- Email: [email protected]
Immune mediated neuropathy following checkpoint immunotherapy
Journal of Clinical Neuroscience xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Review article
Immune mediated neuropathy following checkpoint immunotherapy Yufan Gu a,⇑, Alexander M. Menzies b,c,d, Georgina V. Long b,c,d, S.L. Fernando c,e,f, G. Herkes a,c a
Department of Neurology, Royal North Shore Hospital, Sydney, NSW, Australia Melanoma Institute of Australia, Sydney, NSW, Australia c The University of Sydney, Sydney, NSW, Australia d Department of Medical Oncology, Royal North Shore Hospital, Sydney, NSW, Australia e Department of Clinical Immunology and Allergy, Royal North Shore Hospital, Sydney, NSW, Australia f ImmunoRhuematology Laboratory, Pathology North, Sydney, NSW, Australia b
a r t i c l e
i n f o
Article history: Received 9 November 2016 Accepted 11 July 2017 Available online xxxx Keywords: Checkpoint immunotherapy CTLA-4, PD-1 Ipilimumab Nivolumab Neuropathy Guillain-Barre
a b s t r a c t Checkpoint immunotherapy has revolutionised cancer therapy and is now standard treatment for many malignancies including metastatic melanoma. Acute inflammatory neuropathies, often labelled as Guillain-Barre syndrome, are an uncommon but potentially severe complication of checkpoint immunotherapy with individual cases described but never characterised as a group. We describe a case of acute sensorimotor and autonomic neuropathy following a single dose of combination ipilimumab and nivolumab for metastatic melanoma. A literature search was performed, identifying 14 other cases of acute neuropathy following checkpoint immunotherapy, with the clinical, electrophysiological and laboratory features summarised. Most cases described an acute sensorimotor neuropathy (92%) with hyporeflexia (92%) that could occur from induction up till many weeks after the final dose of therapy. In contrast to Guillain-Barre syndrome, the cerebrospinal fluid (CSF) analysis often shows a lymphocytic picture (50%) and the electrophysiology showed an axonal pattern (55%). Treatment was variable and often in combination. 11 cases received steroid therapy with only 1 death within this group, whereas of the 4 patients who did not receive steroid therapy there were 3 deaths. In conclusion checkpoint immunotherapy – induced acute neuropathies are distinct from and progress differently to GuillainBarre syndrome. As with other immunotherapy related adverse events corticosteroid therapy should be initiated in addition to usual therapy. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction
2. Methods
Checkpoint immunotherapies are now standard treatment for many cancers including advanced melanoma, lung cancer, kidney cancer, head and neck cancer, and bladder cancer [1]. These therapies remove T-cell inhibitory pathways resulting in upregulation of the anti-tumour immune response, with the potential consequence being immune-related adverse events (irAEs) in normal tissues [2]. Peripheral neuropathy is uncommon but cases of severe ‘‘GuillainBarre syndrome” (GBS) resulting in death have been reported.
We provide a detailed description of a case of acute sensorimotor and autonomic polyneuropathy following administration of a single dose of combination ipilimumab and nivolumab for metastatic melanoma. A MEDLINE search was performed of previous publications in English for text word and MeSH headings of ‘‘neuropathy”, ‘‘Guillain-Barre Syndrome” and ‘‘ipilimumab”, ‘‘nivolumab”, ‘‘pembrolizumab”, ‘‘CTLA-4” or ‘‘PD-1”. The references and citations of retrieved articles were all reviewed with relevant cases included for analysis of reported clinical, investigational and treatment findings.
⇑ Corresponding author at: Department of Neurology – Royal North Shore Hospital, Reserve Rd, St Leonards, 2065 New South Wales, Australia. E-mail addresses: [email protected] (Y. Gu), alexander.menzies@ sydney.edu.au (A.M. Menzies), [email protected] (G.V. Long), suran. [email protected] (S.L. Fernando), [email protected] (G. Herkes).
