Prognostic Effect of Age in Resected Pancreatic Cancer Patients: A Propensity Score Matching Analysis

Front Oncol. 2022 Mar 31:12:789351. doi: 10.3389/fonc.2022.789351. eCollection 2022.

Abstract

Background: While the elderly population account for an indispensable proportion in pancreatic ductal adenocarcinoma (PDAC), these patients are underrepresented in clinical trials. Whether surgery offered the same benefit for elderly patients as that for younger cohort and which factors affected long-term outcome of elderly population remained unclear.

Aims: This study aims to evaluate long-term prognosis of elderly PDAC patients (≥70 years old) after surgery and to investigate potential prognostic factors.

Methods: This retrospective study included PDAC patients receiving radical resection from January 2012 to July 2019 in Zhongshan Hospital Fudan University. Patients were divided into young (<70) and old groups (≥70). Propensity score matching (PSM) was conducted to eliminate the confounding factors. We investigated potential prognostic factors via Cox proportional hazards model and Kaplan-Meier estimator. Nomogram model and forest plot were constructed to illustrate the prognostic value of age.

Results: A total of 552 PDAC patients who received radical resection were included in this research. Elderly patients showed poorer nutritional status and were less likely to received adjuvant treatment. After matching, although age [hazard ratio (HR)=1.025, 95%CI 0.997-1.054; p=0.083] was not statistically significant in the multivariate cox regression analysis, further survival analysis showed that patients in the old group had poorer overall survival (OS) when compared with young group (p=0.039). Furthermore, reception of adjuvant chemotherapy (HR=0.411, 95%CI 0.201-0.837; p=0.014) was the only independent prognostic factor among elderly patients and could significantly improve OS. Subgroup analysis indicated that age had better prognostic value in PDAC patients with good preoperative nutritional status and relative low tumor burden. Finally, a prognostic prediction model contained age, reception of adjuvant chemotherapy, American Joint Committee on Cancer (AJCC) 8th T and N stage was constructed and presented in nomogram, whose Harrell's concordance index was 0.7478 (95%CI, 0.6960-0.7996). The calibration curves at 1 and 3 years indicated an optimal conformity between actual and nomogram-predicted survival probability in the PDAC patient who received surgery.

Conclusion: The elderly PDAC patients were associated with worse OS survival after radical resection, and the noticeable negative effect of age was observed among PDAC patients with better preoperative nutritional status and less aggressive tumor biology. Adjuvant chemotherapy was essential to improve survival outcome of elderly PDAC patients following radical resection.

Keywords: adjuvant chemotherapy; pancreatic ductal adenocarcinoma; prognosis; radical resection; the elderly patients.