Autoimmune storm following alemtuzumab
- 1 Department of Medicine, University of British Columbia, Vancouver, British Columbia, Canada
- 2 Department of Neurology, University of British Columbia, Vancouver, British Columbia, Canada
- 3 Neurologie, Faculte de medecine - Universite Laval, Quebec City, Quebec, Canada
- 4 Division of Neurology, Department of Medicine, The University of British Columbia, Vancouver, British Columbia, Canada
- 5 Rheumatology, Arthritis Research Canada, Vancouver, British Columbia, Canada
- Correspondence to Dr Mollie Carruthers; drmolliecarruthers@gmail.com
Abstract
Alemtuzumab has been associated with the emergence of secondary autoimmune diseases. We report a case of a patient with relapsing–remitting multiple sclerosis who developed a refractory immune thrombocytopaenia associated with vasculitis, myelofibrosis and later Guillain-Barré syndrome following alemtuzumab. The medical community should be aware of unusual and unexpected adverse events that may be associated with alemtuzumab, especially when occurring simultaneously in the same patient.
Background
Multiple sclerosis (MS) is a chronic, autoimmune inflammatory disease of the central nervous system. It predominantly manifests in early adulthood and has a significant impact on motor, sensory and cognitive function and, consequently, on quality of life. Dysregulated immune cells invade the central nervous system causing demyelination and axonal and neuronal loss.1
Alemtuzumab is an anti-CD52 monoclonal antibody that was approved for remitting–relapsing multiple sclerosis (RRMS) in 2014. Alemtuzumab induces a prolonged lymphopaenia with B cells repopulating first, after 6 months on average, followed by the CD8+ T cells (10 months) and CD4+ T cells (36 months).2 Antibody-mediated autoimmune diseases are frequent after alemtuzumab, peaking in the second year post-treatment.3 Thyroid autoimmunity is the most prevalent followed by immune thrombocytopaenia (ITP) and nephropathies. ITP is mediated by autoantibodies that accelerate platelet destruction resulting in platelet count less than 100×109/L.4
During the drug clinical development, 34 (2.3%) ITP cases occurred, with a high response to first-line treatments (intravenous immunoglobulin (IVIg) and corticosteroids).5 About one-third of patients required second-line therapies.5 Compared with usual drug-related ITP, alemtuzumab-associated ITP is characterised by a delayed presentation after drug exposure, a good response to therapy and prolonged remission.6
Due to the occurrence of life-threatening complications, such as cerebral haemorrhage and cervical artery dissection, in 2019, European Medicines Agency (EMA) restricted the use of alemtuzumab to patients with severe MS who failed many disease-modifying therapies. New, independent, adverse events have been described during its postmarketing phase, such as Guillain-Barré syndrome, Nocardia infection and cytomegalovirus reactivation.7–9 Though it remains unclear whether all reactions were drug induced, we report a first case of a patient developing refractory ITP associated with vasculitis and myelofibrosis occurring after alemtuzumab exposure and later, Guillain-Barré syndrome.
Case presentation
A young woman in her mid-30s was diagnosed with RRMS after presenting with left arm weakness. She had initially been treated with interferon-β-1a, but due to relapses, the medication was changed to mitoxantrone, natalizumab, glatiramer acetate and fingolimod. The plan to escalate MS treatment due to recurrent transverse myelitis attacks, previous internuclear ophthalmoplegia, previous optic neuritis and cognitive impairment was discussed with and agreed on by the patient. She received two cycles of alemtuzumab (12 mg/1.2 mL infusion for 5 consecutive days followed by 12 mg/1.2 mL infusion for 3 consecutive days 1 year later) 6 months before presentation but her disease remained active with three relapses after the second cycle (two transverse myelitis and one brainstem relapse) and an increase in her Expanded Disability Status Scale from 2.0 to 3.0. This level of disease activity is unusual after alemtuzumab, but reflects the aggressive nature of her MS. In the context of high MS activity level after potent treatment, aquaporin-4 and myelin oligodendrocyte glycoprotein (MOG) antibodies were both tested, but found to be negative. Her previous medical history was also significant for chronic fatigue syndrome and her family history included polycythemia vera in her father. Prior to the suspected alemtuzumab-related complications she was in her normal state of health, living independently, walking her dog twice a day, going to school and hiking.
