Original Article
Pancreatobiliary
Cost-Effectiveness of a Risk-Tailored Pancreatic Cancer Early Detection Strategy Among Patients With New-Onset Diabetes

https://doi.org/10.1016/j.cgh.2021.10.037Get rights and content

Background & Aims

Screening for pancreatic ductal adenocarcinoma (PDAC) in asymptomatic adults is not recommended, however, patients with new-onset diabetes (NoD) have an 8 times higher risk of PDAC than expected. A novel risk-tailored early detection strategy targeting high-risk NoD patients might improve PDAC prognosis. We sought to evaluate the cost effectiveness of this strategy.

Methods

We compared PDAC early detection strategies targeting NoD individuals age 50 years and older at various minimal predicted PDAC risk thresholds vs standard of care in a Markov state–transition decision model under the health care sector perspective using a lifetime horizon.

Results

At a willingness to pay (WTP) threshold of $150,000 per quality-adjusted life-year, the early detection strategy targeting patients with a minimum predicted 3-year PDAC risk of 1% was cost effective (incremental cost-effectiveness ratio, $116,911). At a WTP threshold of $100,000 per quality-adjusted life-year, the early detection strategy at the 2% risk threshold was cost effective (incremental cost-effectiveness ratio, $63,045). The proportion of PDACs detected at local stage, costs of treatment for metastatic PDAC, utilities of local and regional cancers, and sensitivity of screening were the most influential parameters. Probabilistic sensitivity analysis confirmed that at a WTP threshold of $150,000, early detection at the 1.0% risk threshold was favored (30.6%), followed by the 0.5% risk threshold (20.4%) vs standard of care (1.7%). At a WTP threshold of $100,000, early detection at the 1.0% risk threshold was favored (27.3%) followed by the 2.0% risk threshold (22.8%) vs standard of care (2.0%).

Conclusions

A risk-tailored PDAC early detection strategy targeting NoD patients with a minimum predicted 3-year PDAC risk of 1.0% to 2.0% may be cost effective.

Section snippets

Methods

All authors had access to the study data and reviewed and approved the final manuscript. This study did not involve any human subjects research and did not require institutional review board review.

Results

There were 89,881 individuals in our published THIN study cohort with NoD diagnosed at 50 years of age or older. The cumulative incidence of pancreatic cancer was 0.42% over the 3 years after a DM diagnosis. The average age of the cohort was 66.4 years old (SD, ±9.6 y), and 53% of the cohort was male. For each of the 6 risk thresholds used to define the high-risk group, we derived the corresponding proportion of NoD patients in the high-risk group as well as the 3-year predicted risks of being

Discussion

Our base-case analysis showed that at a WTP threshold of $150,000/QALY, early detection for pancreatic cancer using a risk threshold of 1% to define a high-risk group was cost effective. At a more conservative WTP threshold of $100,000/QALY, early detection at the 2% risk threshold was cost effective. For both WTP thresholds at 150,000 and 100,000/QALY, local cancer stage shift (defined as the percentage of patients diagnosed with local cancer at the time of screening), had the greatest impact

CRediT Authorship Contributions

Louise Wang (Conceptualization: Equal; Data curation: Lead; Formal analysis: Lead; Investigation: Lead; Writing – original draft: Lead; Writing – review & editing: Equal)

Frank Scott (Conceptualization: Supporting; Formal analysis: Supporting; Writing – review & editing: Supporting)

Ben Boursi (Conceptualization: Supporting; Writing – review & editing: Supporting)

Kim Reiss (Conceptualization: Supporting; Writing – review & editing: Supporting)

Sankey Williams (Conceptualization: Supporting;

References (23)

  • S.R. Puli et al.

    How good is endoscopic ultrasound-guided fine-needle aspiration in diagnosing the correct etiology for a solid pancreatic mass? A meta-analysis and systematic review

    Pancreas

    (2013)
  • Conflicts of interest These authors disclose the following: Dr Louise Wang has received research support from the NIH T32DK007740 Training grant in GI Epidemiology. Frank I. Scott has received grants from Takeda Pharmaceuticals USA, Janssen Pharmaceuticals, the National Institutes of Health, and the Crohn’s and Colitis Foundation, and has received personal fees from PRIME, Inc, Janssen Pharmaceuticals, Takeda Pharmaceuticals, and Merck Pharmaceuticals; and Kim Reiss has received research support from Lilly Oncology, BMS, GSK, and Clovis Oncology. The remaining authors disclose no conflicts.

    View full text