2123P - Value of immuno-nutritional scores in the prognosis of hospitalised patients with lung cancer

Background

Immuno-nutritional scores like neutrophil-lymphocyte ratio (NLR), prognostic nutritional index (PNI), modified Glasgow prognostic score (mGPS) and controlling nutritional status (CONUT) have been developed to reflect the chronic inflammatory state and malnutrition induced by cancer, and to predict clinical outcomes. This study aims to investigate the prognostic value of these scores in hospitalized patients with lung cancer as we know that unscheduled admission in patients with cancer may imply organ fragility.

Methods

In this retrospective study, we examined patients with lung cancer and unscheduled hospitalizations that took place between June 2021 and May 2022 within the Medical Oncology Service of La Paz University Hospital in Madrid, Spain.

The optimal cutoff for continuous variables was evaluated with Youden index. Survival since admission was estimated with the Kaplan-Meier method. Univariate and multivariate analyses were performed by Cox regression model to determine independent prognostic factors. Areas under the curve (AUCs) or ROCs were calculated to evaluate the discriminatory ability.

Results

We admitted 158 patients with a median age of 68 years, majority of male (66.5%) and stage IV predominance (84.8%). The median follow-up was 14.9 months. The median overall survival since admission was 3.3 months (95%CI: 1.86-7.74). The optimal cut-off value for NLR was calculated to be 6 (Youden index=0.2) and for PNI was 40 (Youden index=0.1). In univariate analysis, poorer survival was associated with NLR>6 (1.87 vs 7.40 months, p=0.006, HR 1.67, 95%CI 1.16-2.42) and PNI<40 (1.67 vs 4.97 months, p=0.044, HR 1.49, 95%CI 1.01-2.20) and was not associated with CONUT (p=0.70) or mGPS (p=0.067). In the multivariate analysis only NLR>6 was an independent prognostic factor (p=0.042, HR 1.61, 95%CI 1.02-2.57). The ROC curve to evaluate the discriminatory ability of 30-day mortality was only significant in the case of NLR>6 (p=0.035, AUC 0.599, 95%CI 0.51-0.69).

Conclusions

In our setting, NLR was the most robust score for estimating prognosis in hospitalized patients with lung cancer.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.