Fluctuations in routine blood count might signal severe immune-related adverse events in melanoma patients treated with nivolumab

G Model DESC 3228 No. of Pages 7

Journal of Dermatological Science xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Journal of Dermatological Science journal homepage: www.jdsjournal.com

Fluctuations in routine blood count might signal severe immune-related adverse events in melanoma patients treated with nivolumab Yasuhiro Fujisawaa,* , Koji Yoshinob , Atsushi Otsukac , Takeru Funakoshid, Taku Fujimurae , Yuki Yamamotof , Hiroo Hatag , Masahiko Goshoh , Ryota Tanakaa , Kei Yamaguchib , Yumi Nonomurac , Ikuko Hiraid, Sadanori Furudatee , Hisako Okuhiraf , Keisuke Imafukug, Megumi Aokii, Shigeto Matsushitai a

Department of Dermatology, University of Tsukuba, Japan Department of Dermatology, Tokyo Metropolitan Cancer and Infectious Disease Center Komagome Hospital, Japan Department of Dermatology, Kyoto University, Japan d Department of Dermatology, Keio University, Japan e Department of Dermatology, Tohoku University, Japan f Department of Dermatology, Wakayama Medical University, Japan g Department of Dermatology, Hokkaido University, Japan h Department of Clinical Trial and Clinical Epidemiology, University of Tsukuba, Japan i Department of Dermato-Oncology/Dermatology, National Hospital Organization Kagoshima Medical Center, Japan b c

A R T I C L E I N F O

A B S T R A C T

Article history: Received 7 April 2017 Received in revised form 17 May 2017 Accepted 11 July 2017

Background: Although nivolumab significantly prolongs survival of metastatic melanoma, about 10% of patients experience severe, even fatal immune-related adverse events (irAEs). Biomarkers to predict irAEs are, therefore, of great interest. Objective: We aimed to correlate changes in routine blood count parameters to the occurrence of serious irAEs (grade 3/4 [G3/4] or lung/gastrointestinal [lung/GI] irAEs) in patients with melanoma who were treated with nivolumab. Methods: We retrospectively analyzed data from 101 patient with melanoma treated with nivolumab from 8 institutes in Japan. We used logistic regression analyses to investigate associations between severe irAEs and fluctuations in routine blood count parameters (total white blood cell [WBC] count, relative neutrophil, monocyte, lymphocyte, and eosinophil count) during the treatment. Receiveroperating characteristic curve was used to determine a cutoff value for the blood count parameters and area under the curve (AUC). Results: Univariate analysis revealed that G3/4 irAEs were associated with increased total WBC count (P = 0.034, cutoff value = +27%, AUC = 0.68, odds ratio [OR] = 1.58) and decreased relative lymphocyte count (RLC, P = 0.042, cutoff value = 23%, AUC = 0.65, OR = 1.65). However, multivariate analysis showed that the same factors, increased WBC count (P = 0.014, cutoff value = +59.1%, AUC = 0.79, OR = 6.04) and decreased RLC (P = 0.012, cutoff value = 32.3%, AUC = 0.81, OR = 5.01) were independent factors associated with lung/GI irAEs. Conclusions: Our results suggest that increased WBC count and decreased RLC are associated with G3/4 and lung/GI irAEs. Our analysis was based on the data point at which irAE occurrence was noticed and, therefore, these factors are not predictive, however, they could be a “signal” of severe irAE occurrence in patients with melanoma treated with nivolumab. © 2017 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved.

Keywords: Melanoma Adverse event Checkpoint inhibitor White blood cell count Blood cell parameters Relative lymphocyte count

* Corresponding author at: 1 1 1 Tennodai, Tsukuba, Ibaraki, Japan. E-mail address: [email protected] (Y. Fujisawa). http://dx.doi.org/10.1016/j.jdermsci.2017.07.007 0923-1811/ © 2017 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved.

