Prognostic value is also difficult to identify for an interim positron emission tomography-computed tomography scan during induction therapy for DLBCL.

Aggressive lymphomas are increasingly seen in HIV-positive patients. Treatment of these patients requires special consideration. For more information, see AIDS-Related Lymphoma Treatment.

In addition to screening for HIV among patients with aggressive lymphomas, active hepatitis B or hepatitis C can be assessed before treatment with rituximab and/or chemotherapy. Patients with detectable hepatitis B viral loads benefit from prophylaxis with entecavir in the context of rituximab therapy. Patients with a resolved hepatitis B virus (HBV) infection (defined as hepatitis B surface antigen-negative but hepatitis B core antibody-positive) are at risk of reactivation of HBV and require monitoring of HBV DNA. Prophylactic nucleoside therapy lowered HBV reactivation from 10.8% to 2.1% in a retrospective study of 326 patients. Similarly, prophylaxis for herpes zoster with acyclovir or valacyclovir and prophylaxis for pneumocystis with trimethoprim/sulfamethoxazole or dapsone are usually applied with rituximab with or without combination chemotherapy. Long-term impaired immune health was evaluated in a retrospective cohort study of 21,690 survivors of DLBCL from the California Cancer Registry. Elevated incidence rate ratios were found up to 10 years later for pneumonia (10.8-fold), meningitis (5.3-fold), immunoglobulin deficiency (17.6-fold), and autoimmune cytopenias (12-fold).

Among 2,508 patients in a Danish registry, the incidence of doxorubicin-induced congestive heart failure increased for 115 NHL survivors with a history of cardiac disease (hazard ratio , 2.71; 95% confidence interval , 1.15−6.36) and/or multiple cardiovascular risk factors (HR, 2.86; 95% CI, 1.56−5.23).

Several unusual presentations of lymphoma occur that often require somewhat modified approaches to staging and therapy. The reader is referred to reviews for a more detailed description of extranodal presentations in the gastrointestinal system, thyroid, spleen, testis, paranasal sinuses, bone, orbit, and skin.

For more information, see Primary CNS Lymphoma Treatment.

Castleman Disease

A biopsy of localized or multifocal collections of lymph nodes may lead to a diagnosis of Castleman disease (CD), although it is an uncommon diagnosis. Strictly speaking, this is not a lymphoma, nor is it even a malignancy. Yet, many patients with CD may be seen and treated by hematologists or oncologists.

Localized or unicentric CD is usually asymptomatic and occurs in the mediastinum, which is the most common presentation for CD. Watchful waiting, surgery, or radiation therapy can be used for this indolent form. Multicentric CD (30% of CD patients) presents with lymphadenopathy in multiple sites; symptoms such as fever, night sweats, weight loss, and fatigue; and laboratory abnormalities such as anemia, low albumin, elevated C-reactive protein, and high fibrinogen. Multicentric CD (MCD) is subdivided into human herpes virus-8–associated MCD (usually with HIV or with severe immunocompromise) or idiopathic MCD. Cytopenias and cytokine storm are attributed to interleukin-6 (IL-6) overproduction. MCD is a feature seen in POEMS (polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy, and skin abnormalities) syndrome and TAFRO (thrombocytopenia, anasarca, fever, reticulin fibrosis, and organomegaly) syndrome. Therapy with siltuximab (an anti–IL-6 monoclonal antibody), rituximab (an anti-CD20 monoclonal antibody), or chemotherapeutic agents has been presented in anecdotal nonrandomized series.

Although localized presentations are uncommon in non-Hodgkin lymphoma (NHL), the goal of treatment is to cure the disease in patients who are shown to have truly localized occurrence after undergoing appropriate staging.

Treatment Options for Indolent Stage I and Indolent, Contiguous Stage II NHL

Treatment options for indolent stage I and indolent, contiguous stage II NHL include the following:

1. Radiation therapy.
2. Rituximab with or without chemotherapy.
3. Watchful waiting.
4. Other therapies as designated for patients with advanced-stage disease.

In a prospective randomized trial, 150 patients with stage I or stage II follicular lymphoma were randomly assigned to 30 Gy of involved-field radiation therapy alone or radiation therapy plus six cycles of R-CVP (rituximab, cyclophosphamide, vincristine, prednisolone). With a median follow-up of 9.6 years, the 10-year progression-free survival (PFS) rate favored combined-modality therapy, at 59% (95% confidence interval , 46%–74%) versus 41% for radiation therapy alone (95% CI, 30%–57%) (P = .033). There was no difference in overall survival (OS) (87% and 95%, P = .40).

The National Lymphocare Study identified 471 patients with stage I follicular lymphoma. Of those patients, 206 were rigorously staged with a bone marrow aspirate and biopsy, and computed tomography (CT) scans or positron emission tomography (PET)-CT scans. Nonrandomized treatments included radiation therapy (27%), rituximab-chemotherapy (R-chemotherapy) (28%), watchful waiting (17%), R-chemotherapy plus radiation therapy (13%), and rituximab alone (12%), although more than one-third of the patients started with expectant therapy. With a median follow-up of 57 months, PFS favored R-chemotherapy or R-chemotherapy plus radiation therapy, but OS was nearly identical, all over 90%. Clinical trials are required to answer questions such as the following:

• If the PET-CT scan is clear after excisional biopsy, is watchful waiting or radiation therapy preferred?
• Should rituximab be added to radiation therapy for stage I follicular lymphoma?
• Is there any role for R-chemotherapy plus radiation therapy?

Radiation therapy

Long-term disease control within radiation fields can be achieved in a significant number of patients with indolent stage I or stage II NHL by using dosages of radiation that usually range from 25 Gy to 40 Gy to involved sites or to extended fields that cover adjacent nodal sites. Almost half of all patients treated with radiation therapy alone will relapse out-of-field within 10 years.

A retrospective review of 512 patients from an international consortium evaluated patients with early-stage follicular lymphoma who received at least 24 Gy of localized radiation therapy at initial presentation. With a median follow-up of 52 months, 29.1% of patients developed recurrent lymphoma at a median of 23 months (range, 1−143 months). With a median follow-up of 33 months after relapse, the 3-year OS rate was 91.4% after patients received subsequent systemic chemotherapy that involved rituximab, usually with chemotherapy.

Very low-dose radiation therapy with 4 Gy (2 Gy × 2 fractions) can result in 50% remission rates for patients who cannot tolerate higher doses. In a multicenter, randomized, prospective trial, 548 patients with follicular or marginal zone lymphoma received radiation therapy, either 4 Gy in 2 fractions or 24 Gy in 12 fractions.

• At a median follow-up of 73.8 months, the 5-year local complete response rate was 89.9% (85.5%–93.1%) after 24 Gy and 70.4% (64.7%–75.4%) after 4 Gy (HR, 3.46; 95% CI, 2.25–5.33, P .0001).
• Although durable local control was superior for patients who received 24 Gy in 12 fractions, the 4 Gy (2 Gy x 2 fractions) regimen was nearly comparable with less radiation exposure, less time undergoing therapy, and less cost.

In situations in which mediastinal radiation would encompass the left side of the heart or would increase breast cancer risk in young female patients, proton therapy may be considered to reduce the radiation dose to organs at risk.

Rituximab with or without chemotherapy

For symptomatic patients who require therapy, when radiation therapy is contraindicated, or when an alternative treatment is preferred, rituximab with or without chemotherapy can be used (as outlined below for more advanced-stage patients). The value of adjuvant treatment with radiation to decrease relapse, plus rituximab (an anti–CD20 monoclonal antibody) either alone or in combination with chemotherapy, has been extrapolated from trials of patients with advanced-stage disease and has not been confirmed.

Watchful waiting

Watchful waiting can be considered for asymptomatic patients. Watchful waiting has never been compared with up-front radiation therapy in a prospective randomized trial; a retrospective analysis of the Surveillance, Epidemiology and End Results (SEER) Program database over 30 years showed improved outcomes for up-front radiation therapy.

Other therapies as designated for patients with advanced-stage disease

Patients with involvement that is not able to be encompassed by radiation therapy are treated as outlined for patients with stage III or stage IV low-grade lymphoma.

Optimal treatment of advanced stages of low-grade non-Hodgkin lymphoma (NHL) is controversial because of low cure rates with the current therapeutic options. Numerous clinical trials are in progress to evaluate treatment issues, and patients are urged to participate. The rate of relapse is fairly constant over time, even in patients who have achieved complete response to treatment. Indeed, relapse may occur many years after treatment. Currently, no randomized trials provide guidance to clinicians about the initial choice of watchful waiting, rituximab, nucleoside analogues, alkylating agents, combination chemotherapy, radiolabeled monoclonal antibodies, or combinations of these options.;

For patients with indolent, noncontiguous stage II and stage III NHL, central lymphatic radiation therapy has been proposed but is not usually recommended as a form of treatment.

