A single institution review of adjuvant therapy outcomes for resectable pancreatic adenocarcinoma: Outcome and prognostic indicators

Introduction: A large single-institution series of patients who recently underwent pancreaticoduodenectomy for resectable pancreatic cancer was analyzed to determine prognostic factors for overall survival, including the impact of adjuvant radiation and chemotherapy. Methods: Medical records were reviewed for 179 consecutive patients treated at The Cleveland Clinic with pancreaticoduodenectomy for resectable pancreatic adenocarcinoma from 1999 to 2006. Clinical data were collected, and Kaplan-Meier method was used to estimate overall survival. Univariate and multivariate analysis was performed. Results: One hundred seventy-nine patients with pT1-3N0-1M0 pancreatic cancer met the above criteria. But analysis was available for 158 patients. Median age at diagnosis was 67 (range 35-93). Peri-operative mortality rate was 0.6%. On univariate analysis, poor prognostic factors for overall survival were poorly differentiated histology, lymph node positive disease, elevated alkaline phosphatase, elevated total bilirubin, elevated AST, age at diagnosis >70, and high T stage. On multivariate analysis, poorly differentiated histology (p =.001), age >70 (p =.007), lymph node involvement (≥3 positive vs <3, p =.03), and elevated LFTs (alkaline phosphatase and/or bilirubin and/or AST; p =.002) were independent predictors of survival. Median survival for patients treated with adjuvant chemo-XRT was 28.4 months (vs. 11.8 months for patients receiving no adjuvant therapy (p <.001) in both univariate analysis and in multivariate analysis after adjusting for the independent prognostic factors described above). Median survival for patients treated with adjuvant chemotherapy alone had not yet been reached (p <.001 compared to no adjuvant therapy, in both univariate and multivariate analysis). Conclusion: In the twenty-first century, curative-intent surgery for pancreatic cancer at large academic institutions can have very low mortality rates. Pathology findings are valuable prognostic markers in resected pancreatic cancer. Few studies have examined the prognostic value of preoperative LFTs or lymph node ratio, and our analysis indicates they may have prognostic value-this should be confirmed in other series. Pts who receive adjuvant therapy (chemo-XRT or chemotherapy) appear to live longer than patients who receive no adjuvant therapy in this retrospective analysis.