Presentation (CTI)

OBJECTIVES: Traditionally, clinical trials have limited enrollment to patients with ECOG Status 0-1, representing patients with very few cancer symptoms at baseline. There has been a debate about expanding enrollment to ECOG Status 0-2, to better serve patients with higher unmet needs, and more accurately reflect real life clinical practice.
METHODS: We conducted an AI-enhanced literature review in pancreatic cancer to test whether the primary clinical outcomes, median progression free survival (PFS) and median overall survival (OS), differed in trials with broader inclusion criteria. Completed trials in advanced/metastatic pancreatic cancer with inclusion criteria of ECOG 0-2 and reported PFS and OS outcomes were identified manually on Clinicaltrials.gov. This was compared to the PFS and OS outcomes that were reported for all advanced/metastatic pancreatic cancer regardless of ECOG inclusion criteria which were identified in PubMed using EvidAI, an AI literature search and data extraction tool. The EvidAI results were quality checked by a manual check of the top 3% of highest and lowest values, a unit test on all values and a random spot check on an additional 3% of values.
RESULTS: A total of 5 studies with 10 cohorts were identified using Clinicaltrials.gov. A total of 375 studies and 617 cohorts were identified on PubMed using EvidAI. The average mPFS reported for ECOG 0-2 studies was 3.5m, and the average mOS was 7.0m. Although the PFS/OS results explicitly including ECOG2 are slightly lower than the general average, there was no significant difference between these values and the corresponding values identified by AI extraction (mPFS=4.9m, p=0.16; mOS=9.1m, p=0.18). No correction for sample size, study characteristics or study quality was made.
CONCLUSIONS: AI literature review and literature extraction can be a useful tool to efficiently filter and analyze large quantities of clinical data, but must be carefully supervised to ensure applicability of results.