Parametrial involvement in cervical cancer is associated with systemic inflammatory indices: a retrospective...

The aim of this study is to evaluate the association between parametrial invasion (PI) and systemic inflammatory indices in patients with cervical cancer, exploring whether routinely available hematological markers reflect local tumor spread.

A retrospective observational study was conducted including patients with cervical cancer treated with surgery and/or radiotherapy and/or chemotherapy between July 2023 and December 2025. Patients were stratified according to the presence or absence of PI, as determined by histopathological or radiological assessment. Systemic inflammatory indices derived from pre-treatment blood tests included neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), platelet-to-lymphocyte ratio (PLR), systemic inflammatory response index (SIR), and systemic inflammation response index (SIRI). Comparisons between groups were performed using non-parametric tests. Receiver operating characteristic (ROC) curves were used to evaluate the standalone discriminative ability of each inflammatory index. Logistic regression models were applied to assess associations with PI. To reduce information loss and limit model instability in this exploratory cohort, inflammatory indices were analyzed as continuous variables. Adjusted analyses were performed using separate clinically predefined models, each including one inflammatory index together with depth of stromal infiltration and fornix involvement.

A total of 79 patients were included in the analysis. PLR, MLR, NLR, and SIRI were significantly higher in patients with PI compared with those without PI, whereas SIR did not differ significantly. In analyses stratified by the extent of PI, significant differences were observed for MLR, NLR, and SIRI. In unadjusted logistic regression models, PLR, MLR, NLR, and SIRI were associated with PI. After adjustment for depth of stromal infiltration and fornix involvement in separate clinically predefined models, only MLR remained significantly associated with PI (adjusted OR per 0.1-unit increase: 1.986; 95% CI: 1.095-3.603; p = 0.024). ROC analysis showed modest standalone discriminative ability, with all AUC slightly below 0.75.

Parametrial invasion in cervical cancer is associated with measurable alterations in systemic inflammatory indices, reflecting the host response to locoregional tumor spread. Among the evaluated indices, MLR retained an association with PI after adjustment for clinically relevant local spread parameters. However, the modest standalone discriminative performance and exploratory nature of the analysis preclude the use of these indices as independent diagnostic tools. Further prospective validation is required before any clinical application.