Old Dog, New Tricks: Targeting CD20 in B Cell Non-Hodgkin Lymphomas

In a year that saw the groundbreaking U.S. Food and Drug Administration (FDA) approval of two different chimeric antigen receptor (CAR) T cell products for the treatment of B cell malignancies, following a year that established the use of PD1 blockade in relapsed Hodgkin lymphoma, it is perhaps surprising that the next best thing to happen in lymphoma therapy in 2017 involves an old target — CD20. Yet, two randomized trials published in the New England Journal of Medicine, one on mantle cell lymphoma (MCL) and one on follicular lymphoma (FL), show that anti-CD20 directed therapy continues to offer significant and meaningful advances for our patients with incurable lymphomas.^1,2 

The treatment of MCL has seen many improvements during the past 10 to 15 years, including the addition of autologous stem-cell transplantation (ASCT) in first remission,³  the incorporation of cytarabine into up-front chemotherapy regimens,^4-6  the establishment of bendamustine as an effective drug,⁷  and the activity of ibrutinib in relapsed/refractory disease.⁸  This has led to an improvement in both progression-free (PFS) and overall survival (OS) when compared to historical controls. However, apart from ASCT, the relative benefit of these therapies has not been validated in a randomized prospective fashion. As a result, there is no consensus standard first-line therapy for this disease, with varied induction therapies and guidelines for the use of ASCT existing across the globe.

This year, the LYSA group published the results of a randomized clinical trial comparing three years of maintenance rituximab to observation following ASCT for patients who had a response to induction chemotherapy.¹  Patients were treated with four cycles of R-DHAP (rituximab plus dexamethasone, cytarabine, and a platinum derivative), with or without four cycles of dose-dense R-CHOP (rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone) pending response, and those achieving a partial (tumor mass reduction by > 75%) or complete response underwent a consolidation ASCT with R-BEAM (rituximab plus armustine, etoposide, ara-c, and melphalan) conditioning chemotherapy. Patients were randomized within the first three months of ASCT (n=240) to observation (n=120), versus maintenance rituximab every two months for a planned three years (n=120). After a median follow-up of 54.4 months, maintenance rituximab was associated with an improvement in PFS (4-year PFS, 83% vs. 64%; p<0.001) and OS (4-year OS, 89% vs. 80%; p=0.04) compared with observation. Three years of maintenance rituximab was administered to 70 percent of the intervention group, and the main reasons for discontinuing rituximab were disease progression (n=16) and neutropenia (n=9). The rates of severe infection after ASCT were similar between each group.

The benefit of maintenance rituximab for MCL has been suggested by other studies. In older patients who are not candidates for intensive induction therapy and/or ASCT, the use of indefinite maintenance rituximab every two months following successful induction therapy with R-CHOP resulted in a significantly increased response duration and OS compared with observation.⁹  Additionally, the Nordic MCL trial allowed for the use of rituximab to treat a minimal residual disease (MRD) relapse prior to clinical relapse.⁶  The LYSA study, however, is the first study to demonstrate that the addition of maintenance therapy following intensive induction chemoimmunotherapy and ASCT improves not only response duration but also OS. This has significant practice-changing implications for the treatment of these patients. Importantly, however, this study did not address whether any treatment can be taken away from these multiphase regimens. The prognostic importance of MRD at the end of induction chemotherapy has been demonstrated, and while patients on this study had MRD testing, the results of this testing were not used to differentiate therapy. To know whether ASCT is necessary for all responding patients with previously untreated MCL, we must await completion of the ongoing ECOG-ACRIN Cancer Research Group study (NCT03267433). This study will randomize patients who are MRD-negative following induction chemotherapy (investigator choice) to either ASCT followed by maintenance rituximab, or to maintenance rituximab alone. With a primary endpoint of OS, we may begin to better understand the relative value of each component of this treatment paradigm.

Just as rituximab has had (and continues to have) important practice-changing effects on MCL, it is the only therapy that has improved OS for FL patients as well.^10,11  Newer anti-CD20 antibodies such as obinutuzumab have been developed recently; they differ from rituximab with respect to epitope binding as well as extent of antibody-dependent cellular cytotoxicity and phagocytosis. Obinutuzumab has been FDA approved in combination with chlorambucil and bendamustine for the treatment of previously untreated chronic lymphocytic leukemia and rituximab-refractory FL, respectively.^12,13  The GALLIUM study published this year is the first to compare obinutuzumab versus rituximab in combination with chemotherapy for the upfront treatment of FL.²  In this study, previously untreated patients were randomized to receive rituximab chemotherapy (n=601) or obinutuzumab chemotherapy (n=601). Chemotherapy could consist of bendamustine, CHOP, or CVP (cyclophosphamide, vincristine and prednisone); bendamustine was most common, being used in slightly less than 60 percent of patients. Responding patients did receive two years of maintenance therapy with the same antibody they had previously received. Although response rates were similar between the two groups, there was a PFS benefit seen with obinutuzumab (3-year PFS, 80% vs. 73%; p=0.001). After a short median follow-up of only 34.5 months, no difference in OS was seen.

Importantly, MRD status at the end of induction therapy correlated with outcome, and MRD-negative patients had an improved PFS.¹⁴  The rates of MRD negativity in the bone marrow, but not the peripheral blood, was improved in the obinutuzumab chemotherapy groups compared with rituximab chemotherapy groups, but this difference was most pronounced with CHOP, and was not seen following bendamustine. The obinutuzumab arms did have higher rates of high grade adverse events. The relationship between MRD negativity and outcome and the lack of significant difference of MRD negativity following either antibody in combination with bendamustine is interesting. The clinical study was not powered to detect PFS differences for each specific chemotherapy regimen, but this does cast doubt as to whether obinutuzumab has the same impact on PFS when used in combination with bendamustine, or whether this difference may be driven by the CHOP and CVP groups. Many oncologists have changed practice and incorporated obinutuzumab in combination with chemotherapy for the upfront treatment of FL, but given the lack of a known survival benefit and impact on response to subsequent therapies, the similar rates of MRD negativity of each antibody in combination with bendamustine, and the increased high grade adverse event rate seen, I and others continue to use rituximab in combination with bendamustine in this setting.

Exactly 20 years following the initial FDA approval of rituximab for the treatment of B cell lymphoma, we are still learning about new ways in which targeting CD20 benefits patients, especially those with previously untreated incurable lymphoma where survival benefits are meaningful and often difficult to demonstrate. The incorporation of post-ASCT maintenance rituximab for the upfront treatment of MCL is now considered by most to be a new standard of care. The adoption of the use of obinutuzumab in combination with chemotherapy for the upfront treatment of FL, with the lack of proven survival benefit and the increased higher-grade toxicity, is instead more variable.