AUTHOR=Zhang Zixi , Yan Xueli , Lu Yan , Guo Xiaowen , Jiao Min , Wang Weizhong , Sun Boqian , Zhou Yi , Hu Qinglin , Chu Dake TITLE=The prognostic impact of BMI on colorectal cancer is stratified by tumor location JOURNAL=Frontiers in Oncology VOLUME=Volume 12 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.987518 DOI=10.3389/fonc.2022.987518 ISSN=2234-943X ABSTRACT=Purpose

Recent studies have revealed the contrasting prognostic roles of body mass index (BMI) and tumor location in colorectal cancer (CRC). Given that right- and left-sided CRC may exhibit inverse effects on outcome and body weight, the present study aimed to examine whether the prognostic value of BMI and tumor location could be reciprocally stratified.

Methods

This prospective, observational study recruited 4,086 patients diagnosed with stage III CRC from five independent clinical centers in China. The association of patients’ outcomes with BMI and tumor location was evaluated hierarchically by Kaplan–Meier and Cox proportional-hazards models.

Results

Although BMI was not associated with overall outcome, the association was significantly modified by tumor location. Among left-sided tumors, obesity and overweight were significantly associated with adverse overall survival (OS) and disease-specific survival (DSS). In contrast, among right-sided tumors, overweight was significantly associated with more favorable OS and DSS compared with the normal-weight group. The association of survival with tumor location did not reach statistical significance. However, hierarchical analysis by BMI revealed that left-sided tumors were associated with more favorable outcomes in the normal-weight group, while there was no statistically significant difference in the overweight or obese group.

Conclusions

BMI and tumor location may have opposing effects on CRC prognosis, when stratified by each other, after adjusting for other known prognostic factors. These findings are the first to show the interactive prognostic impact of BMI and tumor location, which could be relevant to the stratification of patient management.