3. Case vignette A 49-year-old woman with NRAS-mutant metastatic melanoma to chest wall, lung, and retroperitoneal and axillary nodes was commenced on combination ipilimumab (3 mg/kg) and nivolumab (1 mg/kg). Five days following induction she developed painful
http://dx.doi.org/10.1016/j.jocn.2017.07.014 0967-5868/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Gu Y et al. Immune mediated neuropathy following checkpoint immunotherapy. J Clin Neurosci (2017), http://dx.doi. org/10.1016/j.jocn.2017.07.014
2
Y. Gu et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx
paresthesiae in the extremities, progressing over four days to a symmetrical sensorimotor and autonomic neuropathy with proximal loss of antigravity power and loss of independent mobility (Fig. 1. Clinical course). Reflexes were absent and sensory loss was present to the upper arms and knees bilaterally. Two days after her developing sensory symptoms, she experienced a fever and a brief episode of diarrhea. Blood and stool cultures were negative and symptoms resolved promptly. Initial electrodiagnostics suggested an acute generalized motor predominant neuropathy with patchy slowing of conduction velocity (Table 1). MRI of the whole spine was normal. Cerebrospinal fluid (CSF) analysis showed increased protein (1.15 g/L), 15 white blood cells (93% lymphocytes), one erythrocyte, no oligoclonal bands, negative cryptococcal antigens, normal culture and cytology. Serum anti-GM1 ganglioside antibodies were negative. Routine full blood count, biochemistry, glucose, B12, folate, renal, liver and thyroid function tests were normal. Serology for HIV and syphilis were negative. Serum immunofixation showed no monoclonal protein. The ANA and ANCA were negative, and levels of CRP, ESR, C3 and C4 were normal. She commenced IV immunoglobulin (IVIG, 0.4 g/kg/d for 5 days) and IV methylprednisolone (1 g/d for 5 days, then 500 mg/d for 3 days) followed by tapering oral prednisone (1 mg/ kg/d). Symptoms stabilized with mild improvement yet one month later she developed worsening weakness and ongoing painful paresthesia. She also developed persistent nausea coupled with postural hypotension and constipation. After the exclusion of structural and metabolic causes, this was thought to represent an autonomic neuropathy. Another course of IVIG and IV methylprednisone were given with slight but non-sustained improvement. Repeat electrodiagnostics demonstrated an axonal predominant sensorimotor neuropathy. A third course of IV methylprednisone and plasma exchange (PLEX, five exchanges over two weeks, followed by weekly exchanges) was commenced, oral corticosteroids were continued and mycophenolate (1 g bd) was added. Immediate improvements were noted in sensory and autonomic symptoms, followed by a gradual improvement in motor strength. By 6 weeks after commencing PLEX (12 weeks after initial treatment) the patient had only mild weakness and could perform her usual daily activities, and at four months she returned to work with her melanoma in deep partial response: complete response in several lesions, and with only a chest wall and lung metastasis remaining (both smaller than at baseline and with a cystic central component). Prednisone was weaned to 10mg daily and mycophenolate was continued at 1 g BD. PLEX was ceased at five months after initial checkpoint immunotherapy after her Vascath was complicated by a Pseudomonas aeruginosa infection necessitating removal. At seven months her cystic chest wall and lung lesions slightly increased in size, with aspiration and partial resection of the chest wall lesion
Table 1 Nerve conduction studies.
Median nerve – Right Motor Latency, APB-Wrist CMAP amplitude, wrist CMAP amplitude, elbow Motor CV SNAP amplitude Ulnar nerve – Right Motor Latency, ADM–Wrist CMAP amplitude, wrist CMAP amplitude, above elbow Motor CV SNAP amplitude Peroneal nerve – Right Motor Latency, EDB-Ankle CMAP amplitude, ankle CMAP amplitude, fibular head Motor CV Tibial nerve – Right Motor Latency, AH-Ankle CMAP amplitude, ankle CMAP amplitude, knee Motor CV Sural nerve Right SNAP amplitude Left SNAP amplitude
Day 4
Day 39
Day 226
5.2 6.8 6.0 42 14
5.1 2.9 2.4 33 2
4.3 6.3 6.3 48 4
3.5 10.4 9.5 50 20
3.4 5.1 4.2 52 NR
3.1 11.1 10.3 53 3
5.6 3.8 2.8 37
4.4 0.3 0.3 40
5.1 2.1 1.3 41
8.0 0.4 0.3 46
11 13
5.0 2.0
4.0
CMAP: Compound muscle action potential, CV: Conduction velocity, SNAP: Sensory nerve action potential.
finding pigment laden macrophages only. Mycophenolate was weaned to 500 mg daily and prednisone maintained at 10 mg daily. At nine months the two remaining melanoma metastases (lung and chest wall) started to enlarge with a cystic central component. Due to this fact, the extended time interval from initial toxicity, and because she was minimally symptomatic with electrodiagnostics showing partial recovery, her remaining mycophenolate was ceased. Within one week of cessation sensory neurological symptoms flared and corticosteroid and mycophenolate was recommenced. This improved symptoms and sequential imaging demonstrated further regression of melanoma. After a few months of clinical stability (prednisone 25 mg daily, mycophenolate 500 mg bd), tapering of her steroid dose was recommenced which soon resulted in increasing pain, weakness and glove and stocking numbness to her elbows and knees. With a further poor response to IVIG and 50 mg per day of prednisone, she was commenced on rituximab infusion (375 mg/m2 weekly for four weeks). There was no clinical response after rituximab infusion and she has since recommenced plasma exchange with an increased dose of mycophenolate (1.5 g bd) and prednisone. In summary we describe a case of acute peripheral neuropathy occurring after the administration of combination checkpoint immunotherapy for metastatic melanoma. The case provides a
Fig. 1. A graphical representation of the patient’s motor polyneuropathy and the clinical response to acute therapies. MRC sum score: Medical Research Council sum score comprised of the sum of the MRC grades of deltoid, biceps, wrist extensor, iliopsoas, quadriceps femoris and tibialis anterior bilaterally (0–60). IVIG: Intravenous Immunoglobulin. PLEX: Plasma exchange. IV MP: Intravenous Methylprednisone. PO: per os; oral administration.
