As approximately 70% of our newly diagnosed people with MS go onto ocrelizumab (a CD20 and thus B cell-depleting monoclonal antibody), a study that helps us understand how ocrelizumab affects the potential to mount (protective) antibody responses to vaccines and new antigens is more than welcome. Especially, as we are all hoping for the COVID-19 vaccine to be under the Christmas tree.
B cells are the immune cells responsible to generate antibodies (= humoral immunity), and these antibodies are, at their turn, necessary to defend ourselves against new and old foes (= viral and bacterial pathogens). However, B cell depletion with ocrelizumab and other CD20-monoclonal antibodies (i.e. rituximab, ofatumumab, ublituximab) decimate valuable B cell subpopulations (i.e. naïve and memory B cells) and jolt the ‘germinal centres’ in the lymph nodes.
This is detrimental to the production of qualitative antibodies because:
- Naïve B cells egress from the bone marrow and scout the peripheral blood for antigens. If they encounter a cognate antigen, they hurry to the lymph nodes and elicit a germinal center response.
- Memory B cells hover in our blood and tissue for decades, and can generate an accelerated antibody-mediated immune respons in case of reinfection with an antigen you have encountered (potentially years) before.
- The germinal centres are the factories for the generation of affinity-matured B cells (= long-lived plasma cells) specialised in producing improved, very specific antibodies that effectively recognise infectious agents, and for the production of memory B cells.
In the VELOCE study, researchers evaluate if people treated with ocrelizumab who are fully B cell depleted are able to raise an antibody response to vaccines and new antigens. The study population is being exposed to four different vaccine/antigenic triggers: tetanoid, pneumococcal and influenza vaccine as well as keyhole limpet hemocyanin (KLH). Importantly, the antigenic triggers require a different sort of immune equipment, which can be subdivided as follows:
For tetanoid, pneumococcal and influenza vaccine it is very likely that people starting with ocrelizumab will already have encountered some of the vaccine epitopes in the past. Tetanoid vaccine requires boosters every ten years, and pneumococcal and influenza strains are omnipresent pathogens. This means that long-lived plasma cells (which do not carry CD20 on their cell surface and are thus not depleted by ocrelizumab) can still exert their function and provide protection with well-targeted antibodies in the event of a recurring infection.
The VELOCE study showed you are half as likely on ocrelizumab compared to controls to mount an antibody response against tetanoid vaccine and two thirds less likely to mount an antibody response to all the included pneumococcal serotypes. Based on these figures, it is fair to say that the antibody response to these vaccines is poor but not non-existent.
On the other hand, the VELOCE study evaluated the response to the keyhole limpet hemocyanin (KLH) which is an entirely new antigen. This means it will have to be recognised by the naïve B cells (or other antigen-presenting cells) and subsequently brought to the germinal centres in the lymph nodes.
The results of the VELOCE study are worrisome as 12 weeks after KLH administration there is a 5-fold difference in IgM antibody levels and an 11-fold difference in IgG antibody levels between ocrelizumab and controls. This means that the immune response against new pathogens (and thus potentially also COVID-19) cannot rely on an efficient antibody response and will have to be dependent on cellular immunity (e.g. cytotoxic T cells).
How to prevent stab wounds?
The VELOCE study has clearly pointed out that an antibody response to vaccines/new antigens while being B cell depleted is unreliable. Moreover, the people included in the VELOCE study only had received a single course of ocrelizumab. This implies that the picture might become more gloomy if you are continuously treated with six-monthly ocrelizumab infusions. We know that this causes hypogammaglobulinemia and prohibits replenishment off the long-lived plasma cell pool. Although this is definitely not good news, there are some ways to mitigate the risk.
First, the results stress the importance of the biggest medical and human flaw: understanding the need for preventive action. The window of opportunity for a protective antibody response is clearly located before people start with ocrelizumab. Although pneumococcal and influenza vaccines are currently recommended rather than obligatory according to the product characteristics, the VELOCE results underscore the potential gain of this recommendation in terms of meaningful humoral immunity. In this context, we should also consider vaccinating our pwMS with the new Shingrix vaccine against the varicella zoster virus. In the phase III ocrelizumab trial, approximately 5 % of the people in the ocrelizumab group were affected with or oral or cutaneous zoster (compared to 3% in the interferon-beta group). Based on the reported efficacy of the Shingrix vaccine, the occurrence of zoster infections could be reduced with more than 90%. Imagine!
Second, in the context of the current pandemic a low-dose high-frequency CD20-regimen with ofatumumab might be of interest. Ofatumumab is a new CD20-depleting antibody that will join the CD20 forces shortly, and is administered subcutaneously every month. If you have been treated with ocrelizumab (six-monthly high dose-infusions), it takes 72 weeks for your total B cell counts to reach the lower level of normal. If you have been treated with ofatumumab, it takes approximately 38 weeks which is twice as fast. Admittedly, this is still not very fast but definitely an advantage in terms of vaccine readiness. Obviously, vaccine readiness is not the full story when it comes to picking a CD20-depleting antibody as high-dose regimens have potential other advantages when it comes to preventing brain volume loss (cfr. DODO-study).
Amit Bar-Or et al. Neurology 2020
Objective The phase IIIb A Study to Evaluate the Effects of Ocrelizumab on Immune Responses in Participants With Relapsing Forms of Multiple Sclerosis (VELOCE) study (NCT02545868) assessed responses to selected vaccines in ocrelizumab (OCR)-treated patients with relapsing multiple sclerosis.
Methods Patients were randomized 2:1 into the OCR group (n = 68; OCR 600 mg) or control group (n = 34; interferon beta or no disease-modifying therapy). All received tetanus toxoid (TT)-containing vaccine, Pneumovax (23-valent pneumococcal polysaccharide vaccine [23-PPV]), and keyhole limpet hemocyanin (KLH). The OCR group was subdivided into OCR1 (n = 33) and OCR2 (n = 35) at randomization. The OCR1 group received Prevnar (13-valent conjugate pneumococcal vaccine) 4 weeks after 23-PPV; the OCR2 and control groups received influenza vaccine. Vaccinations started 12 weeks after OCR initiation (OCR group) or on day 1 (control group).
Results Positive response rate to TT vaccine at 8 weeks was 23.9% in the OCR vs 54.5% in the control group. Positive response rate to ≥5 serotypes in 23-PPV at 4 weeks was 71.6% in the OCR and 100% in the control group. Prevnar did not enhance response to pneumococcal serotypes in common with Pneumovax. Humoral response to KLH was decreased in the OCR vs control group. Seroprotection rates at 4 weeks against 5 influenza strains ranged from 55.6% to 80.0% in the OCR2 group and 75.0% to 97.0% in the control group.
Conclusion Peripherally B-cell–depleted OCR recipients mounted attenuated humoral responses to clinically relevant vaccines and the neoantigen KLH, suggesting that use of standard nonlive vaccines while on OCR treatment remains a consideration. For seasonal influenza vaccines, it is recommended to vaccinate patients on OCR because a potentially protective humoral response, even if attenuated, can be expected.
Classification of evidence This study provides Class II evidence confirming that the humoral response to nonlive vaccines in patients with relapsing multiple sclerosis after OCR treatment is attenuated compared with untreated or interferon beta–treated patients, but they can still be expected to be protective.
Clinicaltrials.gov identifier NCT02545868.