2023 Volume 46 Issue 3 Pages 427-431
Studies have reported an association between elevated neutrophil-to-lymphocyte ratio (NLR) and poor prognosis in patients with melanoma treated with ipilimumab. However, it remains unclear whether NLR is useful in Japanese patients with melanoma, and if so, what is the optimal cut-off value. We retrospectively examined 38 patients who received ipilimumab from August 2015 to November 2021 at Nagoya University Hospital. We divided patients into two groups: 1–2 versus 3–4 cycles of ipilimumab. In univariate analysis, baseline neutrophil count and NLR were significantly higher in patients who discontinued ipilimumab within 2 cycles. With receiver operating characteristic analysis, the optimal NLR cut-off value was found to be 3.4 (area under the curve, 0.75; 95% confidence interval, 0.58–0.92). In multivariate logistic regression analysis, baseline NLR >3.4 was an independent risk factor for ipilimumab discontinuation (odds ratio, 15.6; 95% confidence interval, 3.0–82) that was significantly associated with shorter progression-free survival (PFS) (p = 0.003, log-rank test). In conclusion, NLR >3.4 is useful for selecting Japanese patients with melanoma who might have better PFS with ipilimumab-containing treatment. Because the optimal NLR cut-off value in this study was lower than values in American and European studies, it possibly differs by race. Hence, it should be extrapolated to Japanese patients with caution.
In recent years, clinical outcomes of patients with advanced melanoma have improved significantly with immune checkpoint inhibitor (ICI) therapy. Treatments that include ICIs are still under active development.1,2) Ipilimumab, an ICI, is an anti-cytotoxic T-lymphocyte protein 4 (CTLA-4) monoclonal antibody. It was the first drug that improved survival in a double-blind randomized study of patients with melanoma.3)
In the comparison of ipilimumab with nivolumab, an anti-programmed cell death protein 1 monoclonal antibody, as adjuvant therapy, grade 3 or 4 adverse events were more frequently reported in patients taking ipilimumab (45.9 vs. 14.4%).4) In the comparison of ipilimumab plus nivolumab with nivolumab monotherapy, a similar result was reported (59 vs. 22%).5) Therefore, in clinical practice, the number of ipilimumab doses is limited to 4. In addition, interprofessional collaboration intended to manage immune-related adverse events (irAE) has been promoted.6,7) However, 4 doses of ipilimumab might not be completed in practice. A significant association between the number of ipilimumab cycles (>2) and prolonged progression-free survival (PFS) has been reported.8) Thus, it is important to find biomarkers for identifying patients who might benefit from ipilimumab.
Neutrophil-to-lymphocyte ratio (NLR), calculated by dividing the neutrophil count by the lymphocyte count, is an indicator of systemic inflammation.9) Systemic inflammation is a hallmark of cancer that substantially contributes to the development and progression of malignancy.10) In patients with melanoma treated with ipilimumab, an association between elevated NLR and poor prognosis has been reported.11–15) NLR cut-off values have also been reported in various studies. For instance, Cassidy et al. reported that NLR >5 is associated with poor clinical outcomes in patients with advanced melanoma treated with ipilimumab in the United States.11) However, it remains unclear whether NLR is also useful as a predictor for prognosis in Japanese patients with melanoma, and if so, what is the optimal cut-off value in Japanese patients. In this study, we investigated the impact of NLR on the number of ipilimumab cycles and evaluated NLR cut-off values in Japanese patients with melanoma.
In this retrospective observational study, participants were Japanese patients with melanoma who had received ipilimumab monotherapy or ipilimumab and nivolumab combination therapy at Nagoya University Hospital from August 2015 to November 2021. Individuals who received investigational drugs were excluded.
TreatmentIn the ipilimumab regimen, 3 mg/kg of ipilimumab was administered intravenously. In the ipilimumab + nivolumab regimen, 3 mg/kg of ipilimumab was administered following nivolumab at a fixed dose of 80 mg. Both regimens were repeated every 3 weeks for up to 4 cycles. Doses and schedule modifications were allowed based on the patient’s condition.
Data CollectionThe following clinical data were collected from medical records: age, sex, body mass index (BMI), treatment regimen, number of ipilimumab cycles, cause of ipilimumab discontinuation, prior ICI treatment, post-ipilimumab treatment, creatinine clearance, lactate dehydrogenase, C-reactive protein, neutrophil count, and lymphocyte count. The data cutoff date was November 30, 2021.
