TPS799 Background: PDAC has a very poor 5-year survival rate of 13% in the US and is projected to become the second leading cause of cancer-related deaths by 2030. It is aggressive and refractory, with an objective response rate to first-line mFOLFIRINOX in mPDAC of ~32%. Despite advances in precision oncology, there remains no validated biomarker to select first-line treatment. Blood-based testing using methylated ctDNA has shown promise as a treatment response biomarker. It does not require tissue, has a fast turn-around time, and high sensitivity with potential to detect subtle changes in tumor burden – the latter being of utmost importance given low ctDNA shedding is a known barrier to exploring ctDNA in PDAC. Methods: This prospective, single-center, two-part study is designed to investigate whether an increase in ctDNA quantity is a predictive biomarker of radiographic progression in mPDAC, and whether clinical implementation to help guide treatment decisions such as an early switch to a potentially more effective and less toxic therapy may lead to improved patient outcomes. The study will use the CLIA-approved Northstar Response assay which quantifies >2000 methylated loci and reports a highly sensitive and quantifiable Tumor Methylation Score. Part 1 is observational and will enroll N=30 with treatment-naïve mPDAC set to receive mFOLFIRINOX. Methylated ctDNA will be drawn at screening, and prior to chemo administration at weeks 0, 2, 4, 6, and 8. Baseline and week 8 imaging will be evaluated by RECIST. Part 2 will be interventional (N=50) with ctDNA drawn at same intervals, however, should the fold-change increase in ctDNA at week 4 ≥ than a determined threshold (to be determined in Part 1), the patient will be switched to SOC gemcitabine/nab-paclitaxel at week 6. Major inclusion criteria include treatment-naïve, ECOG ≤ 1, detectable screening methylated ctDNA, adequate organ function. The primary endpoint of Part 1 will be ctDNA fold-change at week 4 relative to week 0 and radiographic response (RECIST) on week 8. The primary endpoint for Part 2 will be 2L mPFS. In Part 1, a two-group logistic regression design, with outcome Y and one continuous explanatory variable X (ctDNA fold-change), will be used to test the AUC against the null value AUC 0.50. The comparison will be made using a two-sided Z-test with a type I error rate of 0.05. For a power of 80% with sample size of 9 non-responders (30%; which is the expected Y=1=no radiographic response), and 21 responders, the detectable AUC under the alternative hypothesis is 0.805. In Part 2, a single-group design will be used to test whether the 2L treatment hazard rate is different from historical control (2.7 months). Comparison will be made using a two-sided, one-sample logrank test, with a type I error rate 0.05. With a sample size of 22 pts, we will have 83% power to detect a 2L mPFS of 5.4 months in the early-switch arm. Clinical trial information: NCT07096362 .
Terrero et al. (Sat,) studied this question.