Please cite this article in press as: Y. Fujisawa, et al., Fluctuations in routine blood count might signal severe immune-related adverse events in melanoma patients treated with nivolumab, J Dermatol Sci (2017), http://dx.doi.org/10.1016/j.jdermsci.2017.07.007

G Model DESC 3228 No. of Pages 7

2

Y. Fujisawa et al. / Journal of Dermatological Science xxx (2017) xxx–xxx

1. Introduction

2.2. Study design

The management of advanced melanoma has been dramatically changed since the introduction of monoclonal antibody (mAb) that target immune checkpoints such as programmed-death receptor-1 (PD-1) or cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) [1,2]. Compared to chemotherapy, PD-1-specific mAb Nivolumab [3,4] and pembrolizumab[5] significantly improve the survival of patients with advanced melanoma, and they are now used as the first-line treatment. However, checkpoint inhibitors can cause unique adverse events that rarely occur in patients who receive conventional chemotherapy. [3,6] Such events, known as immunerelated adverse events (irAEs), are caused by activation of the autoimmune response and subsequently damage healthy tissue. IrAEs commonly affect the skin, gastrointestinal tract (GI), liver and endocrine systems [3,5,6], and, although rare, irAEs may affect the lungs [3], uvea [7], nerve tissue system [8], pancreas, [9] kidneys [10], myocardium [11], and hematopoietic system [12]. The current clinical consensus is that iatrogenic irAEs can be safely managed through standardized protocols, making them an acceptable tradeoff for increased efficacy [13]. However, even though treatment using checkpoint inhibitors is considered to be well tolerated, 5 to 12% of clinical trial patients who received nivolumab/pembrolizumab developed G3/4 irAEs [4,14,15]. G3/4 irAEs force discontinuation of the checkpoint inhibitor treatment followed by treatment with systemic steroids to control the autoimmune reactions. Although the occurrence of GI irAEs with nivolumab/pembrolizumab treatment is reported to be lower than ipilimumab, 4 fatal cases of ipilimumab-related GI irAEs[16,17] and 3 fatal cases of nivolumab-related pneumonitis in patients with melanoma have been reported [3]. Therefore, caution is warranted with regard to possible checkpoint inhibitor-associated life-threatening irAEs and, as there are no clear biomarkers to predict irAEs occurrence[18], close monitoring of patients’ symptoms is necessary. Recently, there have been reports of several baseline peripheral blood biomarkers associated with the clinical outcome of patients with melanoma treated with either ipilimumab[19–21] or pembrolizumab [22]. However, no study has evaluated the correlation between irAE occurrence and peripheral blood markers. Thus, the aim of this study was to identify factors associated with irAEs by examining changes in peripheral blood count parameters in patients with melanoma during their treatment course with nivolumab.

The correlation between the occurrence of irAEs and fluctuations in the peripheral blood count was investigated. The WBC count and relative count of each cell fraction were compared with the baseline values and the percentage change was calculated as follows: % change = {(value of each data point)/(baseline value) 1} *100. For patients with irAEs, laboratory data obtained on the same day of irAE occurrence were used except for 3 patients whose same-day data were unavailable. For these 3 patients, data from the closest point in time prior to irAE occurrence was used (1, 2, and 6 weeks). No patient received corticosteroids or immunosuppressant at the time of data acquisition, which may affect the laboratory data. The average value of each parameter from the day of first administration to the end of the follow-up period in patients without irAEs was used as a control to decrease data variability.

2. Patients and methods 2.1. Patients Data from patients who were treated with nivolumab were collected from 8 clinical sites in Japan. Inclusion criteria were unresectable or stage IV melanoma treated with at least 1 dose of nivolumab. All patients received nivolumab at 2 mg/kg followed by a 3-week rest, which is the approved dosage schedule in Japan. Clinical data including primary tumor, site of metastasis, previous treatment, treatment course, occurrence of irAEs, and peripheral blood count data (total white blood cell count [WBC] and relative counts of neutrophils [RNC], monocytes [RMC], lymphocytes [RLC], and eosinophils [REC]) were collected. Blood count data taken 0– 28 days before the first dose of nivolumab was considered as baseline. CTCAE Version 4.0 was used to score the irAEs. This study was approved by the ethics committee of all 8 institutions participating in this study.

2.3. Statistical analysis To examine the association between the irAEs and blood count parameters, univariate logistic regression analyses were used to calculate odds ratios (OR) per 10% change and P-values. For the factors shown to be significant by the univariate analysis, multivariate logistic regression analyses, in a stepwise procedure, were performed to determine the independent factors associated with the irAEs. In the procedure, a 0.15 significance level for entering and eliminating explanatory variables was used. ReceiverOperating Characteristic (ROC) curve was applied to calculate cutoff values for the blood count parameters and area under the curves (AUC). The cutoff point was determined using Youden’s index [23]. OR for each of the cutoff points was calculated using the results from each of the corresponding logistic regression analyses. Throughout the analyses, P-values <0.05 were considered statistically significant. Statistical tests were 2-sided and carried out using Stat Flex version 6.0 (Artec, Osaka, Japan) and Prism version 6 (Graph Pad software, CA, USA).

