Brain metastases are the most common intracranial tumors in adult accounting for about 20-40 percent (1). Lung cancer is the most common primary malignant tumor that results in the brain metastases, with non-small cell lung cancer (NSCLC) accounting for more than 60 percent of the lung tumors (1, 2). The prognoses of patients with brain metastases arising from NSCLC varies greatly with the median survival time ranging from 6.9 to 46.8 months (3). Several high-technique models, such as diagnosis-specific graded prognostic assessment (DS-GPA), Graded Prognostic Assessment for Lung Cancer Using Molecular Markers (Lung-molGPA), score index for radiosurgery (SIR), and basic score for brain metastases (BSBM), are used to evaluate the prognosis of NSCLC patients (3–6), but the techniques are ineffective in evaluating the prognosis of NSCLC patients with brain oligo-metastasis treated with fractionated stereotactic radiotherapy (FSRT). FSRT has a higher local control rate and fewer side effects than stereotactic radiosurgery (SRS) therapy and has thus been widely used in the clinic. Compared with SRS therapy, FSRT showed a different biological effectiveness. For example, hypoxic tumor cells may survive after SRS, but FSRT, which is based on the principle of reoxidation, has better control rate in tumor (7), suggesting that the SIR model might not apply to the FSRT patients. There is need to identify prognostic factors for NSCLC patients with brain oligo-metastasis receiving FSRT therapy. Current studies on prognostic factors for oligo-metastasis patients treated with FSRT focus on the size of metastatic tumor and radiation dose. However, there is still no standard evaluation method for tumor size, radiation dose and fractionation scheme, with different studies suggesting different radiotherapy biologically effective dose (BED) (8–10). Therefore, there is need to develop and enhance predictive indexes indicating the efficacy and survival of NSCLC patients with brain oligo-metastasis receiving FSRT.

Since the discovery of the relationship between inflammation and cancer in 1863 by R Virchow (11), inflammatory indicators such as neutrophil-lymphocyte ratio (NLR), platelet to lymphocyte ratio (PLR), lymphocyte to monocyte ratio (LMR), regulatory T cells and peripheral memory CD4+ T cell, have been used in predicting efficacy and survival in different kinds of cancer (12–16). Inflammatory indicators have also been associated with prognosis of surgery, chemotherapy, targeted therapy, immunotherapy (17–20), and curative effect of radiotherapy (21, 22). In our previous study, we demonstrated that NLR, PLR and LMR constitute a simple and effective prediction index, in locally advanced esophageal cancer treated with surgery, and radiotherapy and in NSCLC treated with anti-vascular targeted therapy (23, 24). Several studies have also demonstrated the predictive value of some inflammatory indicators in NSCLC patients with brain metastasis receiving SRS therapy (25, 26). Although FSRT is similar to radiosurgery, the predictive value of inflammatory indicators in NSCLC patients with brain oligo-metastasis treated with FSRT and their effect on radiotherapy dose is unknown.

This study is a retrospective analysis of the association between inflammatory indicators (NLR, PLR and LMR) and the local control rate and survival of NSCLC patients with brain oligo-metastasis treated with FSRT, and their effect on radiotherapy dose.

To our knowledge, this is the first study to evaluate the predictive value of inflammatory indicators in NSCLC patients with brain oligo-metastasis treated with FSRT. Our results showed that, in addition to tumor diameter and BED 10, LMR (an inflammatory indicator), was also an independent prognostic factor for i-LCS. Tumor diameter and BED10 had previously been identified as prognostic factors for i-LCS (9, 30, 31), but our study is the first to identify LMR value as an independent prognostic factor. Anna Cho et al. (25) reported that NLR, PLR and LMR were independent prognostic factors of overall survival in NSCLC patients with brain metastases after Gamma Knife Radiosurgery. However, Aijie Li et al. (32) proposed that LMR was the only independent prognostic factor of overall survival in NSCLC patients with brain metastases. The differences in these finding could be due to differences in the study population. The participants in our study were all NSCLC patients with oligometastatic brain metastases who had undergone FSRT treatment. We found that LMR as the only inflammatory indicator that acted as an independent prognostic factor of overall survival, although PLR was identified as prognostic factor during univariate analysis but was negative in multivariable analysis. Therefore LMR ≥ 4 is an excellent independent prognostic factor for i-LCS and overall survival in NSCLC patients with oligometastatic brain metastases treated with FSRT. We also found that positivity for driver gene mutations was an independent prognostic indicator of overall survival, which was consistent with the findings of Aijie Li et al. (32). The reason why driver gene mutations were independent prognostic indicators of overall survival but not i-LCS may be attributed to the overall poor extracranial control rate in patients negative for driver gene mutations, which leads to the death from extracranial lesions.