Statistical AnalysisWe divided the patient population into two groups. One group included patients who received 1 or 2 ipilimumab cycles. The other group included patients who received 3 or 4 ipilimumab cycles. Risk factors for ipilimumab discontinuation within 2 cycles were statistically analyzed. Univariate analysis was performed with the Mann–Whitney U test or Fisher’s exact test. Multivariate logistic regression was used to estimate adjusted odds ratios. We limited the number of explanatory variables to 2 in order to avoid overfitting. Receiver operating characteristics (ROC) analysis was used to calculate the NLR cut-off value. Kaplan–Meier analysis was used to estimate differences in PFS with the log-rank test. All statistical analyses were conducted using IBM SPSS version 28 (IBM Japan, Tokyo, Japan). A p-value less than 0.05 was considered significant.
Ethical ConsiderationThis study was conducted in accordance with the principles of the Declaration of Helsinki, in compliance with the “Ethical Guidelines for Medical and Health Research Involving Human Subjects,” and with the approval of the ethics committee of Nagoya University Hospital (Approval No. 2022-0056)
Patient characteristics are shown in Table 1. A total of 38 Japanese patients with melanoma who received ipilimumab were included, of which 17 (45%) were treated with the ipilimumab + nivolumab regimen. Seventeen patients (45%) discontinued ipilimumab therapy within 2 cycles. The most common cause of discontinuation was an adverse event.
| Patients (n = 38) | |
|---|---|
| Age (years) | |
| Median | 66 |
| Range | 42–85 |
| Sex | |
| Male | 23 |
| Female | 15 |
| Body mass index (kg/m2) | |
| Median | 22 |
| Range | 14–36 |
| Treatment regimen | |
| Ipilimumab | 21 |
| Ipilimumab + nivolumab | 17 |
| Number of ipilimumab cycles | |
| 1 | 10 |
| 2 | 7 |
| 3 | 7 |
| 4 | 14 |
| Cause of ipilimumab discontinuation | |
| Adverse events | 19 |
| Progressive disease | 4 |
| Other | 1 |
The comparison of baseline characteristics between the two groups, which were classified based on the number of ipilimumab cycles, is summarized in Table 2. In the univariate analysis, baseline neutrophil count and NLR were significantly higher in patients who discontinue ipilimumab within 2 cycles. The ROC analysis showed that the optimal NLR cut-off value was 3.4. The area under the ROC curve was 0.75 (95% confidence interval, 0.58–0.92) (Fig. 1).
| Number of ipilimumab cycles | p-Value | ||
|---|---|---|---|
| 1 or 2 (n = 17) | 3 or 4 (n = 21) | ||
| Age (years) | |||
| Median | 67 | 62 | 0.89a) |
| Range | 42–85 | 46–83 | |
| Sex | |||
| Male | 9 | 14 | 0.39b) |
| Female | 8 | 7 | |
| Body mass index (kg/m2) | |||
| Median | 22 | 21 | 0.32a) |
| Range | 16–36 | 14–32 | |
| Treatment regimen | |||
| Ipilimumab | 9 | 12 | 0.80b) |
| Ipilimumab + nivolumab | 8 | 9 | |
| Prior ICI | |||
| Yes | 8 | 12 | 0.54b) |
| No | 9 | 9 | |
| Post-ipilimumab treatment | |||
| ICI | 6 | 10 | |
| BRAF inhibitor | 2 | 1 | |
| Radiation | 2 | 7 | |
| Best supportive care | 8 | 6 | |
| Creatinine clearance (mL/min) | |||
| Median | 53.2 | 61.1 | 0.50a) |
| Range | 24.3–117 | 20.0–98.9 | |
| Lactate dehydrogenase (U/L) | |||
| Median | 255 | 201 | 0.07a) |
| Range | 138–913 | 151–441 | |
| C-reactive protein (mg/dL) | |||
| Median | 0.12 | 0.11 | 0.66a) |
| Range | 0.01–6.96 | 0–6.03 | |
| Neutrophil count (/µL) | |||
| Median | 4300 | 3100 | 0.04a) |
| Range | 2000–7700 | 1900–6700 | |
| Lymphocyte count (/µL) | |||
| Median | 1100 | 1400 | 0.38a) |
| Range | 700–2100 | 500–2700 | |
| NLR | |||
| Median | 3.8 | 2.4 | < 0.01a) |
| Range | 1.1–8.0 | 1.0–6.2 | |
a) Mann–Whitney U test, b) Fisher’s exact test. ICI immune checkpoint inhibitor, NLR neutrophil-to-lymphocyte ratio.
The optimal NLR cut-off value for ipilimumab discontinuation within 2 cycles was 3.4. AUC area under the curve, CI confidence interval, NLR neutrophil-to-lymphocyte ratio, ROC receiver operating characteristic.