Please cite this article in press as: Gu Y et al. Immune mediated neuropathy following checkpoint immunotherapy. J Clin Neurosci (2017), http://dx.doi. org/10.1016/j.jocn.2017.07.014
Y. Gu et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx
compelling example of an upregulated immune response providing efficacious cancer therapy but also resulting in autoimmunity necessitating ongoing immunosuppression. With long term follow-up we also observe a striking response to steroids, plasma exchange and mycophenolate over other therapies such as IVIG and rituximab. 4. Discussion There are many barriers to an effective immune response against a tumour, complicated by the tumours ability to ‘‘co-opt” certain interacting pathways to its advantage [1,2]. CTLA-4 is expressed on activated T cells and functions as an inhibitory checkpoint receptor early in the immune response at the time of antigen presentation. It outcompetes the costimulatory molecule CD28 for the ligands, CD80 and CD 86, resulting in the downregulation of immune responses by T cells. CTLA 4 also is expressed on T regulatory cells and drives their suppressor functions. Costimulatory molecules that result in T cell activation are also down regulated by tumour cells. In addition, T cells that migrate into the tumour stroma encounter regulatory T cells leading to upregulation of inhibitory receptors, apoptosis, anergy, or exhaustion. PD-1 works predominantly within the tumour microenvironment. It is highly expressed on tumour infiltrating T cells and its ligands are upregulated in many cancers including melanoma. Importantly, it directly inhibits T cell receptor-mediated effector functions. While the benefits of immunotherapy are improved antitumour immunity, with high response rates and durable survival, the downside is the frequent occurrence of irAEs which can occur in any organ. These irAEs appear histopathologically most often as lymphocytic infiltrates [3–5] and are best managed with systemic corticosteroids, often at high doses and with a prolonged wean over many weeks [3]. Occasionally additional immunosuppressants e.g. TNFa inhibitors, mycophenolate, antithymocyte globulin are also required [3,6]. Ipilimumab, a monoclonal antibody that blocks CTLA-4, and the anti-PD-1 antibodies nivolumab and pembrolizumab, all have activity in melanoma, both as monotherapy, and more recently in combination [7–10]. Ipilimumab is associated with irAEs in up to 60% of patients (15% grade 3), that mostly affect the skin and gastro-intestinal tract, however any organ system can be affected. Inflammatory neurological toxicities have been reported, including cases that appear similar to idiopathic Guillain-Barre syndrome, granulomatous inflammation of the CNS, aseptic meningitis, chronic inflammatory demyelinating polyneuropathy, transverse myelitis, myositis and myasthenia gravis [11,12]. The anti-PD-1 antibodies are less likely to cause irAEs but combination immunotherapy with ipilimumab and nivolumab, which appears to have greater efficacy than monotherapy, correspondingly results in greater toxicity (55% grade 3 toxicity) [7]. Our PubMed search identified 12 articles describing 14 cases of neuropathy. Cases were excluded if clinical and/or investigational characteristics were unavailable for comparison. Due to variable levels of detail in the reported cases, results were calculated from those available for analysis.
3
this case series (6/11, 55%) presented with an axonal pattern on EDX. Two studies described conduction block despite an axonal pattern, and a ‘pseudo-conduction block’ was hypothesized to be an explanation [4]. Hence conduction block in the absence of other EDX features of demyelination warrants repeat testing to determine the pattern of nerve injury. CSF studies showed that in 92% (11/12) protein was elevated and in 50% (6/12) a pleocytosis was seen, with five specifically described as lymphocyte predominant. 6. Treatment The estimated incidence of neuropathies as a complication is less than 1% and in most cases transient [3]. The recommendation for grade 2 neuropathy is to hold a dose of treatment, or a threefour week tapering dose of corticosteroids. For grade 3 or 4 complications a longer course of systemic corticosteroids (1–2 mg/kg daily) has been recommended [3]. Our literature search is biased towards the more severe cases of neuropathy and in many cases a combination of therapies for the neuropathy was utilised. Eleven patients (73%) received some form of steroid therapy, reflecting current recommendations, and generally had favourable outcomes with only one death (9%). Mortality in the four who did not receive steroid therapy was 75% with two cases receiving IVIG alone and another receiving a combination of IVIG and PLEX. The remaining mortality case had received a combination of IV methylprednisone, PLEX, infliximab and tacrolimus. Our case which experienced limited response despite several doses of IVIG and a prolonged course of high dose steroids, seemingly demonstrating a durable response following initiation of PLEX. We also describe the successful use of mycophenolate as a corticosteroid-sparing agent, further evidenced by recurrence of symptoms when mycophenolate was ceased. The response to therapy thus appears distinct from idiopathic Guillain-Barre syndrome where firstly there is no evidence for the use of steroids [13]. IVIG is also often used first line for acute inflammatory neuropathies as with GBS it is thought to be equally effective as plasma exchange [14]. We speculate whether the underlying mechanism of autoimmune neuropathy following checkpoint immunotherapy relates to the upregulated T cell functions and therefore demonstrated limited clinical response to IVIG and rituximab. It was however more amenable to PLEX, perhaps due to bulk removal of pathologic substances such as autoantibodies, immune complexes, and cytokines. PLEX may also have an immunomodulatory effect beyond the removal of the immunological molecules such as suppression of IL-2 and IFN-c production and enhanced regulatory T cell function [15]. Autoantibodies could be formed, especially with CTLA-4 therapy, as a result of T cell dependent-activation of B cells. Thus far CTLA-4 blockade in mouse models has demonstrated cellular infiltration and demyelination [16]. Future studies in animal models and patients with presentations similar to our case might elucidate the cytokines present in the plasma and if detected, the correlation between these levels, PLEX treatments and clinical status can be examined. Flow cytometry to T cell phenotypes including regulatory T cells or the detection of an autoantibody may reveal more information.
