Impact on survival by age stratification for elderly gastric cancer patients

How to Cite

Kai. (2017). Impact on survival by age stratification for elderly gastric cancer patients. American Journal of Translational Medicine, 1(3), 173–182. Retrieved from


Elderly patients have a worse prognosis in many cancers. However, the impact on survival by age stratification for elderly patients with resectable gastric cancer (RGC) has not been thoroughly elaborated. Our purpose was to compare the prognostic outcomes for different age subgroups of elderly patients with RGC. A surveillance, epidemiology, and end results (SEER) database was used to identify 10, 931 elderly patients of known age (≥ 40) with GC between 2004 and 2011 treated with surgery. The patients were divided into four groups: group 1 (40-55 years), group 2 (56-70 years), group 3 (71-85 years), and group 4 (85+ years). Kaplan–Meier plots were used to analyze elderly GC-specific survival (GCSS) data. Multivariable Cox regression models were built for the analysis of long-term survival outcomes and risk factors. There were significant differences between the four groups in the year of diagnosis, sex, race, primary site, grade, histologic type, neoadjuvant radiotherapy, AJCC stage, mean number of lymph nodes (LNs) examined, mean number of positive LNs, and mean number of current standards for lymphadenectomy (all p < 0.05). In all elderly patients with RGC, the 5-year cancer-specific survival (CSS) in group 1 was similar to that in group 2 (52.9% vs. 50.3%, p >0.05), but higher than that in group 3 (52.9% vs. 48.4%, p < 0.05) and group 4 (52.9% vs. 42.6%, p < 0.05). Similar prognostic outcomes were observed in AJCC stage II; while in AJCC stage I and III, with age increasing, the survival outcomes by age stratification were worse and worse (group 1 vs. group 2, 3, or 4; group 2 vs. group 3 or 4; group 3 vs. group 4) (all p < 0.05). Cox regression models showed that age was an independent prognostic factor for elderly patients with RGC (p < 0.001). After adjustment of co-variables, stratified analysis for age yielded consistent results (group 1 as ref., group 2: HR, 0.401, 95 %CI: 0.349~0.460; group 3: HR, 0.470, 95 %CI: 0.413~0.535; group 4: HR, 0.685, 95 %CI: 0.603~0.777). Prognostic outcomes by age stratification (40-55, 56-70, 71-85, and 85+) for elderly patients with RGC were essentially heterogeneous. Elderly patients in the subgroups of ages 71-85 and 85+ had poorer CSS than patients 40-70 years old, although they had better clinicopathological characteristics (more with grade I/II, adenocarcinoma, no neoadjuvant radiotherapy, and AJCC I/II). Further studies are warranted to verify our findings. (Am J Transl Med 2017. I:173-182)