Prognostic factors and effect modifiers in patients with relapsed or refractory follicular lymphoma who failed at least two lines of therapy: a systematic literature and expert clinical review

Abstract

To enhance the quality of real-world external comparative studies, it is essential to systematically identify, prespecify, and account for prognostic variables and effect measure modifiers (EMMs), especially between emulated target trial and real-world control arms. These factors can then be utilized to evaluate cohort comparability and perform covariate adjustments, such as in propensity score models. A systematic literature review (SLR)-based identification of prognostic factors, coupled with expert clinical review, offers a comprehensive approach to evaluating and ranking the level of evidence, while also assisting in selection of prognostic factors to assess imbalances between cohorts in single-arm trials and real-world data studies. We performed an SLR followed by a clinical review and ranking by subject-matter experts to identify prognostic factors and EMMs in patients with relapsed or refractory (r/r) follicular lymphoma (FL) who failed at least two lines of therapy (LoTs). Across 13 included studies, the SLR identified 28 prognostic factors that were significantly associated with clinical outcomes, including overall survival, progression-free survival, and objective response rate. Notably, our review did not identify any statistically significant EMMs. Based on expert ranking of the SLR-derived list, the 5 most important prognostic variables in descending order are: progression of disease within 24 months of first LoT (POD24), chemo-immunorefractory/chemoresistant, refractory to last LoT, number of prior LoTs, and serum lactate dehydrogenase. This comprehensive SLR and expert review highlight critical prognostic factors in r/r FL. The identified prognostic variables can inform future research, emphasizing the need for continued investigation into factors affecting outcomes in this challenging and heterogeneous patient population.