Longitudinal trajectory and variability of skeletal muscle as predictors of prognosis after curative surgery for non-metastatic colorectal cancer

Background The impact of dynamic skeletal muscle alterations on colorectal cancer (CRC) prognosis remains unclear. This study evaluates the prognostic significance of longitudinal skeletal muscle index (SMI) trajectories and their variability in CRC. Methods We analyzed 2,560 consecutively enrolled patients with non-metastatic CRC from February 2012 to December 2019. Temporal SMl trajectories and variability were assessed based on serial CTscans spanning the preoperative to postoperative continuum. Latent class mixed models identified SMI trajectoy patterns and variability independent of mean (VIM _SIM ) was used to assess SMI variability. The outcomes included overall survival (OS) and recurrence-free survival (RFS). The STROBE Protocol was applied. Results Three distinct SMI trajectories were identified: Stable state, Declining fluctuations, and Stable increasing. Compared to the Stable State group, the Declining Fluctuations group exhibited significantly higher mortality risk, with adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs) of 2.91 (1.81–4.69) for ≥ 2 measurements, 3.11 (1.85–5.23) for ≥ 3 measurements, and 2.48 (1.32–4.68) for ≥ 4 measurements. In contrast, the Stable increasing group showed no significant survival difference versus Stable State. Higher SMI variability was independently associated with increased mortality risk, with aHRs [95% (CI)] per 0.01-unit increase in VIM _SIM of 1.21 (1.10–1.32) for ≥ 2 measurements, 1.59 (1.15–2.21) for ≥ 3 measurements, and 1.31 (1.06–1.62) for ≥ 4 measurements. Similar associations were observed for RFS. Conclusions SMI trajectory patterns and variability are independently associated with CRC prognosis. We recommend dynamic assessments of muscle health and personalized management strategies for CRC follow-up care.