Patients with a resolved hepatitis B virus (HBV) infection (defined as hepatitis B surface antigen-negative but hepatitis B core antibody-positive) are at risk of reactivation of HBV and require monitoring of HBV DNA. Prophylactic nucleoside therapy lowered HBV reactivation from 10.8% to 2.1% in a retrospective study of 326 patients.

Treatment Options for Indolent, Noncontiguous Stage II/III/IV NHL

Treatment options for indolent, noncontiguous stage II/III/IV NHL include the following:

1. Watchful waiting for asymptomatic patients.
2. Rituximab alone or in combination with cytotoxic agents used in front-line therapy.
3. Lenalidomide and rituximab.
4. Maintenance rituximab.
5. Obinutuzumab alone or in combination with cytotoxic agents used in front-line therapy.
6. Phosphatidylinositol 3-kinase (PI3K) inhibitor.
   • Copanlisib.
7. EZH2 inhibitor.
   • Tazemetostat.
8. Radiolabeled anti-CD20 monoclonal antibodies.
9. Intensive therapy with chemotherapy with or without total-body irradiation or high-dose radioimmunotherapy followed by autologous or allogeneic bone marrow transplantation or peripheral stem cell transplantation (under clinical evaluation).
10. Phase III clinical trials comparing chemotherapy alone versus chemotherapy followed by anti-idiotype vaccine.
11. Extended-field radiation therapy (stage III patients only) (under clinical evaluation).
12. Ofatumumab—human anti–CD20 monoclonal antibody (under clinical evaluation).
13. Short-course low-dose, palliative radiation therapy (2 × 2 Gy) (under clinical evaluation).

Because none of the therapies listed above are curative for advanced-stage disease, innovative approaches are under clinical evaluation.

Watchful waiting for asymptomatic patients

The rate of relapse is fairly constant over time, even in patients who have achieved complete responses (CR) to treatment. Indeed, relapse may occur many years after treatment. In this category, deferred treatment (i.e., watchful waiting until the patient becomes symptomatic before initiating treatment) can be considered. The Follicular Lymphoma International Prognostic Index (FLIPI) and the revised FLIPI-2 can predict progression-free survival (PFS) and overall survival (OS), but the scores cannot be used to establish the need for therapy in asymptomatic patients.

Evidence (watchful waiting):

1. Three randomized trials compared watchful waiting with immediate chemotherapy.;
   • All three trials showed no difference in cause-specific or OS.
   • For patients randomly assigned to watchful waiting, the median time to require therapy was 2 to 3 years and one-third of patients receiving watchful waiting never required treatment with watchful waiting (half died of other causes and half remained progression free after 10 years).
2. A selected group of 107 patients with advanced-stage follicular lymphoma were managed with initial watchful waiting; with a median delay of 55 months, subsequent therapy resulted in equivalent freedom from treatment failure and OS compared with a similar cohort treated immediately with rituximab. This implies that watchful waiting remains a relevant approach even in the rituximab era.

Rituximab alone or in combination with cytotoxic agents used in front-line therapy

Standard therapy includes rituximab, an anti–CD20 monoclonal antibody, either alone, as was shown in the ECOG-E4402 trial (NCT00075946), or in combination with purine nucleoside analogues, such as fludarabine or cladribine, alkylating agents (with or without steroids), or combination chemotherapy. Rituximab may be considered as first-line therapy, either alone or in combination with other agents. Rituximab may be given intravenously or subcutaneously, and biosimilar versions, such as CT-P10 and GP2013, have shown equivalent efficacy and safety. Combinations include the following:

• R-bendamustine: rituximab + bendamustine.
• R-F: rituximab + fludarabine.
• R-CVP: rituximab + cyclophosphamide + vincristine + prednisone.
• R-CHOP: rituximab + cyclophosphamide + doxorubicin + vincristine + prednisone. A Cochrane meta-analysis could not identify any OS benefit of adding doxorubicin to chemotherapy regimens with rituximab or to chemotherapy regimens without rituximab.
• R-FM: rituximab + fludarabine + mitoxantrone.
• R-FCM: rituximab + fludarabine + cyclophosphamide + mitoxantrone.

Evidence (rituximab with or without chemotherapy):

1. A prospective, randomized trial of 534 patients with previously untreated, advanced-stage follicular lymphoma compared R-CHOP, R-FM, and R-CVP.
   • With a median follow-up of 84 months, there was no difference in OS (8-year OS rate, 83%; 95% confidence interval (CI), 79%–87%), but the 8-year PFS rates favored R-CHOP (52%) and R-FM (49%) over R-CVP (42%) (P for the three regimens = .037).
2. Four randomized prospective studies of previously untreated patients (involving more than 1,300 patients) and one Cochrane meta-analysis that included both untreated and previously treated patients (involving almost 1,000 patients) have compared rituximab plus combination chemotherapy with chemotherapy alone.;
   • Rituximab plus chemotherapy was superior in terms of event-free survival (EFS) or PFS (ranging from 2–3 years) in all of the studies and in terms of OS in all but one study (absolute benefit ranging from 6%–13% at 4 years, P .04 and hazard ratio = 0.63 for the meta-analysis).
   • All of these trials were performed in symptomatic patients who required therapy. These results do not negate watchful waiting when appropriate.
   • Fluorine F 18-fludeoxyglucose positron emission tomography–computed tomography (18F-FDG PET-CT) scan status at the completion of rituximab plus chemotherapy induction therapy is strongly predictive of outcome. It is not yet known whether acting on the results of the scans translates into better outcomes.
3. In a prospective randomized trial (NCT00991211), 527 patients with indolent and mantle cell lymphoma were randomly assigned to a bendamustine-and-rituximab arm versus an R-CHOP arm.
   • With a median follow-up of 45 months, the median PFS favored the bendamustine arm (69 months vs. 31 months ) but with no difference in OS.
   • The bendamustine arm was associated with significantly lower rates of alopecia, hematologic toxicity, stomatitis, peripheral neuropathy, and infections than was the R-CHOP arm.
4. In a similar prospective randomized trial, 447 patients with indolent and mantle cell lymphoma were assigned to bendamustine and rituximab versus R-CHOP or R-CVP.
   • With a median follow-up of 65 months, the 5-year PFS rate favored bendamustine and rituximab, 65.5% versus 55.8% (HR, 0.61; 95% CI, 0.45–0.85; P = .0025), but with no difference in OS.
   • Increased deaths in the bendamustine-and-rituximab arm from cardiovascular causes (seven versus one) and from secondary malignancies other than lymphoma (five versus three) may have contributed to the lack of OS advantage.

Lenalidomide and rituximab

The combination of the immunomodulating agent lenalidomide with rituximab (the so-called R2 regimen) has been proposed as an alternative regimen to combinations involving cytotoxic agents and their subsequent short- and long-term toxicities.

Evidence (lenalidomide and rituximab):

1. In a randomized prospective trial (RELEVANCE) of 1,030 patients with previously untreated follicular lymphoma, rituximab plus lenalidomide for 18 months was compared with rituximab plus chemotherapy (usually R-CHOP). All patients received rituximab maintenance for up to 2 years.
   • With a median follow-up of 72 months, the 6-year PFS (60% and 59%) and 3-year OS (89%) were identical (HR, 1.03; 95% CI, 0.84–1.27; P = .78 for PFS) (HR for OS was not reported).
   • This trial established that the R2 regimen is as effective as rituximab plus cytotoxic chemotherapy options. The transformation rate to aggressive lymphoma per year was 0.68% in the R2 group and 0.45% in the R-chemotherapy group. With a median follow-up of 72 months, there were no new safety signals.
2. In a randomized prospective trial of 358 patients with resistant/refractory indolent lymphoma (usually follicular lymphoma), the R2 regimen was compared with rituximab alone.
   • With a median follow-up of 28 months, the median PFS was 39.4 months for R2 and 14.1 months for rituximab alone (P .0001), with no difference in OS.

Maintenance rituximab

After induction therapy with rituximab only or with rituximab plus chemotherapy, rituximab can be used once every 2 to 3 months. Several studies have evaluated this approach.

Evidence (maintenance rituximab for previously untreated patients):

1. In the PRIMA study (NCT00140582), 1,018 high-risk, previously untreated, symptomatic patients achieved CR or partial response (PR) after induction therapy with immunochemotherapy (usually R-CHOP) and were then randomly assigned to 2 years of maintenance rituximab versus no maintenance.
   • With a median follow-up of 9.0 years, median PFS favored rituximab maintenance (10.5 years) compared with observation (4.1 years) (HR, 0.61; 95% CI, 0.52−0.73; P .001), but with no difference in OS.
2. In the United Kingdom/International Study (NCT00112931), 379 previously untreated patients with asymptomatic, low-burden disease were randomly assigned to watchful waiting versus rituximab induction only versus rituximab induction followed by 2 years of rituximab maintenance.
   • Although OS and histological transformation rates were no different at 3 years, maintenance rituximab was favored based on quality-of-life (QOL) studies (Mental Adjustment to Cancer Scale P = .0004 at 7 months; Illness Coping Score P = .0012 at 7 months) and time-to-initiation of new treatment by 3 years (54% for watchful waiting vs. 12% for rituximab maintenance ).
   • This study suggested that for some patients, watch and wait resulted in watch and worry. However, from the perspective of OS and histological transformation rates, no benefit could be seen with rituximab maintenance.
3. In the RESORT study (NCT00075946), 289 previously untreated patients with asymptomatic, low-burden disease were randomly assigned to rituximab induction alone, with a re-treatment strategy that used rituximab at relapse, or rituximab induction plus maintenance rituximab every 13 weeks until treatment failure.
   • With a median follow-up of 4.5 years, there was no difference in median time-to-treatment failure (defined as failing rituximab alone) or in health-related QOL. A re-treatment strategy achieved comparable disease control using significantly fewer doses of rituximab.