She presented to our hospital with acute, severe abdominal pain. Initial abdominal CT showed left adrenal infarct with no obvious cause. Laboratory results revealed mild hyponatraemia (131 mmol/L, normal=135–145 mmol/L), neutrophilia (19.3×109/L, normal=4–11×109/L) and liver enzyme abnormalities. Platelets were normal at 388×109/L (normal=150–400×109/L).
Ten days later, the patient presented with severe headache, right-sided weakness and aphasia. Head CT was unremarkable, and a lumbar puncture only showed a mild cerebrospinal fluid (CSF) protein elevation at 662 mg/L (normal=150–450 mg/L). CSF cell count, glucose level and culture were normal as well as CSF PCR for John Cunningham virus (JCV), Herpes Simplex virus (HSV) 1 and 2, Varicella Zoster virus (VZV) and enterovirus. A brain MRI showed a new left frontal haemorrhage surrounded by oedema (figure 1).
A focal haemorrhage (arrow) was visualised in the left frontal lobe on the axial susceptibility-weighted imaging (A). It was not visualised on the previous MRI done 6 months earlier. The fluid attenuated inversion recovery sequence (B) showed an increased hypersignal, suggestive of oedema (arrow). It improved 1 month later (not shown). We can also observe many juxtacortical lesions associated with her multiple sclerosis.
A suspected vasculitis was confirmed on conventional cerebral angiogram demonstrating discrete narrowing of proximal brain arteries (figure 2). C reactive protein was elevated at 203 mg/L (normal<7.5 mg/L), and there was 1.17 g of protein on the 24-hour urine collection. The remainder of the vasculitis work-up was negative, including negative anti-neutrophilic cytoplasmic autoantibody (ANCA) and anti-GBM (glomerular basement membrane) antibody. Due to the worsening of the liver enzyme abnormalities, a liver biopsy was performed that showed mild non-specific lobular inflammation and extramedullary haematopoiesis.
On this anteroposterior view of the right carotid territory, a focal narrowing of the proximal right anterior cerebral artery is shown (arrow). The conventional cerebral angiogram also demonstrates discrete narrowing of the right and left internal carotid artery, right M1 and right A1 (not shown).
Two weeks after the adrenal infarct and 1 week after the brain haemorrhage, her platelet count dropped abruptly, with a nadir of <1×109/L. Heparin-induced thrombocytopaenia was excluded by a negative Platelet Factor Antibody 4 (PFA-4) assay. No haemolysis or schistocytes were seen on blood smears. A preliminary diagnosis of immune-mediated thrombocytopaenia was made. Bone marrow biopsy showed severe myelofibrosis with clustering of megakaryocytes and intrasinusoidal haematopoiesis. Molecular analysis showed no mutations in genes implicated in myeloproliferative disorders (JAK2, CALR and MPL). Figure 3 shows the longitudinal evolution of the thrombocytopaenia. The initial treatment was prednisone (1 mg/kg), leading to a transient improvement. After the second drop in platelets, pulsed intravenous steroids were given for 3 days simultaneously with rituximab (1000 mg) and, 2 weeks later, two doses of IVIg (1 g/kg) 1 week apart. When the platelets were below 5×109/L, she developed epistaxis, purpura and severe upper gastrointestinal (GI) bleeding requiring massive blood product transfusions. She received a second dose of rituximab (1000 mg), cyclophosphamide (1.3 g) and eltrombopag. Within a week, the platelet count increased, the GI bleeding stopped and the purpura resolved. After 3½ months of in-patient treatment, she was discharged to an outpatient rehabilitation clinic.
Two weeks after the left adrenal infarct and 1 week after the brain haemorrhage, the platelets suddenly dropped below 100×109/L over 4 days. She was treated with 65 mg of oral prednisone (P), and the platelets went up within 2 weeks but dropped again. When she started being symptomatic of the thrombocytopaenia (epistaxis), she was treated with intravenous methylprednisolone (M) for 3 consecutive days, simultaneously with a first dose (1000 mg) of rituximab (R). Rituximab led to a minimal and transient rise of the platelet count that dropped again 2 weeks later. She received two doses of IVIg (G) (1 g/kg) 1 week apart, without any effect. While the platelet blood level was undetectable, she developed a severe upper gastrointestinal (GI) bleeding that required massive transfusions. Then, she sequentially received a second 1000 mg of rituximab, 1.3 g of cyclophosphamide (C) and eltrombopag 75 mg/day (E). Within a week, her platelet level rose. Prednisone and eltrombopag are being tapered since. ITP, immune thrombocytopaenia.