5. Clinical features 7. Toxicity-response relationship Most cases were male (10/15), six (40%) were diagnosed as GBS and three as chronic inflammatory demyelinating polyneuropathy (CIDP). Mean onset of symptoms was 6.5 weeks (range from one to seventeen weeks) from induction of therapy. The neuropathy was usually sensory and motor (12/13, 92%) and deep tendon reflexes were usually absent or reduced (11/12, 92%). Whilst classical GBS is associated with demyelination, a high proportion of patients in
Notably the cancer response is remarkable with eight patients with tumour regression or whom survived to attend long term follow-up, despite early cessation of treatment. Only one case documented cancer progression. This results from pro-inflammatory effects abrogating the actions of inhibitory checkpoint molecules causing both anti-tumour and autoimmune effects. This
Please cite this article in press as: Gu Y et al. Immune mediated neuropathy following checkpoint immunotherapy. J Clin Neurosci (2017), http://dx.doi. org/10.1016/j.jocn.2017.07.014
4
Y. Gu et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx
‘toxicity-response’ relationship is supported by data demonstrating that those with grade 3/4 and grade 1/2 irAEs had better response rates (28% and 22%, respectively) compared to the 2% response rate in those without any irAE [17].
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jocn.2017.07.014.
8. Conclusion
References
Given the growing importance of checkpoint immunotherapy in cancer therapeutics, it is essential that clinicians are able to recognise the uncommon neurological adverse events, especially as they are immune related and amenable to treatment. Our literature review is limited by the variable amounts of detail with which published cases have been described, yet already identifies that a lymphocytic CSF and axonal EDX pattern are common, which is distinct from idiopathic GBS. Larger analysis or case series are needed to characterize this neuropathy with more certainty, particularly in regards to optimal treatment. Nevertheless, based on this series of 15 cases in the literature, we would recommend the use of corticosteroid therapy. The next choice of immunosuppressive agent is more difficult, but we would suggest that PLEX may be more effective than IVIG, and mycophenolate was successfully used as a steroid sparing agent in this case. Additionally, the review supports findings in the literature that irAEs are associated with a more favourable cancer prognosis, arguing for early and active treatment.
[1] Drake CG, Lipson EJ, Brahmer JR. Breathing new life into immunotherapy: review of melanoma, lung and kidney cancer. Nat Rev Clin Oncol 2014;11:24–37. [2] Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer 2012;12:252–64. [3] Weber JS, Kahler KC, Hauschild A. Management of immune-related adverse events and kinetics of response with ipilimumab. J Clin Oncol 2012;30:2691–7. [4] Manousakis G, Koch J, Sommerville RB, et al. Multifocal radiculoneuropathy during ipilimumab treatment of melanoma. Muscle Nerve 2013;48:440–4. [5] Beck KE, Blansfield JA, Tran KQ, et al. Enterocolitis in patients with cancer after antibody blockade of cytotoxic T-lymphocyte-associated antigen 4. J Clin Oncol 2006;24:2283–9. [6] Chmiel KD, Suan D, Liddle C, et al. Resolution of severe ipilimumab-induced hepatitis after antithymocyte globulin therapy. J Clin Oncol 2011;29:e237–40. [7] Larkin J, Chiarion-Sileni V, Gonzalez R, et al. Combined nivolumab and ipilimumab or monotherapy in untreated melanoma. N Engl J Med 2015;373:23–34. [8] Robert C, Long GV, Brady B, et al. Nivolumab in previously untreated melanoma without BRAF mutation. N Engl J Med 2015;372:320–30. [9] Robert C, Schachter J, Long GV, et al. Pembrolizumab versus Ipilimumab in Advanced Melanoma. N Engl J Med 2015;372:2521–32. [10] Hodi FS, O’Day SJ, McDermott DF, et al. Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med 2010;363:711–23. [11] Voskens CJ, Goldinger SM, Loquai C, et al. The price of tumor control: an analysis of rare side effects of anti-CTLA-4 therapy in metastatic melanoma from the ipilimumab network. PloS One 2013;8:e53745. [12] Liao B, Shroff S, Kamiya-Matsuoka C, et al. Atypical neurological complications of ipilimumab therapy in patients with metastatic melanoma. Neuro-oncology 2014;16:589–93. [13] Hughes RA, van Doorn PA. Corticosteroids for Guillain-Barre syndrome. The Cochrane database of systematic reviews. 2012:Cd001446. [14] Hughes RA, Swan AV, van Doorn PA. Intravenous immunoglobulin for GuillainBarre syndrome. The Cochrane database of systematic reviews. 2014: Cd002063. [15] Winters JL. Plasma exchange: concepts, mechanisms, and an overview of the American Society for Apheresis guidelines. Hematology 2012;2012:7–12. [16] Zhu J, Zou L, Zhu S, et al. Cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) blockade enhances incidence and severity of experimental autoimmune neuritis in resistant mice. J Neuroimmunol 2001;115:111–7. [17] Downey SG, Klapper JA, Smith FO, et al. Prognostic factors related to clinical response in patients with metastatic melanoma treated by CTL-associated antigen-4 blockade. Clin Cancer Res 2007;13:6681–8.
Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Conflicts of interest Dr. Gu, has received travel and registration support from Teva. Dr. Menzies is an advisory board member for MSD, Chuga and has received speaker honoraria from BMS and Novartis. Dr. Long is an advisory board member for Amgen, BMS, Merk, Novartis, Provectus, Roche and has received speaker honoraria from BMS, Merk and Novartis. Dr. Fernando and Dr. Herkes declare they have no conflicts of interest.
