Cost-Effectiveness of Biomarker-Associated Early Pancreatic Cancer Detection in New-Onset Diabetes.

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is often diagnosed late, resulting in a dismal survival. Only 1 in 5 patients is eligible for potentially curative surgery. Although earlier detection of PDAC could significantly improve prognosis, population-wide screening is not currently feasible. Individuals at high risk, such as those with new-onset diabetes, among whom approximately 10% have type 3c diabetes (T3cD) and 1% have undiagnosed PDAC, could represent a target population for PDAC screening. To compare the cost-effectiveness of 3 PDAC screening strategies in new-onset diabetes with the standard of care. For this economic evaluation, a Markov state-transition decision model was developed to assess the downstaging benefits associated with screening policies over a 5-year period. The model allowed for diagnosis of resectable, borderline resectable, locally advanced, or metastatic disease. Model parameters were subject to 1-way and multiway sensitivity analyses. Mean treatment costs were extracted from the UK National Healthcare Service cost-collection data (2021-2022), and the screening strategies were applied to a simulated cohort of individuals aged 50 years or older with new-onset diabetes in the UK. Analyses were performed from August 2024 to January 2025. Pancreatic ductal adenocarcinoma screening strategies included the use of a T3cD biomarker, a cancer-specific biomarker, or a combination of a T3cD biomarker followed by a cancer-specific test in the population with new-onset diabetes. Biomarker cohorts were compared with the standard-of-care cohort undergoing diagnostic investigations on the development of clinical symptoms. Per-person incremental cost-effectiveness ratios (ICERs) and net benefits were calculated, allowing for a willingness-to-pay threshold per quality-adjusted life-year (QALY) of £30 000 (US $40 156), as accepted by the UK's National Institute for Health and Care Excellence. Among individuals aged 50 years or older with new-onset diabetes, neither the cancer-specific nor T3cD biomarker strategy was cost-effective compared with the standard of care, with an ICER per QALY of £71 906 (US $96 249) for the cancer-specific biomarker test and an ICER per QALY of £46 371 (US $62 070) for the T3cD biomarker strategy. Sequential use of a T3cD biomarker and a cancer-specific biomarker approached cost-effectiveness (ICER per QALY, £34 223 [US $45 809]), with 2.4 scans performed to detect 1 PDAC case among 200 000 people with new-onset diabetes. Sensitivity analyses highlighted biomarker specificity as a critical determinant of cost-effectiveness. In this economic evaluation of early PDAC detection among individuals with new-onset diabetes, the application of a primary T3cD biomarker followed by a secondary PDAC-specific test was more cost-effective than the sole use of either biomarker, although no strategy was cost-effective compared with standard of care. These findings can be used to inform biomarker-associated early detection strategies for PDAC.