Customize your JAMA Network experience by selecting one or more topics from the list below.
Author Affiliations: International Dermatology Research, Inc (Dr Gratton), and H ôpital de Verdun, Universit é de Montr éal (Drs Germain and Saltiel), Montreal, Canada; Centocor Research & Development, Inc (Dr Szapary), and Johnson & Johnson Pharmaceutical Research & Development, LLC (Dr Goyal), Malvern, Pennsylvania; and Centocor Ortho Biotech Services, LLC, Horsham, Pennsylvania (Dr Fakharzadeh).
Background Reversible posterior leukoencephalopathy syndrome (RPLS) is a rare, generally reversible neurologic syndrome that is diagnosed based on characteristic clinical and radiologic findings.
Observations We describe the first case of RPLS in a 65-year-old woman who underwent ustekinumab therapy for psoriasis. Approximately 2 ½ years after the patient began ustekinumab therapy, she experienced an acute onset of confusion, headache, nausea, vomiting, and seizures. Computed tomographic scans and magnetic resonance images of her head revealed characteristic findings, including white matter abnormalities consistent with edema in the absence of infarction. There was no evidence of vasospasm, thrombosis, or infection. Cerebrospinal fluid tests were negative for the JC virus. The patient improved clinically and was discharged 6 days after she presented to the emergency department. She made a full neurologic recovery, with a reversal of the radiologic findings.
Conclusions Reversible posterior leukoencephalopathy syndrome is an increasingly recognized neurologic disorder that has been reported with the use of systemic and biologic agents to treat moderate to severe psoriasis. Although the relationship between RPLS and ustekinumab therapy remains unclear, this case emphasizes the need for dermatologists to recognize the syndrome's signs and symptoms and to refer patients promptly for evaluation and appropriate treatment if the clinical features of RPLS are suspected.
Reversible posterior leukoencephalopathy syndrome (RPLS) is a rare, generally reversible neurologic syndrome that was first described in a case series by Hinchey et al1 in 1996. The term RPLS is somewhat of a misnomer because the syndrome is neither always reversible nor confined to the white matter or posterior regions of the brain.2 It is synonymous with the frequently used term posterior reversible encephalopathy syndrome, although other terms have also been used for this syndrome in the literature. Clinical presenting features of RPLS commonly include acute onset of headache, visual disturbances, altered mental status, and seizures.1,2 Because the clinical presentation of RPLS is frequently nonspecific, both characteristic clinical findings and confirmatory radiologic studies are needed to diagnose RPLS. Transient cerebral edema without infarction is usually seen in the posterior regions of the brain (most frequently in the parietal and occipital lobes) on computed tomographic (CT) scans and magnetic resonance images (MRIs).3 However, changes in a more anterior distribution1 and involvement of the brainstem, thalamus, and cerebellum also commonly occur.2,4 Furthermore, despite being described as a form of leukoencephalopathy, neuroimaging studies have demonstrated gray matter (cortical) involvement in as many as 94% of RPLS cases.2 Reversible posterior leukoencephalopathy syndrome is often associated with hypertension, autoimmune diseases, and treatment with cytotoxic or immunosuppressive agents (Table).
Ustekinumab is a first-in-class, fully human, immunoglobulin G1 κ monoclonal antibody. It binds with high affinity and specificity to the p40 subunit shared by human interleukins 12 and 23, which are important cytokines in the pathogenesis of psoriasis.34 Over the course of the clinical development program for the use of ustekinumab across multiple indications, a single case of RPLS was observed in a psoriasis clinical trial and is described herein.
Report of a case
A 65-year-old woman with plaque psoriasis was treated with ustekinumab in the phase 3, PHOENIX 2 study35 (clinicaltrials.gov Identifier: NCT00307437). She was diagnosed as having psoriasis at 36 years of age, had a baseline Psoriasis Area and Severity Index score of 19.2 at the time of enrollment in the study, had previously only used topical steroids and acitretin for the treatment of psoriasis before study entry, and had not received any other systemic immunomodulatory agents before ustekinumab. She received 2 initial doses of ustekinumab, 45 mg 4 weeks apart, and then 1 dose every 12 weeks thereafter over the course of 2 ½ years. Six weeks after her 12th dose of ustekinumab, she developed headache, nausea, and vomiting over a period of a few hours. The onset of these symptoms was followed by a probable seizure, as she was found unconscious and with upward deviation of her eyes. By the time medical help arrived, she had regained consciousness but was confused. According to a family member, she had experienced some episodes of head tremor over the previous 1 to 2 weeks. Her medical history was remarkable for hypercholesterolemia (treated with rosuvastatin), alcoholic hepatitis diagnosed 9 years earlier (with no history of alcohol withdrawal), and headaches for several years. Her social history included heavy alcohol use (confirmed by a family member) and ongoing cigarette smoking (1 ½-2 packs per day). Her body mass index was 28.7 (calculated as weight in kilograms divided by height in meters squared).
