"Never doubt that a small group of thoughtful, committed citizens can change the world. Indeed, it is the only thing that ever has."

Margaret Mead

Original article
peer-reviewed

Prognostic Impact of Baseline Neutrophil-to-Eosinophil Ratio in Patients With Metastatic Renal Cell Carcinoma Treated With Nivolumab Therapy in Second or Later Lines



Abstract

Background

Inflammation is a crucial component in carcinogenesis. The neutrophil-to-eosinophil ratio (NER) has been studied as a biomarker of prognosis and predictive of response in metastatic renal cell carcinoma (mRCC). In the present study, we evaluated the relevance of baseline NER on the progression-free survival (PFS) and overall survival (OS) outcomes in real-world patients with mRCC treated with nivolumab in second or subsequent lines. We also assessed the association of baseline NER with objective response, as well as with toxicity and histology.

Methods

In this multicenter retrospective analysis of patients with mRCC treated with nivolumab, the last systemic absolute neutrophil and eosinophil count before treatment with nivolumab was used to calculate the NER. An additive Cox proportional hazards model was used to identify the cut-off point for NER considering PFS and the patients were allocated into low and high NER groups. Median OS and median PFS were estimated using the Kaplan-Meier estimator, and survival curves of groups were compared using the log-rank test. Univariable and multivariable Cox regression models were used to study OS and PFS and Fisher’s exact test was performed to evaluate the association of NER with the response, toxicity, and histology.

Results

The 49 analyzed patients had a median follow-up of nine months. The NER cut-off was established at 48, locating 29 patients in the low NER group (NER < 48) and 20 in the high NER group (NER ≥ 48). Median PFS and median OS were significantly shorter in patients with high NER versus low NER (3 vs. 30 months (p < 0.001) and 6 vs. 24 months (p = 0.002), respectively). Multivariable analyses showed that NER (HR 3.92 (95% CI: 1.66-9.23), p = 0.002) was an independent factor for PFS and that NER (HR 3.85 (95% CI: 1.33-11.17), p = 0.013) and progressive disease (HR 5.62 (95% CI: 1.88-16.83), p = 0.002) were independent factors for OS. NER was significantly associated with objective response rate (ORR) (NER ≥ 48-12.5% vs. NER < 48-87.5%, p = 0.003), immune-related adverse events (irAEs) (NER ≥ 48-10.0% vs. NER < 48-42.9%, p = 0.014), and tumor’s histology as patients of high NER group had more non-clear cell carcinoma than low NER group (35.0% vs. 7.4%, p = 0.017).

Conclusion

Our real-world data analysis of NER in patients with mRCC confirmed the prognostic value of this biomarker, supporting clinical utility in predicting survival. Results also suggested an association between lower NER and better ORR, and that irAEs occur more frequently in patients with a lower NER. However, further large-scale prospective studies are needed to confirm these findings and to validate this biomarker.

Introduction

Renal cell carcinoma (RCC) accounts for 3%-5% of all malignancies, representing the seventh most common cancer in men, and the 10th most common cancer in women [1]. In 2020, around 431,000 new cases have been diagnosed around the world [2]. There is an increasing incidence, especially in developed countries. Given the usual absence of symptoms, RCC is diagnosed at a metastatic stage in 30-50% of patients, with a poor prognosis due to estimated five-year survival of 10% [2].

RCC is a chemo-resistant disease and was also considered radiotherapy-resistant before the emergence of stereotactic radiation therapy [3,4]. Treatment for a long time was exclusively surgical. It was only later on that immunotherapy or targeted therapies showed survival benefits, the latter with a good toxicity profile [3,4]. A variety of targeted therapies, such as vascular endothelial growth factor (VEGF) monoclonal antibodies, tyrosine kinase inhibitors (TKIs), and mammalian target of rapamycin pathway inhibitors (mTORis), have been approved as systemic therapy in metastatic renal cell carcinoma (mRCC) [5]. Currently, immunotherapy has gained an important place in the treatment of this disease.

