Search

Cytologic and histopathologic diagnosis of non-ductal pancreatic neoplasms can be challenging in daily clinical practice, whereas it is crucial for therapy and prognosis. The cancer methylome is successfully used as a diagnostic tool in other cancer entities. Here, we investigate if methylation profiling can improve the diagnostic work-up of pancreatic neoplasms. DNA methylation data were obtained for 301 primary tumors spanning 6 primary pancreatic neoplasms and 20 normal pancreas controls. Neural Network, Random Forest, and extreme gradient boosting machine learning models were trained to distinguish between tumor types. Methylation data of 29 nonpancreatic neoplasms (n = 3708) were used to develop an algorithm capable of detecting neoplasms of non-pancreatic origin. After benchmarking 3 state-of-the-art machine learning models, the random forest model emerged as the best classifier with 96.9% accuracy. All classifications received a probability score reflecting the confidence of the prediction. Increasing the score threshold improved the random forest classifier performance up to 100% with 87% of samples with scores surpassing the cutoff. Using a logistic regression model, detection of nonpancreatic neoplasms achieved an area under the curve of >0.99. Analysis of biopsy specimens showed concordant classification with their paired resection sample. Pancreatic neoplasms can be classified with high accuracy based on DNA methylation signatures. Additionally, non-pancreatic neoplasms are identified with near perfect precision. In summary, methylation profiling can serve as a valuable adjunct in the diagnosis of pancreatic neoplasms with minimal risk for misdiagnosis, even in the pre-operative setting. [Display omitted] [ABSTRACT FROM AUTHOR]

Objective: To report the mutational profile and clinical outcomes of a cohort of patients with KIT-mutant seminomas and nonseminomatous germ-cell tumors (SGCT/NSGCTs).Patients and Methods: Retrospective cohort study of all patients with KIT-mutant GCTs sequenced at Memorial Sloan Kettering between March 2014 and March 2020. Tumors were assessed with MSK-IMPACT, a DNA next-generation sequencing assay for targeted sequencing of up to 468 key cancer genes.Results: Among 568 patients with GCTs, 8.1% had somatic KIT mutations, including 28 seminomas and 18 mixed/NSGCTs. Exons 17 (67.3%), 11 (22.4%), and 13 (6.1%) were most commonly affected. KIT-mutant cases were enriched for oncogenic RAS/MAPK pathway alterations compared to KIT-wildtype cases (34.8% vs 19.2%, P = .02). Among KIT-mutant cases, concurrent mutations were noted in KRAS (21.7%), RRAS2 (11.8%), CBL (6.5%), NRAS (4.3%), MAP2K1 (2.2%), and RAC1 (2.2%). Mutations in KRAS, RRAS2, and NRAS were mutually exclusive. In all, 73.9% of patients developed metastases and 95.7% received chemotherapy. No patients received KIT-directed tyrosine kinase inhibitors (TKIs). Classification as a NSGCT rather than a SGCT was associated with an increased risk of death (hazard ratio 9.1, 95% confidence interval 1.1-78.4, P = .04) while the presence of a concurrent RAS/MAPK pathway alteration was not (hazard ratio 0.8, 95% confidence interval 0.1-4.3, P = .76).Conclusion: Mitogenic driver alterations can co-occur with activating KIT mutations, which may explain the lack of efficacy of KIT-directed TKIs in prior trials. Novel KIT-directed TKIs that target exon 17 mutations may benefit chemotherapy-refractory patients with KIT-mutant GCTs without RAS/MAPK alterations. Dual MEK/KIT inhibitor therapy in KIT-mutant GCTs with concurrent RAS/MAPK alterations could also be a plausible therapeutic strategy. [ABSTRACT FROM AUTHOR]