Patient characteristics and treatment outcomes with Pola-R-CHP: A real-world study of patients with previously untreated diffuse large B-cell lymphoma (DLBCL) in the United States (US)

Abstract

Abstract Background: The Phase III POLARIX study reported improved progression-free survival (PFS) of polatuzumab vedotin (Pola) in combination with rituximab, cyclophosphamide, doxorubicin and prednisone (Pola-R-CHP) compared with rituximab in combination with cyclophosphamide, doxorubicin, vincristine and prednisone (R-CHOP) in patients with previously untreated, intermediate- or high-risk DLBCL (Tilly et al. NEJM 2022). Following regulatory approvals, Pola-R-CHP has become a new standard of care therapy for first-line (1L) DLBCL globally, including in the US. Real-world evidence on the use of Pola-R-CHP in the US has been limited and has primarily included patients treated at academic cancer centers only, demonstrating that Pola-R-CHP has a manageable safety profile and is effective (with robust response rates and high survival rates) in patients with previously untreated DLBCL (Iyengar et al. ASH 2024; Thiruvengadam et al. EHA 2025; Thiruvengadam et al. ICML 2025). This study aimed to describe patient characteristics, treatment patterns, effectiveness, and safety of Pola-R-CHP in patients with previously untreated DLBCL in the US, including representation of patients from community cancer centers. Methods: This study used the electronic health record (EHR)-derived, US nationwide, de-identified real-world Flatiron Health Research Database. Data were abstracted from structured (e.g. treatment administration) and unstructured (e.g. clinical notes, radiology/pathology reports) parts of the EHR. Patients diagnosed with non-Hodgkin lymphoma, with ≥2 documented clinic visits and evidence of DLBCL within the EHR, and who had initiated 1L Pola-R-CHP between January 1, 2023 and April 30, 2025, were included. Efficacy was described using PFS, time to next treatment or death (TTNT-D), and overall survival (OS). Disease progression was assessed using the earliest evidence from radiography/pathology reports, physical exams, or other clinical evidence. Rates of pre-selected safety events documented in the EHR during 1L Pola-R-CHP treatment were described. Reasons for discontinuation of Pola were also abstracted from the EHR (available up to a cutoff date of February 28, 2025). Summary statistics and Kaplan-Meier estimates based on non-missing data were provided. Results: The study included 216 patients treated with 1L Pola-R-CHP. Median age was 70 years (range 21–85), 14% were aged ≥80 years, 42% were female, 83% were White, and 65% had an International Prognostic Index risk score of ≥2. Additionally, most patients (91%) had DLBCL not otherwise specified, 12% had DLBCL transformed from a prior indolent malignancy, 7% had double-hit lymphoma, and 30% had double-expressor lymphoma. The proportion of patients that had a cell of origin of germinal center B-cell (GCB) and non-GCB were similar (41% and 40%, respectively). Overall, 76% of patients received care from a community setting, 66% had commercial insurance, and 30% had a low socioeconomic status score (SES) of 1 or 2 (i.e. 30% of patients were in the lowest two SES quintiles). Median follow-up was 10 months. In terms of patient disposition, 81% of patients were in ongoing follow-up (with no documented second-line [2L] therapy), 12% had documented 2L therapy, and 7% had died with no documented 2L therapy. Median number of cycles for Pola-R-CHP was 6 (range 1–14); 63% had completed ≥6 cycles. Median PFS, TTNT-D, and OS were not reached. Event-free survival rates for PFS, TTNT-D, and OS were 92%, 87%, and 92% at 6 months, and 83%, 77%, and 87% at 12 months, respectively. Time-to-event estimates were consistent between the overall cohort, in patients with ≥6 months of follow-up (n=160), and in patients with ≥12 months of follow-up (n=111). Time-to-event estimates were also similar between subgroups of SES scores (1–2 vs 3–5) and cell of origin (GCB vs non-GCB). Among 139 patients with available data, 84% completed therapy and 7% discontinued Pola due to toxicity. Other reasons for discontinuation of Pola included progression, withdrawal by patient, and unrelated medical issues. Safety events after initiation of Pola-R-CHP included nausea/vomiting (39%), neutropenia (36%), diarrhea (35%), anemia (31%), and peripheral neuropathy (9%). Conclusions: This study adds to the evidence for the effectiveness of Pola-R-CHP as a standard of care therapy in clinical practice for patients with previously untreated DLBCL primarily originating from community cancer centers in the US.