Immune checkpoint inhibitors (ICIs) have changed the treatment strategy for mRCC [6]. Nivolumab is an anti-programmed cell death protein 1 (PD-1) antibody, which is the standard second-line treatment in mRCC, after previous VEGF receptor TKI (VEGFR-TKI), and was approved based on the CheckMate 025 study, a phase III randomized clinical trial, which demonstrated improved overall survival (OS) compared to everolimus, regardless of programmed death-ligand 1 (PD-L1) expression with good toxicity profile [7]. The survival benefit was independent of PD-L1 expression [8]. The CheckMate 214 trial has demonstrated that the combination of nivolumab plus ipilimumab as first-line treatment in mRCC is associated with better OS and objective response rate (ORR), compared to sunitinib, among patients with intermediate and poor prognosis, according to the International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) risk category [9]. Other studies have evaluated combinations of immunotherapy and TKIs as the first-line of treatment in mRCC, namely, the combination of nivolumab plus cabozantinib in CheckMate 9ER or pembrolizumab plus axitinib in KEYNOTE-426 [10,11]. Both clinical trials highlighted improvements in progression-free survival (PFS) and OS for the combinations [10,11]. These associations have now earned a place in standard first-line treatment in mRCC [12].

Consistent and long-lasting responses are seen in a group of patients treated with nivolumab. However, for another group of patients, the benefit of this therapy is limited (20-35%) [13]. The IMDC risk categories were developed using data from patients who received targeted therapy [14]. Thus, novel prognostic biomarkers or models are required in the ICIs era and it is essential to define biomarkers predicting the clinical outcome of ICIs treatment that can be routinely used in clinical practice [15].

Inflammation impacts each step of carcinogenesis, including tumor initiation, promotion, and metastatic progression, and ongoing studies are being carried out to find and select predictive markers, such as gene expression signatures, tumor mutational burden, or tumor-infiltrating lymphocytes [16-19]. However, serum markers are more practical and accessible biomarkers in routine clinical practice; many biomarkers of inflammation, including C-reactive protein (CRP) levels, neutrophil-to-lymphocyte ratio (NLR), eosinophil, and platelet count, have been investigated as prognostic factors [5,16,20].

A retrospective study demonstrated that an increase in eosinophils and relative eosinophil change at six weeks of nivolumab was associated with a good response to immunotherapy [5]. The baseline neutrophil-to-eosinophil ratio (NER) has been reported to be associated with outcomes of immuno-oncology based on combination treatment in mRCC, and a post hoc analysis of a phase III randomized control study, JAVELIN Renal 101, showed that a lower NER was associated with a better ORR and PFS with the combination of avelumab (anti-PD-L1) and axitinib in mRCC [21]. The fact that the early eosinophilia may lead to a better response to immunotherapy can be explained by the results presented by Cheng et al., who showed that the increase in eosinophils was involved with the recruitment of cluster of differentiation 8 (CD8) + T lymphocytes to the tumor microenvironment (TME) and that eosinophils contribute to the cytotoxic antitumor response [22].

The aim of the present study was to evaluate the relevance of baseline NER on the PFS and OS outcomes in real-world patients with mRCC treated with nivolumab in second or subsequent lines. Secondary purposes were to study the association of baseline NER to objective response through imaging exams, according to the Response Evaluation Criteria in Solid Tumors (RECIST) 1.1, as well as with toxicities and histology.

Materials & Methods

Patients

This multicenter retrospective cohort study considered patients with mRCC who were treated with nivolumab as second or subsequent lines at the medical oncology units of Centro Hospitalar Universitário Lisboa Central, Centro Hospitalar de Lisboa Ocidental, Centro Hospitalar Barreiro Montijo, and Hospital Beatriz Ângelo in Portugal, between June 2017 and April 2021. Patients who received immunotherapy as the first line and with other concomitant primary neoplasia(s) were excluded. Baseline patient and tumor characteristics, evidence of metastasis at diagnosis, history of surgical resection of metastases or primary tumor, first-line treatment, and response were all retrospectively collected from the hospital’s electronic database and medical records. The last follow-up data were collected as of 28 November 2021. The treatment response was evaluated by computed tomography (CT) and/or bone scintigraphy at least once every 12 weeks and classified according to RECIST 1.1. The ORR was defined as the percentage of patients with confirmed complete or partial responses, and clinical benefit was defined as the percentage of patients with confirmed complete or partial responses or stable disease among all treated patients. The study protocol was approved by the Research Ethics Committee of all the institutions and applied in conformity with the Declaration of Helsinki. A formal informed consent waiver was accepted due to the retrospective observational nature of this study.

