POLARIX: Is R-CHOP in the Rearview Mirror?

Diffuse large B-cell lymphoma (DLBCL) is the most common lymphoma subtype. While more than half of newly diagnosed patients can anticipate being cured with R-CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone, in combination with rituximab), nearly 40 percent of patients will succumb to their disease within the first few years following diagnosis. Numerous approaches have been explored to improve upon R-CHOP for patients with previously untreated DLBCL, including intensifying the chemotherapy^1,2  or the delivery of rituximab,³  replacing rituximab with an enhanced glycoengineered antibody,⁴  or adding targeted therapy to exploit the underlying biology.^5,6  Large, randomized phase III studies supported by strong rationale and promising early-phase studies failed to surpass R-CHOP. This inspired efforts to revisit the underlying cell of origin and re-classify DLBCL into five molecular subtypes that may inform precision medicine approaches. We also learned that introducing barriers to trial participation tends to exclude patients with urgent need for therapy and enrich studies with good-risk patients who tend to do well with R-CHOP, failing to address the unmet need. Despite the important lessons learned, we were left with R-CHOP as the standard of care in previously untreated DLBCL.

Polatuzumab vedotin (pola) is an antibody drug conjugate (ADC) composed of a CD79b monoclonal antibody conjugated to monomethyl auristatin E (MMAE). Pola provides targeted delivery of a potent microtubule inhibitor to CD79b, ubiquitously expressed on the surface of DLBCL cells. In the relapsed DLBCL setting, pola combination with bendamustine-R resulted in a significant improvement in progression-free survival (PFS) and overall survival (OS) over bendamustine-R alone.⁷  After demonstrating adequate safety with replacement of vincristine with polatuzumab vedotin (due to overlapping peripheral neuropathy toxicity) in the frontline pola-R-CHP phase Ib study,⁸  the phase III randomized study POLARIX set out to challenge R-CHOP with pola-R-CHP in previously untreated adult patients with DLBCL and an International Prognostic Index (IPI) of 2 or higher. More than 1,000 patients were enrolled; 879 underwent 1:1 randomization. Baseline characteristics were well balanced. Despite small numbers, the study enrolled 26 (7.9%) double- or triple-hit patients on the pola-R-CHP arm and 19 (5.7%) on the R-CHOP arm.

With a median follow up of 28.2 months, the study met its primary end point: pola-R-CHP resulted in superior two-year PFS estimates compared to R-CHOP (76.7% vs. 70.2%), which in turn resulted in a reduction in the risk of progression, relapse, or death by 27 percent. There was no OS difference observed. Safety was similar between the two groups, as was dose intensity, with low rates of dose delay, dose reduction, or discontinuation across the study. Febrile neutropenia was higher in the pola-R-CHP group (13.8% vs. 8%). The rates of grade 3 or higher infections were similar (15.2% vs. 12.6%). Peripheral neuropathy of any grade was similar, with lower rates of grade 2 or higher (13.8%. vs. 16.7%) and dose reduction due to peripheral neuropathy (4.4% vs. 8%) in the pola-R-CHP group. Though not powered to address this, the subgroups that tended to benefit from pola-R-CHP were those with male patients older than 60 years, with an IPI of 3 or higher and activated B-cell subtype.