To identify factors associated with ipilimumab discontinuation, we performed multivariate analyses. Neutrophil count was correlated with NLR. Thus, we did not use neutrophil count as an explanatory variable. In addition to NLR cut-off value, combination with nivolumab was used as an explanatory variable because it was known to increase ipilimumab toxicity.5,16) Baseline NLR >3.4 was identified as an independent risk factor for ipilimumab discontinuation within 2 cycles, but combination with nivolumab was not (Table 3). Furthermore, baseline NLR >3.4 was significantly associated with worse PFS (Fig. 2).
| Odds ratio | 95% CI | p-Value | |
|---|---|---|---|
| NLR >3.4 | 15.6 | 3.0–82 | < 0.01 |
| Combination with nivolumab | 1.70 | 0.34–8.5 | 0.52 |
NLR neutrophil-to-lymphocyte ratio, CI confidence interval.
NLR >3.4 was significantly associated with worse PFS in Japanese patients with melanoma treated with an ipilimumab-containing regimen. PFS progression-free survival, NLR neutrophil-to-lymphocyte ratio.
To examine the impact of prior ICI treatments, which possibly affected the response to ipilimumab, we performed subgroup analysis. No significant difference in baseline NLR was observed between patients who had received prior ICI and those who had not (Table 4). Baseline NLR >3.4 was significantly associated with worse PFS in patients who had not received prior ICI, but it was not significant in patients received prior ICI (Fig. 3).
| Prior ICI | p-Value | ||
|---|---|---|---|
| Yes (n = 20) | No (n = 18) | ||
| Drugs | |||
| Nivolumab | 15 | 0 | |
| Pembrolizumab | 5 | 0 | |
| NLR | |||
| Median | 2.5 | 3.5 | 0.17a) |
| Range | 1.1–7.0 | 1.0–8.0 | |
a) Mann–Whitney U test. ICI immune checkpoint inhibitor, NLR neutrophil-to-lymphocyte ratio.
NLR >3.4 was significantly associated with worse PFS in patients who had not received prior ICI (A), but it was not significant in patients received prior ICI (B).
In this study, we found that baseline NLR can be used as a predictor for the efficacy of ipilimumab-containing regimens in Japanese patients with melanoma. Our finding was consistent with previous results from the United States and Europe. Since adverse events were the main reasons for ipilimumab discontinuation, we hypothesize that higher NLR is correlated with severe adverse events, leading to the discontinuation of ipilimumab, which results in worse PFS.
We found that the optimal NLR cut-off value is 3.4 in Japanese patients. Interestingly, this value was relatively lower than those reported in American and European patients.11–15) Most studies on patients with melanoma conducted in the United States and Europe have reported NLR cut-off values in the range of 4–5.17) This difference might be due to genetic polymorphisms and differences in physique. In a study based on next-generation sequencing of patients with melanoma treated with ICIs, the occurrence of irAE was significantly associated with several genetic markers, including human leukocyte antigen homozygosity.18) Moreover, Tazeh et al. reported that African patients with bladder cancer had lower NLR than European patients. They mentioned that ethnicity should be taken into account when interpreting NLR in patients with bladder cancer.19) These previous findings suggest that regional differences in NLR cut-off values might be due to genetic differences. Furthermore, an association between NLR and obesity has been reported.20,21) Considering that Japanese individuals generally have lower BMI, a lower NLR cut-off value in Japanese patients might be reasonable. Therefore, extrapolation of NLR cut-off values from other populations to Japanese individuals should be performed with caution. Research to explore race-specific cut-off values is important for improving the predictive ability of NLR.
In subgroup analysis stratified by prior ICI treatments, no significant difference in baseline NLR was observed, but NLR > 3.4 was not significant predictor in patients who had received prior ICI. It has been known that ICI treatments occasionally induce excessive immune response including irAE. Possibly, NLR cut-off value is influenced by excessive immune response and systemic inflammation caused by prior ICI treatments. However, further studies are needed to identify this possibility because of smaller population sizes in subgroup analysis.
Several limitations of this study should be noted. The retrospective nature of this study limits the interpretations of results. Heterogeneity of patient background, information bias, and data quality might have influenced the results. In addition, the data presented should be interpreted with caution because the number of subjects was relatively small and the study was performed at one university hospital in Japan.
In conclusion, NLR >3.4 was a useful indicator for selecting Japanese patients with melanoma for treatment with an ipilimumab-containing regimen. The optimal NLR cut-off value might differ across racial groups. Thus, NLR cut-off values should be extrapolated to Japanese patients with caution.
The authors are grateful to the members of the Department of Hospital Pharmacy, Nagoya University Hospital for their support on this study.
YM, YK, and MM developed the main concept and study design. YM, YK, and MM collected and checked clinical data. YM and YK performed data analysis. YM and YK drafted the manuscript. K Yokota, HM, MA, and K Yamada contributed to critical revision of the manuscript. HM and K Yamada supervised the study. All authors approved the manuscript.
The authors declare no conflict of interest.
