The neutrophil-lymphocyte ratio has a role in predicting the effectiveness of nivolumab in Japanese patients with metastatic renal cell carcinoma: a multi-institutional retrospective study

Over the last decade, the treatment of metastatic renal cell carcinoma (mRCC) has dramatically changed. Vascular endothelial growth factor (VEGF) and mammalian target of rapamycin (mTOR)-targeting agents have superseded cytokine therapies. More recently, immune checkpoint inhibitors (ICIs) have emerged as a new treatment option for mRCC. Nivolumab, a programmed cell death protein-1 (PD-1) antibody, induces restoration of the anticancer T cell mediated immune response by blocking PD-1 [1]. In a global multicenter phase 3 trial (CheckMate 025), nivolumab treatment was associated with longer overall survival (OS) in mRCC patients who were previously treated with tyrosine kinase inhibitors (TKIs), in comparison to everolimus [2]. On the basis of these results, the major guidelines recommend nivolumab as a second-line treatment for mRCC patients [3, 4].

Although ICIs, including nivolumab, have resulted in changes to the treatment strategies for mRCC, there are unmet needs for biomarkers to predict the effectiveness of ICIs. The Memorial Sloan Kettering Cancer Center (MSKCC) risk category and International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) risk category are widely used for predicting the prognosis of patients [5, 6]. However, the MSKCC and IMDC risk categories were developed using data from patients who received cytokine therapy and targeted therapy, respectively. Thus, novel prognostic biomarkers or models are required in the ICI era.

Many studies have reported the role of the systemic inflammation response, including C-reactive protein (CRP), albumin, neutrophil, and related variables, for predicting the prognosis of patients with various cancers [7‐9], including mRCC [10]. Recently, the neutrophil-lymphocyte ratio (NLR) has been recognized as a predictive marker for both non-metastatic RCC [11‐13] and for mRCC treatment with TKI [14‐17]. Although some studies [18, 19] revealed that a lower NLR was a significant predictor of favorable progression-free survival (PFS) and OS in mRCC patients treated with ICIs, these studies analyzed data from various lines and ICIs. Thus, there are no robust data about the NLR as a prognostic factor from studies focused on sequential treatment with nivolumab. The aim of this study was to evaluate whether the NLR is a predictor of poor oncological outcomes in mRCC patients undergoing sequential treatment with nivolumab.

The patients’ characteristics are shown in Table 1. Of the 52 patients, 18 (35%) and 34 (65%) received one and two or more prior therapeutic treatments, respectively. Eight, 37 and 7 patients had a favorable, intermediate and poor risk IMDC classification, respectively. The median duration of nivolumab therapy was 7.1 months (IQR: 2.9–24.4). The best responses during nivolumab therapy were a complete response in 2 (4%), partial response in 11 (21%), stable disease in 22 (42%) and progressive disease (PD) in 17 (33%) patients. The objective response rate was 25%.

The median NLR at baseline was 3.7 (IQR: 2.7–5.1). At baseline, 32 (61%) had an NLR of ≥3 and 20 (39%) patients had an NLR of < 3. The median NLR at 4 weeks after the initiation of nivolumab treatment was 3.3 (IQR: 2.4–5.7). At 4 weeks, 31 (59%) patients had an NLR of ≥3 and 20 (39%) patients had an NLR of < 3. The NLR at 8 weeks after the initiation of nivolumab treatment could not be analyzed in 11 (21%) of the patients because of discontinuation due to progression disease or adverse events. The median NLR at 8 weeks was 3.2 (IQR: 2.3–5.7). The NLR at baseline and that at 4 weeks did not differ to a statistically significant extent (P = 0.351). The NLR at 4 weeks and at 8 weeks did not differ to a significant extent (P = 0.688). In non-responders (defined as a best response of PD), the NLR at 4 weeks tended to be higher than that at baseline; however, the difference did not reach statistical significance (Fig. 1). Supplementary Table 1 shows predictive factors for patients with NLR of < 3 and NLR of ≥3 at baseline and at 4 weeks after the initiation of nivolumab treatment. There were no significant difference between the proportion of predictive factors and NLR.

This study showed that the NLRs of non-responders at 4 weeks after the initiation of nivolumab tended to be higher than at baseline. We also found that an NLR of ≥3 at 4 weeks was significantly associated with poor PFS and OS. These results suggested that a decline in the NLR may predict a favorable clinical outcome in mRCC patients treated with nivolumab as a sequential therapy.

Cancer-related inflammation is well recognized to be associated with cancer development and a poor prognosis. Tumor-associated neutrophils (TANs) are key regulators of cancer-related inflammation. TANs have been shown to induce genetic instability through the release of reactive oxygen species (ROS) and to produce tumor necrosis factor, IL-1, IL-6 and VEGF, which contribute to tumor proliferation and immune escape [21, 22]. TANs also activate and form neutrophil extracellular traps (NETs), which play a pro-tumor role in tumor progression. TANs are an important factor in the tumor microenvironment, and an increased number of TANs might play a pivotal role in treatment resistance [23].

Boissier et al. performed a meta-analysis to evaluate the prognostic role of the NLR in RCC and demonstrated that a high NLR was associated with a poor prognosis. In patients with both mRCC and localized RCC, an NLR of < 3 predicted better OS and PFS [14]. In non-metastatic clear cell RCC, recurrence-free survival in patients with an NLR of ≥2.7 was significantly shorter than that in patients with an NLR of < 2.7 [11]. Another study on mRCC showed that an NLR of ≥4.0 was an independent predictor of OS [24]. Kobayashi et al. reported that patients with an NLR of < 3.32 had longer PFS in comparison to those with an NLR of ≥3.32 after 1st-line targeted therapy [22].

Several reports have evaluated the role of NLR in the clinical outcomes of mRCC patients were treated with ICI. Jeyakumar et al. showed that an NLR of ≥3 was a predictor of a poor response, PFS and OS in patients with mRCC who were treated with nivolumab, ipilimumab, avelumab, pembrolizumab, or atezolizumab [18]. Lalani et al. reported that a high NLR at baseline and at 6 weeks after the initiation of therapy were associated with short PFS, short OS and a poor response [19]. In our study, the NLR at baseline was not a significant predictor of the response to nivolumab. This may be because all of the patients had received two or more TKI treatments and 63% of them had received three or more such therapies before the initiation of nivolumab treatment. Kobayashi et al. also reported that the NLR significantly decreased, even after treatment with TKIs, especially after sunitinib [22]. However, the NLR before nivolumab treatment was not a significant factor in this study. These results suggested that a decline in the NLR after ICI treatment—rather than TKI treatment—is useful information for predicting a response to nivolumab treatment.

Numerous studies have investigated different cutoff values for the NLR. In mRCC patients, most studies have used a cutoff value of around 3 [15‐19, 22, 24‐26]. We adopted NLR ≥3 as a significant predictor of poor PFS and OS in patients undergoing sequential treatment with nivolumab, based on the results of the meta-analysis by Hu et al. [15, 18].

The present study is associated with some limitations. First, this was a multi-institutional retrospective study. The patients’ characteristics and treatment strategies might have differed among the institutions. The second limitation was the small number of patients included in this study. Despite these limitations, we believe that these results are clinically informative and will be useful for predicting the clinical outcomes of patients receiving nivolumab as a sequential treatment. However, a prospective study using a larger cohort is needed to confirm this.

Although the NLR at baseline was not associated with PFS or OS, an NLR of ≥3 at 4 weeks after the initiation of nivolumab therapy could predict poor PFS and OS. However, a larger prospective study is needed to validate these results.