Anatomically resectable versus biologically borderline resectable pancreatic cancer definition: refining the border beyond anatomical criteria and biological aggressiveness

Background: The anatomically resectable pancreatic ductal adenocarcinoma treatment sequence is still debated. Heterogeneity in patient characteristics within this group may explain literature discrepancies. To overcome these limits, a biologically borderline resectable pancreatic ductal adenocarcinoma category has been analysed according to institutional criteria. The aim of this study was to examine the characteristics and outcomes of patients with biologically borderline resectable pancreatic ductal adenocarcinoma and determine whether they represent a distinct clinical and prognostic subgroup. Methods: Data from all consecutive patients who underwent surgical resection for pancreatic ductal adenocarcinoma between 2015 and 2022 were retrospectively analysed. Biologically borderline resectable disease was classified by the presence of one or more of the following: carbohydrate antigen 19-9 ≥200 U/ml, cancer-related symptoms lasting >40 days, and radiological suspicion of regional lymph node metastases at diagnosis. Results: In total, 886 patients were included in the study and divided into anatomically borderline resectable (266 patients (30%)) and anatomically resectable (620 patients (70%)), which was further divided into resectable (R; 397 patients (64%)) and biologically borderline resectable (223 patients (36%)). Neoadjuvant treatment was administered in 245 patients (92.1%) in the anatomically borderline resectable group, 82 patients (20.7%) in the R group, and 135 patients (60.5%) in the biologically borderline resectable group. After a median follow-up of 45 (95% c.i. 42 to 48) months, the median disease-specific survival in the biologically borderline resectable group was 40 months compared with 59 months in the R group (P < 0.001) and 40 months in the anatomically borderline resectable group (P = 0.570). In the upfront surgery cohort, the median disease-specific survival was worse for biologically borderline resectable patients compared with R patients (27 versus 54 months respectively, P < 0.001). Biologically borderline resectable was also independently associated with worse disease-specific survival, together with age, tumour size at diagnosis, and anatomically borderline resectable. The same, except for age, were also predictors of worse event-free survival. Conclusion: Despite their identical anatomical appearance, resectable and biologically borderline resectable pancreatic ductal adenocarcinoma represent two distinct prognostic entities, warranting separate evaluation and, potentially, different treatment approaches.