Introduction

The impact of cancer-related inflammation pathway as a proxy of aggressiveness is well known in several cancers. In the field of renal cell carcinoma (RCC), it has been already proposed the effect of pre-treatment neutrophil-lymphocyte ratio (NLR), lymphocyte-monocyte ratio (LMR) and platelet-lymphocyte ratio (PLR) on survival after nephrectomy. No data is available as regards the potential impact of platelet to haemoglobin ratio (PHR). We aimed to systematically test and compare the prognostic effect of all those preoperative haematological parameters on long-term survival in renal cancer patients.

Methods

We evaluated 870 patients diagnosed with cM0 RCC and treated with partial or radical nephrectomy at a single institution. Routine laboratory measurements were performed preoperatively the day before surgery. Univariable and multivariable Cox proportional hazards regression models were used to predict cancer specific mortality (CSM) and overall mortality (OM). Covariates included age at surgery, gender, pathological T and N stage, Fuhrman grade, lymph vascular invasion, tumor size, Charlson Comorbidity Index, presence of symptoms and tumour necrosis.

Results

Median age was 63 years (range 21-89 years). Overall, preoperative median NLR, LMR, PLR and PHR were respectively 2.23 (range 0.3 - 75.2), 3.5 (range 0.2 - 24), 121 (range 16.26 - 946) and 15.88 (range 1.79 - 79.21). Median follow-up was 72 months (range 10 - 90). At univariate analysis NLR (HR 1.30, p=0.01, AUC 66%), and PHR (HR 1.07, p<0.001, AUC 77%) were associated with higher risk of CSM; conversely, LMR (HR 0.30, p<0.001, AUC 62%) and PLR (HR 0.92, p<0.001, AUC 60%) were inversely associated with CSM. Considering OM, NLR (HR 2.06, p<0.001, AUC 60%) and PHR (HR 1.07, p<0.001, AUC 66%) were directly associated; conversely, LMR (HR 0.38, p<0.001, AUC 63%) was inversely associated with OM. No correlation was found between PLR and OM. At multivariate analyses, NLR (HR 1.59, p<0.001), PHR (HR 1.1, p<0.001) and LMR (HR 0.25, p<0.001) resulted independent predictors of OM. With respect to CSM, NLR (HR 1.77, p<0.001), PHR (HR 1.11, p<0.001) LMR (HR 0.15, p<0.001) and PLR (HR 0.31, p<0.001) resulted independent predictors.

Conclusions

We validated the role of cancer-related inflammation pathway in predicting CSM and OM according to different haematological parameters. Based on previous investigation supporting the prognostic role of anaemia and thrombocytosis, we tested the effect of PHR that emerged the most accurate predictor of CSM.

Funding

None