Appendix A. Supplementary data
Please cite this article in press as: Gu Y et al. Immune mediated neuropathy following checkpoint immunotherapy. J Clin Neurosci (2017), http://dx.doi. org/10.1016/j.jocn.2017.07.014
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Review article
Immune mediated neuropathy following checkpoint immunotherapy Yufan Gu a,⇑, Alexander M. Menzies b,c,d, Georgina V. Long b,c,d, S.L. Fernando c,e,f, G. Herkes a,c a
Department of Neurology, Royal North Shore Hospital, Sydney, NSW, Australia Melanoma Institute of Australia, Sydney, NSW, Australia c The University of Sydney, Sydney, NSW, Australia d Department of Medical Oncology, Royal North Shore Hospital, Sydney, NSW, Australia e Department of Clinical Immunology and Allergy, Royal North Shore Hospital, Sydney, NSW, Australia f ImmunoRhuematology Laboratory, Pathology North, Sydney, NSW, Australia b
a r t i c l e
i n f o
Article history: Received 9 November 2016 Accepted 11 July 2017 Available online xxxx Keywords: Checkpoint immunotherapy CTLA-4, PD-1 Ipilimumab Nivolumab Neuropathy Guillain-Barre
a b s t r a c t Checkpoint immunotherapy has revolutionised cancer therapy and is now standard treatment for many malignancies including metastatic melanoma. Acute inflammatory neuropathies, often labelled as Guillain-Barre syndrome, are an uncommon but potentially severe complication of checkpoint immunotherapy with individual cases described but never characterised as a group. We describe a case of acute sensorimotor and autonomic neuropathy following a single dose of combination ipilimumab and nivolumab for metastatic melanoma. A literature search was performed, identifying 14 other cases of acute neuropathy following checkpoint immunotherapy, with the clinical, electrophysiological and laboratory features summarised. Most cases described an acute sensorimotor neuropathy (92%) with hyporeflexia (92%) that could occur from induction up till many weeks after the final dose of therapy. In contrast to Guillain-Barre syndrome, the cerebrospinal fluid (CSF) analysis often shows a lymphocytic picture (50%) and the electrophysiology showed an axonal pattern (55%). Treatment was variable and often in combination. 11 cases received steroid therapy with only 1 death within this group, whereas of the 4 patients who did not receive steroid therapy there were 3 deaths. In conclusion checkpoint immunotherapy – induced acute neuropathies are distinct from and progress differently to GuillainBarre syndrome. As with other immunotherapy related adverse events corticosteroid therapy should be initiated in addition to usual therapy. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction
2. Methods
Checkpoint immunotherapies are now standard treatment for many cancers including advanced melanoma, lung cancer, kidney cancer, head and neck cancer, and bladder cancer [1]. These therapies remove T-cell inhibitory pathways resulting in upregulation of the anti-tumour immune response, with the potential consequence being immune-related adverse events (irAEs) in normal tissues [2]. Peripheral neuropathy is uncommon but cases of severe ‘‘GuillainBarre syndrome” (GBS) resulting in death have been reported.
We provide a detailed description of a case of acute sensorimotor and autonomic polyneuropathy following administration of a single dose of combination ipilimumab and nivolumab for metastatic melanoma. A MEDLINE search was performed of previous publications in English for text word and MeSH headings of ‘‘neuropathy”, ‘‘Guillain-Barre Syndrome” and ‘‘ipilimumab”, ‘‘nivolumab”, ‘‘pembrolizumab”, ‘‘CTLA-4” or ‘‘PD-1”. The references and citations of retrieved articles were all reviewed with relevant cases included for analysis of reported clinical, investigational and treatment findings.
⇑ Corresponding author at: Department of Neurology – Royal North Shore Hospital, Reserve Rd, St Leonards, 2065 New South Wales, Australia. E-mail addresses: [email protected] (Y. Gu), alexander.menzies@ sydney.edu.au (A.M. Menzies), [email protected] (G.V. Long), suran. [email protected] (S.L. Fernando), [email protected] (G. Herkes).
3. Case vignette A 49-year-old woman with NRAS-mutant metastatic melanoma to chest wall, lung, and retroperitoneal and axillary nodes was commenced on combination ipilimumab (3 mg/kg) and nivolumab (1 mg/kg). Five days following induction she developed painful
http://dx.doi.org/10.1016/j.jocn.2017.07.014 0967-5868/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Gu Y et al. Immune mediated neuropathy following checkpoint immunotherapy. J Clin Neurosci (2017), http://dx.doi. org/10.1016/j.jocn.2017.07.014
2
Y. Gu et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx
paresthesiae in the extremities, progressing over four days to a symmetrical sensorimotor and autonomic neuropathy with proximal loss of antigravity power and loss of independent mobility (Fig. 1. Clinical course). Reflexes were absent and sensory loss was present to the upper arms and knees bilaterally. Two days after her developing sensory symptoms, she experienced a fever and a brief episode of diarrhea. Blood and stool cultures were negative and symptoms resolved promptly. Initial electrodiagnostics suggested an acute generalized motor predominant neuropathy with patchy slowing of conduction velocity (Table 1). MRI of the whole spine was normal. Cerebrospinal fluid (CSF) analysis showed increased protein (1.15 g/L), 15 white blood cells (93% lymphocytes), one erythrocyte, no oligoclonal bands, negative cryptococcal antigens, normal culture and cytology. Serum anti-GM1 ganglioside antibodies were negative. Routine full blood count, biochemistry, glucose, B12, folate, renal, liver and thyroid function tests were normal. Serology for HIV and syphilis were negative. Serum immunofixation showed no monoclonal protein. The ANA and ANCA were negative, and levels of CRP, ESR, C3 and C4 were normal. She commenced IV immunoglobulin (IVIG, 0.4 g/kg/d for 5 days) and IV methylprednisolone (1 g/d for 5 days, then 500 mg/d for 3 days) followed by tapering oral prednisone (1 mg/ kg/d). Symptoms stabilized with mild improvement yet one month later she developed worsening weakness and ongoing painful paresthesia. She also developed persistent nausea coupled with postural hypotension and constipation. After the exclusion of structural and metabolic causes, this was thought to represent an autonomic neuropathy. Another course of IVIG and IV methylprednisone were given with slight but non-sustained improvement. Repeat electrodiagnostics demonstrated an axonal predominant sensorimotor neuropathy. A third course of IV methylprednisone and plasma exchange (PLEX, five exchanges over two weeks, followed by weekly exchanges) was commenced, oral corticosteroids were continued and mycophenolate (1 g bd) was added. Immediate improvements were noted in sensory and autonomic symptoms, followed by a gradual improvement in motor strength. By 6 weeks after commencing PLEX (12 weeks after initial treatment) the patient had only mild weakness and could perform her usual daily activities, and at four months she returned to work with her melanoma in deep partial response: complete response in several lesions, and with only a chest wall and lung metastasis remaining (both smaller than at baseline and with a cystic central component). Prednisone was weaned to 10mg daily and mycophenolate was continued at 1 g BD. PLEX was ceased at five months after initial checkpoint immunotherapy after her Vascath was complicated by a Pseudomonas aeruginosa infection necessitating removal. At seven months her cystic chest wall and lung lesions slightly increased in size, with aspiration and partial resection of the chest wall lesion
Table 1 Nerve conduction studies.
Median nerve – Right Motor Latency, APB-Wrist CMAP amplitude, wrist CMAP amplitude, elbow Motor CV SNAP amplitude Ulnar nerve – Right Motor Latency, ADM–Wrist CMAP amplitude, wrist CMAP amplitude, above elbow Motor CV SNAP amplitude Peroneal nerve – Right Motor Latency, EDB-Ankle CMAP amplitude, ankle CMAP amplitude, fibular head Motor CV Tibial nerve – Right Motor Latency, AH-Ankle CMAP amplitude, ankle CMAP amplitude, knee Motor CV Sural nerve Right SNAP amplitude Left SNAP amplitude
Day 4
Day 39
Day 226
5.2 6.8 6.0 42 14
5.1 2.9 2.4 33 2
4.3 6.3 6.3 48 4
3.5 10.4 9.5 50 20
3.4 5.1 4.2 52 NR
3.1 11.1 10.3 53 3
5.6 3.8 2.8 37
4.4 0.3 0.3 40
5.1 2.1 1.3 41
8.0 0.4 0.3 46
11 13
5.0 2.0
4.0
CMAP: Compound muscle action potential, CV: Conduction velocity, SNAP: Sensory nerve action potential.
finding pigment laden macrophages only. Mycophenolate was weaned to 500 mg daily and prednisone maintained at 10 mg daily. At nine months the two remaining melanoma metastases (lung and chest wall) started to enlarge with a cystic central component. Due to this fact, the extended time interval from initial toxicity, and because she was minimally symptomatic with electrodiagnostics showing partial recovery, her remaining mycophenolate was ceased. Within one week of cessation sensory neurological symptoms flared and corticosteroid and mycophenolate was recommenced. This improved symptoms and sequential imaging demonstrated further regression of melanoma. After a few months of clinical stability (prednisone 25 mg daily, mycophenolate 500 mg bd), tapering of her steroid dose was recommenced which soon resulted in increasing pain, weakness and glove and stocking numbness to her elbows and knees. With a further poor response to IVIG and 50 mg per day of prednisone, she was commenced on rituximab infusion (375 mg/m2 weekly for four weeks). There was no clinical response after rituximab infusion and she has since recommenced plasma exchange with an increased dose of mycophenolate (1.5 g bd) and prednisone. In summary we describe a case of acute peripheral neuropathy occurring after the administration of combination checkpoint immunotherapy for metastatic melanoma. The case provides a
Fig. 1. A graphical representation of the patient’s motor polyneuropathy and the clinical response to acute therapies. MRC sum score: Medical Research Council sum score comprised of the sum of the MRC grades of deltoid, biceps, wrist extensor, iliopsoas, quadriceps femoris and tibialis anterior bilaterally (0–60). IVIG: Intravenous Immunoglobulin. PLEX: Plasma exchange. IV MP: Intravenous Methylprednisone. PO: per os; oral administration.