Table 1 Background characteristics of the study. N Data collected Data omitted Data in analysis Age Range Average Median Sex Male Female Stage Unresectable tumor Distant metastasis Primary site Extremities Mucosa Trunk Head/Neck Ocular Other/Unknown Average follow-up period (days) No irAE Without irAE Average no. of nivolumab injection No irAE With irAE

106 5 101 30–89 66.7 69 63 38 25 76 46 27 11 10 2 5 176 214 8 7

Please cite this article in press as: Y. Fujisawa, et al., Fluctuations in routine blood count might signal severe immune-related adverse events in melanoma patients treated with nivolumab, J Dermatol Sci (2017), http://dx.doi.org/10.1016/j.jdermsci.2017.07.007

G Model DESC 3228 No. of Pages 7

Y. Fujisawa et al. / Journal of Dermatological Science xxx (2017) xxx–xxx Table 3 Profile of immune-related adverse events.

3. Results 3.1. Patients and irAEs

N (%)

We collected data for 106 patients treated with nivolumab (Table 1). Of those, 101 patients were included in the analysis; 5 patients were omitted from the analysis due to the lack of blood count information. The mean age was 66.7 years (range: 30 to 89 years) and 62% were female. The most common site of the primary tumor was in the extremities (46%) followed by mucosal origin (27%), trunk (11%), head/neck (10%), and ocular origin (2%). Of the 101 patients, 75% had distant metastasis. Nivolumab dosing was 2 mg/kg followed by a 3-week rest. The median follow-up period for patients with and without irAE was 172 and 126 days, respectively (Table 2) and the median number of doses received was 6 doses for both groups. During the treatment, 38 patients (38%) experienced 1 or more irAEs (Table 3). Of the 38 patients with irAEs, 6 experienced multiple irAEs, 2 with concomitant irAEs, and 4 with sequential irAEs. Of the 44 irAE incidents, 15 were Grade 3 or 4 (G3/4) irAEs. The most commonly affected areas were the endocrine system (36%), skin (34%), lung (16%), colon (7%), liver (7%), kidney (5%), and hematopoietic system (5%). The distribution of grade for each affected organ is listed in Table 3. Because the number of participants was small we could not see any statistical difference between the affected organ and days to develop irAE (Fig. 1a). Additionally, Grade 1 or 2 (G1/2) events mostly occurred within 200 days, while 33% (5/15 patients) of G3/4 irAEs occurred after 200 days (P = 0.06, Chi-square test).

Total Patients irAE Patients with irAE Patients with multiple irAEs Concomitant Sequential

101

Total AE incidents Grade Grade1 Grade2 Grade3 Grade4 Affected organ Endocrine Thyroid Pituitary Skin Vitiligo Psoriasiform dermatitis Dermatitis Lung Colon Liver Kidney Hematological Thrombocytopenia Others Synovial bursitis Rheumatoid arthritis Nervous system Arrhythmia

44

38 (38%) 6 (6%) 2 4

16 (36%) 13 (30%) 8 (18%) 7 (16%) 15 (34%) 13 2 8 (18%) 5 2 1 7 3 3 2 2

(16%) (7%) (7%) (5%) (5%) 2

4 (9%) 1 1 1 1

irAE: immune-related adverse event.

3.2. Identification of factors associated with G3/4 irAEs According to the current consensus, the occurrence of a G3/4 irAE, drugs should be discontinued temporarily or even permanently. Thus, we investigated the association between G3/4 irAEs and the fluctuation of each cell fraction in WBC counts. Fig. 2 shows the comparison of each cell fraction between patients with or without irAEs using the Mann-Whitney test and taking into account total WBC count and relative count of each cell fractions (RNC, RMC, RLC, and REC). As shown in Fig. 2A, no statistical difference was found in any factors when comparing patients with or without irAEs. On the other hand, patients with G3/4 irAEs had significantly increased WBC counts and a trend towards a decrease in RLC (Fig. 2B). Next, we used multivariate logistic regression analysis to investigate the impact of each factors on irAE occurrence. As shown in Table 4, although increased WBC count (OR = 1.02, P = 0.034) and decreased RLC (OR = 1.02, P = 0.042) were associated with G3/4 irAE according to the univariate analysis, multivariate analysis revealed no significant factors (Table 4). For G3/4 irAEs, the cutoff values of the parameters and their AUCs, as determined by ROC analysis, are shown in the upper row of Table 5. The AUCs of both WBC count and RLC were higher than Table 2 Total follow-up period in patients with or without adverse event.

