The Association of Abdominal Adiposity With Mortality in Patients With Stage I–III Colorectal Cancer

Justin C. Brown; Bette J. Caan; Carla M. Prado; Elizabeth M. Cespedes Feliciano; Jingjie Xiao; Candyce H. Kroenke; Jeffrey A. Meyerhardt

Disclosures

J Natl Cancer Inst. 2020;112(4):377-383. 

In This Article

Discussion

In this population-based cohort study of 3262 patients with stage I–III colorectal cancer, abdominal adipose tissue quantity and distribution were prognostic of all-cause mortality. The shapes of these associations were generally nonlinear and modified by patient sex. Among men, larger quantities of abdominal subcutaneous adipose tissue were associated with a higher risk of mortality, whereas among women, larger quantities of abdominal visceral adipose tissue were associated with an increased risk of mortality. Conversely, among men, moderate quantities of visceral adipose tissue were associated with a lower risk of mortality, whereas among women, moderate quantities of subcutaneous adipose tissue were associated with a lower risk of mortality.

Previous studies have modeled adipose tissue using categories, which has led to inconsistent findings. In a retrospective cohort study of 62 patients with stage I–III colorectal cancer, a visceral adipose tissue area greater than 130 cm2 was associated with a statistically significantly higher risk of all-cause mortality (hazard ratio [HR] = 7.0, 95% CI = 2.0 to 24.6).[12] In another study of 219 patients with stage I–III colorectal cancer, visceral adipose tissue in the higher 50th percentile was associated with statistically significantly higher risk of disease recurrence and mortality in patients with stage II disease (HR = 2.72, 95% CI = 1.21 to 6.10) and a pattern of lower risk of disease recurrence and mortality in patients with stage III disease, but this did not reach statistical significance (HR = 0.50, 95% CI = 0.23 to 1.06).[13] The large sample size of the current study, combined with the application of restricted cubic splines, afforded us greater statistical power and model flexibility to characterize the associations of abdominal adipose tissue with mortality. Unlike the prior study,[13] no effect modification by cancer stage was identified in the current study. The prognostic associations of adiposity were independent of muscle area, which has been previously reported as a predictor of mortality in cancer patients.[18] Notably, moderate amounts of visceral adiposity among men and moderate amounts of subcutaneous adiposity among women were not associated with mortality; the nadir risks of mortality for visceral adiposity in men were 200–250 cm2 and for subcutaneous adiposity in women were 275–325 cm2.

Prior studies that investigated the association between BMI and mortality reported effect modification by patient sex for early-stage colon cancer (Pinteraction = .034)[2] and metastatic colorectal cancer (Pinteraction < .001).[26] The underlying biological mechanism of effect modification by patient sex is hypothesized to relate to adipose tissue distribution and resultant metabolic perturbations.[9,10] Biopsy studies of patients with colorectal cancer demonstrated that visceral adipose tissue exhibits altered inflammatory, lymphocytic, and fatty acid secretory properties compared with subcutaneous adipose tissue.[5–7] In the current study, correlational analyses between these tissues demonstrated that patients with higher areas of subcutaneous adipose tissue did not necessarily also have higher areas of visceral adipose tissue. This counters the hypothesis that patients who have poor outcomes are those who have both excess visceral adipose tissue and excess subcutaneous adipose tissue when measured at the third lumbar vertebra.

Our data are consistent with the hypothesis that dysregulated deposition of adiposity is prognostic of mortality: Men who preferentially store excess adiposity subcutaneously and women who preferentially store excess adiposity viscerally are at a statistically significantly higher risk of mortality than their counterparts who store adiposity in the regions expected for their sex. Sex steroid hormones are the principal factor that influences adipose tissue deposition.[29] Testosterone suppression in healthy young men increases subcutaneous adipose tissue deposition,[30] whereas ovarian hormone suppression in healthy premenopausal women increases visceral adipose tissue deposition.[31] Similarly, the decline in estrogen that naturally occurs during menopause in women is associated with increases in visceral adipose tissue deposition.[32] Sex steroid hormones that regulate adipose tissue deposition, including estradiol and testosterone, are associated with cancer risk and prognosis.[33,34]

Study limitations include the observational design and the potential for residual confounding. Because this study relied on routine clinically acquired data, measures of socioeconomic status, diet, or menopausal status were not available. The inclusion of these variables may modestly influence our effect size estimates, but it is unlikely that the overall shape of these prognostic associations would substantively change. Body composition was measured at the third lumbar vertebra at a solitary time point. Although the third lumbar vertebra is strongly correlated with visceral and subcutaneous adipose tissue volumes,[20] anatomic differences in the distribution of adiposity between men and women may have an influence on our findings;[35] nevertheless our data demonstrate that the third lumbar vertebra is prognostic in both sexes. We did not have complete measures of body mass or composition before the diagnosis of cancer; therefore, we cannot rule out the possibility that some patients may have experienced tumor-induced changes in adiposity. However, we did not identify any interactions with clinical cancer stage.

There are several strengths of this study. The main strengths are the large sample size and population-based, racially and ethnically diverse sample. The large sample size offered sufficient statistical power to evaluate associations by sex-specific subgroups. The use of clinically acquired CT images, coupled with recently developed semi- and fully automated radiologic techniques to quantify adiposity using clinical imaging, make the integration of body composition measures into clinical practice cost effective and offers the opportunity to personalize oncology care.

In conclusion, abdominal visceral and subcutaneous adipose tissue quantity was prognostic of all-cause mortality in patients with stage I–III colorectal cancer, and the strength and shape of these prognostic associations varied by patient sex. Measurements of body composition using CT can be seamlessly integrated into clinical care and used to identify those at risk for poor outcome.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....