Nationwide Impact of Centralization, Neoadjuvant Therapy, Minimally Invasive Surgery, and Standardized Pathology Reporting on R0 Resection and Overall Survival in Pancreatoduodenectomy for Pancreatic Cancer.

BACKGROUND: Surgeons aim for R0 resection in patients with pancreatic cancer to improve overall survival. However, it is unclear whether recent changes in pancreatic cancer care such as centralization, increased use of neoadjuvant therapy, minimally invasive surgery, and standardized pathology reporting have influenced R0 resections and whether R0 resection remains associated with overall survival. METHODS: This nationwide retrospective cohort study included consecutive patients after pancreatoduodenectomy (PD) for pancreatic cancer from the Netherlands Cancer Registry and the Dutch Nationwide Pathology Database (2009-2019). R0 resection was defined as > 1 mm tumor clearance at the pancreatic, posterior, and vascular resection margins. Completeness of pathology reporting was scored on the basis of six elements: histological diagnosis, tumor origin, radicality, tumor size, extent of invasion, and lymph node examination. RESULTS: Among 2955 patients after PD for pancreatic cancer, the R0 resection rate was 49%. The R0 resection rate decreased from 68 to 43% (2009-2019, P < 0.001). The extent of resections in high-volume hospitals, minimally invasive surgery, neoadjuvant therapy, and complete pathology reports all significantly increased over time. Only complete pathology reporting was independently associated with lower R0 rates (OR 0.76, 95% CI 0.69-0.83, P < 0.001). Higher hospital volume, neoadjuvant therapy, and minimally invasive surgery were not associated with R0. R0 resection remained independently associated with improved overall survival (HR 0.72, 95% CI 0.66-0.79, P < 0.001), as well as in the 214 patients after neoadjuvant treatment (HR 0.61, 95% CI 0.42-0.87, P = 0.007). CONCLUSIONS: The nationwide rate of R0 resections after PD for pancreatic cancer decreased over time, mostly related to more complete pathology reporting. R0 resection remained associated with overall survival.

Imaging-based Machine-learning Models to Predict Clinical Outcomes and Identify Biomarkers in Pancreatic Cancer: A Scoping Review.

OBJECTIVE: To perform a scoping review of imaging-based machine-learning models to predict clinical outcomes and identify biomarkers in patients with PDAC. SUMMARY OF BACKGROUND DATA: Patients with PDAC could benefit from better selection for systemic and surgical therapy. Imaging-based machine-learning models may improve treatment selection. METHODS: A scoping review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses-scoping review guidelines in the PubMed and Embase databases (inception-October 2020). The review protocol was prospectively registered (open science framework registration: m4cyx). Included were studies on imaging-based machine-learning models for predicting clinical outcomes and identifying biomarkers for PDAC. The primary outcome was model performance. An area under the curve (AUC) of ≥0.75, or a P-value of ≤0.05, was considered adequate model performance. Methodological study quality was assessed using the modified radiomics quality score. RESULTS: After screening 1619 studies, 25 studies with 2305 patients fulfilled the eligibility criteria. All but 1 study was published in 2019 and 2020. Overall, 23/25 studies created models using radiomics features, 1 study quantified vascular invasion on computed tomography, and one used histopathological data. Nine models predicted clinical outcomes with AUC measures of 0.78-0.95, and C-indices of 0.65-0.76. Seventeen models identified biomarkers with AUC measures of 0.68-0.95. Adequate model performance was reported in 23/25 studies. The methodological quality of the included studies was suboptimal, with a median modified radiomics quality score score of 7/36. CONCLUSIONS: The use of imaging-based machine-learning models to predict clinical outcomes and identify biomarkers in patients with PDAC is increasingly rapidly. Although these models mostly have good performance scores, their methodological quality should be improved.

Histopathological tumour response scoring in resected pancreatic cancer following neoadjuvant therapy: international interobserver study (ISGPP-1).

