Glioma is one of the most prevalent types of malignant brain tumor, and approximately 20,000 new gliomas were diagnosed in the US in 2015 alone (McNeill, 2016[11]). Glioma is characterized by high invasiveness and results in disappointing survival outcomes. Glioblastoma, in particular, is associated with median survival of one year and five-year mortality > 95 %, despite significant progress in neurosurgery, chemotherapy, and radiotherapy for treating and managing glioma (Zeng et al., 2015[20]; Bush et al., 2017[4]). The prognosis of Chinese patients with glioma is similarly poor (Yang et al., 2013[18]). Thus, simple and reliable prognostic markers are needed in order to predict clinical outcomes of patients with glioma and tailor their therapy accordingly.

As in other malignant cancers, inflammation may contribute to glioma progression (Michelson et al., 2016[12]; Conti et al., 2010[5]). Biomarkers based on inflammation, such as the classical example of C-reactive protein, have been used to assess prognosis of glioma patients (Strojnik et al., 2014[14]). Neutrophil-to-lymphocyte ratio (NLR), a novel marker in many systemic inflammatory disorders (Faria et al., 2016[8]), reflects immune response arising from various stress stimuli. It has shown prognostic potential in several cancers, including lung, breast, and kidney cancer (Templeton et al., 2014[16]). Another inflammation biomarker, the platelet-to-lymphocyte ratio (PLR), has shown prognostic potential for patients with solid tumors (Templeton et al., 2014[15]). The present study sought to investigate whether pre-operative NLR and PLR are significantly associated with survival in Chinese patients with glioma. These results may help provide an evidence-based approach to stratifying patients by poor survival risk, thereby guiding the tailoring of treatment.

Accurate prognostic evaluation of patients with glioma plays a critical role in their management, but it remains a clinical challenge. Here we provide evidence that, at least among Chinese patients with glioma, NLR ≥ 4.00 or PLR ≥ 200.00 may be associated with worse OS, and NLR may be an independent prognostic biomarker. This study is, to our knowledge, the first to assess the prognostic value of NLR and PLR in glioma in Chinese patients.

Abnormal inflammation is a characteristic of malignant cancers and a driver of malignant transformation of low-grade gliomas. Inflammation-induced activation of transcription factors contributes to the survival and rapid growth of glioma cells (Michelson et al., 2016[12]; Sen, 2011[13]). This systemic inflammation leads to a simultaneous increase in neutrophil count and decrease in lymphocyte count, which may reflect increased lymphocyte margination and apoptosis. The net result is inhibition of the cytolytic activity of immune cells and immunosuppression (el-Hag and Clark, 1987[7]). Increased NLR has been described as a novel inflammation biomarker with prognostic potential in several cancers (Templeton et al., 2014[16]). Consistent with this, NLR in our patients increased with WHO grade, suggesting that preoperative NLR may be useful for predicting glial tumor grade. Another study found that NLR correlates with glial brain tumor grade, and that NLR > 2.579 may be a discriminative parameter for predicting glioma grade (Zadora et al., 2015[19]). Our results show that Chinese patients with NLR ≥ 4.00 had significantly shorter OS based on Kaplan-Meier OS analysis, and multivariate Cox regression identified NLR as an independent prognostic factor in glioma. These results are similar to another study that found NLR > 4 to be an independent prognostic marker for poor outcome in patients with glioblastoma multiforme (Bambury et al., 2013[3]). However, our results contrast with a study in which multivariate Cox regression did not identify NLR as a prognostic biomarker in patients with glioma (Auezova et al., 2016[1]). More studies should be performed to validate our findings and assess the role of NLR in predicting the treatment response of patients with glioma, which would help guide their effective management and follow-up.

Platelet-mediated inflammation contributes to cancer progression, so researchers have tested various platelet-based markers of inflammation in several cancers (Bambace and Holmes, 2011[2]). In glioblastoma patients, platelets induce angiogenesis of tumor endothelial cells and secretion of vascular endothelial growth factor, helping promote glioma progression. Platelet-derived growth factor contributes to oncogenic transformation and tumorigenesis (Di Vito et al., 2017[6]; Westermark et al., 1995[17]). Consistent with these studies on platelet-based indices in cancer, we found in the present study that Chinese glioma patients with PLR ≥ 200 had significantly shorter OS. However, Cox regression failed to identify PLR as an independent prognostic biomarker for glioma (HR 1.030, 95 % CI 0.559 to 1.896). This negative finding may reflect the influence of confounding factors, which should be explored in larger, preferably prospective studies aimed at clarifying the potential prognostic relevance of PLR in glioma.

Our data strongly suggest exploring NLR as a simple basis for assessing prognosis of patients with glioma. NLR can be calculated from inexpensive, routine blood analysis conducted on all patients. At the same time, studies should examine the reliability of NLR calculated from blood tests in the presence of potential interference from non-steroidal anti-inflammatory drugs, taken by patients with glioma to relieve headache (Imtiaz et al., 2012[9]). Future studies should also examine the possibility of monitoring NLR changes during glioma treatment in order to detect or predict early recurrence after surgery.

There were several limitations in our study. First of all, we included only 112 patients, so our results require verification in larger studies. Second, we included only Chinese patients at a single center, so multi-site studies with patients of various ethnicities should be performed. Third, we did not have access to complete post-surgical treatment histories in this retrospective study, so other factors that we did not control may have contributed to the observed OS. To avoid these limitations, we suggest large, multi-site prospective studies to comprehensively evaluate the prognostic value of NLR and PLR in glioma.

This study found that among Chinese glioma patients, high NLR and PLR were associated with significantly worse OS, and that NLR may be an independent prognostic biomarker. Our study warrants larger trials involving multiple medical centers and patients of various ethnic groups in order to confirm the prognostic role of NLR in glioma.

Junli Wang, Wenjing Xiao and Wanyi Chen contributed equally as first authors.

This work was supported by the National Natural Science Foundation of China (Grant No. 81402507).