Phase 2 trial of polatuzumab vedotin, rituximab and dose attenuated CHP in patients 75 years and older with large B-cell lymphoma

Abstract Introduction Older patients (pts) with large B-cell lymphoma (LBCL) are underrepresented on prospective clinical trials. Dose attenuated rituximab, cyclophosphamide, doxorubicin, vincristine and prednisone (R-miniCHOP) is an effective, well-tolerated regimen in this population (Peyrade et al, 2011) and is recommended by guidelines as a standard of care. The addition of polatuzumab vedotin (pola) to standard R-CHOP (Pola-R-CHP) backbone provides a progression-free survival (PFS) benefit over R-CHOP in younger pts. We present the results of a phase 2 clinical trial of the addition of pola to dose attenuated chemoimmunotherapy (R-pola-miniCHP) in an older patient population with prospective geriatric assessment (GA). Methods We enrolled pts with LBCL aged 75 years (yrs) and older who were not candidates for standard R-CHOP. Key eligibility included EF≥45%, ECOG ≤3, and creatinine clearance of >30 ml/min. Cell of origin (germinal center (GCB)- vs non-GCB) was determined by immunohistochemistry. Pts were able to receive a cycle of anthracycline containing therapy prior to trial enrollment. Pts received six, 3-week cycles of pola 1.8 mg/kg, and R-miniCHP. Pts received pola and prednisone as a prephase beginning one week prior to cycle 1; however, prephase initiation could be deferred at the investigator's discretion. Pegfilgrastim was required for all pts. All pts had GA at screening, following cycle 1 and at the end of treatment. This included the Vulnerable Elders Survey (VES-13). The primary endpoint was 2-year PFS. Secondary endpoints included overall survival (OS), toxicity, objective response rate (ORR) and complete response (CR). Response assessments used the Lugano Criteria (Cheson et al, 2014). PFS and OS were estimated using the Kaplan-Meier method. Results Thirty-nine pts were enrolled and 37 received components of study treatment. Two pts died prior to starting study treatment due to rapidly progressive disease and were not included in the analysis. The median age was 80 yrs and there was a slight female predominance (54%). The most common histology was DLBCL, NOS (78.4%), followed by high grade B cell lymphoma (18.9%) and follicular large cell (2.7%). The majority of pts had GCB subtype (70.3%) and had an international prognostic index of ≥3 (62%). Twenty-nine (78.4 %) pts completed 6 cycles of therapy. Baseline VES-13 was available in 35 pts and 54.3% of pts were vulnerable. With a median follow-up of 1.4 yrs, the median PFS was 3.1 yrs and the median OS was not reached. The 2-yr PFS estimate was 55.0%, 95% CI=(33.6%, 72.1%). There were no differences in PFS or OS between GCB and non-GBC pts or between vulnerable and non-vulnerable pts by VES-13. Five pts were not evaluable for response due to toxicity prior to response assessment. Of the 32 evaluable pts, 25 had CR (78.1%), 4 had PR (12.5%), 2 had PD (6.3%), and 1 had SD (3.1%). ORR was 90% in non-GCB and 90.9% in GCB (p=1). CR rate was 90% in non-GCB and 72.7% in GCB (p=0.39). Thirty-seven pts were evaluable for safety. The most common adverse events (AEs) were fatigue (59.4%), constipation (43.2%), alopecia (40.5%), nausea (35.1%), diarrhea (32.4%) and dyspnea (32.4%). Approximately one third (35.1%) of pts experienced grade ≥3 AEs. Grade ≥3 AEs seen in more than 2 pts included lung infection (16.2%), atrial fibrillation (8.1%), skin infection (8.1%), and hyponatremia, syncope, hypoxia, urinary tract infection (5.4% each). Peripheral sensory neuropathy was seen in 18.9% of pts and was grade 1 in all cases. There were 5 deaths in pts during treatment. These included lung infection in 3 pts and 1 pt each with multiorgan failure and anorexia/dehydration. Three additional pts have died in follow up, 2 due to disease progression and 1 due to unknown reasons after treatment completion. Conclusions This is the first prospective study to report outcome data for the R-pola-miniCHP regimen. It is active in older pts with LBCL and toxicities are consistent with reported data with R-miniCHOP. The estimated 2-yr PFS did not meet our targeted increase of 20% over the historical 2-yr PFS of 47% with R-miniCHOP, but our cohort was highly enriched for pts with GCB subtype. While not statistically significant, the increased CR rate in the non-GCB group is consistent with the previously demonstrated benefit of pola in this population. Additional follow up is required to evaluate the durability of these responses.