On presentation, her blood pressure was 152/92 mm Hg, although she had no history of hypertension, and she was afebrile. She was awake but disoriented, and she had no focal neurologic signs. While in the emergency department, she had 2 additional witnessed generalized seizures with focal features. The first seizure involved her eyes turning to the left, flexion of her left arm, stiffening of her other limbs, and a slight tremor. She was treated with lorazepam. Postictally, she did not regain normal consciousness but remained stuporous, dysarthric, and unable to follow commands. There were subtle bilateral neurologic signs (minimal motor signs on the right, equivocal left plantar response, and perhaps left gaze preference). After the next seizure, left eye deviation was again briefly observed, and Babinski signs were then present bilaterally. The patient was treated with lorazepam, phenytoin, and propofol, and she was intubated for airway protection. She was subsequently admitted to the hospital for further evaluation and treatment. She later developed a fever of 39.3 °C and was treated empirically with piperacillin and acyclovir. A complete blood cell count and electrolyte levels, including magnesium (1.5 mEq/L), were normal; aspartate aminotransferase and alanine aminotransferase levels were mildly elevated. A lumbar puncture was performed; cerebrospinal fluid (CSF) showed no white blood cells and a mildly elevated protein level of 0.87 g/L. Polymerase chain reaction studies of the CSF samples were negative for herpes simplex virus and JC virus. Serologic tests and cultures of blood samples were negative for West Nile virus and other infectious organisms.
Head CT scans with and without contrast revealed bilateral white matter changes in the cerebellum, left thalamic hypodensity, and no evidence of mass effect, bleeding, or thrombosis. An MRI without gadolinium showed T2 and fluid-attenuated inversion recovery hyperintensities in the white matter of both cerebellar hemispheres (Figure, A), the left superior thalamus (Figure, B), and the right posterior parietal periventricular white matter. There was no evidence of infarction on the diffusion study. An MRI with contrast was not obtained because an attempt to inject gadolinium was unsuccessful when the patient became agitated. An electroencephalogram revealed mild to moderate generalized slowing and occasional right central spikes. The CT angiogram results were normal and did not demonstrate vasospasm or a vertebrobasilar anomaly.
Figure. Magnetic resonance images obtained at presentation showing several abnormalities in a posterior distribution. T2-weighted images show hyperintensities in the cerebellar white matter bilaterally (A, arrows) and in the left superior thalamus (B, arrows).
The features of this case were considered to be consistent with a diagnosis of RPLS. The patient improved rapidly and was extubated within 24 hours, although she remained confused for a few days. She was discharged from the hospital 6 days after admission, and she continued taking phenytoin for 1 month. She recovered completely; her electroencephalogram was normal 1 month after discharge, and an MRI showed no abnormalities 1 month later. Steady state trough ustekinumab serum concentrations (1.06 μg/mL) were detected approximately 6 weeks before the onset of signs and symptoms of RPLS and approximately 6 weeks after they resolved (0.80 μg/mL). The ustekinumab serum levels were within the therapeutic range throughout the patient's treatment and were also within the therapeutic range reported in all patients treated with 45 mg of ustekinumab in the PHOENIX 2 study.35 Ustekinumab therapy was permanently discontinued.
