Comparison of first-line chemotherapy regimens in unresectable locally advanced or metastatic pancreatic cancer: a systematic review and Bayesian network meta-analysis

Background: In advanced pancreatic ductal adenocarcinoma (PDAC), first-line chemotherapy is the standard of care. Due to the absence of head-to-head comparisons in clinical trials, we performed this systematic review and network meta-analysis to compare treatment options for PDAC in terms of their efficacy and toxicity. Methods: PubMed, the Cochrane Central Register of Controlled Trials, Embase, and oncological meetings websites were searched until Nov 15, 2023. We included phase 2–3 randomised controlled trials published after Jan 1, 2000, evaluating first-line treatments in patients with previously untreated, unresectable, locally advanced or metastatic PDAC. Primary endpoints assessed were progression-free survival and overall survival. Summary data were extracted from published reports. The deviance information criterion was used to choose between a random-effects or fixed-effects model. Hazard ratios (HRs) with 95% credible intervals were estimated using a Bayesian approach. The risk of bias was evaluated using the Cochrane Risk of Bias 2 (RoB 2) tool and studies were graded as low, some concerns, or high risk of bias. The quality of evidence was evaluated using the Grading of Recommendations Assessment, Development, and Evaluation approach. This systematic review and network meta-analysis is registered with PROSPERO, CRD42023450330. Findings: 6050 records were screened and 79 randomised controlled trials (22 168 patients) were included in the analysis. Gemcitabine was the most frequent comparator (in 50 [63%] of 79 trials) and was considered as the reference treatment. A fixed-effect model was used to analyse the primary outcomes. Regarding progression-free survival (71 trials, 19 479 patients), the most effective treatments were gemcitabine plus nab-paclitaxel alternating folinic acid, fluorouracil, and oxaliplatin ([FOLFOX] HR 0·32, 95% credible interval 0·22–0·47), cisplatin, nab-paclitaxel, capecitabine, and gemcitabine ([PAXG] 0·35, 0·22–0·55), and liposomal irinotecan in combination with fluorouracil, leucovorin, and oxaliplatin ([NALIRIFOX] 0·43, 0·34–0·54), followed by fluorouracil, leucovorin, irinotecan, and oxaliplatin ([FOLFIRINOX] 0·55, 0·47–0·65) and gemcitabine plus nab-paclitaxel (0·62, 0·54–0·72). Similar results were observed for overall survival (79 trials, 22 104 patients): PAXG (HR 0·40, 95% credible interval 0·25–0·65), gemcitabine plus nab-paclitaxel alternating FOLFOX (0·46, 0·32–0·66), and NALIRIFOX (0·56, 0·45–0·70) had the highest benefit, followed by FOLFIRINOX (0·66, 0·56–0·78) and gemcitabine plus nab-paclitaxel (0·67, 0·59–0·77). The overall risk of bias was low to some concerns. Certainty of evidence was low. Interpretation: Our findings suggest that NALIRIFOX and FOLFIRINOX should be the preferred options for patients who can tolerate these regimens, with gemcitabine plus nab-paclitaxel remaining a viable alternative, particularly in patients unfit for triplet therapy. Phase 3 randomised controlled trials investigating concomitant or sequential quadruplets are warranted. Funding: None.