Presentation (CTI)

OBJECTIVES: There is increasing emphasis on including RWE in health technology assessments and market access strategies. However, the identification of RWE through SLRs adds time and resources required for traditional SLRs focused on clinical evidence. Using a case-study in NSCLC, this study shows how a REAL-SLR approach incorporating both clinical and real-world evidence enhances the analysis of evolving treatment landscapes and improves efficiencies compared to traditional SLRs.
METHODS: We created a daily-updated REal-time AI-assisted Living SLR (REAL-SLR) in NSCLC, compliant with PRISMA and Cochrane guidelines. Workflows, staffing requirements, and timelines were compared with a traditional SLR.
RESULTS: As of January 5, 2025, our REAL-SLR reviewed over 18,000 publications and selected 1526 clinical trials and 845 RWE studies for inclusion. While a traditional SLR involving two junior and one senior researcher requires >3 months of full-time evidence curation, accessing and retrieving studies from the REAL-SLR can be achieved in <5 minutes. Currently there are limited treatment options for patients progressing on first-line immune checkpoint inhibitors (ICIs). Evidence from RWE is emerging that continued ICI or ICI-rechallenge may benefit select patient populations. A REAL-SLR identified 48 clinical trials evaluating ICI mono-or combination therapies in patients previously treated with ICIs. While no clinical trial addressed if ICI-rechallenge improved outcomes, 14 RWE studies provided insights. 4/4 studies showed ICI rechallenge leads to longer OS or PFS compared to non-ICI subsequent treatment; 3/3 studies showed that patients with acquired ICI-resistance were more likely to respond to continued ICI or ICI-rechallenge compared to those with primary resistance.
CONCLUSIONS: Leveraging efficiencies from REAL-SLR, incorporation and assessment of RWE in an NSCLC dataset demonstrates how RWE can augment clinical SLRs by identifying treatment gaps. Shifting from periodic manual reviews to a daily-updated model significantly reduces time and cost while maintaining decision-ready evidence.