No. of patients Follow up periods (days) Range Averange Median No. of Nivolumab injection Range Averange Median

3

No irAE

With irAE

63

38

21–490 176 126

14–573 214 172

1–26 8 6

1–24 7 6

other factors (0.68 and 0.65, respectively), which was consistent with the result shown in Table 4. Taken together, these results suggest that increased WBC count (+27%, calculated OR = 1.58) and decreased RLC ( 23%, calculated OR = 1.65) could be useful markers to detect G3/4 irAEs. 3.3. Identification of factors associated with lung or GI irAEs Next, we used a similar approach to assess the association of lung or GI irAEs (lung/GI) and the fluctuation of each cell fraction in WBC count. We considered these irAEs to be a particular importance because there have been reports of both lung and GI irAE-related deaths. Lung/GI irAEs showed significantly increased WBC count and RNC and decreased RLC which was similar to G3/4 irAEs. (Fig. 2C). As shown in the lower half of Table 4, multivariate logistic regression analysis revealed that both increased WBC count (OR = 1.02, P = 0.014) and decreased RLC (OR = 1.05, P = 0.012) were associated with lung/GI irAE. Although univariate analysis showed that increased RNC associated with lung/GI irAE (OR = 1.07, P = 0.0053), it was not significant in the multivariate analysis. The cutoff values and AUC for each factor determined by ROC analysis are shown in the lower row of Table 5. Similar to the G3/4 irAE results, AUC of both WBC count and RLC were higher than other factors (0.79 and 0.81, respectively), which was consistent with the logistic regression analysis results (Table 4). These results suggest that an increased WBC count (+59.1%, OR = 6.04) and decreased RLC ( 32.3%, OR = 5.01) status is a more useful indictor for detecting lung/GI irAE than it is for detecting G3/4 irAE.

Please cite this article in press as: Y. Fujisawa, et al., Fluctuations in routine blood count might signal severe immune-related adverse events in melanoma patients treated with nivolumab, J Dermatol Sci (2017), http://dx.doi.org/10.1016/j.jdermsci.2017.07.007

G Model DESC 3228 No. of Pages 7

4

Y. Fujisawa et al. / Journal of Dermatological Science xxx (2017) xxx–xxx

Fig. 1. Development of irAEs after initial nivolumab dose. A) There was no difference between the affected organ and days to develop irAE. B) Most of the mild irAEs occurred within 200 days, 5 out of 15 severe irAEs occurred after 200 days.

Fig. 2. Comparison of each cell fraction between patients with or without irAEs. The WBC and relative numbers of each cell fraction were compared with the baseline values. A) There was no statistical differenc found in any factors when comparing patients with or without irAEs. B) Patients with severe irAEs had significantly increased WBC and a trend towards a decrease in rLym. C) Patients with lung/GI irAEs had significantly increased BC and rNeu and decreased rLym. *: P < 0.05, **: P < 0.01 by Mann-Whitney test. (absolute white blood cell count, WBC; relative neutrophil count, RNC; relative monocyte count, RMC; relative lymphocyte count, RLC; relative eosinophil count, REC)

4. Discussion In this study, we examined changes in peripheral blood count parameters during a nivolumab treatment course to identify factors associated with irAEs. To the best of our knowledge, this is the first study to evaluate the association between the occurrence of irAEs and the fluctuation in routine blood cell count during the course of treatment. In this study, 2 factors were shown by univariate analysis to be associated with G3/4 irAEs: (1) increased WBC count and (2) decreased RLC (shown in Fig. 2B and Table 4).

However, subsequent multivariate analysis failed to show significance, perhaps due to a multicollinearity between WBC count and RLC the relationship between these two factors being close. Similar trends were seen in lung/GI irAE cases, but in a more exacerbated manner; typical G3/4 irAE-associated status also presented with significant increases in RNC (Fig. 2C). Interestingly, increased RNC was only observed in lung/GI irAE cases and not in other irAEs. Histological analysis of active colitis in patients treated with ipilimumab is reported to have marked neutrophilic infiltration into the lamina propria [24]. Moreover, another study has shown that during ipilimumab treatment there are increased

