Background: Prospective cohort studies have shown a positive association between body mass index (BMI) and the risk of colorectal cancer (CRC). However, limitations inherent in traditional observational studies, such as reverse causality and residual confounding, might explain the association. To overcome these limitations, Mendelian randomization (MR) studies of the BMI-CRC association have been conducted in European and U.S. groups, but the association remains to be clarified in East Asians. Purpose: We performed MR analyses to investigate the causal association between BMI and CRC in Japanese populations.

Methods: Our study design consisted of 4 steps. (1) Based on a previous Genome-Wide Association Study in Japanese populations, we selected 68 BMI-associated single nucleotide polymorphisms (SNPs), which explained about 2.0% of the BMI variance, as instruments. (2) We examined the associations between 68 SNPs and BMI among general Japanese populations in the Japanese Consortium of Genetic Epidemiology studies (N=36,253). (3) We performed a fixed-effect meta-analysis to investigate associations between 68 SNPs and CRC using individual-level data and publicly available summary-statistic data of Japanese populations (cases=7,473, controls=33,322). (4) Finally, we used the inverse-variance weighted (IVW) method to calculate MR estimates. Several sensitivity analyses were applied to assess robustness or horizontal pleiotropy using weighted median, weighted mode, MR-Egger regression, and MR-Pleiotropy Residual Sum and Outlier (PRESSO) methods.

Results: In the main analysis using the IVW method, a one-unit increase in BMI was associated with an odds ratio of 1.12 (95% confidence interval [CI]: 1.06-1.20) for CRC. Sensitivity analyses consistently showed increased odds ratios for CRC per one-unit increase in BMI. The odds ratios for weighted median, weighted mode, and MR-Egger regression were 1.16 (95% CI: 1.06-1.27), 1.14 (95% CI: 1.05-1.24), and 1.10(95% CI: 0.98-1.23), respectively. The MR-Egger intercept P-value was 0.63. No outlier was detected using the MR-PRESSO method.

Conclusions: Our MR analyses provide evidence that BMI is positively associated with CRC in Japanese populations. Our findings seem to suggest that MR estimates for the BMI-CRC association may be consistent across different ethnicities.

Citation Format: Shiori Suzuki, Atsushi Goto, Masahiro Nakatochi, Akira Narita, Taiki Yamaji, Norie Sawada, Ryoko Katagiri, Tsuyoshi Hachiya, Yoichi Sutoh, Isao Oze, Yuriko Koyanagi, Yumiko Kasugai, Hidemi Ito, Hiroaki Ikezaki, Keitaro Tanaka, Takashi Tamura, Haruo Mikami, Toshiro Takezaki, Sadao Suzuki, Nagato Kuriyama, Kiyonori Kuriki, Yoshikuni Kita, Kokichi Arisawa, Kenji Takeuchi, Kozo Tanno, Atsushi Shimizu, Gen Tamiya, Atsushi Hozawa, Kengo Kinoshita, Kenji Wakai, Makoto Sasaki, Masayuki Yamamoto, Keitaro Matsuo, Shoichiro Tsugane, Motoki Iwasaki. Body mass index and colorectal cancer risk in Japanese populations: a Mendelian randomization study [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 3486.