Neutrophil to Lymphocite Ratio Predicts Overall Survival in Newly Diagnosed Hodgkin Lymphoma Patients-Single Centre Experience

This retrospective analysis included 83 newly diagnosed patients with HL followed up from 2001-2015. ROC curve analysis was used to determine the optimal cut-off values for overall survival. Univariate and multivariate analysis was performed using the Cox proportional hazards regression to identify significant independent prognostic factors.

clinical risk factors to predict the risk for disease progression, relapse and death of patients with HL. The International Prognostic Score (IPS) uses seven prognostic factors to predict clinical outcomes in newly diagnosed HL patients [2]. Adverse prognostic factors in classic HL are age >45 years, stage IV disease, hemoglobin < 10 g/l, lymphocyte count < 0.6 × 10 9 /L, male sex, albumin < 40 g/L, white blood count > 15 × 10 9 /l. However, the IPS does not offer risk stratification for HL patients with limited disease (i.e., stages I and IIA, without constitutional symptoms and no bulky disease [≥ 10 cm in diameter]).
Inflammation has been identified to be a critical component of tumor progression, highlighting the role of the microenvironment, which is largely orchestrated by inflammatory cells as an indispensable participant in the neoplastic process, fostering proliferation, survival and migration [3]. For different solid tumors, as well as lymphomas, inflammation parameters, including leukocytes, neutrophils, lymphocytes and C-reactive protein have been associated with higher mortality rates [4][5][6]. Neutrophil to lymphocyte ratio (NLR) may reflect and clarify the immune response in systemic inflammatory response. NLR has been identified as an independent prognostic factor for overall survival (OS) and progression free survival (PFS) in various types of cancer, including colorectal cancer [7], pancreatic cancer [8], renal cell,sarcoma [9] carcinoma [10], and in diffuse large B cell lymphoma (DLBCL) [6,11]. We studied whether NLR at diagnosis can predict treatment response and overall survival in patients with classic HL. We then correlated the results of this analysis with other well known risk stratification systems.

Materials and Methods
This retrospective study included 83 consecutive patients diagnosed with HL at Medical Center "Bezanijska kosa", Belgrade, Serbia from 2001-2015. All patients met the following criteria: Pathohistologicaly confirmed HL without previous treatment, no previous history of malignancy, negativity for HIV, HBV, HCV or TBC infection with availability of laboratory and follow up data. Clinical data were retrieved from medical records. Treatment options and response criteria were based on standard guidelines. The ANC and ALC were obtained from the routine complete blood count (CBC) with 4-part differential counts (lymphocytes, monocytes, eosinophils, and neutrophils) using ABX Pentra DX-DF 120-HORIBA and NLR was calculated for each patient at diagnosis. The study was conducted according to the ethical guidelines following the Declaration of Helsinki. The institutional review committee approved our study protocol thereby following local biomedical research regulations. Statistical analysis Categorical variables were compared by using χ 2 test (2-sided Pearson or linear-By-linear association). Continuous variables were compared with the Mann-Whitney test.
The determination of optimal cut-off value for the ANC and ALC ratio in predicting overall survival was performed using the Receiver Operating Curve (ROC) method with determination of sensitivity and specificity of all cut-off values. OS was defined as time between the first day of diagnosis and the date of death from any cause or the last follow-up. Spearman's rank correlation coefficient was used to describe the correlation between quantitative variables. The Kaplan-Meier method was used to estimate overall survival (OS). The log-rank test was used to compare the survival distributions. A multivariate analysis (Cox regression method) was performed to examine the effect of presumed prognostic factors on the overall survival. All statistical analyses were performed using the IBM SPSS version 22.0 statistical software program (SPSS, Chicago, IL). The results were considered to be statistically significant when the p value was less than 0.05

Determining the cut-off value of the NLR
An ROC curve of the NLR (ANC/ALC ratio) according to