Steroids were tapered and, 1 month later, she presented with worsening abdominal pain and distention, shortness of breath and confusion. Encephalopathy, oliguria and ascites were determined to be associated with severe kidney failure. Her creatinine reached a maximum of 412 mmol/L (n=40–90 mmol/L). Anti-GBM antibodies were normal on three separate occasions. Acute thrombotic microangiopathy and acute interstitial nephritis were suggested on kidney biopsy. Drug hypersensitivity was thought to contribute to this pathological presentation. Due to the absence of haemolysis and a normal ADAMTS13 activity level, a thrombotic thrombocytopaenic purpura was excluded, and a drug hypersensitivity reaction was considered the most likely cause of the kidney injury. Her clinical status and kidney function improved following dialysis.
A month later, she developed progressive dyspnoea and had patchy ground-glass opacities and centrilobular ground-glass nodules on chest CT. Samples from a bronchoscopy were not conclusive for an infectious process and a granulomatous drug reaction was diagnosed on a lung biopsy. Diffuse lymphadenopathy was also seen progressing on repeated CT. A lymphoproliferative disorder was ruled out by an excisional biopsy, showing extramedullary haematopoiesis. Due to the recurrent nature of the systemic reaction and the positive initial response, cyclophosphamide was repeated at 750 mg intravenously monthly for 6 months.
Eight months later, the patient was self-admitted to neuropsychiatry for mood disturbance related to significant stress. During this stay, she developed bilateral leg weakness, facial numbness, nausea and vomiting. She also had a reduction in her haemoglobin requiring two units of packed red cells. On examination, she was globally areflexic. Electrophysiological studies demonstrated severe demyelinating sensory motor polyneuropathy and CSF studies revealed an albuminocytological dissociation, confirming a diagnosis of Guillain-Barré syndrome. This was complicated by respiratory failure requiring intubation and mechanical ventilation and eventually tracheostomy. The rapid clinical worsening and the 8-month latency from her prolonged hospital stay make a critical illness neuropathy unlikely. The demyelinating nerve involvement was not suggestive of a nerve involvement seen in vasculitis. She was treated with high-dose steroid and IVIg followed by plasmapheresis.
Differential diagnosis
The differential for this case presentation includes ANCA vasculitis, systemic lupus and primary myelofibrosis. The patient had a negative ANCA, antinuclear antibody and JAK2, ruling them out, respectively. Sarcoidosis and intravascular lymphoma were also considered and ruled out by imaging and lymph node biopsy. The timing of the patient’s symptoms was suggestive of secondary immune disease related to alemtuzumab use.
Outcome and follow-up
The patient was maintained on a gradual taper of monthly cyclophosphamide infusions and prednisone for over 2 years. She initially received rituximab treatment for her MS and to prevent the recurrence of the suspected alemtuzumab-related autoimmune complications; however, this was switched to mycophenolate mofetil due to infusion reaction (late-onset urticarial rash which worsens with repeated rituximab infusions). Mycophenolate mofetil was used to treat the vasculitic process and she did not require specific MS disease-modifying treatment after. Three years later, she reported no relapses or progression of her MS symptoms. She has returned to school to pursue an engineering degree and continues to have regular neurology and rheumatology follow-up.
Discussion
Previously well-documented cases of secondary immune disease associated with alemtuzumab treatment indicate thyroid pathology being the most prevalent ranging from 34% to 41.1%.10 ITP is second most common with incidence of 2.3% followed by nephropathies at 1%. Typical drug-induced ITP occurs within hours or days of exposure and resolves shortly after discontinuing the medication; however, alemtuzumab-associated ITP is distinct from other drug-induced ITP in that it has a delayed onset ranging from 1 to 2 years after exposure.11 This case draws its uniqueness from the constellation of autoimmune complications experienced by the patient. It serves to increase awareness of possible unexpected and concurrent alemtuzumab complications.