Blood sample analysis

Values were obtained from clinical practice blood samples analysis. Peripheral blood parameters were collected during the week before the first administration of nivolumab as second or subsequent lines. The baseline NER was calculated from the counts obtained using that sample, dividing the absolute neutrophil count by the absolute eosinophil count.

Statistics

Descriptive statistics were performed with categorical variables being described as frequencies (percentages), and the remaining variables by the median and range (minimum and maximum). An additive Cox proportional hazards model was used to identify the cut-off point for NER considering PFS. Based on this cut-off value, patients were allocated into two groups. The OS and PFS times were calculated from the start of the treatment with nivolumab to the death from any cause, or to the progression of disease or death, respectively. Median OS (mOS) and median PFS (mPFS) were estimated using the Kaplan-Meier estimator, and survival curves of groups were compared using the log-rank test. Univariable and multivariable Cox regression models were used to study OS and PFS. For the multivariable models, all variables that in the univariable analysis attained a p-value ≤ 0.25 were selected. Adjusted hazard ratios (HRs) were estimated with corresponding 95% confidence intervals (95% CI). The proportional hazards assumption of Cox regression was tested with a formal significance test based on standardized Schoenfeld residuals. Fisher’s exact test was performed to evaluate the association of the baseline binary NER with the response, toxicity, and histology. A level of significance α = 0.05 was considered. Data analysis was performed using SPSS version 25.0 (IBM Corp., Armonk, NY) and R (R Foundation for Statistical Computing, Vienna, Austria).

Results

Patients characteristics

The study included 49 patients with mRCC treated between June 2017 and April 2021 in the four institutions included in this study. Median follow-up was nine months (range: 1-57 months). Table 1 details patient demographics and baseline characteristics. Briefly, the median age at diagnosis of metastatic disease in this cohort was 61 years (range: 28-85 years) and 42 patients (85.7%) were male. Thirty-eight patients (77.6%) were diagnosed with a clear cell histotype. Based on the IMDC risk categories [23], nine (18.4%), 34 (69.4%), and six (12.2%) patients were categorized as favorable, intermediate, and poor risk, respectively. Nivolumab was administered in the third or subsequent lines in eight patients (16.3%), and all previously targeted therapies in first-line and most (n = 6) in second-line therapies were TKIs. The median NER was 33 with a range of 3-803. The NER cut-off in our cohort was 48 for the time to disease progression in patients with nivolumab therapy (sensitivity, 53.8%; specificity, 71.4%; Supplemental Figure). The NER was 48 or higher in 20 patients (40.8%, high-risk group) and it was less than 48 in 29 patients (59.2%, low-risk group).

Characteristic All (n = 49) (100%)
Age (years)  
Median 61
Range 28-85
Sex, n (%)  
Men 42 (85.7)
Women 7 (14.3)
PS ECOG, n (%)  
0 31 (63.3)
1 17 (34.7)
2 1 (2.0)
Previous nephrectomy, n (%)  
Yes 39 (79.6)
No 10 (20.4)
Histopathology, n (%)  
Clear cell carcinoma 38 (77.6)
Non-clear cell carcinoma 9 (18.4)
Missing 2 (4.1)
Fuhrman's grade, n (%)  
1, 2 20 (40.8)
3 10 (20.4)
4 5 (10.2)
Missing 14 (28.6)
IMDC risk, n (%)  
Favorable 9 (18.4)
Intermediate 34 (69.4)
Poor 6 (12.2)
Number of previously targeted therapies, n (%)  
1 42 (85.7)
≥2 7 (14.3)
Previously targeted therapies, n (%)  
TKIs 49 (100)
Sunitinib 29 (59.2)
Pazopanib 20 (40.8)
Axitinib 7 (14.3)
mTORis 2 (4.1)
Everolimus 2 (4.1)
Number of metastatic organs, n (%)  
1 25 (51.0)
≥2 24 (49.0)
NER  
<48 29 (59.2)
≥48 20 (40.8)
Occurrence of irAE, n (%)  
Yes 14 (28.6)
No 35 (71.4)

Treatment response and survival outcomes

The response assessment to nivolumab was available in 48 of 49 patients because one patient died before the response assessment from a cause unrelated to the mRCC. Of these, one patient achieved a complete response, and 15 achieved a partial response. The ORR was 32.7% and the clinical benefit was 42.9% (Table 2). Twenty-seven patients (55.1%) experienced disease progression during treatment with nivolumab. The median PFS of nivolumab was eight months (95% CI: 0.58-15.42 months), and the median OS was 13.0 months (95% CI: 7.11-18.89 months) during the median nine months of follow up in this study (Figures 1A, 1B).

