Gaurav Sharma1, Edward E Whang2,3, Daniel T Ruan2, Hiromichi Ito4. 1. Department of Surgery, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA. gsharma1@partners.org. 2. Department of Surgery, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA. 3. Department of Surgery, VA Boston Healthcare System, West Roxbury, MA, USA. 4. Department of Surgery, Michigan State University, College of Human Medicine, East Lansing, MI, USA. Hiromichi.Ito@hc.msu.edu.
Abstract
BACKGROUND: Neoadjuvant therapy-based protocols for potentially resectable pancreatic adenocarcinoma (PAC) have not been directly compared with adjuvant protocols in large prospective randomized trials. This study aimed to compare the efficacy of neoadjuvant versus adjuvant therapy-based management by using a formal decision analytic model. METHODS: A decision analytic model was created with a Markov process to compare neoadjuvant and adjuvant chemo- and/or chemoradiation therapy-based strategies for simulated cohorts of patients with potentially resectable PAC. Base-case probabilities were derived from the published data of 21 prospective phases 2 and 3 trials (3708 patients) between 1997 and 2014. The primary outcome measures determined in an intent-to-treat fashion were overall and quality-adjusted survival rates. One- and two-way sensitivity analyses were performed to assess the effects of model uncertainty on outcomes. RESULTS: The median overall survival and 2-year survival rates for the patients in the standard adjuvant therapy arm of the study were 20 months and 42.2 % versus 22 months and 46.8 % for those in the neoadjuvant strategy arm. Quality-adjusted survival was 18.4 and 19.8 months, respectively. Sensitivity analysis demonstrated that when recurrence-free survival after completion of neoadjuvant therapy and resection is less than 13.9 months or when the rate for progression of disease precluding resection during neoadjuvant therapy is greater than 44 %, the neoadjuvant strategy is no longer the favored option. CONCLUSIONS: The decision analytic model suggests that neoadjuvant therapy-based management improves the outcomes for patients with potentially resectable pancreatic cancer. However, the benefits in terms of overall and quality-adjusted survival are modest.
BACKGROUND: Neoadjuvant therapy-based protocols for potentially resectable pancreatic adenocarcinoma (PAC) have not been directly compared with adjuvant protocols in large prospective randomized trials. This study aimed to compare the efficacy of neoadjuvant versus adjuvant therapy-based management by using a formal decision analytic model. METHODS: A decision analytic model was created with a Markov process to compare neoadjuvant and adjuvant chemo- and/or chemoradiation therapy-based strategies for simulated cohorts of patients with potentially resectable PAC. Base-case probabilities were derived from the published data of 21 prospective phases 2 and 3 trials (3708 patients) between 1997 and 2014. The primary outcome measures determined in an intent-to-treat fashion were overall and quality-adjusted survival rates. One- and two-way sensitivity analyses were performed to assess the effects of model uncertainty on outcomes. RESULTS: The median overall survival and 2-year survival rates for the patients in the standard adjuvant therapy arm of the study were 20 months and 42.2 % versus 22 months and 46.8 % for those in the neoadjuvant strategy arm. Quality-adjusted survival was 18.4 and 19.8 months, respectively. Sensitivity analysis demonstrated that when recurrence-free survival after completion of neoadjuvant therapy and resection is less than 13.9 months or when the rate for progression of disease precluding resection during neoadjuvant therapy is greater than 44 %, the neoadjuvant strategy is no longer the favored option. CONCLUSIONS: The decision analytic model suggests that neoadjuvant therapy-based management improves the outcomes for patients with potentially resectable pancreatic cancer. However, the benefits in terms of overall and quality-adjusted survival are modest.