Advertisement

The development of sarcoidosis in patients receiving daclizumab: A case series from multiple clinical trials

Open ArchivePublished:February 01, 2019DOI:https://doi.org/10.1016/j.rmed.2019.01.015

      Abstract

      Introduction

      Several drugs have been associated with druginduced sarcoidosis-like reactions (DISRs) that are clinically indistinguishable from sarcoidosis. Daclizumab is a humanized monoclonal IgG1 antibody that binds to CD25 that has been studied for the treatment of multiple sclerosis (MS). During MS clinical trials of daclizumab, 12 subjects developed clinical conditions potentially consistent with sarcoidosis. Therefore, an independent adjudication committee of individuals with expertise in sarcoidosis was organized to determine the likelihood of these cases representing sarcoidosis.

      Methods

      The adjudication committee consisted of a pulmonologist, pathologist, and radiologist with clinical experience in sarcoidosis. The committee had access to the subjects' laboratory data, narratives of all suspect adverse reaction reports, radiographic imaging and histology from biopsies. A priori, a grading system was developed to determine criteria to establish the likelihood that the patient had developed sarcoidosis.

      Results

      The adjudication confirmed sarcoidosis in 11/12 subjects. The committee's decisions were unanimous in all cases. Biopsies were available in 7/11 of these. In the 4 subjects who did not have a biopsy, they all had presentations, clinical findings, and/or laboratory findings that were highly specific for sarcoidosis. Alternative causes for these clinical findings were reasonably excluded in all cases. The lung (8/11) and skin (6/11) were the most common organs involved. The mean daclizumab dose given when signs or symptoms of sarcoidosis occurred was 5413 ± 2704 mg and the median time from first daclizumab dose was 996 days. The incidence rate of developing sarcoidosis in those participating in these daclizumab trials was 154/100,000 patient-years compared with incidence rates of sarcoidosis in the United States of 3.2–17.8/100,000/year. These data suggest that these sarcoidosis cases may have represented DISRs related to daclizumab therapy.

      Conclusions

      Given the clinical presentation and subsequent evaluation of these 11 subjects, we suspect that they had DISRs from daclizumab.

      Keywords

      1. Introduction

      Sarcoidosis is a multisystem granulomatous disease of unknown cause. Several drugs have been associated with the development of drug-induced sarcoidosis-like reactions (DISRs) that are clinically indistinguishable from sarcoidosis [
      • Chopra A.
      • Nautiyal A.
      • Kalkanis A.
      • Judson M.A.
      Drug-induced sarcoidosis-like reactions.
      ]. Since the immunopathogenesis of sarcoidosis is unknown, it is not clear if these drugs are truly causing sarcoidosis, causing conditions that are distinct from sarcoidosis, rendering the immune system more susceptible to the development of sarcoidosis, or exacerbating subclinical cases of sarcoidosis [
      • Chopra A.
      • Nautiyal A.
      • Kalkanis A.
      • Judson M.A.
      Drug-induced sarcoidosis-like reactions.
      ]. Unravelling the mechanisms involved in the development of DISRs may yield important insights into the mechanisms involved in the development of sarcoidosis.
      Daclizumab is a humanized monoclonal IgG1 antibody that binds to CD25, the alpha subunit of the interleukin-2 receptor on activated T-lymphocytes [
      • Lycke J.
      Monoclonal antibody therapies for the treatment of relapsing-remitting multiple sclerosis: differentiating mechanisms and clinical outcomes.
      ]. Daclizumab blocks IL-2 binding, increasing the number of CD56bright natural killer cells [
      • Bielekova B.
      • Richert N.
      • Herman M.L.
      • et al.
      Intrathecal effects of daclizumab treatment of multiple sclerosis.
      ]. The developmental international birth date of Daclizumab was 3 July 2006 and first marketing authorization was granted on 27 May 2016. It has been used for the treatment of relapsing forms of multiple sclerosis (MS) where it has been demonstrated to decrease the annual relapse rate by 45% and reduce the number of new number of new MS lesions by more than 50% [
      • Kappos L.
      • Wiendl H.
      • Selmaj K.
      • et al.
      Daclizumab HYP versus interferon beta-1a in relapsing multiple sclerosis.
      ]. Given the nature and complexity of adverse events being reported with daclizumab and the limited number of patients being treated, in March 2018 it was determined that characterizing the evolving benefit/risk profile of daclizumab beta would not be possible going forward and daclizumab was voluntarily withdrawn from the market.
      During clinical trials of daclizumab for the treatment of MS, 12 subjects developed clinical conditions potentially consistent with sarcoidosis. Therefore, an independent adjudication committee of individuals with expertise in sarcoidosis was convened to determine the likelihood of these cases representing sarcoidosis. In this report, we describe the approach of this adjudication committee in assessing the probability that sarcoidosis developed in these subjects as well as their clinical characteristics.

