#MSCOVID19 two swallows don’t make a summer


As you know I have had to backpedal with my SARS-CoV-2/COVID-19 advice in relation to anti-CD20 therapies. I have now had to reinterpret data on the role of immunoglobulins in protecting people from developing COVID-19 and severe COVID-19.

I have been using the two Italian cases of X-linked agammaglobulinaemia who got COVID-19 and make a recovery as an argument that you don’t need B-cells and immunoglobulins to recover from SARS-CoV-2 infection. I now think I am wrong. Both these patients were being managed with immunoglobulin replacement therapy or IVIG (intravenous immunoglobulin therapy). IVIG is essentially a mix of immunoglobulins from blood donors from the general population. The assumption I made was that as SARS-CoV-2 was a new virus there would be no antibodies in the general population that would neutralize the SARS-CoV-2. However, the study below shows that pooled immunoglobulin therapies are able to neutralise SARS-CoV-2 (see Díez et al. below). Therefore I can’t assume, based on these two cases, that you don’t need immunoglobulins to make a recovery from COVID-19. This observation of neutralizing anti-SARS-CoV-2 activity in IVIG is also compatible with the emerging data that rituximab and by implication other anti-CD20 therapies now appear to increase your chances of getting COVID-19 and severe COVID-19. 

I suspect anti-CD20 therapies increase your chances of getting COVID-19 by reducing your chances of having an asymptomatic SARS-CoV2 infection. The immune responses to other human coronaviruses, the ones that cause the common cold, cross-react and help neutralise SARS-CoV-2, which explains why some people get asymptomatic or mild SARS-CoV-2 infections (see Shen et al. below). However, if you are B-cell depleted when getting the common cold, due to coronavirus, your immune system isn’t able to make the necessary high-quality or high-affinity cross-reactive antibodies that you now need to protect yourself from getting COVID-19. 

In support of this the case study published last week of a person with MS on ocrelizumab who failed to seroconvert to having anti-SARS-CoV-2 antibodies despite having confirmed SARS-CoV-2 COVID-19 (see Conte. case report below). This particular patient had mild hypogammaglobulinaemia as a result of ocrelizumab treatment. The failure to seroconvert could be an assay problem, i.e. low sensitivity, or is more likely to be due to the blunted B-cell response from being treated with ocrelizumab. Clearly this case report is going to be very important in shaping our thinking in terms of doing further anti-SARS-CoV-2 seroprevalence studies in people with MS and preparing our patients for a potential SARS-CoV-2 vaccine. The latter gets more pressing as most countries have now abandoned herd immunity as a strategy to deal with COVID-19.

Díez et al. Cross-neutralization activity against SARS-CoV-2 is present in currently available intravenous immunoglobulins. BioRxiv doi: https://doi.org/10.1101/2020.06.19.160879

Background: There is a crucial need for effective therapies that are immediately available to counteract COVID-19 disease. Recently, ELISA binding cross-reactivity against components of human epidemic coronaviruses with currently available intravenous immunoglobulins (IVIG) Gamunex-C and Flebogamma DIF (5% and 10%) have been reported. In this study, the same products were tested for neutralization activity against SARS-CoV-2, SARS-CoV and MERS-CoV and their potential as an antiviral therapy. 

Methods: The neutralization capacity of six selected lots of IVIG was assessed against SARS-CoV-2 (two different isolates), SARS-CoV and MERS-CoV in cell cultures. Infectivity neutralization was measured by determining the percent reduction in plaque-forming units (PFU) and by cytopathic effects for two IVIG lots in one of the SARS-CoV-2 isolates. Neutralization was quantified using the plaque reduction neutralization test 50 (PRNT50) in the PFU assay and the half maximal inhibitory concentration (IC50) in the cytopathic/cytotoxic method (calculated as the minus log10 dilution which reduced the viral titer by 50%). 

