Neutrophil-to lymphocyte ratio as a predictor of local recurrence of feline injection site sarcoma

Feline Injection Site Sarcoma (FISS) is a highly aggressive neoplasm and local recurrence (LR) represents a major concern. Although achieving histologically tumor-free surgical margins is crucial for long term control of the disease, LR has been reported in 19-40% of cases with clean margins.1 Thus, it is desirable to identify further variables that can aid in prognostication and treatment planning. Pretreatment Neutrophil-to-lymphocyte ratio (NLR) is a marker of systemic inflammatory response that has been reported as a prognostic tool for several tumors in humans, and for canine mastocytoma.2,3 The aim of this study is to explore the prognostic impact on LR of NLR in cats with surgically excised FISS. Eighty-two cats with surgically excised, histologically confirmed FISS at first presentation, without distant metastasis were retrospectively enrolled from Veterinary Teaching Hospitals of Milan and Turin. We included cats with available preoperative hematological analysis (within 45 days before surgery). Statistical analysis explored the relationship between NLR and: age, glycemia, tumor dimension, ulceration, concomitant diseases, necrosis, histotype. The impact of NLR on LR and survival time (ST) was then estimated with Cox regression model. Receiver operating characteristic curves (ROC) were made; the area under the curve (AUC) was calculated and Youden index was used to determine the best cut-off values. At the end of the study, 58 cats were death of which 24 for tumor related causes; of these cats, 23/24 experienced LR. Median ST was 975 days. The only variables that were significantly correlated with NLR were tumor size (p=0,004), and histotype (p=0,029), with cats with bigger tumors and histotype other than fibrosarcoma having a higher NLR. No correlation was found between NLR and potential causes of alterations in white blood cell populations, such as concomitant diseases, glycemia (as an indicator of stress), ulceration, necrosis, age. Higher NLR was significantly associated with higher risk of LR (p=0.015; HR=1.066) and shorter ST (p=0.02; HR=1.045). The optimal estimated cut-off for LC was 1.82 at 1 year (Se=95%; Sp=30%) and 3.65 at 2-3 years (Se=52% ; Sp=66%). Preoperative NLR is a readily available variable that may aid the clinician in identifying cats at higher risk of tumor recurrence and poorer outcome; this variable does not seem to be influenced by confounding factors that can alter white blood cell counts. The mechanism underlying this finding remains unclear, although in human and canine medicine it has been hypothesized that a predominant neutrophil response and relative lymphopenia may promote tumor growth and dissemination