Dynamic monitoring of immunoglobulin G as a prognostic indicator after curative resection in high-risk stage II-III colorectal cancer: a retrospective cohort study.

BACKGROUND

Reliable markers for predicting postoperative recurrence in high-risk stage II-III colorectal cancer remain limited. Dynamic changes in immunoglobulin G (IgG) may provide prognostic information beyond carcinoembryonic antigen (CEA).

METHODS

This single-centre retrospective cohort study included 192 patients with high-risk stage II or III colorectal cancer who underwent curative resection between January 2021 and June 2023. The study evaluated whether dynamic postoperative monitoring of serum IgG predicts 2-year disease-free survival (DFS) compared with CEA. Serial IgG and CEA measurements within 24 months were categorised as favourable or unfavourable trajectories based on predefined thresholds and temporal trends. Patients were further stratified into four groups according to combined IgG and CEA dynamics. Survival was assessed using Kaplan-Meier analysis and Cox regression models.

RESULTS

Among 192 eligible patients, 60 (31.3%) experienced recurrence or death within 2 years. Unfavourable IgG trajectories (n=82) were associated with significantly lower 2-year DFS than favourable trajectories (55% [95% CI 44-65] vs 82% [95% CI 73-88], log-rank P<0.01). CEA trajectories showed only borderline separation (67% [95% CI 55-77] vs 79% [95% CI 71-85], log-rank P = 0.06). Patients with both unfavourable IgG and CEA trajectories had the poorest outcomes (2-year DFS 31% [95% CI 16-47]). In multivariable analysis, an unfavourable IgG trajectory independently predicted recurrence or death (HR 2.05, 95% CI 1.32-3.18), whereas CEA trajectory was not significant.

CONCLUSION

Dynamic postoperative IgG monitoring is independently associated with disease recurrence in high-risk stage II-III colorectal cancer and offers incremental prognostic value beyond CEA. Incorporating serial IgG measurements may enhance postoperative risk stratification, although confirmation in prospective multicentre studies is warranted.