Physician Treatment Preferences in Relapsed/Refractory Follicular Lymphoma: A Discrete Choice Experiment

Introduction: Follicular lymphoma (FL) is an indolent form of non-Hodgkin lymphoma with limited treatment options at disease progression. Novel treatments, like bispecific antibodies and chimeric antigen receptor T-cell therapy, have recently emerged, each with novel safety and management profiles. Yet, contemporary research evaluating treatment preferences among physicians treating relapsed or refractory (r/r) FL is lacking. The objective of this study was to identify key treatment attributes that influence physicians' treatment decisions when treating patients with r/r FL.

Methods: We fielded a multi-country survey of physicians from the United States (US), United Kingdom (UK), Brazil, Japan, Germany and France, with up to 50 participants in each country. Data collected as of June 13, 2023 are reported with the remaining data from all countries expected by July 31st. Physicians were recruited through a healthcare research panel for an online, cross-sectional survey that included two discrete choice experiment (DCE) exercises: one for 2 ^nd line (2L) and one for 3 ^rd line (3L) therapy. The DCEs prompted physicians to choose between 2 treatment profiles that varied on 7 attributes associated with treatment for r/r FL: progression-free survival (PFS), overall survival (OS), serious adverse events (AE), cytokine release syndrome (CRS) events, neurological events, fatigue and administration/impact on functioning. Attributes and levels of interest were informed by a targeted literature review, clinical data on r/r FL treatments and clinical input. Sociodemographic and practice characteristics were summarized using descriptive statistics. We generated attribute level preference weights using hierarchical Bayesian modeling. Relative attribute importance (RI) was computed based on differences in preference weights between the most and least favorable level of each attribute. Bivariate statistics were used to evaluate differences between physicians. Interim results from the UK and US are reported with plans to include additional countries at time of presentation.

Results: This analysis included 98 physicians (49 US; 49 UK) who averaged 13.7 years treating patients with FL. Most physicians specialized in medical oncology or hematology oncology (US: 100.0%; UK: 83.7%). Overall, 55.1% of physicians practiced in academic settings whereas 44.9% practiced in community settings; the proportion of academic-based physicians was higher in the UK vs the US (79.6% vs 30.6%; p<0.001). Roughly 60% of patients managed by physicians were >50 years of age. The mean number of FL patients treated over the past six months among physicians from the UK was greater than physicians from the US [65.0 vs 43.0, respectively]. Yet, the mean proportion of patients receiving 2L and 3L therapy over the past six months was roughly equal between physicians from the US and UK [2L: 27.2% vs 26.1%; 3L: 15.9% vs 15.6%]. In the context of 2L (Figure 1) and 3L (Figure 2) therapy, improved PFS was selected as most important (2L RI=20.4%; 3L RI=27.5%) whereas reducing the risk of serious AEs was least important (2L=5.6%; 3L RI=6.7%). In 2L, reducing the risk of neurological events was the second most important attribute (RI=19.1%) followed by increasing OS (RI=16.2%). Conversely, in 3L therapy, increasing OS was the second most important attribute (RI=18.9%) followed by reducing the risk of neurological events (RI=14.3%). Reducing the risk of neurological events and the risk of CRS were relatively more important to physicians in the 2L vs 3L setting. In general, physicians from the US and UK assigned similar levels of importance to each attribute evaluated, regardless of treatment line. However, the RI of fatigue was higher among physicians from the UK compared to those from the US (13.5% vs. 10.0%; p=0.047).

Conclusion: Assessing physician treatment preference facilitates alignment between what matters most to patients and physicians' treatment decisions in r/r FL, particularly amidst an evolving treatment landscape. PFS was the most important attribute influencing physicians' treatment decisions for patients in 2L and 3L settings, which suggests that preserving a longer treatment-free period is important to physicians at both lines of therapy. Future treatments for r/r FL may need to demonstrate PFS benefits to influence physicians' willingness to make trade-offs between efficacy and side effects in earlier treatment lines.