The Natural History of Pancreatic Cancer and the Effectiveness of Screening a Model-Based Analysis
Author(s)
Robert Wittram, MSc, Christian Brettschneider, PhD, Hans-Helmut Koenig, MD.
Department of Health Economics and Health Services Research, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.
Department of Health Economics and Health Services Research, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.
OBJECTIVES: Early detection of pancreatic cancer is crucial for reducing mortality, as it enables patients to access curative treatments before metastasis. Blood-based tests could improve screening, which will be modelled in this study. The aim is to assess average life-years-gained per individual detected compared to no screening in a hypothetical population.
METHODS: A microsimulation model simulates the progression of pancreatic cancer in 100,000 individuals at 1.9% lifetime risk. The full natural history from low-grade precancerous lesions to metastatic pancreatic cancer in male and female individuals can be assessed. Cumulative proportions from the literature are used to calculate pancreatic cancer-related survival. Individuals are screened four times starting at age 55 in 1-year intervals. Stage-specific test parameters and full screening adherence were assumed. The comparator was pancreatic cancer diagnosis due to e.g., symptoms.
RESULTS: Screening resulted in an average life-year gain of 1.6 years per true-positively screened individual. The 5-year survival rate was 29% vs. 5% when true-positively screened versus not screened. Overall, 44 individuals had a true-positive test result to receive subsequent cancer treatment. Among those patients, 36 received an early diagnosis at stage 1 or 2. Sensitivity analysis was conducted to test the impact of the screening start age on the outcome. Starting screening later than 55 increased the average life-year gain up until age 67, before decreasing this gain. A specificity of 90% or 95% would lead to 30,000 or 17,500 false-positive results.
CONCLUSIONS: Pancreatic cancer screening in the normal population would result in an unacceptable number of false-positive and few true-positive patients. This model will investigate which threshold risk, test parameters, and optimal screening protocols, could enable early detection under certain circumstances. Evaluating the cost-effectiveness of blood-based testing for pancreatic cancer will be the ultimate purpose of the model.
METHODS: A microsimulation model simulates the progression of pancreatic cancer in 100,000 individuals at 1.9% lifetime risk. The full natural history from low-grade precancerous lesions to metastatic pancreatic cancer in male and female individuals can be assessed. Cumulative proportions from the literature are used to calculate pancreatic cancer-related survival. Individuals are screened four times starting at age 55 in 1-year intervals. Stage-specific test parameters and full screening adherence were assumed. The comparator was pancreatic cancer diagnosis due to e.g., symptoms.
RESULTS: Screening resulted in an average life-year gain of 1.6 years per true-positively screened individual. The 5-year survival rate was 29% vs. 5% when true-positively screened versus not screened. Overall, 44 individuals had a true-positive test result to receive subsequent cancer treatment. Among those patients, 36 received an early diagnosis at stage 1 or 2. Sensitivity analysis was conducted to test the impact of the screening start age on the outcome. Starting screening later than 55 increased the average life-year gain up until age 67, before decreasing this gain. A specificity of 90% or 95% would lead to 30,000 or 17,500 false-positive results.
CONCLUSIONS: Pancreatic cancer screening in the normal population would result in an unacceptable number of false-positive and few true-positive patients. This model will investigate which threshold risk, test parameters, and optimal screening protocols, could enable early detection under certain circumstances. Evaluating the cost-effectiveness of blood-based testing for pancreatic cancer will be the ultimate purpose of the model.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
MSR200
Topic
Clinical Outcomes, Epidemiology & Public Health, Methodological & Statistical Research
Disease
Gastrointestinal Disorders, Oncology