Regional variation of remyelination with bexarotene 1 of 4
Experimental work before the trial Huang et al 2 of 4
Primary endpoint failed but MTR of demyelinated lesions is improved by bexarotene 3 of 4
Bexarotene restores conduction in demyelinated optic nerves 4 of 4
Background
Scientific rationale
The rationale for this trial rests on the work of Professor Robin Franklin and colleagues at the University of Cambridge. Read the key paper here: : Huang Franklin Nat Neurosci bexarotene.
Huang and colleagues discovered that the retinoid acid receptor RXR-gamma was expressed during remyelination in the rat; and that human oligodendrocyte lineage cells expressed RXR-gamma in active and remyelinated multiple sclerosis lesions. Antagonism of RXR-gamma severely inhibited oligodendrocyte differentiation in culture, and in RXR-gamma knock-outs differentiation of oligodendrocyte precursor cells after demyelination was delayed. The key finding, that informs this trial, was that administration of the RXR agonist 9-cis-retinoic acid to demyelinated cerebellar slice cultures, and to aged rats after demyelination, caused increased remyelination of axons. The scientific motivation behind this trial was to assess whether RXR-agonists promote remyelination in people with multiple sclerosis. An important observation from the Franklin group is that remyelination is normally effective and rapid in young animals but declines in efficiency with age.
Bexarotene is an agonist of the retinoid X receptor (RXR) (acting on the alpha, beta and gamma subtypes), which is licensed for human use. Whilst there is recognition that more selective agonists might have fewer adverse effects, none of the newly synthesised compounds are licensed. We therefore proposed bexarotene as the most suitable agent to test the role of RXR agonists in human remyelination.
Funded by the MS Society, with additional support from the Horne Family Trust and the Grand Charity of the Freemasons.
Conclusions: Despite a negative primary efficacy outcome, we found both magnetic resonance imaging [MTR] and visual evoked potential evidence that RXR agonism promotes remyelination in people with multiple sclerosis. Although bexarotene’s safety profile precludes its widespread use, these data support efforts to develop a selective RXR-gamma agonist.
Trial results
Presented at ECTRIMS on Septemebr 26th 2020
Methods: In a double-blind, placebo-controlled, phase 2a trial (Cambridge Centre for Myelin Repair: CCMR-One), participants aged 18-50 years with relapsing remitting multiple sclerosis, stable on dimethyl fumarate for at least 6 months, were randomised to bexarotene 300mg/m2 or placebo for 6 months. The primary efficacy outcome was change in mean lesional magnetisation transfer ratio (MTR) for lesions whose baseline MTR was below the median lesional MTR for that patient. The secondary efficacy outcome was change in full-field visual evoked potential (VEP) latency in eyes with electrophysiological evidence of optic neuropathy (baseline latency >118ms). We analysed by intention to treat.
Results: 52 patients were randomised 1:1 to receive six months of bexarotene or placebo. Two placebo patients withdrew before receiving study drug and one bexarotene patient withdrew consent during the trial. All bexarotene patients experienced adverse effects, notably central hypothyroidism (26 [100%]), hypertriglyceridaemia (24 [92%]), rash (13 [50%]) and neutropenia (10 [38%]). Two discontinued placebo because of adverse events and five discontinued bexarotene because of rash [2], neutropenia, triglyceridaemia and deterioration in MS). The primary efficacy outcome was negative (mean submedian lesion MTR change was 0.25pu in the bexarotene group versus 0.09pu in the placebo group, p=0.54), but in an exploratory, lesion-level analysis, though treatment difference in submedian lesions was too small to achieve significance, it was significantly greater than in supermedian lesions (p=0·007). The secondary efficacy endpoint was met, with bexarotene treatment reducing full field visual evoked potential latency compared to placebo, in those with optic neuropathy by 4·66 ms/eye (95% CI -8·38 -0·93; p=0·014).
We are committed to open access of trial datasets to researchers.