A survival nomogram involving nutritional-inflammatory indicators for cervical cancer patients receiving adjuvant radiotherapy.

ABSTRACT

Objective: The combined impact of nutritional and inflammatory status on survival of cervical cancer patients remained unclear. This study aimed to construct a survival nomogram involving both nutritional and inflammatory indicators and evaluate their potential correlation. Methods: This retrospective study included 325 cervical cancer patients who received adjuvant radiotherapy between September 2010 and September 2020. Baseline nutritional indicators such as body mass index (BMI), controlling nutritional status (CONUT) and serum albumin were assessed. Inflammatory indicators of platelet/lymphocyte ratio (PLR), neutrophil/lymphocyte ratio (NLR), systemic immune inflammation index (SII) and system inflammation response index (SIRI) were evaluated respectively. The LASSO regression and Cox regression models were applied for variable selection and nomogram building. The predictive accuracy and superiority of prognostic model were assessed by area under curve (AUC), C-index, decision curve analysis (DCA), integrated discrimination improvement (IDI) and net reclassification improvement (NRI). Results: Patients with high inflammatory indicators (PLR, NLR and SII) and poor nutritional status (CONUT scores > 2) suffered poorer prognosis compared to these with well nutritional status and lower inflammation levels. Our study unveiled a positive correlation between malnutrition and hyperinflammation. Even after accounting for baseline inflammatory level, malnutrition remained a significant risk factor for patients. Notably, the inflammatory level and nutritional status were further modulated by the clinical features of patients. Patients with poorer nutritional status exhibited higher levels of PLR, NLR, SII and SIRI, particularly for those in advanced clinical stages and with non-squamous cell carcinoma. In addition, our study found elevated levels of circulating basophil and serum carbohydrate antigen 125 (CA125) were associated with the poor prognosis. The prognostic nomogram which incorporated the nutritional-inflammatory indicators of PLR and CONUT showed a favorable performance with the AUC value of 0.76 at 5-year survival prediction. The DCA, IDI and NRI consistently demonstrated the favorable superiority of the model. Moreover, the nomogram-based risk stratification system could effectively classify patients into three mortality risks subgroups. Conclusions: Poorer nutritional and high inflammatory status collectively contributed to the poorer prognosis. The prognostic nomogram which incorporated nutritional-inflammatory indicators significantly improved the prediction of long-term outcomes of cervical cancer patients undergoing adjuvant radiotherapy.