New Drug Option Fails for Rare Polio-Like Virus in Kids

Damian McNamara

November 14, 2018

Despite initial hope, the antidepressant fluoxetine (multiple brands) failed to improve outcomes in children with enterovirus (EV) D68–associated acute flaccid myelitis (AFM), and in some cases was associated with a worsening of the rare disorder. AFM can cause permanent paralysis in children.

Encouraged by preclinical research suggesting the drug may have some benefit in AFM, investigators found that the medication was well tolerated but that there was no signal of efficacy.

"The lack of an efficacy signal for the treatments for acute flaccid myelitis evaluated in this study emphasizes the need for development and prospective evaluation of more effective treatment and prevention strategies for this potentially devastating condition," study investigator Kevin Messacar, MD, Children's Hospital Colorado in Aurora said in a statement.

The study was published online November 9 in Neurology.

In 2014, 2016, and 2018, clusters of AFM were reported in the United States in association with a widespread outbreak of EV-D68 respiratory disease.

"We quite early knew we didn't have any effective treatment. Steroids didn't really work, plasma exchange didn't work, and some people tried antivirals, and they didn't really work," study coauthor Jay Desai, MD, an attending physician in the Division of Neurology at the Keck School of Medicine of the University of Southern California and a child neurologist at Children's Hospital Los Angeles, told Medscape Medical News.

Further complicating matters is the fact that the cause of AFM has not been absolutely established. "We suspect it is enterovirus D68," Desai said. However, he added, presence of enterovirus D68 has not been confirmed in the spinal fluid of those affected. The researchers also note that there could be more than one cause of AFM.

A selective serotonin reuptake inhibitor, fluoxetine is the only available medication approved by the US Food and Drug Administration that has in vitro antiviral activity against circulating 2014 EV-D68 strains.

"Given the long-term, potentially permanent paralysis associated with AFM, the lack of effective alternative therapies and the possibility of antiviral activity against EV-D68, fluoxetine was proposed as a possible therapeutic agent for AFM," the investigators write.

The attitude among the clinician-investigators, said Desai, was, "we should try fluoxetine because we have to try something."

To determine the safety, tolerability, and efficacy of fluoxetine for proven or presumptive EV-D68-associated AFM, the investigators conducted a retrospective observational cohort study of 56 children with AFM in 2015-2016 from 12 centers across the United States. The children ranged in age from 2.5 years to 9 years (median age, 3.8 years).

Study participants met clinical criteria for acute-onset limb weakness or cranial nerve dysfunction. Participants also had MRI evidence of lesions in the gray matter of the spinal cord or motor nuclei of the brainstem. Sign and symptom onset occurred between January 1, 2015, and November 1, 2016.

The researchers identified an enterovirus in 24 of 56 patients (43%), most commonly EV-D68 (n = 20, 36%), in respiratory or stool specimens.

At initial examination, the patients' summative limb strength scores (SLSSs) were similar, but the 28 patients who underwent treatment with more than one dose of fluoxetine were more likely to have EV-D68.

In the total cohort, 30% presented with asthma or another underlying medical condition. Almost all, 91%, had a prodromal illness; 71% had fever, and 73% had respiratory symptoms.

Neurologic weakness followed onset of the prodromal condition by a median of 8.5 days. Fever, meningeal signs, and limb pain often accompanied the weakness.

A median of two limbs were involved; 84% of participants had upper extremity weakness, 55% developed lower limb weakness, and 36% had cranial nerve dysfunction.

Patients received all therapies, including fluoxetine, at the discretion of clinical care providers. A total 82% received intravenous immunoglobulin, 59% received corticosteroids, and 14% underwent plasmapheresis.

Clinicians preferentially administered fluoxetine to EV-D68-positive patients.

Those who were more severely affected also received fluoxetine because, said Desai, "people were more desperate when kids were getting really ill."

