Characterizing the US Patient Population Receiving Rituximab with Gemcitabine and Oxaliplatin (R-GemOx) for Relapsed or Refractory Diffuse Large B-Cell Lymphoma Using Real-World Data

Background: Outcomes among transplant-ineligible patients (pts) with relapsed or refractory (R/R) diffuse large B-cell lymphoma (DLBCL) are suboptimal. Access to newer treatment (tx) options, such as chimeric antigen receptor (CAR) T-cell therapies, may be limited due to cost, geographic location, patient's fitness, or rapidly progressing disease. R-GemOx has been a conventional tx option for R/R DLBCL and is used as a control arm for several ongoing global phase 3 trials. Despite this, the understanding of practice patterns and outcomes of R-GemOx in the US is limited which in turn limits the US benchmarking of global trials. FIatiron Health has developed a nationwide, longitudinal, geographically diverse real-world (RW) database for DLBCL by aggregating information derived from electronic health records (EHR) of pts within its network, which is composed primarily of community cancer practices. This retrospective observational study summarizes tx patterns and outcomes of R-GemOx for R/R DLBCL using this database. Methods: Pts in the deidentified RW database who were diagnosed withDLBCL from 2011 to 2023, and with evidence on the earliest use of R-GemOx in 2nd or later line of therapy (LOT) were included in the study cohort. Data were abstracted from structured (e.g., tx administration) and unstructured (e.g., clinician notes, radiology/pathology reports) parts of the EHR with a cut-off date of September 30, 2023. Demographic and clinical pt characteristics were described at the start of R-GemOx tx along with prior and next anti-lymphoma therapy. Response assessments were based on recorded radiographic assessments and corresponding information documented in the EHR. Progression date was assessed using earliest evidence of disease progression from radiographic/pathologic reports, physical exam, or other clinical evidence. Outcomes were described for overall response rate (ORR), complete response (CR) rate, progression-free survival (PFS), and overall survival (OS) among pts with non-missing data for each individual endpoint. Results: The study cohort included 281 pts who had received R-GemOx for R/R DLBCL, including 66% (186/281) treated only at community cancer centers. The majority of pts were male (63%, 176/281) and white (78%, 207/266). DLBCL, not otherwise specified was documented for 86% (242/281) of pts. DLBCL transformed from a prior indolent malignancy was noted in 22% (61/281) and 13% (36/281) had double-hit lymphoma. Median age at R-GemOx initiation was 71 years (range: 22-85) and 39% (109/281) of pts were ≥75 years. Among those with available data, Eastern Cooperative Oncology Group performance status was 0/1 in 71% (105/148) and 66% (77/117) had elevated lactate dehydrogenase. The majority of pts (69%, 194/281) had primary refractory disease and 77% (215/281) were refractory to their last prior therapy. R-GemOx was received as the 2nd LOT by 52% (146/281) of pts with usage ranging from 2nd to 8th LOT. Prior therapy included anti-CD20 therapy (99%, 279/281), anthracyclines (89%, 251/281), autologous stem cell transplantation ([ASCT] 10%, 27/281), and CAR T-cell therapy (1%, 4/281). R-GemOx was immediately followed by ASCT in the same LOT in 7% (19/281) of pts. Use of R-GemOx as bridging to CAR T-cell therapy was explicitly noted in 10% (26/273) of pts. CAR T-cell therapy, captured as a LOT of its own, was received by 22% of 158 pts with recorded next LOT. In this cohort, ORR was 45% (95% confidence interval [CI]: 38‒53) and CR rate was 22% (95% CI: 16‒29). Median PFS and OS were 2.8 months (95% CI: 2.6‒3.7) and 12.7 months (95% CI: 10.6‒17.1), respectively. Compared with the overall cohort, pts with primary refractory disease had worse outcomes (ORR: 38%, CR: 16%, median PFS: 2.7 months, median OS: 10.1 months). Pts who received R-GemOx in the 2nd LOT had better outcomes compared with those who received R-GemOx in the 3rd LOT or later (ORR: 46% vs 45%, CR: 26% vs 17%, median PFS: 3.5 vs 2.7 months, median OS: 19.6 vs 10.0 months, respectively). Discussion: This cohort of pts with R/R DLBCL characterizes the RW use of R-GemOx in the US. These data demonstrate that R-GemOx has been used as a viable tx option in the heterogeneous R/R DLBCL setting and offers a RW benchmark for the efficacy of this regimen.