A Transcriptomic-Based Tool to Predict Gemcitabine Sensitivity in Advanced Pancreatic Adenocarcinoma

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is most often diagnosed at an advanced stage that is not amenable to surgery. Therefore, systemic therapy is the only treatment able to improve patients’ outcomes. FOLFIRINOX triychemotherapy is the most effective regimen but comes at the expense of high toxicity. Gemcitabine-based regimens are the typical alternative and have varying survival and response rates. Therefore, establishing a potent stratification method to categorise chemotherapeutic efficacy would improve prognosis by allowing treatment to be tailored. All consecutive patients treated for a locally advanced or metastatic PDAC with first-line gemcitabine monotherapy were retrospectively recorded and included after archived FFPE tumour block recovery. RNA-seq data were obtained from 65 primary tumour and 36 metastasis samples. Patients were stratified by gemcitabine response profile using four RNA-based gemcitabine response signatures on the overall survival (OS) and progression-free survival (PFS). The GemCore signature achieved the best performance in discriminating gemcitabine-sensitive patients, defined as GemCore+, in both OS (P<0.0001) and PFS (P<0.0001). GemCore+ patients had a median OS of 13.9 months (95% CI: 9.51–17.18) and a median PFS of 4.85 months (95% CI: 4.29–8.07) versus a GemCore- OS of 3.1 months (95% CI: 2.33–4.79) and PFS of 1.15 months (95% CI: 0.49–1.87). We also explored the ability of GemCore to stratify biopsies from metastasis. Here, GemCore+ patients had an OS of 6.6 months (95% CI: 4.72–16.13) and PFS of 2.95 months (95% CI: 1.38–4.36) versus a GemCore- OS of 2.1 months (95% CI: 1.64–3.48) and PFS of 0.36 months (95% CI: 0.00–1.34). The RNA-based GemCore stratification predicted the benefit of gemcitabine in patients with advanced PDAC in samples obtained from primary tumours and metastatic sites.