Please cite this article in press as: Y. Fujisawa, et al., Fluctuations in routine blood count might signal severe immune-related adverse events in melanoma patients treated with nivolumab, J Dermatol Sci (2017), http://dx.doi.org/10.1016/j.jdermsci.2017.07.007

G Model DESC 3228 No. of Pages 7

Y. Fujisawa et al. / Journal of Dermatological Science xxx (2017) xxx–xxx

5

Table 4 Factors associated with Grade 3/4 or lung/GI-related immune-related adverse events by logistic regression. Type of irAE

Factors

Univariate Odds ratio

P-value

95% CI

Odds ratio

P-value

95% CI

Grade3/4 irAE

WBC count increase by 10% RNC increase by 10% RMC decrease by 10% RLC decrease by 10% REC decrease by 10%

1.18

0.034

1.012  1.38

1.13

0.074

0.99  1.29

1.27

0.21

0.88  1.86

1.11

0.27

0.92  1.33

1.24

0.042

1.01 1.53

1.18

0.15

0.94  01.48

1.00

0.98

0.99  1.01

WBC count increase by 10% RNC increase by 10% RMC decrease by 10% RLC decrease by 10% REC decrease by 10%

1.36

0.0024

1.11 1.64

1.20

0.014

1.04  1.39

1.91

0.0053

1.21  3.02

1.23

0.073

0.98  1.55

1.65

0.0023

1.19  2.27

1.65

0.012

1.12  2.45

1.07

0.20

0.97  1.17

Lung/GI irAE

Multivariate

GI: gastrointenstinal, RNC: relative neutrophil count, RMC: relative monocyte count, RLC: relative lymphocyte count, REC: relative eosinophil count

Table 5 Cutoff values and association with irAE. Sensitivity (%)

Specificity (%)

AUC of ROC (95%CI)

Odds ratioa (95%CI)

20 43 27 36 45 18 17 46 6 57

67

68

73

57

93

29

60

73

33

90

0.68 0.53  0.83 0.60 0.42  0.78 0.57 0.40  0.73 0.65 0.48  0.82 0.60 0.42  0.77

1.58 1.03  2.40 1.08 0.98  1.23 1.18 0.88  1.59 1.65 1.02  2.67 1.00 0.93  1.07

6 57 10 53 23 40 11 52 37 26

60

95

60

84

70

64

80

83

90

41

0.79 0.62  0.97 0.77 0.61  0.92 0.68 0.46  0.89 0.81 0.63  0.99 0.65 0.46  0.83

6.04 2.10  19.28 2.58 1.33  5.02 1.25 0.98  1.59 5.01 1.78  14.17 1.44 0.83  2.49

irAE

Factor

Category (%change)

No. of patients with irAE

without irAE

Grade 3/4 irAE

WBC count

>+27 +27 >+3.3 +3.3 <+16.3 +16.3 < 23  23 < 49.1  49.1

10 5 11 4 14 1 9 6 5 10

> +59.1  +59.1 > +14.6  +14.6 < 10.7  10.7 < -32.3  32.3 < 57.4 57.4

6 4 6 4 7 3 8 2 9 1

RNC RMC RLC REC

Lung/GI irAE

WBC count RNC RMC RLC REC

a

Odds ratio of the upper category to the lower category. AUC: area under the curve, ROC: receiver-operating characteristic curve.

expression levels of the neutrophil activation markers CD177 and CEACAM1 in the whole blood of patients with GI irAEs [25]. Although these results indicate that our observation could be explained partly by neutrophilic expansion within the affected organs, the difficulty of effective biopsy at these irAE sites (liver, lung, endocrine, etc) remains a barrier to fully understanding of the immune processes at work. Several clinical studies have focused on the neutrophil/ lymphocyte ratio in patients treated with ipilimumab and have shown a correlation between neutrophil/lymphocyte ratio and clinical outcome [20,21,26]. In these studies, however, there was no correlation with the neutrophil/lymphocyte ratio and the occurrence of irAEs. This discrepancy with our findings might be explained by the different drugs that the studies focused on ipilimumab as opposed to nivolumab and thus, different

pathological causes of the irAEs. Moreover, previous studies have been based on the neutrophil/lymphocyte ratio at either a certain point or the baseline value and not fluctuations during the treatment, as was the case in our study. To clarify this, we are now planning to collect data on patients who were treated with ipilimumab and investigate fluctuations in blood counts and occurrence of irAEs. The model onset of irAEs is usually at week 24 of therapy and progression follows a cascade where skin-related irAEs occur earliest, followed by colitis, hepatitis, and endocrinopathy [4,16,18]. However, as shown in Fig. 1A, we could not see a similar pattern of irAE onset in our study population. This discrepancy might be explained by the differences between clinical trial and routine practice. For this report, the enrolled participants were not examined and recorded in a clinical trial setting and so meaning