The investigators compared 28 AFM patients who had been treated with more than one dose of fluoxetine to 26 patients who had not receive it. Two children who only received a single dose of the drug were considered part of the untreated group.

The study's primary outcome was change in SLSS in all four limbs between initial presentation and latest follow-up. Possible scores range from 20 (normal strength) to 0 (complete quadriparesis).

The researchers found similar muscle strength at baseline between the group that received fluoxetine and the untreated group, with mean SLSS scores of 12.9 vs 14.3 (P = .31).

A Worsening Effect?

However, the fluoxetine cohort experienced more severe paralysis at nadir (time of maximum muscle weakness: mean SLSS, 9.3 vs 12.8; P = .02) and at the latest follow-up (mean SLSS, 12.5 vs 16.4; P = .005).

In adjusted analyses, the mean change in SLSS at latest follow-up compared with initial examination was 0.2 lower (95% confidence interval [CI], −1.8 to +1.4) in fluoxetine-treated patients compared to 2.5 higher (95% CI, 0.7 - 4.4) in untreated patients (P = .02).

The researchers controlled for age, sex, additional therapies, and strength at baseline examination using propensity-weighted adjustments.

"We did an analysis to control for confounders. The scores were lower, ultimately, for kids treated with fluoxetine. Did this treatment actually make it worse? think it's a possibility," said Desai.

The fluoxetine group experienced a longer length of stay (median, 14 vs 7 days; P = .007). Treated patients also were more likely to require care in the intensive care unit (ICU), rehabilitation services, and ventilator and supplemental feeding support compared to untreated patients.

Fluoxetine was well tolerated, with no serious adverse events reported.

However, for two participants, the drug was discontinued after a single dose &mdash one because of perceived anxiety, and another because of weakness not severe enough to warrant further treatment. One participant in the fluoxetine group died. There were no deaths among the untreated patients.

The study had several potential limitations, said Desai. Fluoxetine treatment was initiated a median of 5 days after the onset of neurologic symptoms. However, at this point, most patients had reached their nadir of muscle weakness.

Because the study used retrospective data and had a nonrandomized, open-label design, the evidence is of class level IV.

"Despite significant limitations, this study has important implications to inform future therapeutic trials in AFM," the researchers note.

Desai noted that at this point there does not appear to be any other candidate agents for AFM on the horizon.

Leading clinicians and scientists are collaborating on accelerating research and clinical advances in AFM. The Acute Flaccid Myelitis Working Group, based at Johns Hopkins Medicine in Baltimore, for example, holds conference calls several times a week with thought leaders to address diagnosis, etiology, and care for affected children in the ICU setting and after hospital discharge.

Consensus articles to help guide clinicians caring for these patients are forthcoming, said Desai, who is a member of this working group.

"If there is going to be another wave of AFM, then we need to be better prepared," he said.

Lost in Translation

Commenting on the findings for Medscape Medical News, Emmanuelle Tiongson, MD, a pediatric neurologist at Children's Hospital Los Angeles, who was not affiliated with the study, said fluoxetine was promising, "especially early on" in the research.

"It had a direct antiviral effect on the enterovirus D68, but as is common with a lot of studies that start in the lab, once you take them to people, it doesn't always translate. A lot of agents fail when they get to human studies, because humans are very different from a petri dish or a mouse in the laboratory," she said.

Tiongson also noted that the doses of fluoxetine used in the study were higher than typical antidepressant doses, and because of the potential for side effects, "it would be difficult to justify it, in children especially, unless it was totally proven."

The study was funded by several grants from the National Institutes of Health. Dr Desai has received research support from the Epilepsy Foundation of Greater Los Angeles and funding from Novartis, Neurelis, and UCB. Dr Tiongson has disclosed no relevant financial relationships.

Neurology. Published online November 9, 2018. Abstract

Follow Damian McNamara on Twitter: @MedReporter. For more Medscape Neurology news, join us on Facebook and Twitter.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: