Resources to address challenges in first-line treatment intensification in metastatic castration-sensitive prostate cancer (mCSPC): A discrete choice experiment.

5087 Background: Althoughfirst-line treatment intensification (TI) (i.e., androgen-deprivation therapy with chemotherapy, novel hormonal therapies, or both) is recommended in mCSPC guidelines, it is used in <1/3 of patients. Phase 1 of the IMPLEMENT study used an implementation science approach to identify barriers to and facilitators of TI. Phase 2 of IMPLEMENT aimed to prioritize these factors and identify resources to increase TI. Methods: A discrete choice experiment was developed for 300 US-based physicians (150 oncologists, 150 urologists) who manage mCSPC. It included 6 descriptions of potential resources to help with TI decisions, based on results from Phase 1. Respondents selected from pairs of resource sets the set they perceived as most helpful across 12 scenarios. We analyzed responses using a mixed-effects logistic regression model and calculated a coefficient of helpfulness for each resource to identify those with the strongest impact on physicians’ decisions. Coefficients were required to have a minimum value of zero, as resources would be optional. Results: Overall, physicians preferred decision support tools and databases of post-treatment options. Urologists found decision support tools most beneficial (coefficient of helpfulness 3.27; 95% CI 2.90-3.64), while oncologists preferred post-treatment databases (2.58; 95% CI 2.29-2.89) and clinical trial summaries (2.41; 95% CI 2.13-2.69). Tools to reduce administrative burden and information on outcomes of earlier vs later TI were not considered as helpful compared to other resources and across specialties. Conclusions: There were key differences between specialties in the helpfulness of resources, indicating substantial variability in clinical practice needs. These findings provide a clear path to develop concrete, specialty-tailored tools to increase guideline-concordant treatment of mCSPC. [Table: see text]