Background: This study aimed to identify predictors for early and very early disease recurrence in patients undergoing resection of pancreatic ductal adenocarcinoma (PDAC) resection with and without neoadjuvant therapy.Methods: Included were patients who underwent PDAC resection (2014)(2015)(2016). Multivariable multinomial regression was performed to identify preoperative predictors for manifestation of recurrence within 3, 6 and 12 months after PDAC resection.Results: 836 patients with a median follow-up of 37 (interquartile range [IQR] 30-48) months and overall survival of 18 (IQR 10-32) months were analyzed. 670 patients (80%) developed recurrence: 82 patients (10%) <3 months, 96 patients (11%) within 3-6 months and 226 patients (27%) within 6-12 months.LogCA 19-9 (OR 1.25 [95% CI 1.10-1.41]; P < 0.001) and neoadjuvant treatment (OR 0.09 [95% CI 0.01-0.68]; P = 0.02) were associated with recurrence <3 months. ; P < 0.001) and 0-90 venous involvement on CT imaging ; P < 0.001) were associated with recurrence within 3-6 months. A Charlson Age Comorbidity Index 4 (OR 1.53 [95% CI 1.09-2.16]; P = 0.02) and logCA 19-9 (OR 1.24 [95% CI 1.14-1.35]; P < 0.001) were related to recurrence within 6-12 months. Conclusion:This study demonstrates preoperative predictors that are associated with the manifestation of early and very early recurrence after PDAC resection. Knowledge of these predictors can be used to guide individualized surveillance and treatment strategies.

Background The prognostic value of four proposed modifications to the 8th American Joint Committee on Cancer (AJCC) TNM staging system has yet to be evaluated. This study aimed to validate five proposed modifications. Methods Patients who underwent pancreatic ductal adenocarcinoma resection (2014–2016), as registered in the prospective Dutch Pancreatic Cancer Audit, were included. Stratification and prognostication of TNM staging systems were assessed using Kaplan–Meier curves, Cox proportional hazard analyses, and C-indices. A new modification was composed based on overall survival (OS). Results Overall, 750 patients with a median OS of 18 months (interquartile range 10–32) were included. The 8th edition had an increased discriminative ability compared with the 7th edition {C-index 0.59 (95% confidence interval [CI] 0.56–0.61) vs. 0.56 (95% CI 0.54–0.58)}. Although the 8th edition showed a stepwise decrease in OS with increasing stage, no differences could be demonstrated between all substages; stage IIA vs. IB (hazard ratio [HR] 1.30, 95% CI 0.80–2.09; p  = 0.29) and stage IIB vs. IIA (HR 1.17, 95% CI 0.75–1.83; p  = 0.48). The four modifications showed comparable prognostic accuracy (C-index 0.59–0.60); however, OS did not differ between all modified TNM stages (ns). The new modification, migrating T3N1 patients to stage III, showed a C-index of 0.59, but did detect significant survival differences between all TNM stages (p  < 0.05). Conclusions The 8th TNM staging system still lacks prognostic value for some categories of patients, which was not clearly improved by four previously proposed modifications. The modification suggested in this study allows for better prognostication in patients with all stages of disease.