First, prior to the ITP, the patient presented with an adrenal infarct and a brain haemorrhage caused by a vasculitis. Later, she developed kidney failure, diffuse lymphadenopathy and lung involvement. No serological or pathological evidence of vasculitis were found despite multiple biopsies, but the recurrences and the response to immunosuppressive drugs favour an autoimmune process. To our knowledge, there have been limited multisystemic reactions of this nature and intensity previously described after treatment with alemtuzumab. Those documented include ANCA vasculitis with simultaneously occurring Grave’s disease and autoimmune haemolytic anaemia occurring with alveolar haemorrhage and ischaemic stroke.12 13
Second, the poor treatment response was unusual compared with previously described alemtuzumab-induced ITP. Treatment effect is expected to start within few days and complete remission occurs within 1 month on average.3 14 With ITP in general, cases unresponsive to corticosteroids have an increased morbidity and mortality.15 This patient had platelets below 5×109/L for over 3 weeks and suffered from life-threatening GI haemorrhage.
While many of the investigations were suggestive of drug-induced ITP, it is a diagnosis of exclusion. It remains unclear whether the abnormal lung biopsy, renal failure and Guillain-Barré syndrome were secondary to alemtuzumab treatment or another underlying condition, given the patient’s complexity. However, her outcome is consistent with other documented cases of alemtuzumab-induced ITP in which sustained remission was achieved with treatment.11
The simultaneous presence of myelofibrosis and vasculitis could explain the refractory nature of the ITP in this case. Although myelofibrosis can cause thrombocytopaenia, the rapid onset of the thrombocytopaenia, the very low nadir and response to therapy confirm the presence of a concomitant ITP. Secondary autoimmune myelofibrosis is most commonly associated with systemic lupus erythematosus but has not previously been reported with alemtuzumab.16 The time course of the development of myelofibrosis is not clear, and the patient was previously treated with many other therapies, but none of those have been reported to cause myelofibrosis. A cumulative effect of previous therapies associated with cytopaenia may have contributed to reducing bone marrow capacity, especially the use of mitoxantrone 17 years before. Interestingly, the patient had evidence of myelofibrosis with extramedullary haematopoiesis on liver biopsy performed prior to the development of severe thrombocytopaenia.
The co-occurrence of autoimmune diseases was previously described with alemtuzumab with the incidence of autoimmunity being higher around the second-year post-treatment.4 17 18 In our case, the timing relative to alemtuzumab treatment is suggestive of autoimmune complications seen potentially with this drug.
Different treatment options exist for ITP.19 Because of the severity of the patient’s presentation, multiple lines of therapy were used concomitantly to achieve remission as quickly as possible. As the autoimmune diseases observed with alemtuzumab are predominantly antibody mediated, they are more likely to respond to B-cell depletion, making rituximab potentially a good choice for refractory alemtuzumab-associated ITP.5 B cells repopulate before T cells and, in the relative absence of T-cell regulatory mechanisms, potentially cause secondary autoimmunity in alemtuzumab-treated patients.4 20 There is a need for more research to discover which patients are predisposed to secondary autoimmune complications when treated with immune suppressive agents for neurological conditions like MS.
Patient’s perspective
One day, I began having abdominal pain, I could not focus or get to sleep. The next day I walked toward my pharmacy to pick my pharmacist’s brain.
Something is wrong. This feels like gallstones. It’s not as if my gallbladder grew back. Something is definitely wrong.
I passed my pharmacy and walked three more blocks to the ER at the hospital. Where I waited for 3–4 hours, was given some blood tests, and given a script for a minimal amount of opioid pain killers and sent home.
It’s fine. The doctor said this kind of thing happens sometimes and just resolves on its own. At least now I can get some sleep, my dog probably has to go out anyway.
I managed to get a few hours of sleep. And woke up in a greater amount of pain I was in before. The medication was only prescribed for a few days. Whatever was happening had not resolved, I went back to the ER.
I am not fine.
I was eventually admitted after my second or third visit to the ER, through a clinic that one of the doctors ran in hospital
I was in hospital in the small city where I lived for about 1 week, then transferred to critical care in a larger hospital.