Please cite this article in press as: Gu Y et al. Immune mediated neuropathy following checkpoint immunotherapy. J Clin Neurosci (2017), http://dx.doi. org/10.1016/j.jocn.2017.07.014
Y. Gu et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx
compelling example of an upregulated immune response providing efficacious cancer therapy but also resulting in autoimmunity necessitating ongoing immunosuppression. With long term follow-up we also observe a striking response to steroids, plasma exchange and mycophenolate over other therapies such as IVIG and rituximab. 4. Discussion There are many barriers to an effective immune response against a tumour, complicated by the tumours ability to ‘‘co-opt” certain interacting pathways to its advantage [1,2]. CTLA-4 is expressed on activated T cells and functions as an inhibitory checkpoint receptor early in the immune response at the time of antigen presentation. It outcompetes the costimulatory molecule CD28 for the ligands, CD80 and CD 86, resulting in the downregulation of immune responses by T cells. CTLA 4 also is expressed on T regulatory cells and drives their suppressor functions. Costimulatory molecules that result in T cell activation are also down regulated by tumour cells. In addition, T cells that migrate into the tumour stroma encounter regulatory T cells leading to upregulation of inhibitory receptors, apoptosis, anergy, or exhaustion. PD-1 works predominantly within the tumour microenvironment. It is highly expressed on tumour infiltrating T cells and its ligands are upregulated in many cancers including melanoma. Importantly, it directly inhibits T cell receptor-mediated effector functions. While the benefits of immunotherapy are improved antitumour immunity, with high response rates and durable survival, the downside is the frequent occurrence of irAEs which can occur in any organ. These irAEs appear histopathologically most often as lymphocytic infiltrates [3–5] and are best managed with systemic corticosteroids, often at high doses and with a prolonged wean over many weeks [3]. Occasionally additional immunosuppressants e.g. TNFa inhibitors, mycophenolate, antithymocyte globulin are also required [3,6]. Ipilimumab, a monoclonal antibody that blocks CTLA-4, and the anti-PD-1 antibodies nivolumab and pembrolizumab, all have activity in melanoma, both as monotherapy, and more recently in combination [7–10]. Ipilimumab is associated with irAEs in up to 60% of patients (15% grade 3), that mostly affect the skin and gastro-intestinal tract, however any organ system can be affected. Inflammatory neurological toxicities have been reported, including cases that appear similar to idiopathic Guillain-Barre syndrome, granulomatous inflammation of the CNS, aseptic meningitis, chronic inflammatory demyelinating polyneuropathy, transverse myelitis, myositis and myasthenia gravis [11,12]. The anti-PD-1 antibodies are less likely to cause irAEs but combination immunotherapy with ipilimumab and nivolumab, which appears to have greater efficacy than monotherapy, correspondingly results in greater toxicity (55% grade 3 toxicity) [7]. Our PubMed search identified 12 articles describing 14 cases of neuropathy. Cases were excluded if clinical and/or investigational characteristics were unavailable for comparison. Due to variable levels of detail in the reported cases, results were calculated from those available for analysis.
3
this case series (6/11, 55%) presented with an axonal pattern on EDX. Two studies described conduction block despite an axonal pattern, and a ‘pseudo-conduction block’ was hypothesized to be an explanation [4]. Hence conduction block in the absence of other EDX features of demyelination warrants repeat testing to determine the pattern of nerve injury. CSF studies showed that in 92% (11/12) protein was elevated and in 50% (6/12) a pleocytosis was seen, with five specifically described as lymphocyte predominant. 6. Treatment The estimated incidence of neuropathies as a complication is less than 1% and in most cases transient [3]. The recommendation for grade 2 neuropathy is to hold a dose of treatment, or a threefour week tapering dose of corticosteroids. For grade 3 or 4 complications a longer course of systemic corticosteroids (1–2 mg/kg daily) has been recommended [3]. Our literature search is biased towards the more severe cases of neuropathy and in many cases a combination of therapies for the neuropathy was utilised. Eleven patients (73%) received some form of steroid therapy, reflecting current recommendations, and generally had favourable outcomes with only one death (9%). Mortality in the four who did not receive steroid therapy was 75% with two cases receiving IVIG alone and another receiving a combination of IVIG and PLEX. The remaining mortality case had received a combination of IV methylprednisone, PLEX, infliximab and tacrolimus. Our case which experienced limited response despite several doses of IVIG and a prolonged course of high dose steroids, seemingly demonstrating a durable response following initiation of PLEX. We also describe the successful use of mycophenolate as a corticosteroid-sparing agent, further evidenced by recurrence of symptoms when mycophenolate was ceased. The response to therapy thus appears distinct from idiopathic Guillain-Barre syndrome where firstly there is no evidence for the use of steroids [13]. IVIG is also often used first line for acute inflammatory neuropathies as with GBS it is thought to be equally effective as plasma exchange [14]. We speculate whether the underlying mechanism of autoimmune neuropathy following checkpoint immunotherapy relates to the upregulated T cell functions and therefore demonstrated limited clinical response to IVIG and rituximab. It was however more amenable to PLEX, perhaps due to bulk removal of pathologic substances such as autoantibodies, immune complexes, and cytokines. PLEX may also have an immunomodulatory effect beyond the removal of the immunological molecules such as suppression of IL-2 and IFN-c production and enhanced regulatory T cell function [15]. Autoantibodies could be formed, especially with CTLA-4 therapy, as a result of T cell dependent-activation of B cells. Thus far CTLA-4 blockade in mouse models has demonstrated cellular infiltration and demyelination [16]. Future studies in animal models and patients with presentations similar to our case might elucidate the cytokines present in the plasma and if detected, the correlation between these levels, PLEX treatments and clinical status can be examined. Flow cytometry to T cell phenotypes including regulatory T cells or the detection of an autoantibody may reveal more information.