      2. Methods

      As per routine pharmacovigilance practice, upon receiving complex cases reported as or having features of sarcoidosis, Biogen undertook a search of the daclizumab Clinical Trial and Global Safety Databases (GSD) for events coded to the MedDRA High Level Term: Acute and Chronic Sarcoidosis. The cases identified were reported between June 2013 and November 2016 by Investigators from 4 clinical trials (205MS203, 205MS301, 205MS302 and 205MS303). No events were reported from the postmarketing setting. Three of the 4 clinical trials were long term extension studies. Patients with a history of malignancy, human immunodeficiency virus, active viral hepatitis, recent severe bacterial, viral or fungal infection, were excluded from participation in the daclizumab clinical trials.
      To adjudicate the likelihood that reported cases represented sarcoidosis, Biogen assembled an independent expert committee. The adjudication committee was comprised of a pulmonologist with clinical experience in sarcoidosis (MAJ), an experienced pathologist (TVC) and radiologist (BE).The clinical data concerning these cases was deidentified. For all cases reported as potentially consistent with “sarcoidosis,” the adjudication committee was provided with a case adjudication summary, demographic information, clinical history of multiple sclerosis, medical history, allergy history, dosing schedule of daclizumab, all adverse events in the trial including dates of onset and resolution, all concomitant medications and dosages used during the trials, all laboratory data, and narratives of all suspect adverse reaction reports. These narratives included information on the diagnostic evaluations performed to assess for the presence of sarcoidosis. The adjudication committee was provided with radiographic reports and pathological reports when available. In addition, whenever possible, the adjudication committee was provided with the actual radiographic imaging studies as well as pertinent histological images.
      The format of case adjudication proceeded as follows. All the aforementioned clinical information was provided to each member of the adjudication committee who reviewed these clinical data independently. Each adjudication committee member determined the likelihood of the case representing sarcoidosis on the basis of the clinical data provided from the perspective of their individual specialty. The pathologists and radiologist were asked to base their likelihood assessment based on the pathological and radiographic data exclusively unless there was no material provided. Each member of the adjudication committee independently ranked the likelihood of sarcoidosis as follows: a) highly specific: sarcoidosis is the most likely diagnosis with very high confidence; b) fairly specific: sarcoidosis is the most likely diagnosis; c) consistent: sarcoidosis is a likely diagnosis but there could be an alternative diagnoses; d) equivocal: the available data neither support nor exclude sarcoidosis; e) unlikely: sarcoidosis is not a likely diagnosis but is possible; f) highly unlikely: sarcoidosis is a highly unlikely diagnosis. After each adjudication committee member had independently assessed the likelihood of a sarcoidosis, a teleconference call was conducted of all committee members concerning two to four of the cases per teleconference. During these teleconferences, the committee chairman summarized the clinical course of each subject. The pathologists discussed their assessment of the pathological data and the radiologist discussed his assessment of the imaging studies. The committee then discussed the likelihood of the case representing sarcoidosis with the following classifications: a) highly probable; b) probable; c) possible; d) unlikely. This grading system had fewer tiers than the grading system for the independent evaluations in order to maximize the chance of reaching a unanimous committee decision. Highly probable or probable was considered adequate to consider the case as representing sarcoidosis. The committee attempted for the consensus to be unanimous, but relied on a majority vote if there was a disagreement in the likelihood classification. Sarcoidosis organ involvement was determined by the World Association of Sarcoidosis and Other Granulomatous Disorders (WASOG) Organ Assessment Instrument [
      • Judson M.A.
      • Costabel U.
      • Drent M.
      • et al.
      The WASOG Sarcoidosis Organ Assessment Instrument: an update of a previous clinical tool.
      ].

