For nearly two decades, efforts to shift the frontline treatment of diffuse large B-cell lymphoma (DLBCL) have remained unsuccessful, including increases in the duration or dose intensity of treatment, and the addition of targeted agents to the backbone of R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone/prednisolone). The POLARIX study, published in the New England Journal of Medicine in January 2022, evaluated polatuzumab vedotin (pola), an anti-CD79b antibody drug conjugate (ADC) linked with monomethyl auristatin E (MMAE), in combination with rituximab, cyclophosphamide, doxorubicin, and prednisone (Pola-R-CHP), essentially replacing vincristine in the CHOP backbone.1 Pola-R-CHP was superior to R-CHOP in adult patients with newly diagnosed DLBCL with an International Prognostic Index (IPI) of 2 or higher. Unlike prior attempts, this trial met its primary endpoint of significant improvement in progression free survival (PFS), albeit a modest improvement (24-month PFS of 76.7% vs. 70.2%). Will the POLARIX study finally achieve what so many other studies failed to do: change the treatment paradigm for frontline DLBCL?
This study provides ample discussion points for both sides of the debate as to the preferred frontline treatment option for DLBCL. Toxicity was mostly comparable, though there were higher rates of febrile neutropenia among those randomized to the pola-containing arm. To date, there is no difference in overall survival (OS), though preventing a relapse could spare patients the morbidity of the intensive, resource-heavy therapies available in the relapsed setting. POLARIX demonstrated fewer patients going on to salvage chemotherapy, stem cell transplant, or chimeric antigen receptor [CAR] T-cell therapy if they received frontline pola-R-CHP. Is the prevention of subsequent therapy in a small percentage of patients a strong enough argument to justify the societal impact of replacing vincristine, a relatively inexpensive chemotherapy, with a much more costly ADC for the majority of untreated patients with DLBCL? Could we identify a subgroup in which treatment tends to have a greater impact to justify these concerns?
Subgroup analysis of the POLARIX study suggested patients over the age of 60, those with activated B-cell–like (ABC) subtype, and those with an IPI of 3 or higher had a greater benefit from pola-R-CHP. However, this study was not designed or powered to compare PFS among these subgroups, heading caution in overinterpreting subgroup analyses. The size of the clinical subgroup is important to consider when weighing the magnitude of the impact. Both age older than 60 and IPI of 3 or higher were well represented subgroups across the study, with nearly two thirds of the population representing each category. ABC DLBCL on the other hand comprised about one third of the population. Does the fact that we spent two decades unsuccessfully attempting to exploit the underlying biology of DLBCL with novel therapy leave us unwilling to accept a positive study agnostic to cell of origin (COO)? A simple clinical prognostic tool, IPI, now serving to identify patients for which we should forego R-CHOP, is unsatisfactory. Surely, there should be more complexity to precision medicine in the modern era.
Early attempts at delineating divergent approaches to DLBCL arose when gene expression profiling revealed distinct subtypes of DLBCL based on COO. Uncovering the distinct genetic derivation of DLBCL subtypes could identify therapeutic vulnerability. These descriptions provided the rationale for several studies targeting the ABC subtype of DLBCL.2, 3 The B-cell receptor signaling pathway has been preferentially targeted for ABC DLBCL given its critical role in survival of these malignant cells. Pola targets the B cell receptor subunit CD79b, it is therefore plausible that it should have a greater effect on ABC than GCB DLBCL. Several studies done prior to the POLARIX trial demonstrated superior efficacy of pola-based therapy in ABC compared with that of GCB DLBCL.4–6 In the POLARIX study, the hazard ratio for PFS was 0.34 (95% CI 0.13-0.85) among the ABC DLBC group compared with 1.18 (95% CI 0.75-1.84) among the GCB DLBCL group. A recent meta-analysis including this data confirmed the magnitude of this differential effect, which is supported by the underlying biology.7 Therefore, while the POLARIX trial subgroup analysis was not powered to evaluate differences in outcomes based on COO, it did report no clear benefit in the GCB subtype, which is both rational and supported by prior studies.
While we wait for technology to evolve to the point that we have a clinically actionable tool that can inform biologically rational treatment selection, we must navigate the current available evidence. The IPI is readily available and user friendly. Therefore, pola-R-CHP is a reasonable choice to consider for patients with an IPI of 2 or higher. Among these patients, there is a strong biologic rationale to consider pola-R-CHP for those with ABC DLBCL.
Dr. Akkad indicated no relevant conflicts of interest. Dr. Nastoupil has received honoraria for participation in advisory boards and research support from Genentech/Roche.