   These three randomized trials in previously untreated patients show no advantage for the use of rituximab maintenance versus observation and reinduction of therapy at the time of relapse. The trials suggest a benefit for maintenance rituximab after reinduction for relapsed disease. Many questions remain about rituximab maintenance, particularly about truncating therapy at 2 years and long-term safety and efficacy. A trial extending rituximab maintenance to 5 years showed similar EFS or OS versus 1 year of maintenance after induction therapy with rituximab in previously untreated patients.

4. The FOLL12 study (NCT02063685) included 807 patients with previously untreated high-tumor burden follicular lymphoma. Patients received rituximab plus chemotherapy induction and were randomly assigned to either standard rituximab maintenance (every 8 weeks for 2 years) or to postinduction treatment (monitoring, rituximab maintenance, or radioimmunotherapy) based on their complete metabolic response and measurable residual disease–negative status.
   • With a median follow-up of 53 months, the 3-year PFS rate was 86% for patients who received standard maintenance and 72% for patients who received response-based treatment (P .001). The 3-year OS was the same in both groups (98% vs. 97%; P = .238).
   • This trial does not support the use of an end-of-treatment PET-CT scan to guide the use of maintenance rituximab.

For previously untreated patients, all of the studies showed improvement of PFS, with no change in OS.

Evidence (maintenance rituximab for previously treated patients):

1. In a prospective randomized trial of 465 patients with relapsed follicular lymphoma, responders to R-CHOP or CHOP were further randomly assigned to receive rituximab maintenance (1 dose every 3 months for 2 years) or no maintenance.
   • At a median follow-up of 6 years, rituximab maintenance was better for median PFS (44 months vs. 16 months, P .001) and borderline for 5-year OS (74% vs. 64%, P = .07).
   • This benefit for maintenance was evident even for patients who received rituximab during induction therapy. Most patients in both arms received extensive rituximab during postprotocol salvage treatment.
2. In a prospective randomized trial of 280 patients with relapsed follicular lymphoma, responders to chemotherapy and autologous stem cell transplantation consolidation were randomly assigned to receive four doses of rituximab maintenance or no maintenance.
   • With an 8.3-year median follow-up, the 10-year PFS rates favored maintenance (54% vs. 37% ), but there was no difference in OS.
3. A meta-analysis of nine randomized clinical trials with a total of 2,586 patients with follicular lymphoma, most of whom had relapsed disease, compared rituximab maintenance with no maintenance and showed improved OS for rituximab maintenance in previously treated patients (HRdeath, 0.72; 95% CI, 0.57–0.91).

For previously treated patients, there is more evidence to suggest an OS advantage with the use of rituximab maintenance.

Obinutuzumab alone or in combination with cytotoxic agents used in front-line therapy

Obinutuzumab is a glycoengineered type II anti–CD20 monoclonal antibody with greater antibody-dependent cellular cytotoxicity than rituximab.

Evidence (obinutuzumab):

1. A prospective randomized trial (NCT01332968) of 1,202 patients with previously untreated follicular lymphoma compared obinutuzumab combined with bendamustine (50%), cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) (33%), or cyclophosphamide, vincristine, and prednisone (CVP) (10%) with rituximab combined with the same chemotherapy regimens (based on investigator choice). After six cycles of combination therapy, patients had 2 years of maintenance therapy, receiving the same antibody every 2 months.
   • With a median follow-up of 34.5 months, the 3-year PFS rate was 80% in the obinutuzumab group and 73.3% in the rituximab group (HR, 0.66; 95% CI, 0.51–0.85, P = .001).
   • There was no difference in OS.
   • There was a high rate of toxic deaths among patients using bendamustine in the obinutuzumab arm (5.6%) and in the rituximab arm (4.4%) compared with what has been seen historically. For patients with indolent low-grade lymphoma, with median survivals exceeding 15 years, the number of toxic deaths during first-line therapy seems excessive. By comparison, the toxic death rate was 1% to 2% when either antibody was combined with CHOP or CVP.

   Several issues have been raised about this study:

   • The side effects were significantly higher with obinutuzumab in terms of infusion reactions and subsequent adverse events.
   • Obinutuzumab costs significantly more than rituximab.

   In summary, in the absence of any change in OS, switching from rituximab to obinutuzumab in combination with chemotherapy for previously untreated follicular lymphoma is a difficult choice. The PFS differences may be attributable to the imbalance in monoclonal antibody dosing, and the increased side effects and costs are mitigating factors. In this trial, bendamustine combined with either antibody led to unacceptable rates of toxic death.

Phosphatidylinositol 3-kinase (PI3K) inhibitor

Copanlisib

Evidence (copanlisib):

1. A double-blind, randomized, placebo-controlled study assigned 307 patients with recurrent indolent lymphoma to copanlisib plus rituximab and 151 patients to placebo plus rituximab (2:1 randomization).
   • With a median follow-up of 19.2 months, the median PFS was 21.5 months for patients in the copanlisib-plus-rituximab arm (95% CI, 17.8–33.0) versus 13.8 months (95% CI, 10.2–17.5) for patients in the placebo-plus-rituximab arm (HR, 0.52; 95% CI, 0.39–0.69; P .0001).
   • Hyperglycemia and hypertension were the most common grade 3 to 4 adverse events, occurring in half of the patients who received copanlisib.

The PI3K inhibitors have significant adverse effects, including pneumonitis, colitis, transaminitis, hypertension, hyperglycemia, rash, and increased risk of infections. These adverse events have affected the use of these agents until confirmatory trial results can establish their efficacy and safety. They are currently approved for treatment of relapsed and refractory follicular lymphoma after two previously received lines of therapy.

EZH2 inhibitor

Tazemetostat

Tazemetostat is an inhibitor of EZH2, a histone methyltransferase essential to the formation of lymph node germinal centers, especially with activating mutations of EZH2.

Evidence (tazemetostat):

1. A phase II study included 99 patients with relapsed or refractory follicular lymphoma, 45 of whom had activating mutations of EZH2, and 54 of whom had normal wild-type EZH2.
   • Treatment with tazemetostat resulted in an objective response rate of 69% (95% CI, 53%–82%) for patients with activating mutations versus 35% (95% CI, 23%–49%) for patients with wild-type EZH2.
   • With a median follow-up of 22 months, the median PFS was 13.8 months (95% CI, 10.7–22.0) for patients with activating mutations and 11.1 months (95% CI, 3.7–14.6) for patients with wild-type EZH2.
   • Grade 3 or 4 treatment-related adverse events were seen in 4% of patients.

Radiolabeled anti-CD20 monoclonal antibodies

Yttrium Y 90 (90Y)-ibritumomab tiuxetan (Zevalin) is available for previously untreated and relapsing patients with minimal (25%) or no marrow involvement with lymphoma (iodine I 131 -tositumomab is no longer available because of commercial disengagement).

Evidence (radiolabeled anti-CD20 monoclonal antibodies):

1. In a randomized, prospective trial, 554 patients with previously untreated advanced-stage follicular lymphoma received either R-CHOP times six cycles or CHOP times six cycles followed by 131I-tositumomab radioimmunotherapy (RIT).
   • With a median follow-up of 10.3 years, there was improvement in PFS for RIT (56% vs. 42%, P = .01) but no significant difference in 10-year OS (OS R-CHOP, 81%; CHOP-RIT, 75%; P = 0.13).
   • In this trial, the cumulative incidence of death from myelodysplastic syndrome or acute myeloid leukemia is higher for the RIT arm (4% vs. 1%, P = .02).

   131I-tositumomab became commercially unavailable in 2013.

2. In a randomized trial of 409 patients with stage III or IV follicular lymphoma who achieved a CR or PR, 90Y-ibritumomab tiuxetan consolidation versus no consolidation was evaluated.
   • The radiolabeled antibody consolidation improved median PFS by 3 years (P .001), and median time to next treatment was improved by 5.1 years (P .001); however, there was no change in OS.

Durable responses to radiolabeled monoclonal antibodies, such as 90Y-ibritumomab tiuxetan (commercially available) and iodine I 131-tositumomab (commercially unavailable), have also been reported before and after cytotoxic chemotherapy. However, the cumulative incidence of death resulting from myelodysplastic syndrome or acute myeloid leukemia is higher (4% vs. 1%; P = .02) in one of the randomized trials versus nonradiolabeled antibody with chemotherapy.

