Screening for pancreatic cancer has the potential to save lives, but is it practical?

Precursor lesions in familial and hereditary pancreatic cancer

Infiltrating ductal adenocarcinoma of the pancreas, referred to here as "pancreatic cancer," is one of the deadliest of all of the solid malignancies. The five-year survival rate in the United States for individuals diagnosed today with pancreatic cancer is a dismal 12%. Many invasive cancers, including pancreatic cancer, however, arise from histologically and genetically well-characterized precursor lesions, and these precancers are curable. Precursor lesions therefore are an attractive target for early detection and treatment. This is particularly true for individuals with an increased risk of developing invasive cancer, such as individuals with a strong family history of pancreatic cancer, and individuals with a germline variant known to increase the risk of developing pancreatic cancer. There is therefore a need to understand the precursor lesions that can give rise to invasive pancreatic cancer in these individuals.

Power-law growth models explain incidences and sizes of pancreatic cancer precursor lesions and confirm spatial genomic findings

Pancreatic ductal adenocarcinoma is a rare but lethal cancer. Recent evidence reveals that pancreatic intraepithelial neoplasms (PanINs), the microscopic precursor lesions in the pancreatic ducts that can give rise to invasive pancreatic cancer, are significantly larger and more prevalent than previously believed. Better understanding of the growth law dynamics of PanINs may improve our ability to understand how a miniscule fraction of these lesions makes the transition to invasive cancer. Here, using artificial intelligence (AI)-based three-dimensional (3D) tissue mapping method, we measured the volumes of >1,000 PanIN and found that lesion size is distributed according to a power law with a fitted exponent of -1.7 over > 3 orders of magnitude. Our data also suggest that PanIN growth is not very sensitive to the pancreatic microenvironment or an individual's age, family history, and lifestyle, and is rather shaped by general growth behavior. We analyze several models of PanIN growth and fit the predicted size distributions to the observed data. The best fitting models suggest that both intraductal spread of PanIN lesions and fusing of multiple lesions into large, highly branched structures drive PanIN growth patterns. This work lays the groundwork for future mathematical modeling efforts integrating PanIN incidence, morphology, genomic, and transcriptomic features to understand pancreas tumorigenesis, and demonstrates the utility of combining experimental measurement of human tissues with dynamic modeling for understanding cancer tumorigenesis.

Automated Artificial Intelligence Model Trained on a Large Data Set Can Detect Pancreas Cancer on Diagnostic Computed Tomography Scans As Well As Visually Occult Preinvasive Cancer on Prediagnostic Computed Tomography Scans

BACKGROUND & AIMS

The aims of our case-control study were (1) to develop an automated 3-dimensional (3D) Convolutional Neural Network (CNN) for detection of pancreatic ductal adenocarcinoma (PDA) on diagnostic computed tomography scans (CTs), (2) evaluate its generalizability on multi-institutional public data sets, (3) its utility as a potential screening tool using a simulated cohort with high pretest probability, and (4) its ability to detect visually occult preinvasive cancer on prediagnostic CTs.

METHODS

A 3D-CNN classification system was trained using algorithmically generated bounding boxes and pancreatic masks on a curated data set of 696 portal phase diagnostic CTs with PDA and 1080 control images with a nonneoplastic pancreas. The model was evaluated on (1) an intramural hold-out test subset (409 CTs with PDA, 829 controls); (2) a simulated cohort with a case-control distribution that matched the risk of PDA in glycemically defined new-onset diabetes, and Enriching New-Onset Diabetes for Pancreatic Cancer score ≥3; (3) multi-institutional public data sets (194 CTs with PDA, 80 controls), and (4) a cohort of 100 prediagnostic CTs (i.e., CTs incidentally acquired 3-36 months before clinical diagnosis of PDA) without a focal mass, and 134 controls.

RESULTS

Of the CTs in the intramural test subset, 798 (64%) were from other hospitals. The model correctly classified 360 CTs (88%) with PDA and 783 control CTs (94%), with a mean accuracy 0.92 (95% CI, 0.91-0.94), area under the receiver operating characteristic (AUROC) curve of 0.97 (95% CI, 0.96-0.98), sensitivity of 0.88 (95% CI, 0.85-0.91), and specificity of 0.95 (95% CI, 0.93-0.96). Activation areas on heat maps overlapped with the tumor in 350 of 360 CTs (97%). Performance was high across tumor stages (sensitivity of 0.80, 0.87, 0.95, and 1.0 on T1 through T4 stages, respectively), comparable for hypodense vs isodense tumors (sensitivity: 0.90 vs 0.82), different age, sex, CT slice thicknesses, and vendors (all P > .05), and generalizable on both the simulated cohort (accuracy, 0.95 [95% 0.94-0.95]; AUROC curve, 0.97 [95% CI, 0.94-0.99]) and public data sets (accuracy, 0.86 [95% CI, 0.82-0.90]; AUROC curve, 0.90 [95% CI, 0.86-0.95]). Despite being exclusively trained on diagnostic CTs with larger tumors, the model could detect occult PDA on prediagnostic CTs (accuracy, 0.84 [95% CI, 0.79-0.88]; AUROC curve, 0.91 [95% CI, 0.86-0.94]; sensitivity, 0.75 [95% CI, 0.67-0.84]; and specificity, 0.90 [95% CI, 0.85-0.95]) at a median 475 days (range, 93-1082 days) before clinical diagnosis.

CONCLUSIONS

This automated artificial intelligence model trained on a large and diverse data set shows high accuracy and generalizable performance for detection of PDA on diagnostic CTs as well as for visually occult PDA on prediagnostic CTs. Prospective validation with blood-based biomarkers is warranted to assess the potential for early detection of sporadic PDA in high-risk individuals.