The pathogenesis of RPLS is unknown, although 2 hypotheses have commonly been accepted.2,36 One hypothesis implicates a defect in the regulation of the cerebral vasculature. Stable circulation is maintained in the cerebral vasculature by local autoregulation through endothelial release of nitric oxide to allow vasodilation when systemic blood pressure decreases or release of thromboxane-A2/endothelin to allow vasoconstriction when systemic blood pressure rises. Because 50% to 70% of patients with RPLS present with moderate to severe hypertension and demonstrate rapid improvement of signs and symptoms with antihypertensive treatment, the first hypothesis suggests that hypertension exceeding the upper limit of autoregulation (ie, 150-160 mm Hg mean arterial pressure) may compromise the blood-brain barrier, leading to hyperperfusion and subsequent vasogenicedema. However, 20% to 30% of patients with RPLS are normotensive at presentation, and most patients with RPLS who present with hypertension do not have mean arterial pressures exceeding the upper autoregulatory limit. Also, cerebral edema has been found to be less prominent in severely hypertensive patients with RPLS, and recent case series have documented hypoperfusion of the brain instead of hyperperfusion.36,37
Because almost all cases of RPLS occur in association with certain comorbid conditions or the use of specific drug classes, and because the clinical presentation and neuroimaging findings of RPLS are similar in the context of any of these associations, the second hypothesis suggests that RPLS may be triggered by the systemic toxic effect caused by such associations.36 In settings associated with RPLS, shared biologic processes, such as T-cell activation or inflammatory cytokine production, may cause endothelial activation or injury, which may result in the subsequent vasoconstriction and hypoperfusion observed in both hypertensive and normotensive patients with RPLS. Toxicity-related vasoconstriction or hypoperfusion may cause localized brain hypoxia, which prompts the upregulation of vascular endothelial growth factor that induces endothelial cells to stimulate angiogenesis and to increase endothelial permeability. Increased endothelial permeability could, in turn, lead to vasogenic edema.36
We report the first case of RPLS observed among more than 4000 patients across all studies in the ustekinumab clinical development program, to date. The patient showed clinical manifestations and neuroimaging findings consistent with RPLS and did not exhibit clinical findings indicative of other neurologic conditions resembling RPLS. On presentation, she had mild hypertension, confusion, headache, nausea, vomiting, and multiple seizures —all typical clinical manifestations of RPLS.1 Computed tomographic scans of her head obtained during hospitalization revealed left thalamic hypodensity and white matter changes in the cerebellum. T2-weighted MRIs showed hyperintensities in the cerebellar hemispheres, left superior thalamus, and right posterior parietal periventricular white matter. These findings are consistent with vasogenic edema, which is characteristic of RPLS. Also, her signs and symptoms resolved within 1 week of presentation, which is consistent with the clinical course of RPLS.1 Furthermore, an MRI obtained 2 months after clinical recovery showed complete reversal of the radiologic findings, which is also typical of RPLS.
Other diagnoses considered included alcohol withdrawal, ischemia, thrombosis, sepsis, encephalitis, paraneoplastic limbic encephalitis, and progressive multifocal leukoencephalopathy (PML). Although alcohol withdrawal could have precipitated this patient's elevated blood pressure levels and seizures, it would not be expected to explain the neuroimaging findings. No evidence of infarction or thrombosis was seen on MRI or CT angiography. Meningitis and sepsis were ruled out based on the CSF findings, blood culture results, and lack of other clinical signs associated with these conditions. Similarly, the patient's presentation was inconsistent with viral encephalitis, given the cerebellar involvement, the absence of white blood cells and herpes simplex virus in her CSF, and the lack of antibodies to West Nile virus in her serum. Paraneoplastic limbic encephalitis was also ruled out owing to the lack of history of known malignancy and the absence of findings suggestive of a lung tumor on chest x-ray films. Finally, the acute presentation, the absence of the JC virus in the CSF, and the achievement of a full recovery eliminated the diagnosis of PML.
Distinguishing RPLS from PML in clinical practice is critically important. In contrast to RPLS, PML is a subacute and progressive demyelinating disorder that is caused by reactivation of latent JC virus in the context of severe immunosuppression. Approximately 85% of PML cases occur in patients who are infected with the human immunodeficiency virus. Patients with PML typically present with a comparatively gradual onset of visual and speech disturbances, develop progressive impairment of mental and motor functions, and manifest characteristic white matter lesions on MRIs.38 Progressive multifocal leukoencephalopathy usually leads to death within weeks to months of the onset of clinical presentation.
