W. Lutfi1, O. Kantor2, C. H. Wang3, E. Liederbach1, D. J. Winchester1, R. A. Prinz1, M. S. Talamonti1, M. S. Baker1 1Northshore University Health System,Department Of Surgery,Evanston, IL, USA 2University Of Chicago,Department Of Surgery,Chicago, IL, USA 3Northshore University Health System,Center For Biomedical Research Informatics,Evanston, IL, USA
Introduction:
There continues to be substantial international debate regarding the efficacy of neoadjuvant chemotherapy (NCT) prior to resection in early stage pancreatic cancer.
Methods:
We queried the National Cancer Data Base to identify patients that underwent pancreaticoduodenectomy (PD) for clinical stage (cStage) I-II pancreatic adenocarcinoma (PDAC) between 2000 and 2011. Multivariate logistic regression was used to analyze treatment trends and outcomes. Cox-modeling was used for survival analysis.
Results:
For the period studied, 8,099 patients underwent PD for cStage I-II PDAC. 3,237 (40.0%) were cStage I and 4,562 (60.0%) were cStage II. A total of 926 of the total recevied NCT; 271 (8.4%) with cStage I disease and 655 (13.5%) patients with cStage II disease received NCT. Use of NCT more than doubled over the period evaluated (6.5% of total patients in 2000 to 15.8% in 2011, p<0.01). Patients were more likely to receive NCT if they had vascular abutment (30.1% vs. 6.9%, p<0.01), were treated at an academic center (13.6% vs. 8.5%, p<0.01), or at a high volume hospital (17.8% vs. 10.2%, p<0.01). On univariate analysis, patients receiving NCT were more likely to have margin negative resection (81.8% vs. 75.3%, p<0.01) and be lymph node negative on final pathology (57.4% vs. 32.0% p<0.01) than those that did not. Multivariate regression adjusting for age, sex, race, comorbidities, insurance, socio-economic status, hospital type, location and volume, tumor grade, and vascular abutment identified patient age <56 years (OR 2.62, CI: 1.88-3.63), African-American race (OR 1.41, CI: 1.08-1.85), tumors that had vascular abutment (OR 5.18, CI: 4.27-6.28), cStage II disease (OR 1.33, CI: 1.13-1.57), and treatment at a facility with high surgical volume (OR 2.28, CI: 1.73-2.99) to be factors independently associated with use of NCT. Cox survival analysis adjusted for age, sex, race, comorbidities, insurance, socio-economic status, hospital type, location and volume, tumor grade, margins, and vascular abutment demonstrated a significant survival benefit for NCT in patients determined to have cStage II disease with median overall survival 21.6 months for those treated with NCT vs. 16.5 months receiving no treatment prior to surgery (p<0.01). Patients with cStage I treated with NCT demonstrated no difference in overall survival compared to those having surgery prior to systemic therapy (median overall survival 19.4 months vs. 19.7 months, p>0.90).
Conclusion:
Patients with cStage I and II disease treated with NCT prior to resection demonstrate higher rates of margin negative and node negative resection than stage-matched patients receiving no chemotherapy prior to resection. NCT appears to provide a statistically relevant survival benefit to patients presenting with clinical stage II but not stage I disease at time of diagnosis.