OBJECTIVES: Real-world evidence (RWE) to support Food and Drug Administration review of oncology drugs has included historical controls to contextualize safety and efficacy in single-arm trials, and RWE describing safety and effectiveness in routine care settings. Development of validated methods to accurately classify oncology treatment effectiveness outcomes using real-world data (RWD) is needed.

METHODS: We developed a novel methodology, real-world Lugano (rwLugano), derived from Lugano 2014 criteria, to classify lymphoma treatment response using RWD. To assess rwLugano performance, we conducted a retrospective multisite chart review study using Cardinal Health’s Practice Research Network (PRN) to identify adult patients with diffuse large B-cell lymphoma (DLBCL) treated with first line chemoimmunotherapy in US clinical practice (01JAN2015-31DEC2022). We compared the proportion of patients classified as complete response (CR) using three methods: 1) physician-charted, 2) rwLugano-derived, and 3) blinded independent central review (BICR) using Lugano 2014. Statistical analyses included percent agreement, concordance [Cohen’s kappa (κ)], and multivariate generalized linear mixed model (GLMM) [Odds Ratio (OR); 95% CI].

RESULTS: We identified 174 DLBCL patients [male n=103 (59%); female n=71 (41%); mean age 66 years] across 6 PRN sites. CR assignment at initial response assessment was proportionately lower for physician-charted [111 (64%)], compared to rwLugano [145 (83%)] and BICR [142 (82%)]. Physician-charted response [agreement: 76%; κ=0.43] underestimated CR compared to BICR [OR=0.22; 95%CI:0.12-0.43]. However, rwLugano [agreement: 86%; κ=0.50] was not statistically different [OR=1.2; 95%CI:0.6-2.4] than BICR.

CONCLUSIONS: BICR per Lugano criteria is the standard for outcome classification in lymphoma clinical trials yet is typically unfeasible in RWE studies. We developed rwLugano to address this issue. In this cohort, charted response, the current standard for RWD, tended to underestimate CR. However, rwLugano classification performed similar to BICR indicating this novel methodology may be a relevant measure of outcome classification in real-world clinical research. Further validation would be beneficial.