      3. Results

      The sponsor provided the adjudication committee with 12 events (12 patients) that coded to the Medical Dictionary for Regulatory Activities (MedDRA) High Level Term Acute or Chronic Sarcoidosis. The onset dates of the events occurred during the period June 2013 to November 2016. In 8 of the 12 cases, histologic material was available for review, and in 5 of the 12 cases imaging studies were available. Table 1 shows the assessments of the likelihood of a DZISR by the pathologist, radiologist, and clinician on the basis of the available histology, imaging studies, and clinical data respectively. Table 1 also shows the consensus assessment of the adjudication committee for the likelihood of each case representing a sarcoidosis. All of these assessments were unanimous. The committee assessed 11 of the 12 cases as highly probable or probable sarcoidosis cases which met the a priori diagnostic criteria for sarcoidosis [
      • Judson M.A.
      • Costabel U.
      • Drent M.
      • et al.
      The WASOG Sarcoidosis Organ Assessment Instrument: an update of a previous clinical tool.
      ].
      Table 1Grading of sarcoidosis probability.
      Adjudicator# cases available
      Case defined as available for the pathologist and radiologist only if there was at least one pathological specimen and at least one chest imaging study to review respectively.
      grading#, % with each grade (6 grades)
      Pathologist8Highly specific0, 0%
      Fairly specific5, 62.5%
      consistent2, 25%
      equivocal1, 12.5%
      Unlikely0, 0%
      Highly unlikely0, 0%
      radiologist5Highly specific0, 0%
      Fairly specific3, 50%
      Consistent1, 17%
      Equivocal1, 17%
      Unlikely1, 17%
      Highly unlikely0, 0%
      clinician12Highly specific1, 8.3%
      Fairly specific7, 58.3%
      Consistent2, 16.7%
      Equivocal2, 16.7%
      Unlikely0, 0%
      Highly unlikely0, 0%
      Consensus12Highly probable3, 25%
      Probable8, 66.7%
      Possible1, 8.3%
      Unlikely0, 0%
      a Case defined as available for the pathologist and radiologist only if there was at least one pathological specimen and at least one chest imaging study to review respectively.
      A total of 2236 patients were participating in daclizumab trials conducted by the sponsor between June 2013 and November 2016, representing 7771 person-years of exposure to daclizumab. Therefore, 11 cases adjudicated to represent sarcoidosis represented 0.49% of those receiving daclizumab. The one case that was ajudicated as possible sarcoidosis was of a patient with inguinal adenopathy where the biopsy of the inguinal lymph node did not reveal granulomatous inflammation (pathologist graded as “equivocal”) and there were no imaging studies available. In the 6 of 11 patients adjudicated as having sarcoidosis, pathological samples were sent for stains and/or cultures for mycobacteria, and all were negative. .
      The demographics and clinical characteristics of these 11 patients is shown in Table 2. There was no obvious gender predilection, and the median age of the patients was in the early 40's. Table 3 shows the specific organs involved in the 11 cases adjudicated to be sarcoidosis and the number of organs per case. The lung and skin were the most common organs involved. A total of 9 biopsies revealing granulomas were performed in 7 of the 11 sarcoidosis cases. Fig. 1 shows a representative histologic specimen in one of the patients. Although 4 of the cases did not have a biopsy performed, the committee classified them as probably representing sarcoidosis. In 2 of these 4 cases, bilateral hilar adenopathy was present on chest imaging and the patients had developed erythema nodosum skin lesions. In the third patient without a biopsy, chest CT scan was available for review and showed lung nodules that were perilymphatic in distribution with symmetric mediastinal and hilar adenopathy that is typical of sarcoidosis (Fig. 2A and B). The fourth patient without a biopsy had nodular opacities and symmetric mediastinal and hilar adenopathy on chest CT, and the radiographic findings were unchanged for a year. In addition, the clinical course of these 4 patients was consistent with sarcoidosis and there was no alternative cause for the clinical findings.