Results: All IVIG preparations showed neutralization of both SARS-CoV-2 isolates, ranging from 79 to 89.5% with PRNT50 titers from 4.5 to >5 for the PFU method and ranging from 47.0%-64.7% with an IC50 ~1 for the cytopathic method. All IVIG lots produced neutralization of SARS-CoV ranging from 39.5 to 55.1 % and PRNT50 values ranging from 2.0 to 3.3. No IVIG preparation showed significant neutralizing activity against MERS-CoV. 

Conclusion: In cell culture neutralization assays, the tested IVIG products contain antibodies with significant cross-neutralization capacity against SARS-CoV-2 and SARS-CoV. However, no neutralization capacity was demonstrated against MERS-CoV. These preparations are currently available and may be immediately useful for COVID-19 management. 

Conte. Attenuation of antibody response to SARS-CoV-2 in a patient on ocrelizumab with hypogammaglobulinemia. MSARDS June 20, 2020.

Shen et al. Delayed Specific IgM Antibody Responses Observed Among COVID-19 Patients With Severe Progression. Emerg Microbes Infect. 2020 Dec;9(1):1096-1101.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has spread rapidly worldwide since it was confirmed as the causative agent of COVID-19. Molecular diagnosis of the disease is typically performed via nucleic acid-based detection of the virus from swabs, sputum or bronchoalveolar lavage fluid (BALF). However, the positive rate from the commonly used specimens (swabs or sputum) was less than 75%. Immunological assays for SARS-CoV-2 are needed to accurately diagnose COVID-19. Sera were collected from patients or healthy people in a local hospital in Xiangyang, Hubei Province, China. The SARS-CoV-2 specific IgM antibodies were then detected using a SARS-CoV-2 IgM colloidal gold immunochromatographic assay (GICA). Results were analysed in combination with sera collection date and clinical information. The GICA was found to be positive with the detected 82.2% (37/45) of RT-qPCR confirmed COVID-19 cases, as well as 32.0% (8/25) of clinically confirmed, RT-qPCR negative patients (4-14 days after symptom onset). Investigation of IgM-negative, RT-qPCR-positive COVID-19 patients showed that half of them developed severe disease. The GICA was found to be a useful test to complement existing PCR-based assays for confirmation of COVID-19, and a delayed specific IgM antibody response was observed among COVID-19 patients with severe progression.

CoI: multiple

About the author

Prof G

Professor of Neurology, Barts & The London. MS & Preventive Neurology thinker, blogger, runner, vegetable gardener, husband, father, cook and wine & food lover.


  • MS 72yrs old on Rituximab. I am treated in. University MS speciality clinic and have sheltered in place for 114 days.
    I was infused last 2/10/20. I am curious about all.continuing research.
    Thank you for your work.

  • Thanks for this interesting information.

    Trying to apply it to my position – if you have been on anti CD20 for less than one year are you in a better position than if you had been on for example 3 years? For instance I have not had a cold at all since I started ocrelizumab 9 months ago. So I find it difficult to apply the cross antibody situation to myself as I wouldn’t have developed any anyway since I’ve been on the medication. Or have I missed a point.

    Also, MD showed us a paper over the weekend in which there were people with asymptomatic covid Infection who did not produce antibodies. So does this show that cross cover antibodies are not the only way you can have mild or indeed asymptomatic disease?

    The vaccine issue is much more pressing and is the reason I plan to ask to delay my next dose and switch to a vaccine compatible DMT. I am very grateful to this blog for bringing all of this up to date info to me.

    • Better position possibly…With time and repeated infusions you more likely to make less antibodies so by 6 years of treatment the IgM levels have dropped by over 50% and this occurs in ~30% of people at 2 years it is ~15-20% and whilst is can occur in year If you become someone who makes low antiboody level then they are about 1.5 times to get a severe infection, is this the reasons for profGs concern.
      ON the next point, well spotted I am sure there are other ways of having asymptomatic disease and my best bet is because you have an innate immune response that is highly effective. Animals get rid of the virus in 3-5days and they have never had a cold and have never seen a virus before as they live in a bubble and there are very few antibody-making cells that are made that quick.
      Vaccine thing..are we sure they are going to work well enough

