Surgery for Locally Advanced Pancreatic Cancer Following Induction Chemotherapy: A Single-Center Experience.

BACKGROUND: The use of surgery in patients with locally advanced pancreatic cancer (LAPC) following induction chemotherapy is increasing. However, most series do not report on the total cohort of patients undergoing surgical exploration; therefore, this single-center study investigates outcomes among all consecutive patients with LAPC who underwent surgical exploration. METHODS: We conducted a retrospective, single-center analysis including all consecutive patients with LAPC (Dutch Pancreatic Cancer Group criteria) who underwent surgical exploration with curative intent (January 2014-June 2023) after induction therapy. Primary outcomes were resection rate and overall survival (OS) from the time of diagnosis. RESULTS: Overall, 127 patients underwent surgical exploration for LAPC, whereby 100 patients (78.7%) underwent resection and 27 patients (21.3%) underwent a non-therapeutic laparotomy due to the extent of vascular involvement (n = 11, 8.7%) or occult metastases (n = 16, 12.6%). The overall in-hospital/30-day mortality rate was 0.8% and major morbidity was 31.3% (in patients after resection: 1.0% and 33.3%, respectively). The overall 90-day mortality rate was 5.5%, which included 3.1% mortality due to disease progression. Resection was associated with longer median OS {29 months (95% confidence interval [CI] 26-43) vs. 17 months (95% CI 11-26); p < 0.001} compared with patients undergoing non-therapeutic laparotomy, with corresponding 5-year OS rates of 28.4% and 7.7%. In Cox proportional hazard regression analysis, only pancreatic body/tail tumors independently predicted OS (hazard ratio 1.788 [95% CI 1.042-3.068]). CONCLUSION: This single-center series found a resection rate of 78.7% in patients with LAPC selected for surgical exploration, with a low risk of mortality and morbidity in all explored patients and a 5-year OS rate after resection of 28.4%.

Transcriptome-based classification to predict FOLFIRINOX response in a real-world metastatic pancreatic cancer cohort.

Pancreatic ductal adenocarcinoma (PDAC) is often diagnosed at metastatic stage and typically treated with fluorouracil, leucovorin, irinotecan and oxaliplatin (FOLFIRINOX). Few patients benefit from this treatment. Molecular subtypes are prognostic in particularly resectable PDAC and might predict treatment response. This study aims to correlate molecular subtypes in metastatic PDAC with FOLFIRINOX responses using real-world data, providing assistance in counselling patients. We collected 131 RNA-sequenced metastatic biopsies and applied a network-based meta-analysis using published PDAC classifiers. Subsequent survival analysis was performed using the most suitable classifier. For validation, we developed an immunohistochemistry (IHC) classifier using GATA6 and keratin-17 (KRT17), and applied it to 86 formalin-fixed paraffin-embedded samples of advanced PDAC. Lastly, GATA6 knockdown models were generated in PDAC organoids and cell lines. We showed that the PurIST classifier was the most suitable classifier. With this classifier, classical tumors had longer PFS and OS than basal-like tumors (PFS: 216 vs. 78 days, p = 0.0002; OS: 251 vs. 195 days, p = 0.049). The validation cohort showed a similar trend. Importantly, IHC GATA6low patients had significantly shorter survival with FOLFIRINOX (323 vs. 746 days, p = 0.006), but no difference in non-treated patients (61 vs. 54 days, p = 0.925). This suggests that GATA6 H-score predicts therapy response. GATA6 knockdown models did not lead to increased FOLFIRINOX responsiveness. These data suggest a predictive role for subtyping (transcriptomic and GATA6 IHC), though no direct causal relationship was found between GATA6 expression and chemoresistance. GATA6 immunohistochemistry should be seamlessly added to current diagnostics and integrated into upcoming clinical trials.

Artificial Intelligence-based Segmentation of Residual Pancreatic Cancer in Resection Specimens Following Neoadjuvant Treatment (ISGPP-2): International Improvement and Validation Study.

Neoadjuvant therapy (NAT) has become routine in patients with borderline resectable pancreatic cancer. Pathologists examine pancreatic cancer resection specimens to evaluate the effect of NAT. However, an automated scoring system to objectively quantify residual pancreatic cancer (RPC) is currently lacking. Herein, we developed and validated the first automated segmentation model using artificial intelligence techniques to objectively quantify RPC. Digitized histopathological tissue slides were included from resected pancreatic cancer specimens from 14 centers in 7 countries in Europe, North America, Australia, and Asia. Four different scanner types were used: Philips (56%), Hamamatsu (27%), 3DHistech (10%), and Leica (7%). Regions of interest were annotated and classified as cancer, non-neoplastic pancreatic ducts, and others. A U-Net model was trained to detect RPC. Validation consisted of by-scanner internal-external cross-validation. Overall, 528 unique hematoxylin and eosin (H & E) slides from 528 patients were included. In the individual Philips, Hamamatsu, 3DHistech, and Leica scanner cross-validations, mean F1 scores of 0.81 (95% CI, 0.77-0.84), 0.80 (0.78-0.83), 0.76 (0.65-0.78), and 0.71 (0.65-0.78) were achieved, respectively. In the meta-analysis of the cross-validations, the mean F1 score was 0.78 (0.71-0.84). A final model was trained on the entire data set. This ISGPP model is the first segmentation model using artificial intelligence techniques to objectively quantify RPC following NAT. The internally-externally cross-validated model in this study demonstrated robust performance in detecting RPC in specimens. The ISGPP model, now made publically available, enables automated RPC segmentation and forms the basis for objective NAT response evaluation in pancreatic cancer.