Treatment response n (%)
Objective response 16 (32.7)
CR 1 (2.0)
PR 15 (30.6)
SD 5 (10.2)
PD  27 (55.1)
Unable to determine 1 (2.0)

Survival outcomes according to baseline NER and risk factors for survival

During the follow-up period, 27 (55.1%) and 26 (53.1%) patients had disease progression and died from any cause, respectively. PFS and OS were compared according to baseline NER. Median PFS was significantly shorter in patients with high NER (NER ≥ 48) than in those with low NER (<48) (30.0 months (95% CI not available)) vs. (3.0 months (95% CI: 1.75-4.25), p < 0.001) (Figure 2A). In the low-risk group for OS (NER < 48), mOS was 24.0 months (95% CI: 4.5-43.5 months) and in the high-risk group (NER≥48), it was six months (95% CI: 3.8-8.2 months, p = 0.002). Therefore, mOS was also significantly shorter in patients with high NER (Figure 2B).

Univariable analysis showed that histopathology, NER, and occurrence of immune-related adverse events (irAEs) were significant factors for PFS (Table 3). Results of the multivariable analysis showed that only NER remained in the final model (HR 3.92 (95% CI: 1.66-9.23), p = 0.002).

Variable Univariable analysis HR (95% CI) P-value
Age (years)    
≥65 vs. <65 0.98 (0.96-1.01) 0.307
Sex    
Male vs. female 0.95 (0.28-3.20) 0.937
PS ECOG    
0, 1 vs. 2 0.61 (0.27-1.40) 0.243
Previous nephrectomy    
Yes vs. no 0.49 (0.20-1.19) 0.113
Histopathology    
Non-clear cell vs. clear cell carcinoma 3.28 (1.36-7.89) 0.008
Fuhrman's grade    
1, 2 vs. 3, 4 1.45 (0.54-3.87) 0.457
IMDC risk    
Poor vs. favorable/intermediate 3.65 (0.86-15.58) 0.080
Number of previous therapies    
1 vs. ≥2 0.64 (0.19-2.14) 0.471
Number of metastatic organs    
≥2 vs. 1 1.09 (0.50-2.34) 0.833
NER    
≥48 vs. <48 3.92 (1.66-9.23) 0.002
Occurrence of irAE    
Yes vs. no 0.33 (0.12-0.91) 0.033
Delay nivolumab    
No vs. yes 0.78 (0.10-5.82) 0.807
Stop nivolumab    
No vs. yes 0.43 (0.10-1.85) 0.259

Univariable analysis showed that previous nephrectomy, irAE, NER, and progressive disease were significant factors for OS (Table 4). Multivariable analysis showed that NER (HR 3.85 (95% CI: 1.33-11.17), p = 0.013) and progressive disease (HR 5.62 (95% CI: 1.88-16.83), p = 0.002) were independent factors for OS. Those patients with NER ≥ 48 had approximately a four-fold higher risk of dying than those with NER < 48 (p = 0.013). Those patients whose disease progressed had approximately a six-fold higher risk of dying than those without disease progression (p = 0.002).

