A REAL-WORLD COST-EFFECTIVENESS ANALYSIS USING TWO HOSPITAL SETTINGS ON REAL-WORLD DOSING OF PEMBROLIZUMAB IN ADVANCED NSCLC
Author(s)
Zainab Al-khayat, MSc1, Nora Franzen, PhD2, Mariska Rijnen, MSc3, Sushil Badrising, PhD4, Anne Bresser, PhD5, Madelon Voets, PhD4, Willem Herbert Van Harten, PhD6, Valesca P. Retel, MSc, PhD4.
1PhD student, Netherlands Cancer Institute (NKI-AVL), Amsterdam, Netherlands, 2Erasmus University Rotterdam, Rotterdam, Netherlands, 3IQVIA, Amsterdam, Netherlands, 4Netherlands Cancer Institute (NKI-AVL), Amsterdam, Netherlands, 5Rijnstate Hospital, Amsterdam, Netherlands, 6Netherlands Cancer Institute, Amsterdam, Netherlands.
1PhD student, Netherlands Cancer Institute (NKI-AVL), Amsterdam, Netherlands, 2Erasmus University Rotterdam, Rotterdam, Netherlands, 3IQVIA, Amsterdam, Netherlands, 4Netherlands Cancer Institute (NKI-AVL), Amsterdam, Netherlands, 5Rijnstate Hospital, Amsterdam, Netherlands, 6Netherlands Cancer Institute, Amsterdam, Netherlands.
OBJECTIVES: Pembrolizumab has improved outcomes in advanced NSCLC. Although early clinical studies evaluated a weight-based dose of 2 mg/kg, fixed dosing regimens were adopted to simplify administration. However, fixed dosing results in variable weight-adjusted dose intensity among patients with different body weights. This study aimed to explore real-world variation in weight-adjusted dose intensity and its association with overall survival (OS) in patients with advanced/metastatic NSCLC and PD-L1 expression ≥50% receiving first-line pembrolizumab monotherapy as a first step toward a weight-based cost-effectiveness analysis using real-world hospital data.
METHODS: This retrospective cohort study used RWD from two Dutch hospitals. Data quality measures included transparency of data collection (manual or NLP-based), missing data assessment, and validation procedures. For each administration, the dose was converted to be expressed as mg/kg q3w using the average (dose administered/weight) every 3 weeks. Patients were categorized into two groups according to their mean administered dose intensity: ≤2 mg/kg q3w or >2 mg/kg q3w.
RESULTS: Data accuracy and validation were primarily required for the NLP-derived cohort and depended strongly on how information was documented in clinical notes, with performance status and disease stage requiring additional manual validation. Eligible patients (N=253) were stratified into two dose intensity groups: ≤2 mg/kg (n=49) and >2 mg/kg (n=204). Significant differences in weight, BMI, and performance status were observed between groups. In the unadjusted analysis, patients receiving a mean dose >2 mg/kg had significantly worse OS than those receiving ≤2 mg/kg (HR 1.74, p=0.011).
CONCLUSIONS: The unadjusted analysis suggests that a pembrolizumab dose intensity of ≤2 mg/kg could be associated with superior OS compared with higher dose intensities, despite lower drug exposure. These findings could support a weight-based dosing cost-effectiveness analysis in this setting to inform decision-making.
METHODS: This retrospective cohort study used RWD from two Dutch hospitals. Data quality measures included transparency of data collection (manual or NLP-based), missing data assessment, and validation procedures. For each administration, the dose was converted to be expressed as mg/kg q3w using the average (dose administered/weight) every 3 weeks. Patients were categorized into two groups according to their mean administered dose intensity: ≤2 mg/kg q3w or >2 mg/kg q3w.
RESULTS: Data accuracy and validation were primarily required for the NLP-derived cohort and depended strongly on how information was documented in clinical notes, with performance status and disease stage requiring additional manual validation. Eligible patients (N=253) were stratified into two dose intensity groups: ≤2 mg/kg (n=49) and >2 mg/kg (n=204). Significant differences in weight, BMI, and performance status were observed between groups. In the unadjusted analysis, patients receiving a mean dose >2 mg/kg had significantly worse OS than those receiving ≤2 mg/kg (HR 1.74, p=0.011).
CONCLUSIONS: The unadjusted analysis suggests that a pembrolizumab dose intensity of ≤2 mg/kg could be associated with superior OS compared with higher dose intensities, despite lower drug exposure. These findings could support a weight-based dosing cost-effectiveness analysis in this setting to inform decision-making.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
EE243
Topic
Economic Evaluation, Health Technology Assessment, Real World Data & Information Systems
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, Oncology