My platelets were below 6000, at points undetectable. I wasn’t allowed out of bed without supervision. I wasn’t allowed to blow my nose; it had started to bleed. I had dark purple bruises all over my arms from the numerous daily blood draws. I had a large bruise on my face from the gentle touch of a doctor’s hand during an exam. I had bruises all over my legs. None of these bruises hurt. There was no trauma—there were just no platelets.
Friends and family would visit, and I would be able to leave my room in a transfer chair; we’d play cards, we’d chat and joke around, they would bring me non-hospital food, occasionally they could bring my dog. They kept me balanced, they kept me hopeful, their mere presence helped me remember what life was like outside of the hospital walls; what it was like to have autonomy over my own body, what it was like to have privacy.
I had my first ever blood, and platelet transfusions, rituximab, and IVIG infusions … rinse and repeat. Just kidding—I did not have access to showers there.
After a couple of months, I was transferred to a larger teaching hospital where the specialists were that were consulting on my case.
I was bounced all over the hospital, and not permitted out of bed.
A medical student from one of my teams told me that I was a complex combination of symptoms; a much more interesting puzzle than the examples provided in her textbooks. I told her that I am a human being—a person, not a puzzle. I was partly reminding myself.
After I’d been there awhile, a consulting oncologist told me he was disappointed with my latest biopsy came back negative. An oncologist had just told me that he was disappointed that I did not have cancer… As a scientist, I understood what he meant—I wanted an explanation too. As a human being, a person with hopes, dreams, plans, fears, interests and passions—that existed entirely outside of the hospital’s walls, I didn’t know what to think. And quietly wept after he’d left my bedside.
I had a GI bleed; at least I had my own room for a while—but no control over my bowel. Once it resolved, I was moved again.
This time into a room with another young woman about my age, we bonded over our failing 30-something year old bodies, we’d talk. She was the first person I’d shared a room with that didn’t insist her problems were worse than mine. Even though it is probable that they were.
I was started on eltrombopag and had another infusion of rituximab.
About 75% of my beautiful, long, brown, hair came out in my brush… I cried. I wanted to shave the rest off; what remained looked ridiculous. My “roommate’s” husband brought in some clippers, our nurse gave us some gowns to use as a cape, and this young woman, whom I’d know just over a week shaved my head. We stopped part way to do a Krusty the clown hair style selfie, and again for a Homer Simpson style selfie. I didn’t quite have enough hair left to pull off the Krusty look; but I nailed the Homer Simpson look—before she shaved off the last few strands of my hair.
I did not know whether or not it would be growing back.
Eventually I was stable enough to be transferred for rehabilitation; among other things I got to learn to walk again.
I am so tremendously grateful, to each of my brilliant doctors; especially the one’s who authored this paper. For their support, care and expertise while I was in hospital, and continued support, monitoring and treatment. I am certain that I would not be here, if it were not for all of you.
Something about spending the better part of 2 years where your most hard-earned accomplishment it not dying, has a very clarifying effect. I have much stronger relationships with my friends, and some family members. And ended relationships with many others. I pursue my interests and goals with abandon. And continually strive to be present and live a life that is authentically my own.
Learning points
-
The patient in this case developed central nervous system vasculitis and intracranial haemorrhage, immune thrombocytopaenia, myelofibrosis and nephropathy, followed by Guillain-Barré syndrome. The combination of these simultaneous conditions made management particularly challenging.
-
Unusual and new autoimmune disorders associated with alemtuzumab could still arise with increased postmarketing experience.
-
Due to the occurrence of life-threatening complications, European Medicines Agency restricted the use of alemtuzumab in November 2019. Its use should be restricted to patients with severe multiple sclerosis who failed many disease-modifying therapies.
Ethics statements
Patient consent for publication
Acknowledgments
We would like to thank Dr Kimberly Ambler, Dr Virginia Devonshire, Dr Alyson Plecash and Dr Mei Lin Bissonnette for their expertise and input throughout this case.
Footnotes
-
Contributors CC and PB were involved in drafting and revision of the report. MC and A-LS were involved in the revision and final approval of the version to be published.
-
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
-
Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.
-
Competing interests None declared.
-
Provenance and peer review Not commissioned; externally peer reviewed.
- © BMJ Publishing Group Limited 2022. No commercial re-use. See rights and permissions. Published by BMJ.
References
Use of this content is subject to our disclaimer