Key points are not available for this paper at this time.
Abstract Background: FOLFIRINOX (FFX) and gemcitabine plus nab-paclitaxel (GA) are two recommended first-line systemic options for pancreatic cancer; however, there is limited evidence to support a particular regimen or provide guidance for treatment selection. We sought to compare outcomes for FFX versus GA across disease stages, and test whether there was an association between clinicogenomic features, treatment response, and survival. Methods: For this retrospective analysis, we utilized AACR’s GENIE PANC v1. 2 database, the first large-scale pancreatic cancer dataset that combines genomic data with clinical annotations. The clinicogenomic profiles of patients were stratified and compared by stage: resectable/borderline resectable (R/BR), locally advanced (LA), or metastatic (MET). Patients who received FFX or GA as first-line therapy were compared across stages, with a focus on MET patients. We evaluated OS and PFS as survival outcomes and performed an adjusted analysis using Cox proportional hazards modeling. Treatment responses were further evaluated by change in serum CA 19-9. To test for features associated with treatment response, the genomic profiles of best and worst responders were compared. Results: Of 987 patients with complete genomic data, there were 452 R/BR, 82 LA, and 453 MET. While there was no difference in KRAS status, patients with later-stage disease accumulated greater common co-alterations: CDKN2A (19%, 23%, 35%), SMAD4 (26%, 29%, 30%), TP53 (72%, 73%, 77%), and ARID1A (6%, 6%, 11). Within MET patients, there was no significant difference in PFS between those treated with FFX (n=239) and GA (n=82), but patients treated with FFX had significantly longer OS, even after multivariable adjustment (median 1. 14 vs. 0. 854 years, p 0. 01; HR: 0. 66, 95% CI: 0. 49- 0. 89, p 0. 01). Despite improved OS in FFX patients, GA patients had a greater reduction in CA 19-9 (median 50. 0 vs. 69. 0 U/mL). In addition, when comparing the genomic profiles of best/worst responders for MET patients treated with FFX and GA, there was no clear signal that one regimen should be favored over another in a specific context. Exploratory analyses in the R/BR and LA groups indicated no significant survival differences between first-line FFX vs. GA patients, nor in first-line treatment among patients who developed metastases post- diagnosis. Conclusion: Common co-alterations appear to accumulate in later-stage disease, and outcomes in patients undergoing systemic therapy is worse generally in patients with more complex genomic profiles. However, we did not observe any specific genomic features suggestive of a particular response to FFX vs. GA. Of note, we found that Met GA patients had a larger magnitude in CA 19-9 response, even though Met FFX was associated with improved OS. This provokes important questions regarding how biochemical response should be gauged in patients treated with these regimens. Citation Format: Kunling Tong, Elishama Kanu, Ethan Agritelley, Jiayin Bao, Peter J Allen, Daniel P Nussbaum. Clinicogenomic determinants of response to FOLFIRINOX versus gemcitabine plus nab-paclitaxel as first-line systemic therapy for pancreatic cancer abstract. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Pancreatic Cancer Research; 2024 Sep 15-18; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2024;84 (17 Suppl₂): Abstract nr C054.
Tong et al. (Sun,) studied this question.