Roche is the first company to publish their data on how patients with MS are doing with COVID-19 who are being treated with ocrelizumab. Please note the data is on the first 100 cases that have been reported to Roche via their adverse event (AE) reporting mechanisms. Twenty-six of the cases have been reported as being severe with 5 being critical (see table below). These figures are pretty much in line with what happens in the general population in relation to COVID-19 outcomes with a similar proportion of severe and critical cases. The one caveat is that ocrelizumab-treated patients may be younger and have fewer comorbidities than the general population.
Some of my colleagues have accused Roche of hiding data and selective reporting. Why would they do that? Pharma’s pharmacovigilance systems are transparent and open to audit by the regulators. It is not in Roche’s interests to not be open and frank about their data. At the end of the day, the data is what it is and we need to use it to help us make decisions about treatment.
I agree the data is full of warts and open to interpretation. It is likely to be biased in that clinicians are more likely to report hospitalised and severe cases as AEs and miss the milder cases. For example, I have collected 9 patients who have likely had COVID-19 over the last 8-10 weeks. Only one of these nine patients was admitted to a hospital with moderate COVID-19. In this case, the COVID-19 syndrome was confirmed on chest x-ray and other clinical features and her nasopharyngeal swab was positive for SARS-CoV-2. She was fortunately discharged after 4 days well and didn’t need any ventilatory support. The other 8 patients with MS all self-isolated and recovered at home and had no swabs taken. Only the admitted patient is part of the UK’s official figures the other 8 cases are not. Patients with COVID-19 who have mild disease, who don’t come to the hospital to get swabbed don’t get counted in the official figures. The same phenomenon is almost certainly happening with ocrelizumab and our other DMTs. Mild cases are not getting swabbed, diagnosed and reported. I, therefore, suspect that the Roche data represents the worse end of the spectrum.
Overall the Roche data is in keeping with the Italian, French and other registry data that people with MS on ocrelizumab don’t appear to be at higher risk of getting severe COVID-19. This also needs to be interpreted in the context of the immunology of COVID-19 and SARS-CoV-2. It looks as if innate immunity (monocytes and macrophages) and T-cells, in particular CD8+ T-cells, are the most important lines of defence against SARS-CoV-2. The fact that two patients with X-linked agammaglobulinaemia recovered from COVID-19 and that anti-CD20 treated patients with no B-cells in the peripheral blood are recovering from COVID-19 tells you that B-cells are not essential for the antiviral response. The latter is also in keeping with the trial data that patients on B-cell depleting therapies don’t seem to have a problem dealing with viral infections.
Could there be an upside to B-cell depletion? Yes, I suspect there may be. As the humoral or antibody response emerges antibody-mediated damage to lung with complement activation may be responsible for some of the delayed tissue damage that occurs in COVID-19. This is why it will be important to get more data and better-defined comparator groups to see if anti-CD20 therapy treated patients may be doing better than expected. The recently presented Swedish data suggests not. However, the Swedish data is on rituximab, and not ocrelizumab, and the rituximab doses used in Sweden may not be high enough to block antibody responses to SARS-CoV-2 and hence the data can’t simply be extrapolated to ocrelizumab or vice versa. For those of you who don’t know ocrelizumab is a much more potent B-cell depleter than rituximab.
Now, what about vaccine readiness? Are people on ocrelizumab less likely to develop an antibody response after COVID-19 and if they do is the antibody response good quality, i.e. capable of neutralizing the virus and preventing reinfection? We are in the process of validating an ultrasensitive assay and doing an antibody study to answer this question. So watch this space.
Even if antibody responses to SARS-CoV-2 are blunted in ocrelizumabers, who have had COVID-19, this doesn’t mean they are not immune to reinfection. Cellular responses are likely to be sufficient to prevent reinfection. The latter has been shown to occur with the measles virus.
Vaccine readiness may become a real issue if a vaccine for SARS-CoV-2 emerges. However, I suggest crossing that bridge when we get there. With anti-CD20 therapies, all you will have to do is miss a dose or two, wait for naive B-cell reconstitution and then have the vaccine. We have very good data that after 3 or 4 courses of ocrelizumab missing infusions for the next 12-18 months is unlikely to affect MS disease activity. The latter observation is why we are still planning to do an adaptive-dosing study in the UK (ADIOS Study).
I personally want to congratulate and thank Roche for putting their data into the public domain so rapidly. I have asked other Pharmaceutical companies to do the same. Having access to this data alongside other real-life data sets is what we and others are using to adjust treatment guidelines and is why we as an MS treatment centre are starting and redosing patients with MS with ocrelizumab. At last, we can say our practice is evidence-based rather than opinion-based.
Hughes et al. COVID-19 in persons with multiple sclerosis treated with ocrelizumab – a pharmacovigilance case series. MSARDs Available online 16 May 2020, 102192.
CoI: Multiple. Importantly I am a steering committee member on the ocrelizumab phase 3 development programme and I am PI on the ORATORIO-HAND study of ocrelizumab in PPMS.