Variable Univariable analysis HR (95% CI) P-value Multivariable analysis HR (95% CI) P-value
Age (years)        
≥65 vs. <65 1.01 (0.98-1.05) 0.555 - -
Sex        
Male vs. female 1.18 (0.35-3.96) 0.787 - -
PS ECOG        
0, 1 vs. 2 1.31 (0.59-2.89) 0.510 - -
Previous nephrectomy        
Yes vs. no 0.36 (0.15-0.85) 0.019 - -
Histopathology        
Non-clear cell vs. clear cell carcinoma 1.55 (0.49-4.87) 0.453 - -
Fuhrman's grade        
1, 2 vs. 3, 4 1.07 (0.34-3.35) 0.904 - -
IMDC risk        
Poor vs. favorable/intermediate 1.81 (0.40-8.24) 0.442 - -
Number of previous therapies        
1 vs. ≥2 1.45 (0.57-3.68) 0.436 - -
Number of metastatic organs        
≥2 vs. 1 1.64 (0.74-3.63) 0.225 - -
NER        
≥48 vs. <48 4.02 (1.49-10.86) 0.006 3.85 (1.33-11.17) 0.013
Disease progression        
Yes vs. no 6.66 (2.26-19.64) 0.001 5.62 (1.88-16.83) 0.002
Occurrence of irAE        
Yes vs. no 0.26 (0.09-0.79) 0.018 - -
Delay nivolumab        
No vs. yes 0.55 (0.07-4.55) 0.576 - -
Stop nivolumab        
No vs. yes 0.39 (0.09-1.71) 0.215 - -

Objective response rate according to baseline NER

An association between baseline NER and ORR was evaluated. Table 5 shows that NER was significantly associated with ORR (NER ≥ 48 (12.5%) vs. NER < 48 (87.5%), p = 0.003).

Variable CR PR ORR P-value
NER       0.003
≥48 0 (0.0%) 2 (13.3%) 2 (12.5%)  
<48 1 (100.0%) 13 (86.7%) 14 (87.5%)  

Immune-related adverse events according to baseline NER

An association between irAE and NER baseline was evaluated, which revealed that NER was significantly associated with irAE, and patients with NER ≥ 48 had fewer irAE than patients with NER < 48 (10.0% vs. 42.9%, p = 0.014) (Figure 3).

Histopathology according to baseline NER

Association between histopathology and baseline NER was also evaluated, which revealed that NER was significantly associated with tumor histology and patients with NER ≥ 48 had more non-clear cell carcinoma than patients with NER < 48 (35.0% vs. 7.4%, p = 0.017) (Figure 4).

Discussion

In recent decades, evidence has been presented establishing the association between inflammation and cancer development and several studies have suggested that inflammatory factors may be useful to predict survival in mRCC [5,13,19]. In this context, many inflammatory factors were investigated including the baseline NER associated with the results of immuno-oncology-based combination treatment in mRCC. A post hoc analysis of a phase III randomized control study, JAVELIN Renal 101, demonstrated that a lower NER was associated with a better ORR and PFS with the combination of avelumab and axitinib [21]. In addition to the association of NER and immuno-oncology-based combination treatment, this study also evaluated the possible value of this biomarker in patients treated with TKI monotherapy as the first line, revealing no significant difference for PFS in this subset of patients, but improvement in OS with lower NER. So, similarly to what has been documented in the NLR studies [13,24,25], NER may have a prognostic value in mRCC, regardless of the treatment type. Nevertheless, it seems to have an additional predictive value for patients treated with immunotherapy [21]. More recently, this was supported by the retrospective study in which Tucker et al. concluded that lower baseline NER was associated with better clinical outcomes (PFS, OS, and ORR) in patients with mRCC treated with nivolumab plus ipilimumab in the first line [26].

Therefore, our current retrospective analysis performed in patients with mRCC treated with nivolumab monotherapy in the second or later line represents an assessment of baseline NER as a possible prognostic biomarker. Our study provides real-world evidence for the impact of baseline NER when estimated prior to treatment with nivolumab in second or subsequent lines and demonstrated that a lower baseline NER was associated with an improvement in OS and PFS in these patients. This study also suggests an association between lower NER and better ORR. This is in agreement with previous studies [21,26]. Patients with a baseline NER ≥ 48 were considered high-risk groups, revealing a lower PFS and OS than the low group with a baseline NER < 48.

