Survival Benefit of Exercise Differs by Tumor IRS1 Expression Status in Colorectal Cancer

Abstract

Background: High-level physical activity is associated with lower colorectal cancer (CRC) mortality, likely through insulin sensitization. Insulin receptor substrate 1 (IRS1) is a mediator of insulin and insulin-like growth factor (IGF) signaling pathways, and its down-regulation is associated with insulin resistance. Therefore, we hypothesized that tumor IRS1 expression status might modify cellular sensitivity to insulin and IGF, and the prognostic association of physical activity.

Methods: We assessed IRS1 expression level in 371 stage I-III rectal and colon cancers in the Nurses' Health Study and the Health Professionals Follow-up Study by immunohistochemistry. In survival analysis, Cox proportional hazards model was used to assess an interaction between post-diagnosis physical activity (ordinal scale of sex-specific quartiles Q1 to Q4) and IRS1 expression (ordinal scale of negative, low, and high), controlling for potential confounders, including microsatellite instability, CpG island methylator phenotype, long interspersed nucleotide element-1 (LINE-1) methylation level, and KRAS, BRAF, and PIK3CA mutation status.

Results: There was a statistically significant interaction between post-diagnosis physical activity and tumor IRS1 expression in CRC-specific mortality analysis (P interaction = 0.005). Multivariable hazard ratio (95% confidence interval) for higher post-diagnosis physical activity (Q3-Q4 vs. Q1-Q2) was 0.15 (0.02-1.38) in the IRS1-negative group, 0.45 (0.19-1.03) in the IRS1-low group, and 1.32 (0.50-3.53) in the IRS1-high group.

Conclusions: The association of post-diagnosis physical activity with colorectal carcinoma patient survival may differ by tumor IRS1 expression level. If validated, tumor IRS1 expression status may serve as a predictive marker to identify subgroups of patients who might gain greater survival benefit from an increased level of exercise.

Figure 1. Flow chart of case selection

Among the incident cases of colorectal cancer, those with one or more of the followings were excluded from this study: no available cancer tissue, stage IV, no tissue in tissue microarray or valid IRS1 data, or no information of post-diagnosis physical activity. (HPFS, Health Professionals Follow-up Study; NHS, Nurses’ Health Study, TMA, tissue microarray)

Figure 2. IRS1 immunohistochemical analysis in colorectal cancer cells

(A) Colorectal cancer cells with no/little staining for IRS1 are classified to be IRS1-negative. (B) Weak staining for IRS1 in cytoplasm of colorectal cancer cells indicates low-level IRS1 expression. (C) Strong staining for IRS1 in cytoplasm of colorectal cancer cells indicates high-level IRS1 expression.