BACKGROUND: Most tumour response scoring systems for resected pancreatic cancer after neoadjuvant therapy score tumour regression. However, whether treatment-induced changes, including tumour regression, can be identified reliably on haematoxylin and eosin-stained slides remains unclear. Moreover, no large study of the interobserver agreement of current tumour response scoring systems for pancreatic cancer exists. This study aimed to investigate whether gastrointestinal/pancreatic pathologists can reliably identify treatment effect on tumour by histology, and to determine the interobserver agreement for current tumour response scoring systems. METHODS: Overall, 23 gastrointestinal/pancreatic pathologists reviewed digital haematoxylin and eosin-stained slides of pancreatic cancer or treated tumour bed. The accuracy in identifying the treatment effect was investigated in 60 patients (30 treatment-naive, 30 after neoadjuvant therapy (NAT)). The interobserver agreement for the College of American Pathologists (CAP) and MD Anderson Cancer Center (MDACC) tumour response scoring systems was assessed in 50 patients using intraclass correlation coefficients (ICCs). An ICC value below 0.50 indicated poor reliability, 0.50 or more and less than 0.75 indicated moderate reliability, 0.75 or more and below 0.90 indicated good reliability, and above 0.90 indicated excellent reliability. RESULTS: The sensitivity and specificity for identifying NAT effect were 76.2 and 49.0 per cent respectively. After NAT in 50 patients, ICC values for both tumour response scoring systems were moderate: 0.66 for CAP and 0.71 for MDACC. CONCLUSION: Identification of the effect of NAT in resected pancreatic cancer proved unreliable, and interobserver agreement for the current tumour response scoring systems was suboptimal. These findings support the recently published International Study Group of Pancreatic Pathologists recommendations to score residual tumour burden rather than tumour regression after NAT.

Amsterdam International Consensus Meeting: tumor response scoring in the pathology assessment of resected pancreatic cancer after neoadjuvant therapy.

Histopathologically scoring the response of pancreatic ductal adenocarcinoma (PDAC) to neoadjuvant treatment can guide the selection of adjuvant therapy and improve prognostic stratification. However, several tumor response scoring (TRS) systems exist, and consensus is lacking as to which system represents best practice. An international consensus meeting on TRS took place in November 2019 in Amsterdam, The Netherlands. Here, we provide an overview of the outcomes and consensus statements that originated from this meeting. Consensus (≥80% agreement) was reached on a total of seven statements: (1) TRS is important because it provides information about the effect of neoadjuvant treatment that is not provided by other histopathology-based descriptors. (2) TRS for resected PDAC following neoadjuvant therapy should assess residual (viable) tumor burden instead of tumor regression. (3) The CAP scoring system is considered the most adequate scoring system to date because it is based on the presence and amount of residual cancer cells instead of tumor regression. (4) The defining criteria of the categories in the CAP scoring system should be improved by replacing subjective terms including "minimal" or "extensive" with objective criteria to evaluate the extent of viable tumor. (5) The improved, consensus-based system should be validated retrospectively and prospectively. (6) Prospective studies should determine the extent of tissue sampling that is required to ensure adequate assessment of the residual cancer burden, taking into account the heterogeneity of tumor response. (7) In future scientific publications, the extent of tissue sampling should be described in detail in the "Materials and methods" section.

Artificial intelligence for assessment of vascular involvement and tumor resectability on CT in patients with pancreatic cancer.

OBJECTIVE: This study aimed to develop and evaluate an automatic model using artificial intelligence (AI) for quantifying vascular involvement and classifying tumor resectability stage in patients with pancreatic ductal adenocarcinoma (PDAC), primarily to support radiologists in referral centers. Resectability of PDAC is determined by the degree of vascular involvement on computed tomography scans (CTs), which is associated with considerable inter-observer variability. METHODS: We developed a semisupervised machine learning segmentation model to segment the PDAC and surrounding vasculature using 613 CTs of 467 patients with pancreatic tumors and 50 control patients. After segmenting the relevant structures, our model quantifies vascular involvement by measuring the degree of the vessel wall that is in contact with the tumor using AI-segmented CTs. Based on these measurements, the model classifies the resectability stage using the Dutch Pancreatic Cancer Group criteria as either resectable, borderline resectable, or locally advanced (LA). RESULTS: We evaluated the performance of the model using a test set containing 60 CTs from 60 patients, consisting of 20 resectable, 20 borderline resectable, and 20 locally advanced cases, by comparing the automated analysis obtained from the model to expert visual vascular involvement assessments. The model concurred with the radiologists on 227/300 (76%) vessels for determining vascular involvement. The model's resectability classification agreed with the radiologists on 17/20 (85%) resectable, 16/20 (80%) for borderline resectable, and 15/20 (75%) for locally advanced cases. CONCLUSIONS: This study demonstrates that an AI model may allow automatic quantification of vascular involvement and classification of resectability for PDAC. RELEVANCE STATEMENT: This AI model enables automated vascular involvement quantification and resectability classification for pancreatic cancer, aiding radiologists in treatment decisions, and potentially improving patient outcomes. KEY POINTS: • High inter-observer variability exists in determining vascular involvement and resectability for PDAC. • Artificial intelligence accurately quantifies vascular involvement and classifies resectability for PDAC. • Artificial intelligence can aid radiologists by automating vascular involvement and resectability assessments.

