Association of preoperative CT-derived visceral adipose tissue index with synchronous metastasis and metastasis-free survival after curative-intent surgery in colorectal cancer.
Level 3 - non-randomized controlled study
Retrospective two-center cohort and matched case-control study
PubMed 42311267 · doi:10.3389/fonc.2026.1830599
What was done
This retrospective two-center study included 468 patients with newly diagnosed colorectal cancer (CRC; 77 metastatic [mCRC], 391 non-metastatic [nmCRC]) who had abdominal CT scans within 1 month before diagnosis or initial treatment between 2019 and 2022. For synchronous metastasis analysis, mCRC and nmCRC patients were propensity score matched 1:1 (77 pairs) on age, sex, CEA, CA19-9, center, and tumor location, followed by multivariable logistic regression. For metastasis-free survival (MFS), 391 surgical nmCRC patients were stratified by sex-specific median visceral adipose tissue index (VATI) into low-VATI (n = 195) and high-VATI (n = 196) cohorts for Kaplan-Meier and multivariable Cox proportional hazards analyses. An exploratory comparison assessed body composition metrics between early (≤ 2 years, n = 76) and late (> 2 years, n = 20) postoperative metastasis.
What was found
After matching, mCRC patients had higher subcutaneous adipose tissue index (SATI), VATI, and visceral-to-subcutaneous fat area ratio (VSR) than nmCRC patients. On multivariable logistic regression, VATI was independently associated with synchronous metastasis (OR 1.110, 95% CI 1.067–1.155, p < 0.001), while SATI and VSR were not. In the surgical nmCRC cohort, high VATI was associated with significantly shorter MFS (log-rank p = 0.001). On multivariable Cox analysis, VATI remained an independent predictor of shorter MFS (HR 1.017, 95% CI 1.003–1.032, p = 0.021), along with CA19-9, CEA, N stage, and lymphovascular invasion. No significant body composition differences were observed between early and late metastasis groups.
Why it matters
CT-derived VATI functions as an opportunistic imaging biomarker that correlates with both synchronous metastasis and recurrence risk following curative-intent CRC resection, potentially aiding risk stratification without extra imaging procedures.
Limits
The study is limited by its retrospective design across two centers, potential selection bias, and unmeasured confounders. The hazard ratio per unit increase in VATI for MFS was small (HR 1.017), and prospective external validation is lacking.