Michelakos, Theodoros, Sekigami, Yurie, Kontos, Filippos, Fernández-del Castillo, Carlos, Qadan, Motaz, Deshpande, Vikram, Ting, David T., Clark, Jeffrey W., Weekes, Colin D., Parikh, Aparna, Ryan, David P., Wo, Jennifer Y., Hong, Theodore S., Allen, Jill N., Catalano, Onofrio, Warshaw, Andrew L., Lillemoe, Keith D., and Ferrone, Cristina R.
Background: Dynamic survival data based on time already survived are lacking for resected borderline resectable/locally advanced (BR/LA) pancreatic ductal adenocarcinoma (PDAC) patients who received total neoadjuvant therapy (TNT) with FOLFIRINOX followed by chemoradiation. Conditional survival, i.e., the probability of surviving an additional length of time after having already survived an amount of time, offers such information. We aimed to determine actuarial and conditional overall (OS, COS) and disease-free survival (DFS, CDFS) among this cohort. Methods: Clinicopathologic data were retrospectively collected for resected BR/LA PDAC patients who received TNT (2011–2019). COS and CDFS rates were calculated for patients being event (death/recurrence)-free at multiple intervals and by recurrence status. Results: After a median follow-up of 32.1 months, the 183 patients had a median OS and DFS of 39.1 months and 16.8 months, respectively. COS and CDFS increased as a function of time already survived. The probability of surviving an additional 24 months if a patient survived 2 years post-operatively was 70%, whereas the 4-year actuarial OS was 47%. Similarly, the probability of surviving disease-free an additional 24 months after 2 years was 66%, while actuarial 48-month DFS was 27%. COS for disease-free patients increased further over time. For patients remaining disease-free 12 months post-operatively, BR vs. LA status at diagnosis, tumor ≤ 4 cm at diagnosis, and R0 resection were independent predictors of favorable additional OS and DFS. Conclusions: For resected TNT-treated BR/LA PDAC patients, the probability of surviving an additional length of time increases as a function of survival already accrued. Dynamic survival estimates may allow personalized follow-up and counseling. [ABSTRACT FROM AUTHOR]