COMPARATIVE EFFECTIVENESS ANALYSIS OF FOLFIRINOX AND GEMCITABINE PLUS NAB-PACLITAXEL IN METASTATIC PANCREATIC ADRENOCARCINOMA; USING MARGINAL STRUCTURAL MODEL
Author(s)
Adebowale S. Adeyemi, BPharm1, Pegah Farrokhi, PharmD1, David Stenehjem, PharmD2.
1Department of Pharmaceutical Care and Health Systems, University of Minnesota, College of Pharmacy, Minneapolis, MN, USA, 2Department of Pharmacy Practice and Pharmaceutical Sciences, University of Minnesota, College of Pharmacy, Duluth, MN, USA.
1Department of Pharmaceutical Care and Health Systems, University of Minnesota, College of Pharmacy, Minneapolis, MN, USA, 2Department of Pharmacy Practice and Pharmaceutical Sciences, University of Minnesota, College of Pharmacy, Duluth, MN, USA.
OBJECTIVES: FOLFIRINOX and Gemcitabine plus nab-paclitaxel (Gem+NP) have demonstrated clinically meaningful survival benefits in randomized clinical trials. However, the comparative effectiveness of these regimens in routine clinical practice remains uncertain. This study applied causal inference approaches to estimate the average treatment effect (ATE) of FOLFIRINOX versus Gem+NP on overall survival(OS) in patients with pancreatic adrenocarcinoma.
METHODS: Using the Flatiron EHR dataset, adults(≥18 years) with metastatic pancreatic adenocarcinoma diagnosed between January 2019 and August 2024 were included. The exposure was first-line chemotherapy (FOLFIRINOX and Gem+NP). OS was the primary outcome. Covariates were balanced using propensity scores. A marginal structural model was used to estimate the average treatment effect, and other supplemental analyses were used to strengthen the analysis.
RESULTS: In adjusted analyses, the inverse probability-weighted MSM estimated HR of 0.861 (95% CI, 0.801-0.926, p = 5.18 × 10⁻⁵), consistent with the adjusted outcome regression Cox model (HR = 0.847, 95% CI 0.8455-0.8482, p < 0.001). Median OS was longest with FOLFIRINOX (10.35 months, 95% CI 9.90 -10.90), followed by Gem+NP (7.48 months, 95% CI 7.03-7.97) and no treatment (5.42 months, 95% CI 4.87-6.00). Supplemental analysis included a conditional Cox analysis and subgroup analysis. After covariate adjustment, the conditional Cox model showed that FOLFIRINOX has a non-significant survival benefit (HR = 0.945, 95% CI 0.887-1.007, p = 0.0819), while Gem+NP remained associated with increased risk (HR = 1.090, 95% CI 1.027-1.157, p = 0.0046). Subgroup comparative analyses indicated that smokers had greater benefit with FOLFIRINOX (MSM HR = 0.81, 95% CI 0.74-0.90; outcome regression HR = 0.80, 95% CI 0.73-0.88) than non-smokers (MSM HR = 0.92, 95% CI 0.82-1.02; outcome regression HR = 0.90, 95% CI 0.81-1.00), showing that treatment worked differently across groups, rather than smoking itself being beneficial.
CONCLUSIONS: FOLFIRINOX was associated with the longest survival, and the findings were consistent across causal models.
METHODS: Using the Flatiron EHR dataset, adults(≥18 years) with metastatic pancreatic adenocarcinoma diagnosed between January 2019 and August 2024 were included. The exposure was first-line chemotherapy (FOLFIRINOX and Gem+NP). OS was the primary outcome. Covariates were balanced using propensity scores. A marginal structural model was used to estimate the average treatment effect, and other supplemental analyses were used to strengthen the analysis.
RESULTS: In adjusted analyses, the inverse probability-weighted MSM estimated HR of 0.861 (95% CI, 0.801-0.926, p = 5.18 × 10⁻⁵), consistent with the adjusted outcome regression Cox model (HR = 0.847, 95% CI 0.8455-0.8482, p < 0.001). Median OS was longest with FOLFIRINOX (10.35 months, 95% CI 9.90 -10.90), followed by Gem+NP (7.48 months, 95% CI 7.03-7.97) and no treatment (5.42 months, 95% CI 4.87-6.00). Supplemental analysis included a conditional Cox analysis and subgroup analysis. After covariate adjustment, the conditional Cox model showed that FOLFIRINOX has a non-significant survival benefit (HR = 0.945, 95% CI 0.887-1.007, p = 0.0819), while Gem+NP remained associated with increased risk (HR = 1.090, 95% CI 1.027-1.157, p = 0.0046). Subgroup comparative analyses indicated that smokers had greater benefit with FOLFIRINOX (MSM HR = 0.81, 95% CI 0.74-0.90; outcome regression HR = 0.80, 95% CI 0.73-0.88) than non-smokers (MSM HR = 0.92, 95% CI 0.82-1.02; outcome regression HR = 0.90, 95% CI 0.81-1.00), showing that treatment worked differently across groups, rather than smoking itself being beneficial.
CONCLUSIONS: FOLFIRINOX was associated with the longest survival, and the findings were consistent across causal models.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
CO22
Topic
Clinical Outcomes
Topic Subcategory
Comparative Effectiveness or Efficacy
Disease
SDC: Gastrointestinal Disorders, SDC: Oncology