Clinical Prediction Models for Pancreatic Cancer in General and At-Risk Populations: A Systematic Review

Abstract

Identifying high-risk individuals using a risk prediction model could be a crucial first stage of screening pathways to improve the early detection of pancreatic cancer. A systematic review was conducted to critically evaluate the published primary literature on the development or validation of clinical risk prediction models for pancreatic cancer risk. METHODS:MEDLINE, Embase, and Web of Science were searched for relevant articles from the inception of each database up to November 2021. Study selection and data extraction were conducted by 2 independent reviewers. The Prediction model Risk Of Bias Assessment Tool (PROBAST) was applied to assess risk of bias. RESULTS:In total, 33 studies were included, describing 38 risk prediction models. Excluding studies with an overlapping population, this study consist of 15,848,100 participants, of which 58,313 were diagnosed with pancreatic cancer. Eight studies externally validated their model, and 13 performed internal validation. The studies described risk prediction models for pancreatic cancer in the general population (n 5 14), patients with diabetes (n 5 8), and individuals with gastrointestinal (and other) symptoms (symptoms included abdominal pain, unexplained weight loss, jaundice, and change in bowel habits and indigestion; n 5 11). The commonly used clinical risk factors in the model were cigarette smoking (n 5 27), age (n 5 25), diabetes history (n 5 22), chronic pancreatitis (n 5 18), and body mass index (n 5 14). In the 25 studies that assessed model performance, C-statistics ranged from 0.61 to 0.98. Of the 33 studies included, 6 were rated as being at a low risk of bias based on PROBAST. DISCUSSION:Many clinical risk prediction models for pancreatic cancer had been developed for different target populations. Although low risk-of-bias studies were identified, these require external validation and implementation studies to ensure that these will benefit clinical decision making.

An Integrative Pancreatic Cancer Risk Prediction Model in the UK Biobank

“…In our multivariable logistic regression models, the significant risk variables were as follows: non-modifiable-male gender and age, genetic predisposition-non-O blood type and a higher PRS quintile (Q4 vs. Q1, OR = 1.67; Q5 vs. Q1, OR = 2.03), lifestyle-related modifiable variables-current smokers and higher alcohol consumption, and medical history-related variables-pancreatitis, DM, and cholecystitis, cholelithiasis, or cholecystectomy. In previously published studies, a number of logistic regression models [20,21, were built to explore pancreatic cancer risks. Of them, nine models [20,53,56,57,[62][63][64]66,68] were developed in the general population, and others were created in populations with DM history or clinical gastrointestinal symptoms.…”

“…There are multiple models [21] that have been published associated with PaCa risk prediction. However, each model concentrated only on selected risk variables, such as symptoms-based, genetic-based, or lifestyle-related.…”

Risk Assessment and Pancreatic Cancer: Diagnostic Management and Artificial Intelligence

“…Beyond familial pancreatic cancer, the most well-recognized risk factor for pancreatic tumor is smoking, followed by chronic pancreatitis, diabetes, and obesity, specifically high body-mass index (BMI) and centralized fat distribution [105][106][107][108][109][110][111][112][113][114].…”

“…In addition, another subgroup of familial pancreatic tumor is due to germline mutations in ATM [94][95][96][97][98][99][100][101][102][103][104]. Beyond familial pancreatic cancer, the most well-recognized risk factor for pancreatic tumor is smoking, followed by chronic pancreatitis, diabetes, and obesity, specifically high body-mass index (BMI) and centralized fat distribution [105][106][107][108][109][110][111][112][113][114].…”

Diabetes of the Exocrine Pancreas: Implications for Pharmacological Management

“…Several observational cohort studies have suggested an antineoplastic effect of metformin. Diabetes and chronic pancreatitis alone are associated with an increased risk of pancreatic cancer (with an especially high risk if both are present) and are integral parts of clinical prediction models for pancreatic cancer [ 22 ]. Therefore, any effect of metformin to prevent pancreatic cancer would be of great value.…”