Please cite this article in press as: Y. Fujisawa, et al., Fluctuations in routine blood count might signal severe immune-related adverse events in melanoma patients treated with nivolumab, J Dermatol Sci (2017), http://dx.doi.org/10.1016/j.jdermsci.2017.07.007

G Model DESC 3228 No. of Pages 7

6

Y. Fujisawa et al. / Journal of Dermatological Science xxx (2017) xxx–xxx

mild irAEs (such as low grade skin rash) could have been overlooked. As for the onset of irAEs, although most of the mild irAEs (Grade 1/2) occurred within 200 days, 5 out of 15 G3/4 irAEs occurred after 200 days (Fig. 1B). Therefore, vigilance against severe irAE occurrence is warranted even 200 days after the initial nivolumab dosing. Several studies reported a positive correlation between the occurrence of irAEs and clinical responses. In spite of a contradictory report [27], the literature tends to favor this concept of immunomodulatory treatment for patients with melanoma using interferon [28,29], ipilimumab [30], and nivolumab [31]. These data indicate a close relationship between self-tolerance and tumor-tolerance. To date, several studies have been published that show an association between outcome and routine blood cell count parameters. High pre-treatment RLC and REC among pembrolizumab patients[22] and low absolute monocyte count [19], high absolute eosinophil count [19], high absolute lymphocyte count, [32] low RLC [19], and increasing absolute lymphocyte and eosinophil count during ipilimumab treatment[33] have been reported to be associated with favorable outcomes. Although our study design differs somewhat, we found that increased absolute lymphocyte and eosinophil count did not correlate with the occurrence of severe irAEs. This discrepancy might be explained partly by the fact that most of the favorable correlations were shown with skin irAEs (skin rash or vitiligo). As discussed above, massive inflammation causes neutrophil expansion and as a consequence, the immune mileu might be greatly affected. There are several limitations in this study that should be considered: low sensitivity of possible factors, small study population, retrospective nature of the study, heterogeneous study population, possibly overlooked mild irAEs (eg G1 skin rash), and possible variability of measurement values between institutes. Moreover, due to the small-sample size, we could not perform more sophisticated analyses to assess the relationship between irAE and blood parameters. For example, age-sex of the individuals and the time points of the observed blood parameters should be matched between the case (ie patients with irAEs) and control (ie patients without irAEs) groups. Thus, our results should be considered as an exploratory investigation. Although the sensitivity is not high enough to detect all events, using easily accessible data may justify the application of these biomarkers in daily practice to detect possible G3/4 or lung/GI irAEs. Further studies are needed to address the limitations of this investigation, but and our proposed factors could provide useful information for managing nivolumab patients as well as establishing a seminal basis for future research. Funding source None. Conflict of interest The authors declare no potential conflicts of interest. Acknowledgement We would like to thank Bryan Mathis and Thomas Mayers of the Medical English Communication Center of the University of Tsukuba, for their excellent English revisions. References [1] A.M. Eggermont, A. Testori, M. Maio, C. Robert, Anti-CTLA-4 antibody adjuvant therapy in melanoma, Semin. Oncol. 37 (2010) 455–459.