  • “I now think I am wrong”. I like your honesty and humility. However, I think the issue is that you are too quick on the draw and don’t take time to breath. You are one of many MSologists in the UK, but as soon as Covid-19 came on the horizon you were like a Duracell Bunny – pumping out posts, taking the lead on the advice provided to MSers on therapies….. Learn to share the load, delegate, let’s other MSologists take on more. No one went to their grave and said on their deathbed “I wished I’d spent more time at work”. I wonder how much of your weight loss is Keto and how much is due to the stress you put yourself under? You need a six month break. Then come back and focus on two MS things (EBV and neuro-protection). Prof G should be a sniper (all about precision and accuracy) not a double barrel shotgun user (you’ll hit the target, but they’ll also be collateral damage). You are a victim of Covid-19 despite not contracting the virus. You need time to breath and recover (I’m sure Mrs G would concur).

    • So which MSologists have responded to the pandmic?, clearly the MS register but who from UK has made valuable comments….may you the readers can enlighten me.

      As for Shot guns it seems shotguns are the norm….HSCT almtuzumab are the punt guns

    • There is more data that will emerge from all the databases that are collecting cases prospectively. I anticipate seeing this data in the next 4-6 weeks.

  • How is hypogammaglobulinemia defined ? What percentage of OCR patients develop it ? How long do they need to be on the drug before it is detected ? I suspect less than 400 is defined as hypogammaglobulemia ? As well, is there a gender or age predilection ? And since CD20 cell death must not, in theory, cause hypogammaglobulinemia per se, why does the gamma globulins diminish with OCR use ?


      Abstract: 65

      Title: Serum immunoglobulin levels and risk of serious infections in the pivotal Phase III trials of ocrelizumab in multiple sclerosis and their open-label extensions

      Type: Scientific Session

      Abstract Category: Therapy – Immunomodulation/Immunosuppression

      Authors: T. Derfuss1, M.S. Weber2,3, R. Hughes4, Q. Wang4, A. Sauter4, H. Koendgen4, S.L. Hauser5, A. Bar-Or6, H.-P. Hartung7

      1University Hospital Basel, University of Basel, Basel, Switzerland, 2Institute of Neuropathology, Universitätsmedizin Göttingen, 3Department of Neurology, Universitätsmedizin Göttingen, Göttingen, Germany, 4F. Hoffmann-La Roche Ltd, Basel, Switzerland, 5University of California, San Francisco, CA, 6Department of Neurology and Center for Neuroinflammation and Experimental Therapeutics, Perelman School of Medicine, University of Pennsylvania, Pennsylvania, PA, United States, 7Department of Neurology, UKD, Center of Neurology and Neuropsychiatry and LVR-Klinikum, Heinrich-Heine University Düsseldorf, Düsseldorf, Germany

      Background: Exposure-dependent decreased serum levels of immunoglobulin (Ig)G, IgM and/or IgA can occur in patients treated with B-cell-depleting therapies. An increased risk of serious infections (SIs) has been observed, mainly with reduced serum IgG levels.

      Objectives: To assess serum Ig levels over 5.5 years and evaluate a potential association between a decrease in IgG, IgM or IgA levels and SIs, in the Phase III OPERA I/OPERA II (NCT01247324; NCT01412333) and ORATORIO (NCT01194570) trials/open-label extensions of ocrelizumab (OCR) in multiple sclerosis (MS).