Reversible posterior leukoencephalopathy syndrome is a rare neurologic syndrome of unknown incidence; however, it is increasingly recognized in association with a variety of heterogeneous clinical conditions and the use of therapeutic agents (Table). As the use of cytotoxic and immunosuppressive agents increases, the incidence of RPLS is also likely to increase.32 The syndrome has been reported in patients aged 2 to 90 years and appears to be more common in women, even when eclampsia-associated cases are excluded.1 Dermatologists should be aware of the signs and symptoms of RPLS because the systemic medications cyclosporine17,18 and methotrexate,19,20 which are commonly used to treat psoriasis, have been associated with RPLS in patients with psoriasis. Furthermore, cases of RPLS have been reported in patients who were treated for disorders other than psoriasis with the anti –tumor necrosis factor biologic agents etanercept22 and infliximab,23,24 which are also approved for the treatment of psoriasis.
Several clinical features of this case are atypical of drug-associated RPLS. The time to onset after the initial drug exposure in this case (2 ½ years) is much longer than has been reported for other cases of drug-associated RPLS. The time to onset of RPLS typically ranges from hours (with intravenous immunoglobulin39) to months (with, eg, cyclosporine,17 tacrolimus,40 and etanercept,22) or even up to 1 year (with bevacizumab41). The patient also achieved complete recovery despite still having steady state therapeutic serum levels of ustekinumab, suggesting that there could be additional or alternative mechanisms accounting for her developing RPLS. No theoretical or mechanistic association between ustekinumab and RPLS has been identified, to our knowledge. Furthermore, signs of toxic effect in the vascular endothelium have not been observed with ustekinumab therapy at doses higher than those used for the treatment of psoriasis (up to 180 mg weekly for 4 weeks and monthly thereafter in the phase 2 clinical study for multiple sclerosis).42
Although the disease mechanism of RPLS has not been defined and the relationship between ustekinumab and RPLS remains unclear, this case emphasizes the need for dermatologists to recognize the signs and symptoms of RPLS and to refer patients promptly for evaluation and appropriate treatment if the clinical features of RPLS are suspected.
Correspondence: David Gratton, MD, FRCPC, International Dermatology Research, Inc, 3550 Cote des Neiges, Ste 740, Montreal, QC H3H 1V4, Canada (firstname.lastname@example.org).
Accepted for Publication: April 10, 2011.
Published Online: June 16, 2011. doi:10.1001/archdermatol.2011.161
Author Contributions: Drs Szapary and Fakharzadeh had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Szapary, Goyal, and Fakharzadeh. Acquisition of data: Gratton, Szapary, Germain, and Saltiel. Analysis and interpretation of data: Szapary, Goyal, Fakharzadeh, Germain, and Saltiel. Drafting of the manuscript: Fakharzadeh and Saltiel. Critical revision of the manuscript for important intellectual content: Gratton, Szapary, Goyal, Fakharzadeh, Germain, and Saltiel. Administrative, technical, or material support: Szapary. Study supervision: Gratton, Szapary, and Goyal.
Financial Disclosure: Dr Gratton served as an investigator in the PHOENIX 2 trial35 and has received honoraria as an investigator, speaker, and advisory board member for Centocor, Abbott, Amgen, and Schering-Plough. Drs Szapary, Goyal, and Fakharzadeh are employed by subsidiaries of Johnson & Johnson.
Funding/Support: The study in which this case was first reported was supported by Centocor Research & Development, Inc, a subsidiary of Johnson & Johnson.
Role of the Sponsors: The sponsor contributed in part to the design and conduct of the study in which this case was first reported; the collection, management, analysis, and interpretation of the data in the manuscript; and the preparation, review, and approval of the manuscript. This case report was prepared jointly by the study investigator, diagnosing physicians, and the sponsor in carrying out due diligence to inform the medical community of this important adverse event.
Additional Contributions: Jennifer Han and Mary Whitman, PhD, of Centocor Ortho Biotech Services, LLC, drafted and revised the manuscript. Drs Julie Gr égoire, Pierre Pascual, Marie Jolivet, H él ène Marinier, and Jean-Dominique Gervais participated in the care of this patient.
Gratton D, Szapary P, Goyal K, Fakharzadeh S, Germain VÉ, Saltiel P. Reversible Posterior Leukoencephalopathy Syndrome in a Patient Treated With Ustekinumab: Case Report and Review of the Literature. Arch Dermatol. 2011;147(10):1197–1202. doi:10.1001/archdermatol.2011.161
Coronavirus Resource Center