Elevated serum homocysteine levels associated with poor recurrence-free and overall survival in patients with colorectal cancer

Abstract

This study aimed to evaluate the significance of homocysteine (HCY) levels in predicting recurrence-free survival (RFS) and overall survival (OS) in colorectal cancer (CRC) patients. This retrospective study involved 1272 CRC patients. The risk of mortality increased with increasing HCY levels in CRC patients. The optimal HCY cutoff value in CRC patients was 15.2 μmol/L. The RFS (45.8% vs. 60.5%, p < 0.001) and OS (48.2% vs. 63.2%, p < 0.001) of patients with high HCY levels were significantly lower than those of patients with low HCY levels. Patients with high HCY levels were older, male, had large tumours, high carcinoembryonic antigen (CEA) levels, and long hospital stays, and incurred high hospitalisation costs. Multivariate analysis showed that when HCY levels exceeded 15.2 μmol/L, the risk of adverse RFS and OS increased by 55.7% and 61.4%, respectively. Subgroup analysis showed that HCY levels could supplement CEA levels and pathological staging. We constructed HCY-based prognostic nomograms, which demonstrated feasible discrimination and calibration values better than the traditional tumour, node, metastasis staging system for predicting RFS and OS. Elevated serum HCY levels were strongly associated with poor RFS and OS in CRC patients. HCY-based prognostic models are effective tools for a comprehensive evaluation of prognosis.

Keywords: Colorectal cancer; Homocysteine; Malnutrition; Nomogram; Prognostic.

Figure 1

Kaplan–Meier curve of HCY levels in patients with colorectal cancer. (A) Kaplan–Meier curve for RFS; (B) Kaplan–Meier curve for OS. RFS recurrence-free survival, OS overall survival.

Figure 2

Stratified Kaplan–Meier curve of HCY levels based on TNM stage subgroup in patients with colorectal cancer. (A) RFS (Stage I–II); (B) OS (Stage I–II); (C) RFS (Stage III–IV); (D) RFS (Stage III–IV). RFS recurrence-free survival, OS overall survival.

Figure 3

Restricted cubic splines to flexibly model and visualize the relation of predicted HCY with mortality in patients with colorectal cancer. (A) Restricted cubic splines for RFS; (B) Restricted cubic splines for OS. Model a: No adjusted. Model b: Adjusted for age, sex, BMI, T stage, N stage, M stage. Model c: Adjusted for sex, age, BMI, T stage, N stage, M stage, tumor location, tumor size, perineural invasion, vascular invasion, differentiation, radiotherapy, chemotherapy, hypertension, diabetes, family history. RFS recurrence-free survival, OS overall survival.

Figure 4

Construction the HCY-based RFS nomograms in CRC patients. Notes: The nomogram is composed of specific clinical features, with each feature corresponding to a specific point. The score for each feature can be calculated by drawing a straight line along the point axis, and the sum of these feature scores is then positioned on the total point axis. The risk probability can be calculated by drawing downward to the predicted axis. RFS recurrence-free survival, CRC colorectal cancer. Normal CEA was defined as < 5 ng/mL, while high CEA was defined as ≥ 5 ng/mL.

Figure 5

Construction the HCY-based OS nomograms in CRC patients. The nomogram is composed of specific clinical features, with each feature corresponding to a specific point. The score for each feature can be calculated by drawing a straight line along the point axis, and the sum of these feature scores is then positioned on the total point axis. The risk probability can be calculated by drawing downward to the predicted axis. OS overall survival, CRC colorectal cancer. Normal CEA was defined as < 5 ng/mL, while high CEA was defined as ≥ 5 ng/mL.