In general, treatment with standard agents rarely produces a cure in patients whose disease has relapsed. Sustained remissions after relapse can often be obtained in patients with indolent lymphomas, but relapse will usually ensue. Favorable survival after relapse has been associated with an age younger than 60 years, complete remission rather than partial remission, and duration of response longer than 2 years. Even the most favorable subset, however, has a tenfold greater mortality compared with age-adjusted U.S. population rates.

Patients who experience a relapse with indolent lymphoma can often have their disease controlled with single agent or combination chemotherapy, rituximab (an anti–CD20 monoclonal antibody), lenalidomide, radiolabeled anti–CD20 monoclonal antibodies, or palliative radiation therapy. Long-term freedom from second relapse, however, is uncommon and multiple relapses will usually occur. Patients with indolent lymphoma may experience a relapse with a more aggressive histology. If the clinical pattern of relapse suggests that the disease is behaving in a more aggressive manner, a biopsy can be performed. Documentation of conversion to a more aggressive histology requires an appropriate change to therapy applicable to that histological type. Rapid growth or discordant growth between various disease sites may indicate a histological conversion.

In a retrospective review of 325 patients between 1972 and 1999, the risk of histological transformation was 30% by 10 years from diagnosis. In this series, high-risk factors for subsequent histological transformation were advanced stage, high-risk Follicular Lymphoma International Prognostic Index, and expectant management. The median survival after transformation was 1 to 2 years, with 25% of patients alive at 5 years and with approximately 10% to 20% of patients alive 10 years after re-treatment.

A prospective trial of 631 patients with follicular lymphoma and with a median follow-up of 60 months in the rituximab era (2002–2009) found a 5-year transformation rate (11%) to a higher-grade histology. The median overall survival (OS) after transformation was 50 months, and the 5-year OS rate was 66%, if the transformation occurred more than 18 months after a diagnosis of follicular lymphoma. This series describes a better prognosis for patients with transformation than was experienced by patients in the prerituximab era.

For descriptions of the regimens used to treat histological conversions, see the Treatment of Aggressive, Recurrent NHL section. The durability of the second remission may be short, and clinical trials can be considered.

Treatment Options for Indolent, Recurrent NHL

Treatment options for indolent, recurrent non-Hodgkin lymphoma (NHL) include the following:

1. Rituximab alone or in combination with cytotoxic agents used in front-line therapy.
2. Obinutuzumab alone or in combination with cytotoxic agents used in front-line therapy.
3. Lenalidomide and rituximab.
4. Phosphatidylinositol 3-kinase (PI3K) inhibitor.
   • Copanlisib.
5. EZH2 inhibitor.
   • Tazemetostat.
6. Palliative radiation therapy.
7. Chemotherapy (single agent or combination).
8. Radiolabeled anti-CD20 monoclonal antibodies.
9. Chimeric antigen receptor (CAR) T-cell therapy.
10. Bispecific T-cell engagers.
11. Stem cell transplantation (SCT).

Rituximab alone or in combination with cytotoxic agents used in front-line therapy

Rituximab results in a 40% to 50% response rate in patients who relapse with indolent B-cell lymphomas. Rituximab can also be combined with combination chemotherapy.

Evidence (rituximab):

1. In three randomized, prospective studies involving previously treated patients with relapsed indolent lymphoma, patients were randomly assigned to rituximab maintenance after re-treatment with combination chemotherapy (with or without rituximab during induction) or rituximab alone.
   • All trials showed prolongation of response duration, and one trial demonstrated improvement in median progression-free survival (PFS) (3.7 years vs. 1.3 years, P .001) and OS rate (74% vs. 64%, P = .07) at 5 years with a median follow-up of 39 months favoring maintenance rituximab.

Obinutuzumab alone or in combination with cytotoxic agents used in front-line therapy

Obinutuzumab is a CD20-binding monoclonal antibody with alternative epitope binding.

Evidence (obinutuzumab):

1. In a randomized prospective trial (NCT01059630) involving 396 patients with rituximab-refractory indolent lymphoma (mostly follicular lymphoma), patients received obinutuzumab plus bendamustine followed by obinutuzumab maintenance therapy for 2 years versus bendamustine alone with no maintenance therapy.
   • With a median follow-up of 31.8 months, the 2-year OS rate favored the obinutuzumab combination (74.5% vs. 65.1%) (hazard ratio , 0.67; 95% confidence interval , 0.47–0.96; P = .027). The median PFS also favored the obinutuzumab combination (25.8 months vs. 14.1 months ) (HR, 0.57; 95% CI, 0.44–0.73; P .001).
   • The contribution of maintenance therapy to the outcome could not be assessed in this design.

Lenalidomide and rituximab

Responses of 20% to 56% have been reported for lenalidomide, especially in patients with follicular lymphoma and small lymphocytic lymphoma, with even higher responses noted for the combination of lenalidomide and rituximab.

Phosphatidylinositol 3-kinase (PI3K) inhibitor

Copanlisib

Evidence (copanlisib):

1. A double-blind, randomized, placebo-controlled study assigned 307 patients with recurrent indolent lymphoma to copanlisib plus rituximab and 151 patients to placebo plus rituximab (2:1 randomization).
   • With a median follow-up of 19.2 months, the median PFS was 21.5 months for patients in the copanlisib-plus-rituximab arm (95% CI, 17.8–33.0) versus 13.8 months (95% CI, 10.2–17.5) for patients in the placebo-plus-rituximab arm (HR, 0.52; 95% CI, 0.39–0.69; P .0001).
   • Hyperglycemia and hypertension were the most common grade 3 to 4 adverse events, occurring in half of the patients who received copanlisib.
2. A phase II study included 142 patients with relapsed and refractory indolent lymphoma.
   • Treatment with intravenous copanlisib resulted in a 59% objective response rate (12% CR), with a median duration of response of 22.6 months.

The PI3K inhibitors have significant adverse effects, including pneumonitis, colitis, transaminitis, hypertension, hyperglycemia, rash, and increased risk of infections. These adverse events have affected the use of these agents until confirmatory trials can establish their efficacy in combinations. They are currently approved for treatment of relapsed and refractory follicular lymphoma after two previously received lines of therapy.

EZH2 inhibitor

Tazemetostat

Tazemetostat is an inhibitor of EZH2, a histone methyltransferase essential to the formation of lymph node germinal centers, especially with activating mutations of EZH2.

Evidence (tazemetostat):

1. A phase II study included 99 patients with relapsed or refractory follicular lymphoma, 45 of whom had activating mutations of EZH2, and 54 of whom had normal wild-type EZH2.
   • Treatment with tazemetostat resulted in an objective response rate of 69% (95% CI, 53%–82%) for patients with activating mutations versus 35% (95% CI, 23%–49%) for patients with wild-type EZH2.
   • With a median follow-up of 22 months, the median PFS was 13.8 months (95% CI, 10.7–22.0) for patients with activating mutations and 11.1 months (95% CI, 3.7–14.6) for patients with wild-type EZH2.
   • Grade 3 or 4 treatment-related adverse events were seen in 4% of patients.

Palliative radiation therapy

Palliation may be achieved with very low-dose (4 Gy) involved-field radiation therapy in two fractions for patients with indolent and aggressive relapsed disease. In a prospective randomized trial, treatment with 4 Gy was inferior to treatment with 24 Gy in 12 fractions in PFS (77% vs. 92%, P .0001).

Chemotherapy (single agent or combination)

Patients may respond to the original induction regimen again, especially if the duration of remission exceeds 1 year. For relapsing patients, rituximab alone or in combination with agents not previously used may induce remissions. For more information, see the Rituximab alone or in combination with cytotoxic agents used in front-line therapy section.

Radiolabeled anti-CD20 monoclonal antibodies

Durable responses to radiolabeled monoclonal antibodies, such as yttrium Y 90 (90Y)-ibritumomab tiuxetan (commercially available) and iodine I 131-tositumomab (commercially unavailable), have also been reported before and after cytotoxic chemotherapy. However, the cumulative incidence of death resulting from myelodysplastic syndrome or acute myeloid leukemia is higher (4% vs. 1%; P = .02) in one of the randomized trials versus nonradiolabeled antibody with chemotherapy.

Evidence (radiolabeled anti-CD20 monoclonal antibodies):

1. In a prospective trial of 409 patients with follicular lymphoma who responded to induction chemotherapy, patients were randomly assigned to 90Y-ibritumomab tiuxetan or no further consolidation.
   • With a median follow-up of 7.3 years, the 8-year PFS rate favored 90Y-ibritumomab tiuxetan (41% vs. 22% ), but there was no difference in OS.

Chimeric antigen receptor (CAR) T-cell therapy

CAR T-cell therapy, with the autologous anti-CD19 therapeutics axicabtagene ciloleucel, lisocabtagene maraleucel, or tisagenlecleucel, has been approved for patients with relapsed follicular lymphoma after two or more lines of prior therapy.