Two studies associated NER with survival and ORR outcomes, placing patients into low- and high-risk groups, using as cut-off values higher or lower than the median NER calculated for the sample around 26-29 [21,26], different from the cut-off obtained in our study. It applied an additive Cox proportional hazards model to identify the cut-off point for NER considering PFS. In multivariable Cox regression, baseline NER was considered an independent factor for PFS and OS in patients with mRCC treated with nivolumab in second or later lines. These results obtained from real-world data suggest that NER may be a potential prognostic factor for survival in patients with mRCC treated with nivolumab and support the hypothesis that this biomarker could be predictive of response to immunotherapy in patients with mRCC and of clinical utility. Looking at the literature, the hazard ratio associated with NER in our analysis is higher than in other published results [21,26]. Possible explanations for this discrepancy are the different statistical methods used to determine the NER cut-off and the fact that we have analyzed baseline NER pre-starting second or later lines with nivolumab. Our retrospective analysis also revealed a significant association between immune-related adverse events and NER baseline, as it showed that patients with higher NER tend to have less irAE than patients with lower NER. One retrospective study by Giorgione et al. published in February 2021 revealed that patients with mRCC who developed irAE had higher baseline eosinophil counts compared to those who did not, and this affects the degree of toxicity, which is higher for patients with higher baseline absolute eosinophils count [27]. Another retrospective study, which involved metastatic melanoma, mRCC, and metastatic non-small cell lung cancer receiving ICIs, showed that baseline eosinophil count may predict irAE and patients with baseline eosinophilia had more irAE [28]. This is related to the regulatory and effector role of eosinophils in multiple immune functions [27,28].

Another important finding from our retrospective analysis was the greatest tendency of patients with non-clear cell carcinoma to have higher NER. Considering the small number of patients with non-clear cell carcinoma histology, we cannot draw major conclusions from this result, so we believe it is pertinent to develop a study with a bigger sample and conduct a subgroup analysis of the different histology according to the NER and evaluating clinical outcomes.

This study has several limitations, including the fact that it is a retrospective study with a relatively short period of time of patient’s inclusion and that both neutrophils and eosinophils can be affected by a variety of situations including infection and medications, which makes the study even more limited to adequately explain the possible influences of these external factors. Another noteworthy limitation of this retrospective analysis is the fact that we did not perform a multivariable analysis to determine the predictive value of NER for the outcome ORR. We believe that the exploration of irAEs stratified by NER baseline in a larger sample and with subgroup analyses for the different types and degrees of toxicities is a hypothesis that should be researched in the future. This could add value to the role of the baseline NER as a predictor of irAE and understand the eventual association with better outcomes in the group of patients with lower baseline NER and who had more irAEs. Furthermore, the characterization of metastasis load, metastasis sites, and types of metastasis was not considered in our study. We recognize that the lack of this information may have led to biased results and should be considered in subsequent studies. Additionally, only about 18% of the patients in this study had non-clear cell histology, so we were limited in how the analysis of histology subgroups for PFS, OS, and ORR stratified by NER could result in an overestimation of the effect in such a small group. Combinations with immunotherapy are now preferred in the first line, therefore, we anticipate more real-world studies to confirm the predictive role of NER of important oncological outcomes also in the first line.

Conclusions

The present real-world analysis using a multi-institutional cohort with mRCC treated with nivolumab in second or later lines confirmed the prognostic value of this biomarker, supporting its clinical utility in predicting survival. Results also suggested an association between lower NER and better ORR, and that immune-related adverse events occur more frequently in patients with a lower NER. Because this biomarker can be easily assessed and monitored in routine clinical practice, its use can contribute to effective treatment and follow-up of patients. However, further large-scale prospective studies are needed to confirm these findings and to validate this biomarker.