[2] C. Robert, L. Thomas, I. Bondarenko, S. O'Day, J. Weber, C. Garbe, et al., Ipilimumab plus dacarbazine for previously untreated metastatic melanoma, N. Engl. J. Med. 364 (2011) 2517–2526. [3] S.L. Topalian, F.S. Hodi, J.R. Brahmer, S.N. Gettinger, D.C. Smith, D.F. McDermott, et al., Safety, activity, and immune correlates of anti-PD-1 antibody in cancer, N. Engl. J. Med. 366 (2012) 2443–2454. [4] S.L. Topalian, M. Sznol, D.F. McDermott, H.M. Kluger, R.D. Carvajal, W.H. Sharfman, et al., Survival, durable tumor remission, and long-term safety in patients with advanced melanoma receiving nivolumab, J. Clin. Oncol. 32 (2014) 1020–1030. [5] O. Hamid, C. Robert, A. Daud, F.S. Hodi, W.J. Hwu, R. Kefford, et al., Safety and tumor responses with lambrolizumab (anti-PD-1) in melanoma, N. Engl. J. Med. 369 (2013) 134–144. [6] J.S. Weber, K.C. Kahler, A. Hauschild, Management of immune-related adverse events and kinetics of response with ipilimumab, J. Clin. Oncol. 30 (2012) 2691–2697. [7] M.R. Robinson, C.C. Chan, J.C. Yang, B.I. Rubin, G.J. Gracia, H.N. Sen, et al., Cytotoxic T lymphocyte-associated antigen 4 blockade in patients with metastatic melanoma: a new cause of uveitis, J. Immunother. 27 (2004) 478– 479. [8] S. Wilgenhof, B. Neyns, Anti-CTLA-4 antibody-induced Guillain-Barre syndrome in a melanoma patient, Ann. Oncol. 22 (2011) 991–993. [9] A.M. Di Giacomo, R. Danielli, M. Guidoboni, L. Calabro, D. Carlucci, C. Miracco, et al., Therapeutic efficacy of ipilimumab, an anti-CTLA-4 monoclonal antibody, in patients with metastatic melanoma unresponsive to prior systemic treatments: clinical and immunological evidence from three patient cases, Cancer Immunol. Immunother. 58 (2009) 1297–1306. [10] H. Izzedine, V. Gueutin, C. Gharbi, C. Mateus, C. Robert, E. Routier, et al., Kidney injuries related to ipilimumab, Invest. New Drugs 32 (2014) 769–773. [11] H. Laubli, C. Balmelli, M. Bossard, O. Pfister, K. Glatz, A. Zippelius, Acute heart failure due to autoimmune myocarditis under pembrolizumab treatment for metastatic melanoma, J. Immunother. Cancer 3 (2015) 11. [12] M. Akhtari, E.K. Waller, D.L. Jaye, D.H. Lawson, R. Ibrahim, N.E. Papadopoulos, et al., Neutropenia in a patient treated with ipilimumab (anti-CTLA-4 antibody), J. Immunother. 32 (2009) 322–324. [13] N.M. La-Beck, G.W. Jean, C. Huynh, S.K. Alzghari, D.B. Lowe, Immune checkpoint inhibitors: new insights and current place in cancer therapy, Pharmacotherapy 35 (2015) 963–976. [14] C. Robert, G.V. Long, B. Brady, C. Dutriaux, M. Maio, L. Mortier, et al., Nivolumab in previously untreated melanoma without BRAF mutation, N. Engl. J. Med. 372 (2015) 320–330. [15] C. Robert, A. Ribas, J.D. Wolchok, F.S. Hodi, O. Hamid, R. Kefford, et al., Antiprogrammed-death-receptor-1 treatment with pembrolizumab in ipilimumab-refractory advanced melanoma: a randomised dose-comparison cohort of a phase 1 trial, Lancet 384 (2014) 1109–1117. [16] J.S. Weber, R. Dummer, V. de Pril, C. Lebbe, F.S. Hodi, Patterns of onset and resolution of immune-related adverse events of special interest with ipilimumab: detailed safety analysis from a phase 3 trial in patients with advanced melanoma, Cancer 119 (2013) 1675–1682. [17] M.E. Lacouture, J.D. Wolchok, G. Yosipovitch, K.C. Kahler, K.J. Busam, A. Hauschild, Ipilimumab in patients with cancer and the management of dermatologic adverse events, J. Am. Acad. Dermatol. 71 (2014) 161–169. [18] F.S. Hodi, S.J. O'Day, D.F. McDermott, R.W. Weber, J.A. Sosman, J.B. Haanen, et al., Improved survival with ipilimumab in patients with metastatic melanoma, N. Engl. J. Med. 363 (2010) 711–723. [19] A. Martens, K. Wistuba-Hamprecht, M. Geukes Foppen, J. Yuan, M.A. Postow, P. Wong, et al., Baseline peripheral blood biomarkers associated with clinical outcome of advanced melanoma patients treated with ipilimumab, Clin. Cancer Res. 22 (2016) 2908–2918. [20] L. Khoja, E.G. Atenafu, A. Templeton, Y. Qye, M.A. Chappell, S. Saibil, et al., The full blood count as a biomarker of outcome and toxicity in ipilimumab-treated cutaneous metastatic melanoma, Cancer Med. 5 (2016) 2792–2799. [21] P.F. Ferrucci, P.A. Ascierto, J. Pigozzo, M. Del Vecchio, M. Maio, G.C. Antonini Cappellini, et al., Baseline neutrophils and derived neutrophil-to-lymphocyte ratio: prognostic relevance in metastatic melanoma patients receiving ipilimumab, Ann. Oncol. 27 (2016) 732–738. [22] B. Weide, A. Martens, J.C. Hassel, C. Berking, M.A. Postow, K. Bisschop, et al., Baseline biomarkers for outcome of melanoma patients treated with pembrolizumab, Clin. Cancer Res. 22 (2016) 5487–5496. [23] W.J. Youden, Index for rating diagnostic tests, Cancer 3 (1950) 32–35. [24] D. Berman, S.M. Parker, J. Siegel, S.D. Chasalow, J. Weber, S. Galbraith, et al., Blockade of cytotoxic T-lymphocyte antigen-4 by ipilimumab results in dysregulation of gastrointestinal immunity in patients with advanced melanoma, Cancer Immunol. 10 (11) (2010). [25] V. Shahabi, D. Berman, S.D. Chasalow, L. Wang, Z. Tsuchihashi, B. Hu, et al., Gene expression profiling of whole blood in ipilimumab-treated patients for identification of potential biomarkers of immune-related gastrointestinal adverse events, J. Transl. Med. 11 (75) (2013). [26] A.M. Di Giacomo, P.A. Ascierto, P. Queirolo, L. Pilla, R. Ridolfi, M. Santinami, et al., Three-year follow-up of advanced melanoma patients who received ipilimumab plus fotemustine in the Italian Network for Tumor Biotherapy (NIBIT)-M1 phase II study, Ann. Oncol. 26 (2015) 798–803. [27] F.P. Madorsky Rowdo, A. Baron, M. Urrutia, J. Mordoh, Immunotherapy in cancer: a combat between tumors and the immune system; you win some, you lose some, Front. Immunol. 6 (127) (2015).