      Table 2Demographics and clinical characteristics (N = 11).
      Age, years42.4 ± 10.7, median age 41
      SexFemale (7, 64%)
      Male (4, 36%)
      Smoking historyCurrent: 2 (18%)
      Not current: 9 (82%)
      Years with multiple sclerosis until onset of sarcoidosis symptoms (N = 10)
      Missing data on one patient.
      median = 11 years (range: 3–21 years)
      a Missing data on one patient.
      Table 3Daclizumab-induced sarcoidosis-like reaction, organ involvement, N = 11.
      OrganClinically involved
      Multiple organ involvement possible in individual patient.
      (N,%)
      Biopsied (N,%)
      Lung8, 73%2, 18%
      Skin6, 55%5, 45%
      Kidney1, 7%1, 7%
      Peripheral lymph node1,7%1,7%
      Number of organ involved (N,%)
      17, 64%
      23, 27%
      31, 9%
      a Multiple organ involvement possible in individual patient.
      Fig. 1
      Fig. 1Transbronchial lung biopsy from one patient in this cohort (Hematoxylin and Eosin, 200× magnification). The histology revealed noncaseating granulomatous inflammation.
      Fig. 2
      Fig. 2Coronal reformatted CT (A) with intravenous contrast on mediastinal windows demonstrates symmetric bilateral hilar (white arrows), right paratracheal (yellow arrow) and aorticopulmonary window (red arrow) lymphadenopathy. Axial image (B) on lung windows demonstrates bilateral lung nodules that are predominantly subpleural in distribution (black arrows). This combination of findings is typical of sarcoidosis. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
      In terms of describing the clinical manifestations of the 11 sarcoidosis cases rigorously, the adjudication committee was hampered by several issues. First, the cases came from disparate medical centers so that the methodology of the radiographic, pathologic, and laboratory testing was not uniform. Second, there was no standard protocol for evaluating these sarcoidosis cases so that patients underwent different clinical evaluations and laboratory testing. Third, some histologic and radiographic data was available by report only without access to imaging. Given these limitations, a brief description of pertinent clinical data concerning these sarcoidosis patients follows. Ten of these patients underwent chest CT scanning, with 7 (70%) showing mediastinal lymphadenopathy and lung nodules, 1 (10%) demonstrating mediastinal adenopathy without lung nodules, 1 (10%) revealing rare scattered nodules (this patient had a skin lesion biopsies showing noncaseating granulomatous inflammation), and 1 (10%) had a normal chest CT scan (this patient also had a skin lesion biopsies showing noncaseating granulomatous inflammation). Two of 11 (22%) of the sarcoidosis patients presented erythema nodosum skin lesions. Two of 11 (22%) underwent a bronchoalveolar lavage (BAL) and both demonstrated a BAL lymphocytosis of greater than 30% (37% and 55%). Serum angiotensin enzyme levels were measured in two patients and was elevated in one of them. One patient developed hypercalciuria.
      Table 4 displays the daclizumab dose and duration of daclizumab therapy prior to the first sign or symptoms of sarcoidosis. The mean daclizumab dose was 5413 ± 2704 mg and the median time from first daclizumab dose to first sign of symptom of sarcoidosis was 996 days. There was inadequate data to assess the outcome of therapy on the development of sarcoidosis in this cohort, as in 8 of 11 of these cases daclizumab was permanently discontinued and the subjects were withdrawn from the studies. Eight of 11 of these patients received corticosteroid therapy (systemic in 6, topical in 2) while the remaining 3 did not receive anti-granulomatous therapy. The outcome for 3 of events was reported as resolved, and for the remaining 8 events the outcome was reported as not resolved.
      Table 4Daclizumab dosing until development of sarcoidosis-like reaction.
      Mean ± SD, (median) [min, max]
      Cumulative dose until symptom onset (mg)5413 ± 2704 (mg) [1,350, 10,800]
      Cumulative time until symptom onset (days)1106 ± 135 (996) [419, 2014]