Pathological Complete Response in Patients With Resected Pancreatic Adenocarcinoma After Preoperative Chemotherapy.

Importance: Preoperative chemo(radio)therapy is increasingly used in patients with localized pancreatic adenocarcinoma, leading to pathological complete response (pCR) in a small subset of patients. However, multicenter studies with in-depth data about pCR are lacking. Objective: To investigate the incidence, outcome, and risk factors of pCR after preoperative chemo(radio)therapy. Design, Setting, and Participants: This observational, international, multicenter cohort study assessed all consecutive patients with pathology-proven localized pancreatic adenocarcinoma who underwent resection after 2 or more cycles of chemotherapy (with or without radiotherapy) in 19 centers from 8 countries (January 1, 2010, to December 31, 2018). Data collection was performed from February 1, 2020, to April 30, 2022, and analyses from January 1, 2022, to December 31, 2023. Median follow-up was 19 months. Exposures: Preoperative chemotherapy (with or without radiotherapy) followed by resection. Main Outcomes and Measures: The incidence of pCR (defined as absence of vital tumor cells in the sampled pancreas specimen after resection), its association with OS from surgery, and factors associated with pCR. Factors associated with overall survival (OS) and pCR were investigated with Cox proportional hazards and logistic regression models, respectively. Results: Overall, 1758 patients (mean [SD] age, 64 [9] years; 879 [50.0%] male) were studied. The rate of pCR was 4.8% (n = 85), and pCR was associated with OS (hazard ratio, 0.46; 95% CI, 0.26-0.83). The 1-, 3-, and 5-year OS rates were 95%, 82%, and 63% in patients with pCR vs 80%, 46%, and 30% in patients without pCR, respectively (P < .001). Factors associated with pCR included preoperative multiagent chemotherapy other than (m)FOLFIRINOX ([modified] leucovorin calcium [folinic acid], fluorouracil, irinotecan hydrochloride, and oxaliplatin) (odds ratio [OR], 0.48; 95% CI, 0.26-0.87), preoperative conventional radiotherapy (OR, 2.03; 95% CI, 1.00-4.10), preoperative stereotactic body radiotherapy (OR, 8.91; 95% CI, 4.17-19.05), radiologic response (OR, 13.00; 95% CI, 7.02-24.08), and normal(ized) serum carbohydrate antigen 19-9 after preoperative therapy (OR, 3.76; 95% CI, 1.79-7.89). Conclusions and Relevance: This international, retrospective cohort study found that pCR occurred in 4.8% of patients with resected localized pancreatic adenocarcinoma after preoperative chemo(radio)therapy. Although pCR does not reflect cure, it is associated with improved OS, with a doubled 5-year OS of 63% compared with 30% in patients without pCR. Factors associated with pCR related to preoperative chemo(radio)therapy regimens and anatomical and biological disease response features may have implications for treatment strategies that require validation in prospective studies because they may not universally apply to all patients with pancreatic adenocarcinoma.

Computer-Aided Detection for Pancreatic Cancer Diagnosis: Radiological Challenges and Future Directions.

