Investigation of pharmacodynamic and prognostic potential of FAP-activity assessed through FAP-mediated cleavage of type III collagen in patients with PDAC undergoing chemotherapy.

Abstract

e15051 Background: The purpose of the study was to investigate the pharmacodynamic and prognostic potential of FAP-cleaved type III collagen (C3F), a serologically assessed biomarker reflecting FAP-activity, in patients with PDAC receiving chemotherapy. Methods: C3F, a fragment of type III collagen released upon FAP cleavage, was quantified serologically using ELISAs with specific monoclonal antibodies to reflect FAP activity. Similarly, PRO-C3, the cleaved pro-peptide of type III collagen, was measured to evaluate type III collagen formation. C3F and PRO-C3 was measured in serum samples from 64 patients with stage 2-4 PDAC prior to chemotherapy, at an early follow-up (median = 21 days after first treatment), and at a late follow-up (median = 89 days after first treatment). Correlations between the two biomarkers and across the three timepoints were assessed using Spearman correlations. Biomarker measurements at the three time points were compared using Tukey’s multiple comparisons test. The prognostic potential of C3F was analyzed using Kaplan-Meier curves and uni- and multivariate Cox proportional hazard models, adjusting for age, sex, disease stage (2-3 vs. 4), pretreatment performance status (PS), and pretreatment PRO-C3 levels (≤median vs > median). Results: C3F and PRO-C3 levels demonstrated a weak correlation before treatment (r = 0.32, p = 0.011), and no correlation observed after treatment (r = -0.21, p = 0.102; r = -0.05, p = 0.723 at the 1 st and 2 nd follow-up, respectively). C3F levels decreased significantly after treatment initiation ( p = 0.0003), while PRO-C3 showed no significant changes. Detectable C3F levels (C3F + ) after treatment were associated with better overall survival (OS) (Hazard ratio (HR) = 0.60, 95% CI 0.33-1.07, p = 0.08), with a median OS of 13.1 months (95% CI 8.2-16.5) for patients with undetectable C3F levels (C3F - ) and 19.1 months (95% CI 13.3-30.5) for post-treatment C3F + patients. After adjusting for age, sex, disease stage, PS, and pretreatment PRO-C3 levels, the presence of detectable C3F post-treatment levels was significantly associated with better OS (HR = 0.53, 95% CI 0.29-0.98, p = 0.04). However, pretreatment C3F levels were not associated with OS (HR = 0.82, 95% CI: 0.47-1.41, p = 0.47), with a median OS of 14.4 months (95%CI: 8.4-17.1) for C3F - patients and 15.5 months (95% CI 8.5-17.4) for pretreatment C3F + patients. Conclusions: The weak to absent correlation between C3F and PRO-C3, along with differences in biomarker response to treatment, indicates that they reflect distinct biological processes. Chemotherapy had a pharmacodynamic effect on C3F, and post-treatment detectability of C3F demonstrated prognostic potential, whereas pretreatment levels did not. These findings indicate the potential utility of C3F as a biomarker for assessing treatment response in patients with PDAC.