THE IMPACT OF CHEMOTHERAPY SEQUENCING ON PANCREATIC CANCER BY STAGE
Eduardo A. Vega*1,4, Onur C. Kutlu2, Omid Salehi1,4, Olga Kozyreva1,5, Beth Herrick1, Christopher Lathan1,5, Ernest R. Camp3, Sandeep Krishnan1, Christopher Stallwood1, Claudius Conrad1,4
1St. Elizabeth's Medical Center, Boston, MA; 2University of Miami Miller School of Medicine, Miami, FL; 3University of South Carolina, Charleston, SC; 4Tufts University School of Medicine, Boston, MA; 5Dana-Farber Cancer Institute/ Harvard Medical School, Boston, MA
Introduction: The optimal sequence of chemotherapy (preoperative, postoperative, or perioperative) for patients with resectable pancreatic adenocarcinoma (PAC) remains an area of active investigation. Further, the utility of neoadjuvant chemotherapy in very early stage patients remains controversial.
Methods: National Cancer Database (NCDB) was queried for patients with confirmed PAC (histology codes 8140, 8500), who underwent pancreaticoduodenectomy (PD) between 2004 through 2016. Patients for whom all of the following parameters were available were included: T category, nodal status, tumor size, tumor grade, M0 disease, Charleson-Deyo comorbidity score, margin status, status at last follow-up, 30 day mortality, details of chemotherapy (sequencing, use of single vs. multiple agent), complete follow-up. Institutional case volumes were calculated and disease stages for all patients were converted to AJCC 8th edition using custom computer script programming. Kaplan Meier Survival analyses for disease specific (DSS) and overall survival (OS) was performed to assess the effect of chemotherapy sequence; preoperative only (NCT), postoperative only (ACT), pre-postoperative (PCT), and no therapy (NoT) Groups were compared pairwise with log-rank test. Cox multivariable regression analyses were performed for both DSS and OS correcting for age, sex, comorbidity status, grade, margin status, chemotherapy sequence, hospital stay, institutional case volume, and type of institution for each AJCC TNM stage separately.
Results: 22975 patients met inclusion criteria. 13944 (60.7%) received ACT, 1793(7.8%) received NCT, and 946 (4.1%) received PCT while 6292 (29.5%) did not receive chemotherapy. Log-rank test showed inferior survival in the NoT group compared to NCT, ACT, and PCT. The only difference among groups receiving chemotherapy was observed between NCT and PCT (p=0.014), with PCT demonstrating superior outcome. After correcting for confounders, compared to NoT group, PCT had the lowest rate of death (HR 0.704, p<0.001) followed by NCT (HR 0.721, p<0.001) and ACT (HR 0.759 p<0.001). Chemotherapy positively affected DSS and OS at all stages except for NCT in stage IA. (p=0.185).
Conclusion: All resectable PAC patients (all stages) should receive chemotherapy. More specifically, NCT should be offered to all resectable PAC-patients independent of their stage and optimally complemented with postoperative chemotherapy (PCT). While there was a trend towards optimal survival for PCT, total neoadjuvant chemotherapy is a reasonable option for stages IB-III. PAC-patients receiving chemotherapy independent of their stage will result in decreased rates of disease specific mortality as well as optimal DSS and OS.
Back to 2020 Abstracts