Please cite this article in press as: Y. Fujisawa, et al., Fluctuations in routine blood count might signal severe immune-related adverse events in melanoma patients treated with nivolumab, J Dermatol Sci (2017), http://dx.doi.org/10.1016/j.jdermsci.2017.07.007

G Model DESC 3228 No. of Pages 7

Y. Fujisawa et al. / Journal of Dermatological Science xxx (2017) xxx–xxx [28] M.G. Bouwhuis, S. Suciu, S. Collette, S. Aamdal, W.H. Kruit, L. Bastholt, et al., Autoimmune antibodies and recurrence-free interval in melanoma patients treated with adjuvant interferon, J. Natl. Cancer Inst. 101 (2009) 869–877. [29] A.A. Tarhini, D. Shin, S.J. Lee, J. Stuckert, C.A. Sander, J.M. Kirkwood, vSerologic evidence of autoimmunity in E2696 and E1694 patients with high-risk melanoma treated with adjuvant interferon alfa, Melanoma Res. 24 (2014) 150–157. [30] S.G. Downey, J.A. Klapper, F.O. Smith, J.C. Yang, R.M. Sherry, R.E. Royal, et al., Prognostic factors related to clinical response in patients with metastatic melanoma treated by CTL-associated antigen-4 blockade, Clin. Cancer Res. 13 (2007) 6681–6688.

7

[31] M. Freeman-Keller, Y. Kim, H. Cronin, A. Richards, G. Gibney, J.S. Weber, Nivolumab in resected and unresectable metastatic melanoma: characteristics of immune-Related adverse events and association with outcomes, Clin. Cancer Res. 22 (2016) 886–894. [32] S. Kelderman, B. Heemskerk, H. van Tinteren, R.R. van den Brom, G.A. Hospers, A.J. van den Eertwegh, et al., Lactate dehydrogenase as a selection criterion for ipilimumab treatment in metastatic melanoma, Cancer Immunol. Immunother. 63 (2014) 449–458. [33] J. Delyon, C. Mateus, D. Lefeuvre, E. Lanoy, L. Zitvogel, N. Chaput, et al., Experience in daily practice with ipilimumab for the treatment of patients with metastatic melanoma: an early increase in lymphocyte and eosinophil counts is associated with improved survival, Ann. Oncol. 24 (2013) 1697–1703.

Please cite this article in press as: Y. Fujisawa, et al., Fluctuations in routine blood count might signal severe immune-related adverse events in melanoma patients treated with nivolumab, J Dermatol Sci (2017), http://dx.doi.org/10.1016/j.jdermsci.2017.07.007