Abstract
BACKGROUND: Systemic inflammation markers have shown prognostic values with variability in rectal cancer. Considering the association of serum calcium with inflammation, we aimed to examine whether it could improve systemic inflammation markers for survival prediction. METHODS: We enrolled 508 patients with stage I to III rectal cancer who underwent curative resection. The cohort was grouped by corrected serum calcium (cCa), platelet-to-lymphocyte ratio (PLR), and CaPLR (a score model combining cCa with PLR) for survival analysis. The LR (likelihood ratio) test and AIC (Akaike information criterion) were applied to compare models in survival prediction. The primary endpoint was disease-free survival (DFS). RESULTS: A total of 26.7% (136/508) patients reached recurrence after curative surgery. Both high cCa (HR 1.486; 95% CI, 1.018-2.171; P=0.040) and high PLR (HR 1.452; 95% CI, 1.059-1.991; P=0.021) were significantly associated with worse DFS. In model comparison, the AIC and LR were improved after cCa was added to PLR model in DFS prediction (AIC: 1,704.83 vs. 1,707.14 vs. 1,707.15; LR: 8.68 vs. 4.37 vs. 4.36; P=0.037). The CaPLR was developed for DFS prediction with adjusted HRs of 2.216 (95% CI, 1.256-3.909; P=0.006) and 1.679 (95% CI, 1.004-2.836; P=0.047) for high and intermediate score group respectively compared to low score group. A nomogram for predicting DFS was generated by using CaPLR and other clinical predictors, with a concordance index of 0.705 (95% CI, 0.620-0.789; P<0.001). CONCLUSIONS: Serum calcium could improve systemic inflammation markers in survival prediction for patients with rectal cancer.