Prognostic value of combined systemic inflammation response index and prognostic nutritional index in colorectal cancer patients

BACKGROUND The prognosis of colorectal cancer (CRC) patients is notably influenced by both inflammation and nutritional status. The prognostic nutritional index (PNI) and systemic inflammatory response index (SIRI) have been reported in prognostic studies of various tumors. However, the efficacy of the combination of the two in predicting the prognosis of CRC patients has not been studied. AIM To evaluate the effectiveness of PNI and SIRI in predicting the prognosis of patients with CRC. METHODS We retrospectively gathered data from 470 CRC patients who underwent feasible radical surgery at Xinjiang Cancer Hospital. The optimal cut-off values for SIRI and PNI, along with their predictive power for survival, were determined through area under the receiver operating characteristic curve using time-dependent receiver operating characteristic analysis. The Kaplan-Meier method and log-rank test were applied to assess prognostic impact, and a multifactorial Cox proportional hazards model was employed for analysis. Additionally, a new model, PSIRI, was developed and assessed for its survival prediction capability. RESULTS The optimal cutoff values for PNI and SIRI were determined to be 47.80 and 1.38, respectively. Based on these values, patients were categorized into high PNI and low PNI groups, as well as high SIRI and low SIRI groups. Significant differences in age, T stage, neutrophil to lymphocyte ratio (NLR), monocyte to lymphocyte ratio (MLR), and platelet-to-lymphocyte ratio (PLR) subgroups were observed between the PNI groups in the baseline profile. In the SIRI group, notable differences were found in gender, T stage, nerve invasion, intravascular tumor emboli, NLR, MLR, and PLR subgroups. Both low PNI and high SIRI were identified as independent risk factors for poor prognosis in CRC patients. When combined into the PSIRI model, it was shown that patients with a PSIRI ≤ 1 had a higher risk of death compared to those with a PSIRI of 2. CONCLUSION We assessed the impact of PNI and SIRI on the prognostic survival of CRC patients and developed a new model, PSIRI. This model demonstrated superior predictive accuracy, with a concordance index of 0.767.