      Methods: Serum IgG, IgM and IgA levels were measured at least every 24 weeks. Lower limit of normal (LLN) for IgG, IgA and IgM was defined as 5.65g/L, 0.70g/L and 0.40g/L, respectively. SI rates during episodes of < LLN were compared with SI rates during episodes of ≥ LLN. Episodes of < LLN were defined as from the day the laboratory value reached < LLN, until normalisation. Analyses are presented using a July 2018 data-cut. Results: Over 5.5 years of OCR treatment, the reduction in serum levels (relative reduction [%]; mean decrease from baseline to 264 weeks [g/L]) in relapsing MS were: IgG, 17.0% (from 10.53 to 8.79); IgA, 21.3% (from 2.13 to 1.74); IgM, 58.1% (from 1.35 to 0.60) and in primary progressive MS were: IgG, 16.9% (from 10.68 to 8.96); IgA, 20.5% (from 2.17 to 1.76); IgM, 56.3% (1.37 to 0.64). At Week 264, the proportions of patients reaching IgG, IgA and IgM levels < LLN were 5.7%, 5.4% and 29.2%, respectively. Overall, 14 SIs occurred during a drop in IgG levels < LLN (14 adverse events [AEs] per 215.5 patient years [PY], equating to a rate of 6.50 AEs/100PY), compared with IgG levels ≥ LLN, (191 AEs/9049.1PY; 2.11/100PY). Five SIs occurred during a drop in IgA levels < LLN (5 AEs/215.8PY; 2.3/100PY) vs IgA levels ≥ LLN (200 AEs/9038.7PY; 2.21/100PY). A total of 64 SIs occurred during a drop in IgM < LLN (64 AEs/1749.1PY; 3.66/100PY) vs IgM levels ≥ LLN (141 AEs/7515.6PY; 1.88/100PY). No pattern of SIs was identified in terms of organ affected, pathogen, duration, latency or severity comparing < LLN/≥ LLN for each Ig subclass. All SIs resolved with standard care, whilst patients continued OCR therapy. Updated analyses (January 2019 data-cut) will be presented. Conclusions: Over 5.5 years of OCR treatment, a reduction in serum Ig levels was observed, with an apparent association with increased rates of SIs. The association was strongest for IgG and less so for IgM or IgA.

    • Some of your questions are answered in our papers and I have a post lined up that will address some of those issues.

      Why does anti-cd20 depletion reduce antibody is a good question, I can make some suggestions but dont know the answer

  • Why not give normal human IG to people on continuous anti CD20 during this pandemic?

    Is this an issue with long-term anti CD20 treatment anyway in that even long lived plasma cells don’t live forever? (words I read here somewhere).

    • There is an indeterminate risk with any blood product. You only have to think about all the haemophiliacs who were infected with HIV before we knew of it’s existence to see that. If we can avoid blood products it seems sensible.
      As we saw from the German study 100% of 903 patients with MS have EBV and that immortalises B-cells.
      I suspect we’ll see that the interval between doses of anti-CD20 drugs is increased with the duration of such treatment.
      I’m keen to see more from the research into hyperbaric oxygen treatment of CoVID-19 as this might render fear of infection unnecessary.
      All this talk about vaccine readiness assumes that creative efforts towards vaccine production are successful, but there is no guarantee of that. If vaccine production is successful and mass vaccination becomes commonplace, herd immunity will protect those of us on anti-CD20 treatment.

  • This is going to sound like a silly question, but is it still innate immunity T cells macrophages etc that are the first line of defence before the antibodies?
    So patients on anti CD20 still have some hope as that portion of their immune system is effective and therefore there is still a chance they can have mild disease even without cross cover antibodies etc?

    • They will be there before the antibodies and the data to date suggusts that most (~75%) re people on anti-CD20 have a mildish course

  • Re “I suspect anti-CD20 therapies increase your chances of getting COVID-19 by reducing your chances of having an asymptomatic SARS-CoV2 infection. ”

    Are there any reported cases that can confirm this suspicion?

    With so many variables involved, it seems easy to relevant factors that could in such predictions

  • Re ’emerging data that rituximab and by implication other anti-CD20 therapies now appear to increase your chances of getting COVID-19 and severe COVID-19′

    How many covid cases reported of patients on rituximab/ocrelizumab?
    How much more likely is severe infection?

    • In todays post you can see the influence on about 400 people and about 15% were hospitalised, reports from China reported about 20% hospitalisation.This will be a biased sample

By Prof G



Recent Posts

Recent Comments