The limited benefit of B cell-directed therapies in progressive MS has been attributed in part to poor penetration of the CNS by monoclonal antibodies. However, two new studies have reported that intrathecal administration of rituximab has little effect on CNS inflammation, suggesting that limited drug entry into the brain is not the main problem.
The first study is the long-term extension of the Intrathecal Treatment Trial in Progressive Multiple Sclerosis (ITT-PMS) study (Bergman et al. J Neurol 2021;268:651-657). In the initial study, 23 PMS patients received three weekly doses of rituximab 25 mg intrathecally via an Ommaya reservoir (Bergman et al. Neurology 2018;91:e1893-e1901). No treatment effect was seen during the one-year follow-up period. Eligible patients could then enrol in a two-year extension phase to evaluate stabilization of neurological worsening, MS symptoms and quality of life. At the end of the three-year study period, there was a significant 33% worsening from baseline in walking speed (median 1 m/sec decreasing to 0.65 m/sec). There was a slight improvement (1-point change) in SDMT scores, but no effect on hand function (9-Hole Peg Test), MS fatigue or other clinical measures. Two cases of bacterial meningitis were reported.
The second study is the phase II EFFRITE study (Bonnan et al. Mult Scler Int 2021;2021:8813498). Ten PMS patients without active disease for >2 years (mean age 62 years) were randomized to intrathecal rituximab 20 mg, intrathecal + intravenous rituximab 375 mg/m2, or placebo All patients were pretreated with paracetamol, dexchlorpheniramine and methylprednisolone. IT rituximab was infused after depletive lumbar puncture and patients were placed in the Trendelenburg position (15 degrees) for one hour. IV rituximab was started one hour after IT rituximab. Endpoints included the change in osteopontin and neurofilament-light (NfL) levels in CSF; IgG synthesis (OCBs, IgG index); and 3T-MRI changes. Clinical endpoints were EDSS, T25FW, SDMT, 9HPT, and the Modified Fatigue Impact Scale (MFIS). CSF sample collection was started early (days 4, 21) since IT rituximab is cleared from CSF within a few days.
There was complete B cell depletion in the IT + IV rituximab group by day 4, which was sustained for six months; B cell recovery remained incomplete at month 12. In the IT-only group, there was complete B cell depletion in 3 of 4 patients by day 4, with some B cell recovery at day 21. There were no significant between-group differences in osteopontin or NfL levels in CSF during treatment, although small decreases in NfL were seen at day 21 and at 6 months. No clinical benefits were seen on EDSS, SDMT, MFIS or 9HPT. There was no difference in lesion load with the addition of IT rituximab and no apparent effect on leptomeningeal enhancement. The authors speculated that rituximab does not adequately diffuse into brain tissue so resident B cells may not have been depleted, an idea first suggested in the RIVITALISE study (Komori et al. Ann Clin Transl Neurol 2016;3:166-179).
These negative results were consistent with those obtained from previous studies. In the RIVITALISE study, IV + IT rituximab failed to deplete B cells in CNS tissue (-10% to -20%) despite significant B cell depletion in CSF (-79%). NfL levels were unchanged. Similar results were seen in a small study of intrathecal rituximab (2 x 25 mg two weeks apart) in MS patients with meningeal inflammation (Bhargava et al. Mult Scler Relat Disord 2019;30:136-140). Rituximab had no demonstrable effect on leptomeningeal enhancing lesions over the 24-week study despite B cell depletion in CSF and peripheral blood. B cell depletion in the CNS was transient, which may have been due to the low dose or rapid drug clearance from CSF. The authors speculated that leptomeningeal enhancement may persist due to drug resistance resulting from higher levels of B cell survival factors (e.g. B cell activating factor, BAFF), and/or ongoing inflammatory activity by other immune cells. It has also been suggested that rituximab-mediated internalization of CD20 by B cells, or inadequate complement for complement-dependent cytotoxicity (CDC), may limit the efficacy of rituximab in the CNS (Rolfes et al. BioDrugs 2020; 34:587-610).