It is generally acknowledged that the cut-off value of LMR is 4 (25, 26), which was confirmed in our study. There is no consensus on the cut-off value of tumor diameter. Some articles indicate that the cut-off value of tumor diameter is 1cm in brain metastases receiving radiosurgery (33), and 2cm or 3cm in brain metastases receiving FSRT (34, 35). The i-LCS was significantly better in patients with lower cut-off value. Our results showed the cut-off value was 2cm demarcated by the median value. The median i-LCS was just 13.7 months in patients with tumors above 2cm but was as long as 31.5 months in patients with tumors below 2 treated with FSRT. The effective radiation dose for clinical application is still controversial. It is generally agreed that the radiation dose cut-off value for SRS treatment in brain metastases is 18Gy, with local control deteriorating significantly in patients receiving the dose below 18Gy (36). The use of BED is determining the curative effect of SRS is debatable (37), but is usually applied in determining the curative effect of FSRT treatment. Several divide-up radiotherapy plans have been suggested (8–10). One review summarized and compared the curative effect of different BED values, and concluded that BED12 values greater than 40Gy (which equals to BED10 greater than 48Gy) achieved a higher local control rate (38). Another study showed the 1 year local control rate was 100% for BED10 greater than 48Gy but was only 33% for BED10 less that 48Gy in treating postoperative metastasis tumor bed using FSRT (39). Thus it is generally believed the cut-off value of BED10 is 48Gy. In our study, we found that the median i-LCS of patients who received BED10 greater than 48Gy was significantly longer than for patient who received less BED10 (17.0 months vs 5.5 months). Results from multivariate analysis indicated the BED10 was an independent prognostic factor of i-LCS. Samuel R et al. (34) assessed if enhancement of BED value improved i-LCS, and found that enhancing BED10 value did not improve i-LCS in patients with a tumor diameter of more than 3cm. However, we found that administration of BED 10 greater than 53.6Gy improved the i-LCS of patients with two independent risk factors (tumor diameter ≥ 2 and LMR < 4), but had no benefit in the patients with 1 or no independent risk factor. The difference in findings may be because the study by Samuel R et al. enrolled patients with only 1 independent risk factor (tumor diameter ≥3cm) and did not screen the patients for obstinate resistance to radiotherapy. Unlike their results, this study analyzed the LMR value below 4 as the other independent risk factor. Thus got benefit by improving BED value in patients simultaneously possessed two independent risk factors (tumor diameter ≥ 2 and LMR < 4) equivalent to possessing obstinate resistance to radiotherapy.

Neutrophils inhibit immune functions and induce resistance to chemoradiotherapy by secreting cytokines and chemokines (13, 40, 41). Platelets are a critical source of cytokines, such as transforming growth factor-β, platelet-derived growth factor, and vascular endothelial growth factor (VEGF), which induce angiogenesis and cell invasion (42, 43). Moreover, lymphocytes can produce several cytokines, including IFN-γ and perforin, to prevent tumor development and induce apoptosis in cancer cells (36). Monocytes are innate immune cells that play important roles in tumor progression, invasion and metastasis and can be grouped into macrophages and myeloid-derived suppressor cells (44, 45). These findings from literature suggest that NLR, PLR and LMR have potential roles as prognostic factors in tumor development and treatment. In our study, we found that LMR but not NLR was a prognostic factor in the NSCLC patients with oligometastatic brain metastases receiving FSRT treatment. The difference between the two factors could be due to the effect of glucocorticoids. Glucocorticoids are usually administered to reduce the intracranial pressure before radiation therapy once the diagnosis of brain metastasis has been confirmed. The use of glucocorticoids can affect neutrophil counts which affects NLR value resulting in a negative result. It is also possible that NLR has no prognostic value in patients receiving FSRT treatment.

There are some limitations in our study. First, this was a retrospective study involving a small sample from a single center, which may have caused analytical bias. Second, the lack of follow-up data for different treatments before and after the radiotherapy may have influenced the analysis. Third, adverse reactions such as acute cerebral edema and radionecrosis are difficult to detect and were not reported during follow-up, thus adverse reactions analysis could not be carried out. Fourthly, selection bias may have been present despite the strict inclusion criteria, and thus the findings need to be validated in future prospective studies. Fifth, a part of patients from our study had been pronounced dead because of the extracranial lesions before the intracranial lesions, and this point was linked to the end of intracranial follow-up which resulted in attrition bias. Therefore, there is need for multi-center prospective randomized clinical trials with large sample size to validate the prognostic value of LMR in NSCLC patients with oligometastatic brain metastases treated with FSRT and its effect on radiotherapy dose.

LMR is a prognostic factor for i-LCS and OS in NSCLC patients with oligometastatic brain metastases treated with FSRT. The basic dose for BED10 was greater that 48Gy, but should be increased to greater than 53.6Gy in patients with tumor diameter ≥ 2 and LMR < 4.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Ningbo Medical Center Lihuili Hospital ethics committee. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements. Written informed consent was not obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

TC and WS contributed to the conception and design of the study. TC and MT wrote the article together. SL, YL, and JW contributed to the acquisition and analysis of the data. YZ, ZW, and HW participated in revising of the article. All authors contributed to the article and approved the submitted version.