Evidence (CAR T-cell therapy):

1. In a phase II trial, 153 patients with relapsed or refractory follicular lymphoma or marginal zone lymphoma received axicabtagene ciloleucel.
   • With a median follow-up of 17.5 months, the overall response rate was 92% (95% CI, 85%–97%) and the CR rate was 74%.
   • The 18-month PFS rate was 64.8% (95% CI, 54.2%–73.5%).
   • Cytokine release syndrome occurred in 78% of patients and was grade 3 or 4 in 6% of patients.
   • Tocilizumab was required in 50% of all patients, and 5% required vasopressors. Grade 3 or 4 neurological events occurred in 15% of patients.
2. In a phase II trial, 98 patients with relapsed or refractory follicular lymphoma after two or more lines of prior therapy received anti-CD19 CAR T-cell therapy with tisagenlecleucel.
   • With a median follow-up of 16.6 months, the complete response rate was 69.1% (95% CI, 58.8%–78.3%), and the overall response rate was 86.2% (77.5%–92.4%).
   • Grade 3 or 4 cytokine release syndrome occurred in 48.5% of patients, and 37.1% had grade 3 or 4 neurotoxicity.

CAR T cells are being used for high-risk patients whose disease has relapsed rapidly after chemoimmunotherapy. Such an approach is considered in the context of numerous other available agents.

Bispecific T-cell engagers

Bispecific T-cell engagers bind to CD20 (or CD19) and to CD3 to direct T cells to eliminate malignant B cells. Similar to CAR T-cell therapy, almost one-half of patients who receive this therapy experience cytokine release syndrome.

Mosunetuzumab

Evidence (mosunetuzumab):

1. In a single-arm, multicenter, phase II study, 90 patients with relapsed or refractory follicular lymphoma after two or more prior lines of therapy (including an anti-CD20 therapy and an alkylating agent) received mosunetuzumab.
   • With a median follow up of 27 months, the objective response rate was 77.8% (95% CI, 67.8%–85.9%), and the complete response rate was 60.0% (95% CI, 49.1%–70.2%), per investigator assessment. The median PFS per investigator assessment was not reached. The 24-month PFS was 51.4 months (95% CI, 39.4–63.3) after receiving mosunetuzumab, compared with 23.5 months (95% CI, 14.5–32.5) for patients' last prior therapy (63% chemoimmunotherapy, 8% PI3K inhibitor, 2% CAR T-cell therapy, 2% anti-CD20 antibody plus lenalidomide, and others).
   • Cytokine release syndrome occurred in 44.4% of patients; 97.2% of cases were grade 1 or 2 in severity.

Stem cell transplantation

In many institutions, autologous or allogeneic SCTs are being used for high-risk patients whose disease has relapsed rapidly after chemoimmunotherapy. Such an approach is considered in the context of numerous other available agents.

Evidence (SCT):

1. The German Low-Grade Lymphoma Study Group treated 307 patients with follicular lymphoma with two cycles of CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone)-like induction chemotherapy and then randomly assigned them to autologous SCT versus interferon maintenance.
   • With a median follow-up of 4.2 years, the 5-year PFS rate was 65% for patients who received transplantation versus 33% for patients who received interferon (P .001). There was no difference in OS.

Patients with aggressive stage I or aggressive, contiguous stage II diffuse large B-cell lymphoma (DLBCL) are candidates for combination chemotherapy with or without involved-field radiation therapy (IF-XRT).

Patients with a resolved hepatitis B virus (HBV) infection (defined as hepatitis B surface antigen-negative but hepatitis B core antibody-positive) are at risk of reactivation of HBV and require monitoring of HBV DNA. Prophylactic nucleoside therapy lowered HBV reactivation from 10.8% to 2.1% in a retrospective study of 326 patients.

Treatment Options for Aggressive Stage I and Aggressive, Contiguous Stage II NHL

Treatment options for aggressive stage I and aggressive, contiguous stage II non-Hodgkin lymphoma (NHL) include the following:

1. R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone) with or without IF-XRT.
2. R-ACVBP (rituximab, doxorubicin, cyclophosphamide, vindesine, bleomycin, prednisone) (under clinical evaluation).

R-CHOP with or without IF-XRT

The confirmation of efficacy for rituximab in advanced-stage disease has suggested the use of R-CHOP with or without radiation therapy but its use is only supported by retrospective comparisons.

• R-CHOP (four to six cycles).
• R-CHOP (three to six cycles) + IF-XRT.

Evidence (R-CHOP with or without IF-XRT):

1. In a randomized prospective trial of 334 patients with nonbulky (≤7 cm) stage I or stage II DLBCL, after receiving four to six cycles of R-CHOP-14 (R-CHOP delivered every 2 weeks), patients were randomly assigned to receive or not receive 40 Gy of radiation therapy.
   • After a median follow-up of 64 months, the 5-year event-free survival (89%–92%, P = .18) and 5-year overall survival (OS) (92%–96%, P = .32) were the same.

   Similar to the results of randomized studies of radiation therapy in the prerituximab era, radiation therapy can be deferred in nonbulky early-stage patients. For patients unable to tolerate prolonged-course chemotherapy, three cycles of R-CHOP plus radiation therapy has produced equivalent results based on single-arm retrospective trials.

2. In a randomized prospective trial (NCT00278421) of 592 patients younger than 60 years with nonbulky (7.5 cm) stage I or stage II DLBCL, patients were randomly assigned to receive either four or six cycles of R-CHOP (with an extra two cycles of rituximab after four cycles).
   • With a median follow-up of 66 months, the 3-year progression-free survival (PFS) rate was 96% (95% CI, 94%−99%)for patients who received four cycles of R-CHOP, which was 3% better (lower limit of one-sided 95% CI was zero) than the PFS rate for patients who received six cycles, establishing noninferiority for the four-cycle regimen.
3. A retrospective analysis at Memorial Sloan Kettering Cancer Center between 2001 and 2015 included 341 patients with stage I disease. The analysis found that 66% of patients had an extranodal presentation. After R-CHOP (or a similar regimen), with or without radiation therapy, the 5-year disease-free survival rate was 77%, and the 5-year OS rate was 94%. A multivariate analysis suggested that radiation therapy may improve outcomes for patients with extranodal disease that is positron emission tomography (PET)–positive at the end of therapy. This hypothesis needs confirmation in a prospective randomized trial.

In summary, for patients with favorable prognosis nonbulky (7 cm) stage I or stage II DLBCL, four cycles of R-CHOP is sufficient. For patients with unfavorable prognosis, six cycles of R-CHOP or three cycles of R-CHOP and 40 Gy of radiation therapy can be used. Early-stage patients with bulky disease (>7.5 cm) have not been studied in randomized trials; combined-modality therapy with R-CHOP for four to six cycles plus radiation therapy is usually chosen. Although a retrospective study suggested that patients with stage I extranodal disease and a positive PET scan at the end of therapy may benefit from radiation therapy, this hypothesis must be confirmed in a prospective randomized trial.

The treatment of choice for patients with advanced stages of aggressive non-Hodgkin lymphoma (NHL) is combination chemotherapy, either alone or supplemented by local-field radiation therapy.

The following drug combinations are referred to in this section:

• R-CHOP: rituximab + cyclophosphamide + doxorubicin + vincristine + prednisone.
• R-ACVBP: rituximab, an anti–CD20 monoclonal antibody, + doxorubicin + cyclophosphamide + vindesine + bleomycin + prednisone.

Treatment Options for Aggressive, Noncontiguous Stage II/III/IV NHL

Treatment options for aggressive, noncontiguous stage II/III/IV NHL include the following:

1. R-CHOP.
2. Other combination chemotherapy.
3. Bone marrow transplantation (BMT) or stem cell transplantation (SCT) (under clinical evaluation).
4. Radiation therapy consolidation to sites of bulky disease (under clinical evaluation).

R-CHOP

The following studies established R-CHOP as the standard regimen for newly diagnosed patients with diffuse large B-cell lymphoma (DLBCL). Dose intensification of R-CHOP by a 14-day versus a 21-day cycle did not result in improved outcomes.

