Abstract
Introduction: More than half of patients (pts) diagnosed with diffuse large B-cell lymphoma (DLBCL) are diagnosed at the age of 65 and older. While R-CHOP is an established standard of care for most pts with DLBCL, older or more frail pts may receive dose-attenuated treatment with R-“mini”-CHOP in an attempt to balance safety and efficacy. Our study sought to characterize the real-world population of pts diagnosed with DLBCL at age 80 or older (≥ 80) and assess differences in first-line (1L) treatment patterns and resulting outcomes.
Methods: This retrospective, observational study utilized the COTA real-world database derived from the electronic health records (EHR) of partnered healthcare centers in the U.S. Eligible pts were ≥ 80 when diagnosed with DLBCL between 1/1/2011 and 12/31/2024 and initiated 1L therapy. Pts were excluded if they had multiple primary cancers at diagnosis (except for non-melanoma skin cancer or lobular carcinoma in situ) or were missing key study dates. Index date was defined as the date of initiation of 1L therapy. Patient characteristics and treatment patterns were summarized descriptively overall and by 1L treatment class, grouped into the following categories: standard dose R-CHOP/R-EPOCH, R-mini-CHOP, or Other Frail Regimens. The window used to assess patient characteristics was +/- 30 days from index or, if a given variable was only assessed at diagnosis, any time prior to diagnosis through +30 days from index. Time-to-event outcomes were evaluated using the Kaplan-Meier method, including real-world progression-free survival (rwPFS) and real-world overall survival (rwOS). Treatment discontinuation rates and reasons for discontinuation were summarized overall and by treatment group.
Results: In the COTA database, 1,391 pts met the study criteria. Median age at diagnosis was 84 years, and pts were predominantly White (77.3%) and treated in the community setting (84.9%). Median follow up time (reverse KM) from diagnosis was 58.5 months (mos) (95% CI: 53.1, 63.3).
A total of 651 (46.8%) pts received 1L R-CHOP/R-EPOCH, 395 (28.4%) received 1L R-mini-CHOP, and 292 (21.0%) received 1L Other Frail Regimens (mostly commonly including R-CVP, single agent R and R-bendamustine). Utilization of R-mini-CHOP increased over time. There were 53 (3.8%) pts who received other therapies (investigational agents or non-anthracycline-based intensive therapy) and were not compared. Median follow up time (reverse KM) from diagnosis by treatment group was: R-CHOP/R-EPOCH: 72.2 mos, R-mini-CHOP: 37.6 mos, and Other Frail Regimens: 57.6 mos. Patient characteristics were relatively similar across the treatment groups, though pts receiving Other Frail Regimens tended to be slightly older with a higher ECOG performance status (PS). Comparing R-CHOP/R-EPOCH, R-mini-CHOP, and Other Frail Regimens, respectively: median age at diagnosis: 82 vs. 84 vs. 85 years, ECOG PS ≥ 2: 20.4% vs. 19.2% vs. 33.2%, CCI score ≥ 3 14.6% vs. 19.2% vs. 19.2%, B symptoms: 32.6% vs. 27.6% vs. 33.2%, bulky tumor: 25.5% vs. 24.8% vs. 21.2%, and extranodal disease: 72.5% vs. 77.0% vs. 76.7%.
Pts who received Other Frail Regimens had a higher rate of treatment discontinuation (44.2%), as compared to R-CHOP/R-EPOCH (29.0%) and R-mini-CHOP (21.5%). Among pts who discontinued treatment, a higher proportion of pts discontinued R-CHOP/R-EPOCH therapy due to toxicity (38.6%), as compared to R-mini-CHOP (28.2%) and Other Frail Regimens (24.8%).
Long-term outcomes were relatively comparable between R-CHOP/R-EPOCH and R-mini-CHOP, but significantly worse for pts who received Other Frail Regimens, respectively: median (95% CI) rwPFS 34.0 mos (27.1, 45.8) vs. 28.5 mos (17.9, 45.6) vs. 6.9 mos (6.0, 8.5), rwOS: 47.3 mos (40.5, 54.9) vs. 45.3 mos (34.7, 55.0), 13.5 mos (11.5, 19.8).
Conclusions: We present one of the largest real-world datasets of pts with DLBCL ≥ 80. Utilization of R-mini-CHOP resulted in comparable outcomes to more intensive regimens, with lower rates of discontinuation due to toxicity. However, use of further intensity-modified frail regimens resulted in worse outcomes compared to R-mini-CHOP, despite similar baseline age and CCI scores. This finding supports continued use of “mini” anthracycline containing regimens in pts ≥ 80 and their use as comparators in randomized trials.