References

  1. Escudier B, Porta C, Schmidinger M, et al.: Renal cell carcinoma: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2019, 30:706-20. 10.1093/annonc/mdz056
  2. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F: Global Cancer Statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021, 71:209-49. 10.3322/caac.21660
  3. De Meerleer G, Khoo V, Escudier B, et al.: Radiotherapy for renal-cell carcinoma. Lancet Oncol. 2014, 15:e170-7. 10.1016/S1470-2045(13)70569-2
  4. Miccio JA, Oladeru OT, Jun Ma S, Johung KL: Radiation therapy for patients with advanced renal cell carcinoma. Urol Clin North Am. 2020, 47:399-411. 10.1016/j.ucl.2020.04.011
  5. Herrmann T, Ginzac A, Molnar I, Bailly S, Durando X, Mahammedi H: Eosinophil counts as a relevant prognostic marker for response to nivolumab in the management of renal cell carcinoma: a retrospective study. Cancer Med. 2021, 10:6705-13. 10.1002/cam4.4208
  6. Suzuki K, Terakawa T, Furukawa J, Harada K, Hinata N, Nakano Y, Fujisawa M: C-reactive protein and the neutrophil-to-lymphocyte ratio are prognostic biomarkers in metastatic renal cell carcinoma patients treated with nivolumab. Int J Clin Oncol. 2020, 25:135-44. 10.1007/s10147-019-01528-5
  7. Escudier B, Sharma P, McDermott DF, et al.: CheckMate 025 randomized phase 3 study: outcomes by key baseline factors and prior therapy for nivolumab versus everolimus in advanced renal cell carcinoma. Eur Urol. 2017, 72:962-71. 10.1016/j.eururo.2017.02.010
  8. Bedke J, Stenzl A: Re: nivolumab versus everolimus in advanced renal-cell carcinoma. Eur Urol. 2016, 69:538-9. 10.1016/j.eururo.2015.12.025
  9. Motzer RJ, Tannir NM, McDermott DF, et al.: Nivolumab plus ipilimumab versus sunitinib in advanced renal-cell carcinoma. N Engl J Med. 2018, 378:1277-90. 10.1056/NEJMoa1712126
  10. Powles T, Plimack ER, Soulières D, et al.: Pembrolizumab plus axitinib versus sunitinib monotherapy as first-line treatment of advanced renal cell carcinoma (KEYNOTE-426): extended follow-up from a randomised, open-label, phase 3 trial. Lancet Oncol. 2020, 21:1563-73. 10.1016/S1470-2045(20)30436-8
  11. Choueiri TK, Powles T, Burotto M, et al.: Nivolumab plus cabozantinib versus sunitinib for advanced renal-cell carcinoma. N Engl J Med. 2021, 384:829-41. 10.1056/NEJMoa2026982
  12. eUpdate - renal cell carcinoma treatment recommendations. (2021). https://www.esmo.org/guidelines/genitourinary-cancers/renal-cell-carcinoma/eupdate-renal-cell-carcinoma-treatment-rec....
  13. Ishihara H, Tachibana H, Takagi T, et al.: Predictive impact of peripheral blood markers and C-reactive protein in nivolumab therapy for metastatic renal cell carcinoma. Target Oncol. 2019, 14:453-63. 10.1007/s11523-019-00660-6
  14. Heng DY, Xie W, Regan MM, et al.: External validation and comparison with other models of the International Metastatic Renal-Cell Carcinoma Database Consortium prognostic model: a population-based study. Lancet Oncol. 2013, 14:141-8. 10.1016/S1470-2045(12)70559-4
  15. Nishiyama N, Hirobe M, Kikushima T, et al.: 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. BMC Urol. 2020, 20:110. 10.1186/s12894-020-00679-2
  16. Gooden MJ, de Bock GH, Leffers N, Daemen T, Nijman HW: The prognostic influence of tumour-infiltrating lymphocytes in cancer: a systematic review with meta-analysis. Br J Cancer. 2011, 105:93-103. 10.1038/bjc.2011.189
  17. Simonaggio A, Epaillard N, Pobel C, Moreira M, Oudard S, Vano YA: Tumor microenvironment features as predictive biomarkers of response to immune checkpoint inhibitors (ICI) in metastatic clear cell renal cell carcinoma (mccRCC). Cancers (Basel). 2021, 13:231. 10.3390/cancers13020231
  18. Miao D, Margolis CA, Gao W, et al.: Genomic correlates of response to immune checkpoint therapies in clear cell renal cell carcinoma. Science. 2018, 359:801-6. 10.1126/science.aan5951
  19. Misra S, Hascall VC, Markwald RR, O’Brien PE, Ghatak S: Inflammation and cancer. Wound Healing: Stem Cells Repair and Restorations, Basic and Clinical Aspects. Turksen K (ed): Wiley, Hoboken, New Jersey; 2018. 239-74.
  20. Ravindranathan D, Master VA, Bilen MA: Inflammatory markers in cancer immunotherapy. Biology (Basel). 2021, 10:325. 10.3390/biology10040325
  21. Tucker MD, Voss MH, Choueiri TK, et al.: Association between neutrophil-to-eosinophil ratio (NER) and efficacy outcomes in the JAVELIN Renal 101 study. Clin Oncol. 2021, 39:4549. 10.1200/JCO.2021.39.15_suppl.4549
  22. Cheng JN, Luo W, Sun C, et al.: Radiation-induced eosinophils improve cytotoxic T lymphocyte recruitment and response to immunotherapy. Sci Adv. 2021, 7:eabc7609. 10.1126/sciadv.abc7609
  23. Ko JJ, Xie W, Kroeger N, et al.: The international metastatic renal cell carcinoma database consortium model as a prognostic tool in patients with metastatic renal cell carcinoma previously treated with first-line targeted therapy: a population-based study. Lancet Oncol. 2015, 16:293-300. 10.1016/S1470-2045(14)71222-7
  24. Na N, Yao J, Cheng C, Huang Z, Hong L, Li H, Qiu J: Meta-analysis of the efficacy of the pretreatment neutrophil-to-lymphocyte ratio as a predictor of prognosis in renal carcinoma patients receiving tyrosine kinase inhibitors. Oncotarget. 2016, 7:44039-46. 10.18632/oncotarget.9836
  25. Santoni M, De Giorgi U, Iacovelli R, et al.: Pre-treatment neutrophil-to-lymphocyte ratio may be associated with the outcome in patients treated with everolimus for metastatic renal cell carcinoma. Br J Cancer. 2013, 109:1755-9. 10.1038/bjc.2013.522
  26. Tucker MD, Brown LC, Chen YW, et al.: Association of baseline neutrophil-to-eosinophil ratio with response to nivolumab plus ipilimumab in patients with metastatic renal cell carcinoma. Biomark Res. 2021, 9:80. 10.1186/s40364-021-00334-4
  27. Giorgione R, Pillozzi S, Mela MM, et al.: Absolute eosinophil count as predictive biomarker of irAEs in patients with metastatic renal cell carcinoma (mRCC) treated with immune checkpoint inhibitors (ICIs). J Clin Oncol. 2021, 39:345. 10.1200/JCO.2021.39.6_suppl.345
  28. Giommoni E, Giorgione R, Paderi A, et al.: Eosinophil count as predictive biomarker of immune-related adverse events (irAEs) in immune checkpoint inhibitors (ICIs) therapies in oncological patients. Immuno. 2021, 1:253-63. 10.3390/immuno1030017