Evidence (R-CHOP):

1. R-CHOP showed improvement in event-free survival (EFS) and overall survival (OS) compared with CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) alone in 399 advanced-stage patients with DLBCL older than 60 years (EFS rate, 57% vs. 38%; P = .002, and OS rate, 70% vs. 57%; P = .007 at 2 years). At a median follow-up of 10 years, the OS rate of patients who received R-CHOP compared with patients who received CHOP was 44% versus 28% (P .0001).
2. Similarly, for 326 evaluable patients younger than 61 years, R-CHOP showed improvement in EFS and OS compared with CHOP alone (EFS rate, 79% vs. 59%, P = .001, and OS rate, 93% vs. 84%, P = .001 at 3 years).
3. A randomized study (DSHNHL-1999-1A ) of 1,222 patients older than 60 years compared R-CHOP given every 2 weeks for six or eight cycles with CHOP given every 2 weeks for six or eight cycles. With a median follow-up of 72 months, the EFS favored R-CHOP given every 2 weeks for six or eight cycles (EFS rate at 6 years, 74% vs. 56%; P .0001). The OS favored R-CHOP for only six cycles because of increased toxicity in the eight-cycle arm (OS rate at 6 years, 90% vs. 80%; P = .0004). There was no comparison with standard R-CHOP or CHOP given every 3 weeks.
4. A trial (NCT00140595) of 380 patients younger than 60 years with DLBCL and an age-adjusted International Prognostic Index (IPI) rating of 1 randomly assigned treatment of patients to ACVBP and R-ACVBP plus consolidation with methotrexate, ifosfamide, etoposide, and cytarabine versus CHOP and rituximab. With a median follow-up of 44 months, 3-year OS rates favored R-ACVBP (92% vs. 84%; hazard ratio , 0.44; 95% confidence interval , 0.28–0.81, P = .007). The significantly worse toxicities with R-ACVBP, the narrow target population (60 years with either elevated lactate dehydrogenase (LDH) or stage III-stage IV disease, but not both), and the lack of a confirmatory trial may inhibit adoption of R-ACVBP as a new standard of care.

Modifications to R-CHOP to achieve improved efficacy continue to be explored in clinical trials. For over 2 decades, no modification has replaced R-CHOP as the worldwide standard induction therapy.

Pola-R-CHP

R-CHOP has been compared to Pola-R-CHP (polatuzumab vedotin, rituximab, cyclophosphamide, doxorubicin, and prednisone). Polatuzumab is an antibody-drug conjugate composed of an anti-CD79B monoclonal antibody attached to vedotin (monomethyl auristatin E), a microtubule inhibitor.

Evidence (Pola-R-CHP):

1. A provocative, prospective, randomized study of 879 patients with previously untreated DLBCL compared R-CHOP with Pola-R-CHP. Polatuzumab vedotin was substituted for vincristine to mitigate neurological toxicity.
   • At a median follow-up of 28.2 months, the 2-year PFS rate was significantly higher in the Pola-R-CHP group than in the R-CHOP group: 76.7% (95% CI, 72.7%–80.0%) versus 70.2% (95% CI, 65.8%–74.6%) (HR, 0.73; 95% CI, 0.57–0.95; P = .02).
   • The 2-year OS rate was 88.7% (95% CI, 85.7%–91.6%) for patients who received Pola-R-CHP and 88.6% (95% CI, 85.6%–91.6%) for patients who received R-CHOP (HR, 0.94; 95% CI, 0.65–1.37; P = .75).

The follow-up interval was too short to establish whether the 6% improvement in the PFS rate will plateau or improve after 2 years, and there is no evidence of OS advantage. Nonetheless, physician advisory panels are considering whether this represents a new standard regimen to replace R-CHOP. The new regimen is more than twice the cost of R-CHOP using acquisition prices in 2022.

There is no validated trial for interim positron emission tomography–based treatment intensification. R-CHOP has curative potential even in patients older than 80 years who are frail and require reduced dosage of R-CHOP components. In a retrospective review of 239 patients, the 5-year cause-specific survival rate was 48% (95% CI, 41%−55%).

Less than 10% of patients with DLBCL present with a concurrent indolent lymphoma at diagnosis, and these are predominantly of germinal center B-cell phenotype. A retrospective review of 1,324 patients showed similar EFS (HR, 1.19) and OS (HR, 1.09). For 847 patients who were treated with R-CHOP and free of disease 24 months after therapy, the rate of indolent lymphoma relapse by 5 years was higher with a concurrent diagnosis of follicular lymphoma (7.4% vs. 2.1% at 5 years, P .01) and with a germinal center B-cell phenotype (3.9% vs. 0.0% at 5 years, P = .02).

Stage IE or IIE gastric DLBCL

Four case series involving more than 100 patients with stage IE or IIE disease (with or without associated mucosa-associated lymphatic tissue) and with positive Helicobacter pylori infection reported that more than 50% of patients attained a durable complete remission after appropriate antibiotic therapy to eradicate H. pylori.

Prognostic factors

The National Comprehensive Cancer Network IPI for aggressive NHL (diffuse large cell lymphoma) identifies the following five significant risk factors prognostic of OS and their associated risk scores:

• Age.
  • 40 years: 0.
  • 41–60 years: 1.
  • 61–75 years: 2.
  • >75 years: 3.
• Stage III/IV: 1.
• Performance status 2/3/4: 1.
• Serum LDH.
  • Normalized: 0.
  • >1x to 3x: 1.
  • >3x: 2.
• Number of extranodal sites ≥2: 1.

Risk scores:

• Low (0 or 1): 5-year OS, 96%; progression-free survival (PFS), 91%.
• Low intermediate (2 or 3): 5-year OS, 82%; PFS, 74%.
• High intermediate (4 or 5): 5-year OS, 64%; PFS, 51%.
• High (>6): 5-year OS, 33%; PFS, 30%.

Age-adjusted and stage-adjusted modifications of this IPI are used for younger patients with localized disease. Shorter intervals of time between diagnosis and treatment appear to be a surrogate for poor prognostic biological factors.

The BCL2 gene and rearrangement of the MYC gene or dual overexpression of the MYC gene, or both, confer a particularly poor prognosis. Patients at high risk of relapse may be considered for clinical trials. Molecular profiles of gene expression using DNA microarrays may help to stratify patients in the future for therapies directed at specific targets and to better predict survival after standard chemotherapy.

Treatment of tumor lysis syndrome

Patients with bulky and extensive lymphadenopathy and elevations of serum uric acid and LDH are at increased risk of tumor lysis syndrome resulting in metabolic derangements such as hyperuricemia, hyperkalemia, hyperphosphatemia, hypocalcemia, and subsequent acute renal failure. Treatment options include: alkaline hydration, allopurinol, and rasburicase, a recombinant urate oxidase.

CNS prophylaxis

The CNS-IPI tool predicts which patients have a CNS relapse risk exceeding 10%. It was developed by the German Lymphoma Study Group and validated by the British Columbia Cancer Agency database. The presence of four to six of the CNS-IPI risk factors (age >60 years, performance status ≥2, elevated LDH, stage III or IV disease, >1 extranodal site, or involvement of the kidneys or adrenal glands) was used to define a high-risk group for CNS recurrence (a 12% risk of CNS involvement by 2 years).

CNS prophylaxis (usually with four to six doses of intrathecal methotrexate) is often recommended for patients with testicular involvement. A retrospective analysis of the German RICOVER studies compared intrathecal methotrexate with no prophylaxis in patients with DLBCL. This study was completed during the R-CHOP treatment era. With the possible exception of patients with testicular involvement, the analysis showed that intrathecal methotrexate did not reduce the risk of CNS disease. Some clinicians employ high-dose intravenous (IV) methotrexate (usually four doses) as an alternative to intrathecal therapy because drug delivery is improved and patient morbidity is decreased. Five retrospective studies and one network meta-analysis evaluating high-dose methotrexate alone in patients with high-risk DLBCL also showed no improvement in CNS relapse rate. Patients deemed at high risk for CNS relapse (e.g., patients with four to six CNS-IPI risk factors) may receive intrathecal methotrexate or high-dose IV methotrexate, but the lack of confirmatory randomized studies calls this standard into question and shows an urgent need for better therapeutics verified in clinical trials. Although patients with testicular involvement or kidney/adrenal involvement are considered an exception, there is only anecdotal benefit from intrathecal or high-dose IV methotrexate in reducing CNS recurrence.

The addition of rituximab to CHOP-based regimens has significantly reduced the risk of CNS relapse in retrospective analyses. Patients with CNS dissemination at diagnosis or at relapse usually receive rituximab and high doses of methotrexate and/or cytarabine followed by autologous stem cell transplantation (SCT), but this approach has not been assessed in randomized trials.

BMT or SCT

Several randomized prospective trials evaluated the role of autologous BMT or SCT consolidation versus chemotherapy alone in patients with diffuse large cell lymphoma in first remission.; Although some of these trials demonstrated significant increases in EFS (by 10% to 20%) among patients who received high-dose therapy, significant differences in OS could not be demonstrated prospectively in any of the series.

Retrospective analyses of high-intermediate (two risk factors) or high-risk (more than three risk factors) patients as defined by IPI suggest improved survival with BMT in two of the trials. These studies do not establish that high-dose consolidation is of value to patients with aggressive lymphoma who are truly at high risk of relapse, and they also demonstrate that EFS may be a poor surrogate for OS for these patients.

Radiation therapy consolidation to sites of bulky disease

After R-CHOP induction chemotherapy (or similar regimens), the addition of involved-field radiation therapy to sites of initial bulky disease (≥5–10 cm) or to extralymphatic sites remains controversial. Increased risks, such as long-term toxicities (e.g., second malignancies), must be considered.

