Baseline neutrophilia, derived neutrophil‐to‐lymphocyte ratio (dNLR), platelet‐to‐lymphocyte ratio (PLR), and outcome in non small cell lung cancer (NSCLC) treated with Nivolumab or Docetaxel

Nivolumab is a novel therapeutic option in NSCLC, associated with a significant survival gain compared with Docetaxel. However, predictive biomarkers are lacking. The presence of systemic inflammation has been correlated with poor outcome in many cancer types. We aimed to evaluate whether there is a correlation between some indicators of inflammation and response to Nivolumab or Docetaxel in pre‐treated NSCLCs. Data of 62 patients receiving Nivolumab or Docetaxel were analyzed. Baseline neutrophilia and thrombocytosis were not associated with response. High dNLR was associated with no response to Nivolumab, but not with Docetaxel, whereas high PLR correlated with low treatment response in both groups. Among refractory patients, a higher incidence of thrombocytosis, neutrophilia, high PLR, and high dNLR levels were observed compared with the overall population. This is one of the first reports in this field and suggests that indicators of inflammation might be included together with other predictive biomarkers in the baseline evaluation of patients candidate for immunotherapy.

2013; Hanahan & Weinberg. 2011). In addition, the presence of systemic inflammation correlates with poor outcome in many cancer types, including NSCLC. The immune cell composition of NSCLC is dominated by neutrophils (Kargl et al., 2017) and recently several authors correlated the presence of high neutrophil-to-lymphocyte ratio (NLR) and/or high absolute neutrophil count (ANC) with poor prognosis and lower response to conventional treatments in NSCLC (Carus et al., 2013;Derman et al., 2017;Kang et al., 2017;Scilla et al., 2017). Similar findings were reported for high platelet-to-lymphocyte ratio (PLR) (Gu et al., 2016;Sanchez-Salcedo et al., 2016).
We aimed to evaluate whether there is a correlation between some indicators of inflammation status, including neutrophilia, high NRL, and high PLR with response in pretreated NSCLC patients receiving Nivolumab or Docetaxel.

| MATERIALS AND METHODS
In this retrospective monocentric study twenty-eight consecutive patients with NSCLC receiving Nivolumab were included. Baseline white cell count (WBC) and absolute neutrophil count (ANC) were collected and correlated with tumor response. Thirty-four NSCLC patients treated with Docetaxel were used as controls. An ANC ≥7500 cell/μl was defined as neutrophilia. Derived neutrophil-to-lymphocyte ratio (dNLR) was calculated as: ANC/(WBC-ANC). Platelet-to-lymphocyte ratio (PLR) was defined as platelet count (PLT)/lymphocyte count. dNLR ≥3 and PLR ≥160 were defined high. PLT ≥450 × 103/µl was defined as thrombocytosis (Ferrucci et al., 2016;Kim et al., 2016). Categorical variables were compared using chi-square or Fisher's exact test. Overall survival (OS) was defined as time from Nivolumab or Docetaxel start to death and Progression Free Survival (PFS) as time from treatment start to Progression Disease (PD) or death for any cause. OS and PFS survival were estimated using the Kaplan-Meier method. Survival curves were compared using the log-rank test. To estimate the hazard ratio (HR), Cox regression analysis was used. Statistical analyses were performed with the R program version 3.3.2.

| RESULTS
Baseline characteristics of patients in our cohort are reassumed in with an equal distribution between the subgroups (p = 0.3572). Among non-squamous patients, 16.2% were EGFR mutated and 8.1% were KRAS-mutated, with an equal distribution in both treatment groups (p = 0.0649 for EGFR mutations and p = 0.9188 for KRAS mutations).
The majority of patients received Nivolumab or Docetaxel as second line therapy (69.3%), with no statistical differences between subgroups (p = 0.205).
ORRs according to baseline inflammation markers are reassumed in Table 2. Baseline neutrophilia was detected in 18% and 26% of patients in the Nivolumab and Docetaxel subgroups, respectively, and thrombocytosis was found in 3.5% and 3% of patients, respectively.
Baseline neutrophilia and thrombocytosis were associated with a significant reduction in the probability of response both in the global study population (p = 0.0003 and p = 0.0018, respectively) and in Nivolumab (p = 0.0001 and p = 0.0004, respectively) and Docetaxel (p = 0.0006 and p = 0.0042, respectively) subgroups. High dNLR was associated with lower ORRs with Nivolumab (p = 0.0001), whereas no statistically significant differences were observed in the total study population (p = 0.0787) and in patients treated with Docetaxel (p = 0.9319). Conversely, a high PLR predicted a poor response in the overall study population (p = 0.0407) and in the two treatment subgroups (p < 0.0004 and p = 0.0094 for Nivolumab and Docetaxel, respectively).
Patients with no baseline neutrophilia tended to have longer PFS than those with neutrophilia in the whole study population (4.0 vs.

| DISCUSSION
Immune checkpoint inhibitors are rapidly re-shaping the therapeutic landscape of NSCLC in both pre-treated and chemotherapy naïve  patients (Borghaei et al., 2015;Brahmer et al., 2015;Herbst et al., 2016;Reck et al., 2016;Rittmeyer et al., 2017) with unprecedented results in terms of overall survival. However, the advent of this innovative class of anticancer agents brings several novel challenges in clinical practice, such as new response criteria (Seymour et al., 2017), novel potential patterns of progression (Champiat et al., 2016), undeniable high costs (Matter-Walstra et al., 2016), and most importantly the need of predictive biomarkers for patients selection.
PD-L1 immunohistochemical expression is the best-studied predictive biomarker to anti-PD1/PD-L1 agents. However, the use of this biomarker for patient selection poses different biological and technical issues (Kerr et al., 2015;Patel & Kurzrock, 2015) and has produced contrasting results in clinical trials, with a significant proportion of patients responding to these agents even in absence of PD-L1 expression (Borghaei et al., 2015;Brahmer et al., 2015;Rittmeyer et al., 2017). Several different predictive biomarkers beyond PD-L1 IHC expression have been studied (Chen et al., 2016;Daud et al., 2016;Prat et al., 2017;Rizvi et al., 2015), but to date none has entered into clinics.
Moreover, in some instances the clinical implementation of these methodologies is difficult due to high costs and high level of expertise required.
Inflammatory reaction results in chronic illnesses, including cancer (Rajendran et al., 2018). Inflammation is a hallmark of cancer and is a known driver of cancer development and progression.

Different markers of inflammation have been investigated in NSCLC
and other solid tumors and correlated with poor outcome and low therapeutic response. dNLR and PLR, in addition to neutrophilia and thrombocytosis are the most well-studied markers of systemic Tumor microenvironment and immune system play a central role in response to immunomodulating agents (Nishino, Ramaiya, Hatabu, & Hodi, 2017) and, recently, it has been reported that neutrophils dominate the immune cell composition in NSCLC (Kargl et al., 2017). Here

| CONCLUSIONS AND FUTURE PERSPECTIVES
The results of this study suggest that indicators of inflammation such as baseline neutrophilia, thrombocytosis, and high dNLR are usually associated with a dismal prognosis and low efficacy of both chemotherapy and immunotherapy in NSCLC. Given their relative easy estimation, baseline evaluation of these indicators may be included together with other predictive biomarkers in the baseline evaluation of patients candidate for immunotherapy.

CONFLICTS OF INTEREST
No potential conflicts of interest declared.