Appendices


Original article
peer-reviewed

Prognostic Impact of Baseline Neutrophil-to-Eosinophil Ratio in Patients With Metastatic Renal Cell Carcinoma Treated With Nivolumab Therapy in Second or Later Lines


Author Information

Lucia Gil Corresponding Author

Medical Oncology, Centro Hospitalar Universitário Lisboa Central, Lisboa, PRT

Fátima R. Alves

Medical Oncology, Centro Hospitalar de Lisboa Ocidental, Lisboa, PRT

Diana Silva

Medical Oncology, Hospital Beatriz Ângelo, Loures, PRT

Isabel Fernandes

Medical Oncology, Centro Hospitalar Barreiro Montijo, Barreiro, PRT

Mário Fontes-Sousa

Medical Oncology, Hospital CUF Tejo, Lisboa, PRT

Marta Alves

Epidemiology and Statistics Unit, Research Centre, Centro Hospitalar Universitário Lisboa Central, Lisboa, PRT

Ana Papoila

Epidemiology and Statistics Unit, Research Centre, Centro Hospitalar Universitário Lisboa Central, Lisboa, PRT

Ricardo Da Luz

Medical Oncology, Centro Hospitalar Universitário Lisboa Central, Lisboa, PRT


Ethics Statement and Conflict of Interest Disclosures

Human subjects: Consent was obtained or waived by all participants in this study. Ethics Committee for Health - CHULC issued approval 1157/2021. The Ethics Committee for Health - CHULC admitted that there are no ethical limitations to carrying out the study in accordance with the Declaration of Helsinki, and a favorable opinion was issued. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.



Original article
peer-reviewed

Prognostic Impact of Baseline Neutrophil-to-Eosinophil Ratio in Patients With Metastatic Renal Cell Carcinoma Treated With Nivolumab Therapy in Second or Later Lines


Figures etc.

SIQ
-
RATED BY 0 READERS
CONTRIBUTE RATING

Scholarly Impact Quotient™ (SIQ™) is our unique post-publication peer review rating process. Learn more here.