5. Clinical features 7. Toxicity-response relationship Most cases were male (10/15), six (40%) were diagnosed as GBS and three as chronic inflammatory demyelinating polyneuropathy (CIDP). Mean onset of symptoms was 6.5 weeks (range from one to seventeen weeks) from induction of therapy. The neuropathy was usually sensory and motor (12/13, 92%) and deep tendon reflexes were usually absent or reduced (11/12, 92%). Whilst classical GBS is associated with demyelination, a high proportion of patients in
Notably the cancer response is remarkable with eight patients with tumour regression or whom survived to attend long term follow-up, despite early cessation of treatment. Only one case documented cancer progression. This results from pro-inflammatory effects abrogating the actions of inhibitory checkpoint molecules causing both anti-tumour and autoimmune effects. This
Please cite this article in press as: Gu Y et al. Immune mediated neuropathy following checkpoint immunotherapy. J Clin Neurosci (2017), http://dx.doi. org/10.1016/j.jocn.2017.07.014
4
Y. Gu et al. / Journal of Clinical Neuroscience xxx (2017) xxx–xxx
‘toxicity-response’ relationship is supported by data demonstrating that those with grade 3/4 and grade 1/2 irAEs had better response rates (28% and 22%, respectively) compared to the 2% response rate in those without any irAE [17].
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jocn.2017.07.014.
8. Conclusion
References
Given the growing importance of checkpoint immunotherapy in cancer therapeutics, it is essential that clinicians are able to recognise the uncommon neurological adverse events, especially as they are immune related and amenable to treatment. Our literature review is limited by the variable amounts of detail with which published cases have been described, yet already identifies that a lymphocytic CSF and axonal EDX pattern are common, which is distinct from idiopathic GBS. Larger analysis or case series are needed to characterize this neuropathy with more certainty, particularly in regards to optimal treatment. Nevertheless, based on this series of 15 cases in the literature, we would recommend the use of corticosteroid therapy. The next choice of immunosuppressive agent is more difficult, but we would suggest that PLEX may be more effective than IVIG, and mycophenolate was successfully used as a steroid sparing agent in this case. Additionally, the review supports findings in the literature that irAEs are associated with a more favourable cancer prognosis, arguing for early and active treatment.
[1] Drake CG, Lipson EJ, Brahmer JR. Breathing new life into immunotherapy: review of melanoma, lung and kidney cancer. Nat Rev Clin Oncol 2014;11:24–37. [2] Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer 2012;12:252–64. [3] Weber JS, Kahler KC, Hauschild A. Management of immune-related adverse events and kinetics of response with ipilimumab. J Clin Oncol 2012;30:2691–7. [4] Manousakis G, Koch J, Sommerville RB, et al. Multifocal radiculoneuropathy during ipilimumab treatment of melanoma. Muscle Nerve 2013;48:440–4. [5] Beck KE, Blansfield JA, Tran KQ, et al. Enterocolitis in patients with cancer after antibody blockade of cytotoxic T-lymphocyte-associated antigen 4. J Clin Oncol 2006;24:2283–9. [6] Chmiel KD, Suan D, Liddle C, et al. Resolution of severe ipilimumab-induced hepatitis after antithymocyte globulin therapy. J Clin Oncol 2011;29:e237–40. [7] Larkin J, Chiarion-Sileni V, Gonzalez R, et al. Combined nivolumab and ipilimumab or monotherapy in untreated melanoma. N Engl J Med 2015;373:23–34. [8] Robert C, Long GV, Brady B, et al. Nivolumab in previously untreated melanoma without BRAF mutation. N Engl J Med 2015;372:320–30. [9] Robert C, Schachter J, Long GV, et al. Pembrolizumab versus Ipilimumab in Advanced Melanoma. N Engl J Med 2015;372:2521–32. [10] Hodi FS, O’Day SJ, McDermott DF, et al. Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med 2010;363:711–23. [11] Voskens CJ, Goldinger SM, Loquai C, et al. The price of tumor control: an analysis of rare side effects of anti-CTLA-4 therapy in metastatic melanoma from the ipilimumab network. PloS One 2013;8:e53745. [12] Liao B, Shroff S, Kamiya-Matsuoka C, et al. Atypical neurological complications of ipilimumab therapy in patients with metastatic melanoma. Neuro-oncology 2014;16:589–93. [13] Hughes RA, van Doorn PA. Corticosteroids for Guillain-Barre syndrome. The Cochrane database of systematic reviews. 2012:Cd001446. [14] Hughes RA, Swan AV, van Doorn PA. Intravenous immunoglobulin for GuillainBarre syndrome. The Cochrane database of systematic reviews. 2014: Cd002063. [15] Winters JL. Plasma exchange: concepts, mechanisms, and an overview of the American Society for Apheresis guidelines. Hematology 2012;2012:7–12. [16] Zhu J, Zou L, Zhu S, et al. Cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) blockade enhances incidence and severity of experimental autoimmune neuritis in resistant mice. J Neuroimmunol 2001;115:111–7. [17] Downey SG, Klapper JA, Smith FO, et al. Prognostic factors related to clinical response in patients with metastatic melanoma treated by CTL-associated antigen-4 blockade. Clin Cancer Res 2007;13:6681–8.
Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Conflicts of interest Dr. Gu, has received travel and registration support from Teva. Dr. Menzies is an advisory board member for MSD, Chuga and has received speaker honoraria from BMS and Novartis. Dr. Long is an advisory board member for Amgen, BMS, Merk, Novartis, Provectus, Roche and has received speaker honoraria from BMS, Merk and Novartis. Dr. Fernando and Dr. Herkes declare they have no conflicts of interest.
Appendix A. Supplementary data
Please cite this article in press as: Gu Y et al. Immune mediated neuropathy following checkpoint immunotherapy. J Clin Neurosci (2017), http://dx.doi. org/10.1016/j.jocn.2017.07.014