ESTRO 2025 - Abstract Book

S1320

Clinical - Lung

ESTRO 2025

1321

Digital Poster NLR does not remain prognostic in sarcopenic lung and oesophageal cancer patients Alan McWilliam 1 , Azadeh Abravan 1 , Kathryn Banfill 2 , Ganesh Radhakrishna 2 , Donal McSweeney 1 1 Division of Cancer Science, University of Manchester, Manchester, United Kingdom. 2 Clinical Oncology, The Christie NHS Foundation Trust, Manchester, United Kingdom Purpose/Objective: Neutrophil-to-lymphocyte ratio (NLR) is an accepted pan-cancer prognostic biomarker. NLR is associated with an individual's inflammatory state, which has multiple associations including the patients burden of cancer and sarcopenic state (degenerative muscle loss). Here, we investigate the prognostic value of NLR in patients with lung and oesophageal cancer when accounting for sarcopenia. Material/Methods: The muscle compartment was automatically segmented on a single CT slice at mid-T12 for 388 lung and 283 oesophageal cancer patients. Segmentations were visually checked and skeletal muscle index (SMI) calculated by dividing cross-sectional muscle area by the individual’s height squared. Multivariable Cox models included the interaction of SMI and NLR and accounted for known prognostic factors common for both cancers and variables that impact muscle: age, sex, T-stage, N-stage, and tumour volume. Contrast plots allowed visualisation of the interaction term for patients with SMI in the lowest, central and highest quartiles. Patients were classified as sarcopenic if they were in the lowest quartile SMI and multivariable models for sarcopenic versus not sarcopenic groups run to identify if NLR remained prognostic. Results: SMI showed similar distribution between patient groups: mean 20.1cm 2 /m 2 (range 8.6–34.6) and 19.8cm 2 /m 2 (8.0– 37.9) for lung and oesophageal patients respectively. The multivariable model showed a significant interaction between SMI and NLR for lung cancer, p=0.007. The interaction term for oesophageal cancer was p=0.09. Tumour volume was significantly associated with survival for both cohorts (p<0.001). The contrast plot demonstrated for patients in the lowest quartile, (SMI<15cm 2 /m 2 ) NLR showed no prognostic value. While NLR was significant in the central and highest quartile in the lung cohort and the highest quartile alone in the oesophagus cohort (figure 1). The multivariable analysis for the sarcopenic groups (n=62 for lung and n=44 for oesophagus) showed NLR had no significant association with survival for either cohort. For the patients with higher SMI (none-sarcopenic) the significance remained, p<0.001. This is despite both the lung and oesophagus patient cohorts showing a higher NLR ratio in the sarcopenic groups: mean NLR 5.52 compared to 3.9 for the oesophagus cohort and 5.2 compared to 4.5 for the lung cohort.