Prognostic Value of the Modified Systemic Inflammation Score in Stage I-III Colorectal Cancer: A Comparison with Other Inflammation-Based Indices

Kuan Wang^1*Boxiang Zhang¹Kejin Li²Ziyi Zhang¹Xiangyue Zeng¹Binlin Ma³Wu Zhimin⁴Yipeng Pan⁵Lucy Yue Lau⁶Zeliang Zhao^1*Yi Chen^1*

• ¹Cancer Hospital of Xinjiang Medical University, Urumqi, China
• ²Southern Medical University Nanfang Hospital, Guangzhou, China
• ³Bortala Mongolian Autonomous Prefecture People's Hospital, Bole, China
• ⁴The Maternal and Child Health Care Hospital of Guizhou Medical University, Guizhou, China
• ⁵Zhejiang University School of Medicine Sir Run Run Shaw Hospital, Hangzhou, China
• ⁶Harvard Medical School, Boston, United States

Background Colorectal cancer (CRC) remains one of the leading causes of cancer-related mortality worldwide, and poor postoperative prognosis significantly compromises patients' quality of life. Numerous inflammation-based biomarkers have been reported to predict survival outcomes in CRC patients. However, the prognostic value of the modified Systemic Inflammation Score (mSIS) in long-term outcome prediction following CRC surgery has not yet been fully established. This study aimed to evaluate the prognostic significance of mSIS—constructed using the neutrophil-to-lymphocyte ratio (NLR) and albumin levels—in predicting postoperative survival in CRC patients and to compare its predictive performance with other commonly used inflammation-related markers. Methods We retrospectively analyzed the clinical data of 489 CRC patients from two hospitals, with one institution serving as the validation cohort. Receiver operating characteristic (ROC) curve analysis was performed to determine the optimal cut-off values for pre-treatment systemic inflammation markers and to compare the AUC among various inflammatory indicators. The association between the mSIS and overall survival (OS) was evaluated using Kaplan–Meier survival curves and multivariate Cox regression analysis. Based on the results of the multivariate analysis, a nomogram model was constructed and externally validated. Results The optimal cut-off values for the NLR and serum albumin were determined to be 2.96 and 38.95 g/L, respectively, using the Youden index. High NLR and low albumin levels were independently associated with poorer OS. By combining NLR and albumin levels, we established a mSIS, which stratified patients into low-, intermediate-, and high-risk groups. In the training cohort, the 5-year OS rates were 93.75%, 61.26%, and 18.18%, respectively. The prognostic accuracy of mSIS surpassed that of either marker used alone. mSIS demonstrated a significantly higher AUC for predicting survival outcomes, and the nomogram model constructed based on multivariate Cox regression analysis exhibited superior prognostic discrimination. Conclusion Our study demonstrates that the mSIS can serve as an effective and accessible prognostic tool for stratifying stage I–III CRC patients after surgery and optimizing individualized treatment strategies.