TABLE 1.
Overview included studies
| Study | Modeling aim | Modeling method | Simulated population | Outcome |
|---|---|---|---|---|
| Rulyak, 2003 12 | Cost‐effectiveness of EUS screening in FPC kindreds | Decision tree | Members of familial pancreatic cancer kindreds | Endoscopic screening was cost‐effective, with an incremental cost‐effectiveness ratio of 16 885/life‐year saved. Screening was more cost‐effective as the probability of dysplasia increased and as the sensitivity of the test increased. Expected gain in LE was 38 years for a cohort of 100 people |
| Weinberg, 2010 15 | Assist patients and physicians with understanding how their decisions concerning pancreatic cysts affect overall survival/QoL | Markov model | Pancreatic cyst (SB‐IPMN located in head) | Initial pancreaticoduodenectomy was the dominant strategy to maximize overall survival for any cyst >2 cm, regardless of age or comorbidities. “Do nothing” maximized quality of life for all cysts <3 cm in patients aged <75, when measuring quality adjusted survival |
| Pandharipande, 2015 9 | Identify when, from the standpoint of RR, one time screening is effective in high‐risk individuals for PC | Markov model | RR for PC from 1 to 70 | One time MR imaging screening in average risk group of 100 000 men at age 50 identifies 2375 low risk cysts, 159 high‐risk cysts and 56 cancers. Resulting in 39 cancer deaths averted and a net LE loss of 3 days). If the PC risk exceeds 2.4 (men) or 2.7 (women) there was a gain in LE |
| Pandharipande, 2015 10 | Compare effectiveness of different PC screening strategies in BRCA2 mutation carriers, from standpoint of life expectancy | Markov model | BRCA2 mutation carriers (with or without FDR with PC) | One time screening at age 50 resulted in a LE gain of 3.9 days for the BRCA2 cohort. The gain was higher with more FDRs with PC. Annual screening resulted in a LE loss of 12.9 days for BRCA2 mutation carriers. BRCA2 carriers with 2 FDRs gained 20.6 days with annual screening |
| Cucchetti, 2016 16 | Verify survival benefit obtained from a hypothetical screening where a 20,30 or 50% reduction of PC stage was obtained | Monte Carlo Simulation | Pancreatic cancer patients | Mean expected LE for PC patients was 13 months. When a hypothetical screening reduced stage III/IV with 30‐50% this LE was: 14‐15.9 months |
| Peters, 2018 11 | Gain insight into the natural history of PanIN and to assess the potential of screening | Markov model | General population | Lifetime probability for PanIN1 to progress to PDAC (1.5%). Duration of this progression: 33.6 years. A hypothetical perfect test for PanIN 3 detection and treatment could provide a maximum, average LE gain of 40 days |
| Raphel, 2018 17 | Determine the effect of patient age and comorbidity on LE benefits associated with SB‐IPMN follow‐up | Markov model | Pancreatic cyst (SB‐IPMN) | The LE benefit of SB‐IPMN follow‐up is 5.3‐6.4 months for healthy 60 year old individuals. The effect on LE was limited in case of 80 year old individuals and coexistence of severe comorbidity |
| Koopmann, 2021 13 | Analyze the impact of relevant uncertainties on the effect of PC screening in high‐risk individuals | Markov model | High‐risk individuals | Screening reduced PC mortality in all modeled scenarios. The reduction depended strongly on natural disease course (progressive vs indolent and faster progressive lesions). The impact of test sensitivity was much smaller. The NNS was impacted most by PC risk |