A pharmacokinetic and biomarker study of delayed-release dimethyl fumarate in subjects with secondary progressive multiple sclerosis: evaluation of cerebrospinal fluid penetration and the effects on exploratory biomarkers.

Background: Biomarkers are a useful and reliable measure of disease activity in many fields of medicine. Axonal and glial biomarkers in multiple sclerosis (MS) are being applied more often as technology is improving and becoming increasingly reliable. Nonclinical studies have shown dimethyl fumarate (DMF) to have cytoprotective and anti-inflammatory effects. The purpose of this study is to explore the pharmacokinetics (PK) of DMF (by measuring MMF, the active compound) in serum and cerebrospinal fluid (CSF) as well as relevant biomarker data for patients with secondary progressive MS (pwSPMS) and whether there is objective evidence for neuroprotection in pwSPMS treated with DMF.

Methods: Sixteen pwSPMS had serum and cerebrospinal fluid (CSF) evaluation for PK studies levels of MMF at various time points after ingestion of DMF. The CSF biomarkers neurofilament light chain (NfL), glial fibrillary acidic protein (GFAP), ubiquitin carboxyo-terminal hydrolase isozyme L1 (UCH-L1) and total tau (T-tau) were measured at baseline, week 6 and week 28 after initiating DMF with a starting dose of 120 mg twice daily for 4 weeks, followed by a maintenance dose of 240 mg twice daily. Clinical correlation of these patients with EDSS and MRI at these same time periods were made with the biomarkers. Four normal human volunteers had CSF studies for biomarkers at baseline.

Results: PK data showed CSF MMF concentration 11% of plasma with Tmax of plasma at 5 hr and Tmax of CSF at 7 hr. Biomarker data showed that CS NfL and to a lesser extent, GFAP, but not UCH-L1 nor T-tau showed relevant changes with clinical data. Some pwSPMS receiving DMF showed clinical improvements in Expanded Disability Status Scale (EDSS). Biomarker changes, but not MRI, correlated with clinical measures in this group of pwSPMS over the observation period.

Conclusions: PK data showed that the Tmax of CSF MMF peaked only 2 hours later than that of plasma with 11% measured in the CSF so that MMF readily crossed the blood brain barrier allowing potential direct penetration into the brain. NfL CSF levels, and to a lesser extent, GFAP CSF levels, showed correlation to disease activity in pwSPMS . These data suggest that DMF may have some benefit in reducing disease activity in pwSPMS if studied for a longer duration and larger well-controlled studies are warranted. DMF was reasonably well tolerated but 3 of the 16 patients did discontinue DMF at 6 weeks due to persistent side effects. NfL appeared to be more clinically relevant biomarker than brain MRI in this this group during the 28-week study period.