      4. Discussion

      The adjudication committee identified 11 subjects treated with daclizumab who developed sarcoidosis, defined by a priori criteria [
      • Judson M.A.
      • Costabel U.
      • Drent M.
      • et al.
      The WASOG Sarcoidosis Organ Assessment Instrument: an update of a previous clinical tool.
      ]. Seven of these cases were confirmed by histologic evidence of noncaseating granulomatous inflammation coupled with a compatible clinical history of sarcoidosis, whereas the remaining cases had either very specific clinical features of sarcoidosis or very specific chest imaging findings. It should be noted that the diagnosis of sarcoidosis cannot be conclusively established by histologic or radiographic findings, and these cases were classified as representing sarcoidosis on the basis of additional information, including the lack of reasonable alternative diagnosis and the duration of follow-up that exceeded the expected time for a clinically significant infection, vasculitis, or malignancy to develop. In addition, although the investigators in these daclizumab trials did not provide conclusive evidence to exclude specific alternative causes of granulomatous inflammation in these cases, patients were excluded in participating in these trials if they had a previous history of malignancy or a recent history of a significant viral, bacterial or fungal infection. In addition, 6 of the 11 cases had stains and or cultures of granulomatous tissue that were negative for mycobacteria. We do acknowledge that case reports from clinical trials of immunomodulator products often lack complete information, making vigilance and a multidisciplinary adjudication approach relevant for accurate evaluation and conclusions.
      Although we cannot exclude that these 11 cases were true cases of sarcoidosis that coincidentally occurred in conjunction with daclizumab therapy, we suspect that these cases represent daclizumab-induced DISRs. The probability that non-daclizumab-related sarcoidosis occurred in 11 separate cases is minimal, as the incidence rate of developing a sarcoidosis in those receiving daclizumab in one of these trials was 142/100,000 patient-years compared with incidence rates of sarcoidosis in the United States varying from 3.2 to 17.8/100,000/year [
      • Baughman R.P.
      • Field S.
      • Costabel U.
      • et al.
      Sarcoidosis in America. Analysis based on health care use.
      ]. In addition, it is possible that there is an association between multiple sclerosis and sarcoidosis. We know of no such association, except in reported cases of sarcoidosis developing in multiple sclerosis patients [
      • Carbonelli C.
      • Montepietra S.
      • Caruso A.
      • et al.
      Sarcoidosis and multiple sclerosis: systemic toxicity associated with the use of interferon-beta therapy.
      ,
      • Petousi N.
      • Thomas E.C.
      Interferon-beta-induced pulmonary sarcoidosis in a 30-year-old woman treated for multiple sclerosis: a case report.
      ,
      • Sahraian M.A.
      • Moghadasi A.N.
      • Owji M.
      • et al.
      Cutaneous and pulmonary sarcoidosis following treatment of multiple sclerosis with interferon-beta-1b: a case report.
      ] which has been postulated to be associated with DISRs [
      • Chopra A.
      • Nautiyal A.
      • Kalkanis A.
      • Judson M.A.
      Drug-induced sarcoidosis-like reactions.
      ].
      The lung and skin were the most common organs involved with these 11 patients, which is similar to what has been found with DISRs from other drugs [
      • Chopra A.
      • Nautiyal A.
      • Kalkanis A.
      • Judson M.A.
      Drug-induced sarcoidosis-like reactions.
      ]. On average, signs or symptoms of sarcoidosis did not occur until after more than 2 years of receiving daclizumab therapy. There was inadequate follow-up data concerning this cohort to assess the effect of discontinuing daclizumab or anti-granulomatous therapy on the outcome of these cases.
      The mechanism by which daclizumab could cause a DISR in these individuals is unclear. As these granulomatous reactions were detected in a very small percentage of those who received daclizumab, it is likely that those who developed this complication were predisposed to do so, perhaps on a genetic basis. Sarcoidosis patients have been shown to have an increased number of natural T-cell fractions [
      • Tondell A.
      • Ro A.D.
      • Borset M.
      • Moen T.
      • Sue-Chu M.
      Activated CD8+ T cells and natural killer T cells in bronchoalveolar lavage fluid in hypersensitivity pneumonitis and sarcoidosis.
      ] and, specifically, an increased number of natural killer cells with CD56 activation markers [
      • Snyder-Cappione J.E.
      • Nixon D.F.
      • Chi J.C.
      • et al.
      Invariant natural killer T (iNKT) cell exhaustion in sarcoidosis.
      ]. As daclizumab is known to increase the number of CD56bright natural killer cells, this is a plausible mechanism by daclizumab could lead to the development of sarcoidosis. It is also theoretically possible that daclizumab could stimulate the immune system by pathways that are currently unknown, resulting in immune-related adverse events (irAEs). irAEs may manifest in a variety of disorders and abnormalities including rheumatoid arthritis, immune thrombocytopenia, Sjogren syndrome, polymyalgia rheumatica, psoriatic arthritis, seronegative polyarthritis and DISRs [
      • Chen T.W.
      • Razak A.R.
      • Bedard P.L.
      • Siu L.L.
      • Hansen A.R.
      A systematic review of immune-related adverse event reporting in clinical trials of immune checkpoint inhibitors.
      ,
      • Le Burel S.
      • Champiat S.
      • Mateus C.
      • et al.
      Prevalence of immune-related systemic adverse events in patients treated with anti-Programmed cell Death 1/anti-Programmed cell Death-Ligand 1 agents: a single-centre pharmacovigilance database analysis.
      ,
      • Postow M.A.
      • Sidlow R.
      • Hellmann M.D.
      Immune-related adverse events associated with immune checkpoint blockade.
      ].
      In summary, we report 11 individuals with multiple sclerosis who developed a syndrome consistent with sarcoidosis while receiving prolonged therapy with daclizumab. We are aware of only one recently published case of a sarcoidosis-like reaction in a 45-year-old man whose clinical presentation was similar to our series of cases [
      • Rhone E.E.
      • Cho P.S.P.
      • Birring S.S.
      • Galloway J.
      • Silber E.
      Pulmonary sarcoidosis in a patient with multiple sclerosis on daclizumab monotherapy.
      ]. On the basis of the incidence rates of sarcoidosis in the United States [
      • Baughman R.P.
      • Field S.
      • Costabel U.
      • et al.
      Sarcoidosis in America. Analysis based on health care use.
      ], it is unlikely that these individuals developed sarcoidosis by chance alone. Therefore, we suspect that these cases could represent DISRs from daclizumab. Understanding the mechanisms causing DISRs may yield important insights into the immunopathogenesis of sarcoidosis.

      Conflicts of interest

      MAJ, BME, TVC, and SK were consultants for Biogen. EdW, SC, CP SK PS are employed by Biogen.
      Maj was has also received grant funding for his institution from Novartis and Mallinckrodt.

      Acknowledgement

      The authors wish to acknowledge the expertise of Dr. James G. Krueger, a dermatologist who consulted specifically on some of the DZISR cases with skin involvement. Biogen funded the adjudication process described in this manuscript.

      References

        • Chopra A.
        • Nautiyal A.
        • Kalkanis A.
        • Judson M.A.
        Drug-induced sarcoidosis-like reactions.
        Chest. 2018; 154: 664-677
        • Lycke J.
        Monoclonal antibody therapies for the treatment of relapsing-remitting multiple sclerosis: differentiating mechanisms and clinical outcomes.
        Therapeutic advances in neurological disorders. 2015; 8: 274-293
        • Bielekova B.
        • Richert N.
        • Herman M.L.
        • et al.
        Intrathecal effects of daclizumab treatment of multiple sclerosis.
        Neurology. 2011; 77: 1877-1886
        • Kappos L.
        • Wiendl H.
        • Selmaj K.
        • et al.
        Daclizumab HYP versus interferon beta-1a in relapsing multiple sclerosis.
        N. Engl. J. Med. 2015; 373: 1418-1428
        • Judson M.A.
        • Costabel U.
        • Drent M.
        • et al.
        The WASOG Sarcoidosis Organ Assessment Instrument: an update of a previous clinical tool.
        Sarcoidosis Vasc. Diffuse Lung Dis. 2014; 31: 19-27
        • Baughman R.P.
        • Field S.
        • Costabel U.
        • et al.
        Sarcoidosis in America. Analysis based on health care use.
        Annals of the American Thoracic Society. 2016; 13: 1244-1252
        • Carbonelli C.
        • Montepietra S.
        • Caruso A.
        • et al.
        Sarcoidosis and multiple sclerosis: systemic toxicity associated with the use of interferon-beta therapy.
        Monaldi Arch. Chest Dis. 2012; 77: 29-31
        • Petousi N.
        • Thomas E.C.
        Interferon-beta-induced pulmonary sarcoidosis in a 30-year-old woman treated for multiple sclerosis: a case report.
        J. Med. Case Rep. 2012; 6: 344
        • Sahraian M.A.
        • Moghadasi A.N.
        • Owji M.
        • et al.
        Cutaneous and pulmonary sarcoidosis following treatment of multiple sclerosis with interferon-beta-1b: a case report.
        J. Med. Case Rep. 2013; 7: 270
        • Tondell A.
        • Ro A.D.
        • Borset M.
        • Moen T.
        • Sue-Chu M.
        Activated CD8+ T cells and natural killer T cells in bronchoalveolar lavage fluid in hypersensitivity pneumonitis and sarcoidosis.
        Sarcoidosis Vasc. Diffuse Lung Dis. 2015; 31: 316-324
        • Snyder-Cappione J.E.
        • Nixon D.F.
        • Chi J.C.
        • et al.
        Invariant natural killer T (iNKT) cell exhaustion in sarcoidosis.
        Eur. J. Immunol. 2013; 43: 2194-2205
        • Chen T.W.
        • Razak A.R.
        • Bedard P.L.
        • Siu L.L.
        • Hansen A.R.
        A systematic review of immune-related adverse event reporting in clinical trials of immune checkpoint inhibitors.
        Ann. Oncol.: official journal of the European Society for Medical Oncology. 2015; 26: 1824-1829
        • Le Burel S.
        • Champiat S.
        • Mateus C.
        • et al.
        Prevalence of immune-related systemic adverse events in patients treated with anti-Programmed cell Death 1/anti-Programmed cell Death-Ligand 1 agents: a single-centre pharmacovigilance database analysis.
        Eur. J. Cancer. 2017; 82: 34-44
        • Postow M.A.
        • Sidlow R.
        • Hellmann M.D.
        Immune-related adverse events associated with immune checkpoint blockade.
        N. Engl. J. Med. 2018; 378: 158-168
        • Rhone E.E.
        • Cho P.S.P.
        • Birring S.S.
        • Galloway J.
        • Silber E.
        Pulmonary sarcoidosis in a patient with multiple sclerosis on daclizumab monotherapy.
        Multiple sclerosis and related disorders. 2018; 20: 25-27