Radiological imaging plays a crucial role in the detection and treatment of pancreatic ductal adenocarcinoma (PDAC). However, there are several challenges associated with the use of these techniques in daily clinical practice. Determination of the presence or absence of cancer using radiological imaging is difficult and requires specific expertise, especially after neoadjuvant therapy. Early detection and characterization of tumors would potentially increase the number of patients who are eligible for curative treatment. Over the last decades, artificial intelligence (AI)-based computer-aided detection (CAD) has rapidly evolved as a means for improving the radiological detection of cancer and the assessment of the extent of disease. Although the results of AI applications seem promising, widespread adoption in clinical practice has not taken place. This narrative review provides an overview of current radiological CAD systems in pancreatic cancer, highlights challenges that are pertinent to clinical practice, and discusses potential solutions for these challenges.

Radiomics preoperative-Fistula Risk Score (RAD-FRS) for pancreatoduodenectomy: development and external validation.

BACKGROUND: Accurately predicting the risk of clinically relevant postoperative pancreatic fistula after pancreatoduodenectomy before surgery may assist surgeons in making more informed treatment decisions and improved patient counselling. The aim was to evaluate the predictive accuracy of a radiomics-based preoperative-Fistula Risk Score (RAD-FRS) for clinically relevant postoperative pancreatic fistula. METHODS: Radiomic features were derived from preoperative CT scans from adult patients after pancreatoduodenectomy at a single centre in the Netherlands (Amsterdam, 2013-2018) to develop the radiomics-based preoperative-Fistula Risk Score. Extracted radiomic features were analysed with four machine learning classifiers. The model was externally validated in a single centre in Italy (Verona, 2020-2021). The radiomics-based preoperative-Fistula Risk Score was compared with the Fistula Risk Score and the updated alternative Fistula Risk Score. RESULTS: Overall, 359 patients underwent a pancreatoduodenectomy, of whom 89 (25 per cent) developed a clinically relevant postoperative pancreatic fistula. The radiomics-based preoperative-Fistula Risk Score model was developed using CT scans of 118 patients, of which three radiomic features were included in the random forest model, and externally validated in 57 patients. The model performed well with an area under the curve of 0.90 (95 per cent c.i. 0.71 to 0.99) and 0.81 (95 per cent c.i. 0.69 to 0.92) in the Amsterdam test set and Verona data set respectively. The radiomics-based preoperative-Fistula Risk Score performed similarly to the Fistula Risk Score (area under the curve 0.79) and updated alternative Fistula Risk Score (area under the curve 0.79). CONCLUSION: The radiomics-based preoperative-Fistula Risk Score, which uses only preoperative CT features, is a new and promising radiomics-based score that has the potential to be integrated with hospital CT report systems and improve patient counselling before surgery. The model with underlying code is readily available via www.pancreascalculator.com and www.github.com/PHAIR-Consortium/POPF-predictor.

Artificial Intelligence-Based Segmentation of Residual Tumor in Histopathology of Pancreatic Cancer after Neoadjuvant Treatment.

BACKGROUND: Histologic examination of resected pancreatic cancer after neoadjuvant therapy (NAT) is used to assess the effect of NAT and may guide the choice for adjuvant treatment. However, evaluating residual tumor burden in pancreatic cancer is challenging given tumor response heterogeneity and challenging histomorphology. Artificial intelligence techniques may offer a more reproducible approach. METHODS: From 64 patients, one H&E-stained slide of resected pancreatic cancer after NAT was digitized. Three separate classes were manually outlined in each slide (i.e., tumor, normal ducts, and remaining epithelium). Corresponding segmentation masks and patches were generated and distributed over training, validation, and test sets. Modified U-nets with varying encoders were trained, and F1 scores were obtained to express segmentation accuracy. RESULTS: The highest mean segmentation accuracy was obtained using modified U-nets with a DenseNet161 encoder. Tumor tissue was segmented with a high mean F1 score of 0.86, while the overall multiclass average F1 score was 0.82. CONCLUSIONS: This study shows that artificial intelligence-based assessment of residual tumor burden is feasible given the promising obtained F1 scores for tumor segmentation. This model could be developed into a tool for the objective evaluation of the response to NAT and may potentially guide the choice for adjuvant treatment.