gms | German Medical Science

Background: Pancreatic cancer is still one of the deadliest malignancies with a five-year survival rate of less than 6%. Complete surgical resection remains the most promising option for cure. If the tumor can be removed completely (R0 resection) and no nodal spread is found, overall five-year survival can rise up to 30%. To achieve final R0 pathological sections (PS), frozen sections (FS) from the resection margins are obtained intraoperatively. However, recent data challenge the benefit of additional resection in case of positive FS to achieve clear PS with respect to patients’ overall survival.

Aim: To investigate the effect of routinely obtained FS results and their management on oncologic outcome in patients undergoing surgery for exocrine pancreatic malignancies.

Materials and methods: We screened the prospectively maintained pancreatic surgery database for patients who underwent surgery for exocrine pancreatic malignancy with curative intent at our institution. Of 1477 patients with pancreatic surgery between 02/1993 and 12/2014, 483 patients met the following inclusion criteria: (1) Malignancy of the exocrine pancreatic tissue, (2) classical or pylorus-preserving partial pancreaticoduodenectomy (cPD, PPPD), distal pancreatectomy (DP), (3) results from pathology (FS and PS) available for analysis. All resection margins were included (pancreas neck, retroperitoneal, bile duct). Rx and R2 documented patients were excluded. Data was stratified by resection margin (group A: FS-R0 to PS-R0; group B: FS-R1 to PS-R0; group C: FS-R1 to PS-R1) and subsequently compared by using the Chi-square test and the t-test. Survival charts were created by using the Kaplan-Meier method and survival was compared using the log-rank test.

Results: In total, 69%, 21% and 9% of patients underwent PPPD, cPD and DP, respectively. 317 (75%) patients were in margin group A, 32 (8%) in group B, 73 (17%) in margin group C. As expected, patients in group B were found to have a significant longer operation time compared to the others. Furthermore, patients in group B and C appeared to have tumors with higher T stage and more frequently positive for lymphovascular invasion (LVI), perineural invasion (PNI), and lymph node involvement (p < 0,05). Additionally these characteristics were a significant predictor of poor OS. Median OS in margin group 1, 2 and 3 was 29, 36 and 12 months, respectively (p < 0,001). Survival analysis revealed no significant difference in survival between patients in Group A and B (p = 0.849), whereas patients in group C had significantly poorer outcome compared to group A (p < 0.001) and group B (p = 0.039).

Conclusion: In our cohort, we found no statistically significant difference in survival for patients with initially negative and positive FS results in case R0 resection is achieved on final pathological examination. We therefore conclude that FS analysis should be performed routinely in patients undergoing pancreatic cancer surgery with the aim to achieve a R0 resection.