Prognostic significance of the NLR (ANC/ALC ratio)
There was significant difference in OS between patients with NLR ≥ 4.3 vs. patients with NLR ≤ 4.3 (p < 0.018) ( Figure 2). Univariate analysis revealed that OS significantly correlated with older age (> 45 years), lymphopenia (< 600/µl vs. ≥ 600/µl), NLR ratio (continuous variable), NLR ≥ 4.3, male sex, advanced clinical stage, poor risk according IPS, GHSG and EORTC staging, and lower than CR as treatment response (Table 2). Multivariate analysis showed that independent overall survival was generated to determine the cut-off value. The area under the curve was recorded as 0.655 ± 0.064 (95% CI, 0.536-0.724) (Figure 1). The NLR value of 4.3 corresponded to the maximum combined sensitivity and specificity on the ROC curve (66% sensitivity and 59% specificity   showed that NLR can reliable identify patients who will have a shorter survival, alone or in respect to other well known staging systems. Second, most of the studies on the significance of NLR in HL reported so far are multi centric studies [12][13][14]. Our study is a single centre study with limitations of retrospective nature of the study designed and relatively small number of patients, but nevertheless informative enough and conclusive. In addition, it seems that NLR improve estimation of prognosis in early (un) favorable stage of the disease according to GHSG and EORTIC staging systems.
Although HL is considered a highly curable disease, 20% of patients cannot be cured with standard first-line chemotherapy and have a dismal outcome. Since current clinical parameters do not allow precise risk stratification prognostic variables for OS were high IPS (≥ 4) (p = 0.013), NLR ≥ 4.3 (p = 0.007) and age ≥ 45 (p = 0.001) ( Table 3). Patients with NLR ≤ 4.3 in comparison to patients with NLR ≥ 4.3, had longer survival considering any IPS (Figure 3 and Figure 4), but high NLR was particularly poor risk factor in patients with IPS 4 (log rank = 0.004) (Figure 3 and Figure 4). Taking in account that IPS was generated for patients with advanced stage HL, we future explored the role of NLR according to both, GSHG and EORTC stage of HL [2] (Figure 4 and Figure 5). Results clearly showed that patients with NLR ≥ 4.3 in comparison to patients with NLR ≤ 4.3 had poorer survival in early (un) favorable in comparison to advanced stage disease.

Discussion
This study has few important aspects. First we clearly  Pathologically, HL is characterized by the presence of a small number of diagnostic Reed-Sternberg cells in a background of bystander reactive cells composed of lymphocytes, neutrophils, macrophages, eosinophils, and plasma cells. Lymphopenia is related to adverse survival outcoma [17], as well as tumor-infiltrating lymphocytes and tumor-infiltrating macrophages that constitute tumor microenvironment [12,18,19]. Tumor-infiltrating lymphocytes and tumor-infiltrating macrophages are reported to be prognostic factors for survival in patients with HL [16]. The tumor microenvironment and, in particular, there is a continuing need for identification of new prognostic factors that may be useful for risk stratification and predicting the response to treatment [15]. To address this issue and better stratify the risk, tests for various prognostic factors have been suggested, including gene expression profiling [16], immunohistochemically analysis of biomarkers, and positron emission tomography [17]. But, not all countries have resources to implement these advanced and sophisticated methods. Current gold standard for risk stratification in HL word wide is the IPS, but this prognostic score is generated for advanced HL.
In addition, as for any mathematical cutoff value applied to biological phenomenon, one has to be cautions when dealing with "borderline cases". For example, hypothetical patients with HL and WBC of 14.9 × 10 9 /l and lymphocyte count of 0.61 × 10 9 /l accounts for zero points in IPS system, regardless of his obviously very high NLR. In addition, NLR should be used for more precise estimation of prognosis in the inflammatory response play an important role in cancer development and progression and may be associated with systemic inflammation. NLR can be used as a marker of systemic inflammation since it has long been known that the baseline raised white blood cell count (≥ 15 × 10 9 /L) and lower lymphocyte count has a prognostic role in classic Hodgkin lymphoma.
The accumulation of neutrophils is related to increased cytokine levels, especially IL-8 [12,20] and it has been suggested that a high number of neutrophils may actually promote tumor growth and metastasis and/or inhibit lymphocyte activity, thereby counteracting the antitumor immune response [13,21]. These observations suggest that an imbalance in the ratio of neutrophil to lymphocyte in the peripheral blood of patients with cancer may be related to tumor development. However, the number of studies related to the clinical significance of NLR, especially on the longterm results of hematological malignancies are quite a few [14], when compared to those carried out with solid tumors [22]. Our single center study clearly showed that high NLR is associated with shorter overall survival in patients with newly diagnosed HL. The quick, accessible, inexpensive, easily applicable parameter is strong indicator of shorter survival in patients with advanced disease. However, comprehensive prospective studies in larger cohorts of HL patients are needed to evaluate prognostic significance of NLR in early stage disease. The authors declare that they have no conflict of interest.