Treatment Options for Aggressive, Recurrent NHL

In a retrospective review of multiple international trials, 636 patients were identified as having refractory diffuse large B-cell lymphoma (DLBCL), which was defined as progression or stable disease during or just at completion of full-course chemotherapy or relapse within 1 year after autologous stem cell transplantation (SCT). With subsequent therapy the objective response rate was 26%, complete response (CR) rate was 7%, median overall survival (OS) was 6.3 months, and only 20% of patients were alive at 2 years. Even with reinduction chemotherapy with planned autologous SCT, outcomes remain poor.

Treatment options for aggressive, recurrent non-Hodgkin lymphoma (NHL) include the following:

1. Chimeric antigen receptor (CAR) T-cell therapy for primary refractory disease or relapse within one year.
2. Bone marrow transplantation (BMT)/SCT consolidation.
3. CAR T-cell therapy for relapse after autologous SCT.
4. Tafasitamab plus lenalidomide.
5. Rituximab plus lenalidomide.
6. Polatuzumab vedotin plus rituximab and bendamustine.
7. Loncastuximab tesirine.
8. Bispecific T-cell engagers.
9. Palliative radiation therapy.

CAR T-cell therapy for primary refractory disease or relapse within one year

Patients with DLBCL who relapse during or within 2 months of receiving R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone) chemotherapy have primary refractory disease. Any patient with disease relapse within 1 year of R-CHOP chemotherapy or with primary refractory disease has a poor prognosis even with reinduction using chemoimmunotherapy followed by autologous SCT. Patients who received CAR T-cell therapy had a 40% to 50% 3-year progression-free survival (PFS) rate with a 40-month follow-up, a result equivalent retrospectively to autologous SCT in bone marrow registries.

Three randomized trials compared chemoimmunotherapy followed by autologous SCT with CAR T-cell therapy with or without bridging chemoimmunotherapy for patients with high-risk relapsed disease defined as primary refractory disease or relapse within 12 months of initial R-CHOP therapy.

Evidence (CAR T-cell therapy):

1. A prospective randomized trial included 359 patients with primary refractory disease or relapse within 12 months of initial R-CHOP chemotherapy. Patients received the CAR T-cell therapy axicabtagene ciloleucel with only bridging steroids or second-line chemoimmunotherapy (usually R-ICE or R-DHAP ) followed by autologous SCT.
   • With a median follow-up of 24.9 months, the 2-year event-free survival (EFS) rate was 40.5% for patients who received axicabtagene ciloleucel and 16.3% for patients who received chemoimmunotherapy followed by autologous SCT (hazard ratio , 0.398; 95% confidence interval , 0.308–0.514; P .0001).
   • The OS was not significantly better at 2 years, with an HR of 0.73 (95% CI, 0.53–1.01; P = .027).
   • On the chemoimmunotherapy arm of the study, 64% of patients never received autologous SCT because of inadequate response, progression, or death.
   • Clinically meaningful and statistically significant differences in quality of life were obtained in the CAR T-cell arm at day 100 and day 150, compared with the standard of care.
   • Grade 3 or 4 cytokine release syndrome occurred in 6% of patients and grade 3 or 4 neurotoxicity occurred in 21% of patients.
2. A prospective randomized trial included 184 patients with primary refractory disease or relapse within 12 months of initial R-CHOP chemotherapy. Patients received the CAR T-cell therapy lisocabtagene maraleucel, with 63% of patients receiving bridging therapy or second-line chemoimmunotherapy followed by autologous SCT.
   • With a median follow-up of 6.2 months, the median PFS was 14.8 months for patients who received lisocabtagene maraleucel and 5.7 months for patients who received chemoimmunotherapy followed by autologous SCT (HR, 0.41; P = .001).
   • On the chemoimmunotherapy arm of the study, 53% of patients never received autologous SCT because of inadequate response, progression, or death.
   • Grade 3 cytokine release syndrome occurred in 1% of patients, and grade 3 neurotoxicity occurred in 4% of patients. There were no grade 4 or 5 occurrences.
3. A prospective randomized trial included 322 patients with primary refractory disease or relapse within 12 months of initial R-CHOP chemotherapy. Patients received the CAR T-cell therapy tisagenlecleucel, with most patients receiving bridging therapy to achieve response, or second-line chemoimmunotherapy followed by autologous SCT.
   • There was no difference in EFS for patients in either arm (HR, 1.07; 95% CI, 0.82–1.40; P = .69).
   • In the CAR T-cell therapy arm, 48% of patients received two or more cycles of chemoimmunotherapy as part of bridging therapy. This approach to bridging therapy may have led to an unacceptable number of cases of progressive disease.

In summary:

• For patients with high-risk relapsing DLBCL with primary refractory disease or relapse within 12 months of R-CHOP-based chemotherapy, axicabtagene ciloleucel and lisocabtagene maraleucel are superior induction regimens compared with chemoimmunotherapy with regimens like R-ICE, R-DHAP, and R-GDP (rituximab, gemcitabine, dexamethasone, and cisplatin).
• The interval until patients receive CAR T cells must be minimized, optimally by using only steroids, eliminating bridging chemoimmunotherapy, and infusing the CAR T-cell product as quickly as possible.
• The preference for CAR T-cell therapy over chemoimmunotherapy followed by autologous SCT does not apply to patients who relapse more than 12 months after R-CHOP therapy.
• The American Society of Clinical Oncology (ASCO) has compiled guidelines for the management of adverse events in patients treated with CAR T-cell therapy.

Bone marrow/stem cell transplantation consolidation

BMT consolidation is a treatment for patients whose lymphoma has relapsed. Preliminary studies indicate that approximately 20% to 40% of patients will have a long-term disease-free status, but the precise percentage depends on patient selection and the specific treatment used. Preparative drug regimens have varied; some investigators also use total-body irradiation. Similar success has been achieved using autologous marrow, with or without marrow purging, and allogeneic marrow.

Evidence (BMT):

1. In a prospective randomized study, (EORTC-PARMA), 215 patients in first or second relapse of aggressive lymphoma, younger than 60 years, and with no bone marrow or central nervous system involvement, were given two cycles of intensive combination chemotherapy. The 109 patients who responded were randomly assigned to receive four more cycles of chemotherapy and involved-field radiation therapy (IF-XRT) versus autologous BMT followed by IF-XRT. With a 5-year median follow-up, the EFS rate was significantly improved with transplantation (46% vs. 12%). The OS rate was also significantly better with transplantation (53% vs. 32%). Salvage BMT was unsuccessful for patients on the nontransplant arm whose disease relapsed.

   In general, patients who responded to initial therapy and who responded to conventional therapy for relapse before the BMT have had the best results.

2. In a prospective trial, patients who relapsed late (>12 months after diagnosis) had better OS than patients who relapsed earlier (the 8-year survival rate was 29% vs. 13%, P = .001).

Peripheral SCT has yielded results equivalent to standard autologous SCT. Even patients who never experienced complete remission with conventional chemotherapy may have prolonged PFS (31% at 5 years) after high-dose chemotherapy and hematopoietic SCT if they retain chemosensitivity to reinduction therapy. Some patients who relapse after a previous autologous SCT can have durable remissions after myeloablative or nonmyeloablative allogeneic SCT.; Reduced-intensity conditioning for allogeneic SCT typically involves fludarabine plus busulfan or fludarabine plus cyclophosphamide, with or without 2 Gy of total-body irradiation.

Evidence (peripheral SCT):

1. In a randomized prospective trial, 396 patients with DLBCL in first relapse or who were refractory to first-line therapy received either R-ICE or R-DHAP followed by autologous SCT.
   • There was no difference in 3-year EFS or OS.
2. In a randomized prospective trial, 619 patients with relapsed or refractory aggressive lymphoma received either R-DHAP or R-GDP followed by autologous SCT.
   • At a median follow-up of 53 months, there was no difference in EFS or OS, but patients who received R-GDP reported less toxicity.

CAR T-cell therapy for relapse after autologous SCT

In the event of disease relapse after autologous SCT, many patients receive consolidation with CAR T-cell therapy.

Multiple trials describe patients with refractory large B-cell lymphoma who underwent an infusion of T cells that were engineered to express a CAR to target the CD19 antigen expressed on the malignant B cells using three different constructs: axicabtagene ciloleucel, tisagenlecleucel, and lisocabtagene maraleucel. Each study reported a CR rate of 50% to 60% and a 2-year OS rate of 40% to 50%, but the long-term durability of response has yet to be determined in these highly-selected patients. This therapy is an option for patients with otherwise refractory or resistant disease. These results have been verified off-study in two reports that included 397 patients treated after U.S. Food and Drug Administration (FDA) approval. The highest risk patients who respond adequately may receive a subsequent allogeneic SCT consolidation in some cases if eligible.

ASCO has compiled guidelines for the management of adverse events in patients treated with CAR T-cell therapy.

