The Effect of Ocrelizumab on JC Virus Antibody Index

Background: The JC virus antibody index is a two-step ELISA to detect antibody to VP1, a JCV structural protein. This test has been validated as a risk factor for progressive multifocal leukoencephalopathy (PML) with natalizumab and often employed with other disease modifying therapies (DMTs) that carry a risk of PML though it has not been validated in the latter population. The long-term effect of anti-CD20 monoclonal antibodies suggests that a decline in anti-JCV antibody index would be anticipated with prolonged therapy as has been reported with rituximab (Friedman-Urevich et al 2017; Baber et al 2018). However, at least one study (Rempe 2020) of 145 patients on ocrelizumab demonstrated no change in JCV antibody index over 2 years of treatment.

Objectives: The primary outcome measure was the magnitude of decline in the anti-JCV antibody index over the course of 2 years of ocrelizumab treatment. A secondary outcome measure was the rate of conversion from JCV antibody seropositive at study entry to indeterminate or seronegative over the 2-year course of the study.

Methods: JCV antibody index were recorded at approximately 6-month intervals over a 2-year period in 645 persons with multiple sclerosis. JCV antibody index levels were available before the initiation of treatment in 560. The initial JCV antibody index level was compared to that after 2-years of treatment. Variables such gender, age at treatment initiation, prior DMT, immunoglobulin levels were also recorded.

Results: Of the 560 patients in whom JCV antibody index using STRATIFY JCV ELISA (Quest Diagnostic, San Juan Capistrano, CA) was assessed prior to the initiation of ocrelizumab, 360 (54.6%) were JCV seropositive (>0.4) and 200 (45.4%) were seronegative (0.0-0.20) or indeterminate (0.2-0.4). Over the course of 2 years, among those individuals who were JCV seropositive, there was a small though statistically significant trend towards lower antibody titers; however, no individuals became persistently seronegative. The average decline in the JCV antibody index was 0.257 (95% CI -0.318 to -0.196). In a small number of individuals, there was a reclassification at repeat testing between seronegative, indeterminate, and seropositive, but only 3 individuals who were seronegative at initial assessment converted to seropositive status at 2 years.

Conclusions: This study confirms a small though statistically significant trend to lower JCV antibody index titers in individuals treated with ocrelizumab over 2 years. The small number of individuals becoming seropositive during the course of treatment is in keeping with the observation in other populations and is unlikely to have been the effect of the therapy. The rarity of PML with ocrelizumab and the results from this study indicate that measuring JCV antibody index in this population is unlikely to be of any prognostic value.

This study was supported by a research grant from Genentech/Roche.