Tafasitamab plus lenalidomide

Evidence (tafasitamab plus lenalidomide):

1. In a phase II study, 80 patients with relapsed or refractory DLBCL were treated with tafasitamab, a CD19-directed humanized monoclonal antibody combined with lenalidomide.
   • The CR rate was 43% and the objective response rate was 60%.

The FDA approved the combination of tafasitamab and lenalidomide for patients with relapsed or refractory DLBCL. It is unclear if using this CD19-directed therapy would interfere with consolidation using CD19-CAR T cells.

Rituximab plus lenalidomide

Evidence (rituximab plus lenalidomide):

1. In two phase II trials, 49 patients showed a 19% to 35% overall response rate to lenalidomide with rituximab.

Polatuzumab vedotin plus rituximab and bendamustine

Polatuzumab vedotin is a CD79b-directed monoclonal antibody conjugated to the cytotoxic agent vedotin (an antibody-drug conjugate).

Evidence (polatuzumab vedotin plus rituximab and bendamustine):

1. In a randomized, prospective trial, 80 patients with relapsed or refractory DLBCL were treated with either polatuzumab vedotin combined with bendamustine and rituximab (BR) or BR alone, with a primary end point of complete response.
   • The CR rate by positron emission tomography−computed tomography scan was 40% for the polatuzumab vedotin-BR combination, compared with 18% for BR alone (P = .026).
   • Similarly, the median PFS was higher for patients who received the polatuzumab vedotin combination (9.5 months) than for the patients who received BR alone (3.7 months) (HR, 0.36; 95% CI, 0.21−0.63; P .001); the OS was 12.4 months for patients who received the polatuzumab vedotin combination versus 4.7 months for the patients who received BR alone (HR, 0.42; 95% CI, 0.24−0.75; P = .002).

The FDA approved the combination of polatuzumab vedotin and BR for patients with relapsed or refractory DLBCL.

Loncastuximab tesirine

Loncastuximab tesirine is a CD19-directed antibody conjugated to a pyrrolobenzodiazepine dimer cytotoxin (an antibody-drug conjugate).

Evidence (loncastuximab tesirine):

1. A phase I and subsequent phase II trial included 184 patients with relapsed or refractory disease.
   • The overall response rate was 48.3% (95% CI, 39.9%–56.7%) and the CR rate was 24%.

Bispecific T-cell engagers

Bispecific T-cell engagers bind to CD20 (or CD19) and to CD3 to direct T cells to eliminate malignant B cells. As with CAR T-cell therapy, almost one-half of patients who receive this therapy experience cytokine release syndrome.

Palliative radiation therapy

In general, patients with aggressive lymphoma who relapse with indolent histology will benefit from palliative therapy. Palliation may be achieved with very low-dose (4 Gy) IF-XRT for patients with indolent and aggressive relapsed disease.

LBL is a very aggressive form of non-Hodgkin lymphoma (NHL), which often occurs in young patients but not exclusively. LBL is the lymphomatous manifestation of ALL. The treatment paradigms are based on trials for ALL because LBL and ALL are considered different manifestations of the same biological disease. LBL is commonly associated with large mediastinal masses and has a high predilection for disseminating to bone marrow and the central nervous system (CNS). Intensive combination chemotherapy with CNS prophylaxis is the standard treatment of this aggressive histological type of NHL. Radiation therapy is sometimes given to areas of bulky tumor masses. Because these forms of NHL tend to progress quickly, combination chemotherapy is instituted rapidly once the diagnosis has been confirmed. For more information, see Adult Acute Lymphoblastic Leukemia Treatment.

The most important aspects of the pretreatment staging workup include careful review of the following pathological specimens:

• Bone marrow aspirate.
• Biopsy specimen.
• Cerebrospinal fluid cytology.
• Lymphocyte marker.

Treatment Options for LBL/ALL

Treatment options for LBL include the following:

1. Intensive therapy.
2. Radiation therapy.

New treatment approaches are being developed by the national cooperative groups. Other approaches include the use of bone marrow transplantation for consolidation.

For more information, see Adult Acute Lymphoblastic Leukemia Treatment.

Intensive therapy

Standard treatment is intensive combination chemotherapy with CNS prophylaxis.

Radiation therapy

Radiation therapy is sometimes given to areas of bulky tumor masses.

Diffuse, small noncleaved-cell/Burkitt lymphoma typically involves younger patients and represents the most common type of pediatric non-Hodgkin lymphoma (NHL). High-grade B-cell lymphoma, not otherwise specified, includes lymphomas with Burkitt-like or blastoid morphology without double hit cytogenetics, and with germinal center B-cell phenotype. Up to one-half of patients have a single MYC rearrangement. Optimal treatment is poorly defined because the diagnosis is rare. Burkitt lymphoma regimens with central nervous system (CNS) prophylaxis are usually chosen.

Treatment Options for Diffuse, Small Noncleaved-Cell/Burkitt Lymphoma

Treatment options for diffuse, small, noncleaved-cell/Burkitt lymphoma include the following:

1. Aggressive multidrug regimens.
2. CNS prophylaxis.

Aggressive multidrug regimens

Treatment for diffuse, small noncleaved-cell/Burkitt lymphoma is usually an aggressive multidrug regimen similar to those used for the advanced-stage aggressive lymphomas (such as diffuse large cell). Adverse prognostic factors include age older than 40 years, high serum lactate dehydrogenase (>3 times normal), Eastern Cooperative Oncology Group performance status of 2 or greater, and CNS involvement. A retrospective review of 641 adult patients with Burkitt lymphoma from 30 U.S. cancer centers found a 3-year progression-free survival (PFS) rate of 64%. Nineteen percent of patients had CNS involvement, 14% had primary refractory disease, and the treatment-related mortality rate was 10%.

Evidence (aggressive multidrug regimens):

• Aggressive combination chemotherapy patterned after that used in childhood Burkitt lymphoma has been very successful for adult patients. More than 60% of advanced-stage patients were free of disease at 5 years.
• Rituximab has been incorporated into these aggressive combination chemotherapy regimens. A nonrandomized, single-arm, prospective, multicenter trial of 363 patients, aged 16 years to 85 years, showed a 5-year PFS rate of 71% and a 5-year overall survival rate of 80%.

CNS prophylaxis

Patients with diffuse, small noncleaved-cell/Burkitt lymphoma have a 20% to 30% lifetime risk of CNS involvement. CNS prophylaxis with methotrexate is recommended for all patients, usually given as four to six intrathecal injections. For more information, see Adult Acute Lymphoblastic Leukemia Treatment.

Evidence (CNS prophylaxis):

• In a series of 41 patients treated with systemic and intrathecal chemotherapy, 44% of those who presented with CNS disease and 13% of those who relapsed with CNS involvement became long-term disease-free survivors. CNS relapse patterns were similar whether or not patients received radiation therapy, but increased neurological deficits were noted among those patients who received radiation therapy.

Treatment Option Overview for NHL During Pregnancy

Table 5. Treatment Options for Non-Hodgkin Lymphoma (NHL) During Pregnancy

Indolent NHL During Pregnancy

Treatment may be delayed for those women with an indolent NHL.

Aggressive NHL During Pregnancy

Immediate therapy

According to anecdotal case series, most NHLs in pregnant patients are aggressive, and delay of therapy until after delivery appears to have poor outcomes. Consequently, some investigators favor immediate therapy, even during pregnancy. In a review of 121 patient case reports from 74 papers, one-half of the patients had very aggressive lymphomas, such as Burkitt lymphoma, and one-half of the patients had involvement of the breast, ovaries, uterus, or placenta. One-half of the patients received therapy antepartum, and the 6-month survival rate was reported at 53%, with a live-birth rate of 83%.

A multicenter retrospective analysis of 50 patients described pregnancy termination in 3 patients, deferral of therapy to postpartum in 15 patients (median 30 weeks gestation), and antenatal therapy applied to the remaining 32 patients (median 21 weeks gestation, all done after the first trimester). With a median follow-up of 41 months, the 3-year progression-free survival rate was 53%, and the overall survival rate was 82%, using R-CHOP (rituximab, cyclophosphamide, doxorubicin hydrochloride, vincristine sulfate, and prednisone) or modifications of this regimen.

Early delivery when feasible

For some women, early delivery, when feasible, may minimize or avoid exposure to chemotherapy or radiation therapy.

Termination of pregnancy

Termination of pregnancy in the first trimester may be an option that allows immediate therapy for women with aggressive NHL.

Evidence (treatment effect on children exposed in utero):

• With follow-up ranging from several months to 11 years, children who were exposed to high-dose doxorubicin-containing combination chemotherapy in utero (especially during the second and third trimester) had normal growth and no evidence of congenital malformations. For most of the chemotherapeutic agents used for the treatment of NHL, there are no data regarding long-term effects on children exposed in utero.
• In one anecdotal case, a newborn exposed to a rituximab-containing regimen in utero was born with no circulating B lymphocytes. The newborn was otherwise healthy and recovered the circulating B lymphocytes by age 6 months with no unusual or persisting intercurrent infections.