Dr Kim discusses recent research he presented at the ACR Convergence demonstrating how patients on various immunosuppressive therapies respond to vaccines—including the vaccine against SARS-CoV-2—and some surprising results regarding anti-tumor necrosis factor agents and corticosteroids.

Alfred Kim, MD, is an assistant professor of medicine and director of the Lupus Clinic at the Washington University School of Medicine in St Louis, Missouri.

TRANSCRIPT:

Dr. Alfred Kim:  Hi, my name is Al Kim. I'm an adult rheumatologist at Washington University School of Medicine in St. Louis, Missouri.

Today, I'll be talking about the plenary presentation that our team had at this year's ACR Convergence examining the immunogenicity of vaccine responses in the immunosuppressed with rheumatic diseases.

The study itself is called COVaRiPAD, or COVID-19 Vaccine Responses in patients with Autoimmune Diseases. This was a prospective observational study that examined both immunogenicity and reactogenicity of participants from the pre-vaccination to 5 or 6 months postvaccination. Then, we also have a substudy examining the effect of the additional dose.

The data that we presented at ACR focused on the acute antibody response, so this is going to be the immediate time point about 2 to 4 weeks after the initial series of vaccinations. We've predominantly had mRNA vaccines in our study, and then we also presented some data looking at cross-variant neutralization, specifically to Delta variant.

The top line results from our study demonstrated that in the acute antibody response phase, that 90% of our participants with chronic inflammatory diseases — this includes rheumatoid arthritis, inflammatory bowel disease, psoriasis, psoriatic arthritis, systemic lupus erythematosus, and vasculitides — had detectable antibodies. This is actually very good news.

Generally speaking, when you're immunosuppressed, we do that in order to attenuate immune responses because of the autoimmune disease state. Obviously, when the immune response is needed, such as in vaccine responses, the immunosuppression can blunt those type of responses.

Having 90% of the participants generate an antibody response was encouraging to us. The lining within that is that we did see, at the population level, about a 3-and-a-half-fold reduction in antibody titers to the spike protein across the board. This is not unexpected that we would have reduced responses in certain populations.

The home run point here is that people are generating responses, although they're just not as robust as those people not on immunosuppression. Of course, we don't know right now whether or not that will confer reduced clinical protection against COVID-19.

These breakthrough infection studies are ongoing, particularly in the era of Delta variant, so this is something we are going to be also looking at in the future. Getting back to the data we presented, we also knew that there's a wide variety of immunosuppressive classes.

We wanted to look and see which class of immunosuppressives conferred the worst outcomes, in terms of antibody responses. Not surprisingly, we found that B-cell-depleting agents, such as rituximab or ocrelizumab, was associated with the greatest reductions in antibody levels.

We found a 57.7-fold reduction of titers in our cohort, which is similar to several other groups globally, demonstrating the dramatic impact of B-cell-depleting therapies on antibody responses. This is to be predicted, though, because B cells are required to be activated to eventually make antibody.

If you don't have B cells when you get vaccinated, you're very likely not to make antibodies. This is something we would have predicted. Another class of medications that we found detrimental to antibody responses was going to be the mycophenolate mofetil or mycophenolic acid.

This is predominantly used in more severe autoimmune diseases, say, severe lupus, for some cases of multiple sclerosis. It's also widely used in the transplant field. Here, we found a 20-fold reduction in antibody levels compared to those who are not on immunosuppression.

Again, this has been recapitulated by several other groups. Laura Boekel in the Netherlands and Julie Paik at Johns Hopkins have found similar observations. This is also something that was observed within the solid organ transplant field.

Again, the first is B-cell-depleting agents, and then it's going to be mycophenolic acid or mycophenolate use. Finally, the third one that had the most dramatic effect on vaccine responses was glucocorticoid use, such as prednisone, prednisolone, methylprednisolone.

Here, we found a 9-fold reduction in antibody titers. This is not necessarily surprising either, because glucocorticoids are such a sledgehammer on the immune system.

One thing that we did find that was surprising was that the dose did not seem to influence the level of antibodies. In other words, what we've found was that even in low-dose prednisone users — this is 5 milligrams or less — we found about half the people were unable to generate good antibody responses.

This is a bit of a confusing story because there's data — again, generated by Laura Boekel and Julie Paik — that showed that prednisone monotherapy users were able to generate very good antibody responses.

Julie Paik further showed that the poor antibody responses seen in prednisone users were only seen in those that had other immunosuppressive medications such as mycophenolate, which we had just discussed has a major impact on reducing antibody responses.

Another explanation of our data is that the people on low-dose prednisone were also on other immunosuppression. We are in the process of doing statistical analyses to confirm this hypothesis.

This will then help us determine whether or not any prednisone use or a certain level of prednisone use is going to be the main factor that drives reduced antibody responses. Keep your ears to the ground regarding those analyses.

Other immunosuppressive medications, such as methotrexate, TNF inhibitors, IL-12/23 inhibitors, generally reduced antibody responses very modestly, somewhere between the 2- to 3-fold reduction.

When you look at those people and the spread of antibody levels in those people, the majority of these people had antibody levels that were consistent with antibody levels within the immunocompetent group. In other words, those that did not have any immunosuppression.

This is very encouraging for the bulk of those users, because the antibody levels, while decreased, were still within the range of antibody levels that are seen in the people who were not immunosuppressed.

This is something that we were very eager to be able to observe, particularly with methotrexate and TNF inhibitors, simply because there's so many people on methotrexate and also TNF inhibition, but there's another issue with TNF inhibition.

While their antibody levels were present, the problem that we ran into when we started looking at later time points was the fact the TNF inhibitors had somewhat of a unique inability to neutralize Delta variant. This was done in collaboration with Michael Diamond, who is an exceptional virologist at Washington University.

Using live SARS-CoV-2 with the Delta variant present in that virus, what we found at 3 months postvaccination, that of the people who were on TNF inhibitor monotherapy, so they were not on any other medications, 64% of the group were unable to neutralize Delta variant.

This is in comparison to 8% of the people who were immunocompetent. About one-third or 35% percent of the patients generally with chronic inflammatory diseases. Almost twice as many people on TNF inhibitors were unable to neutralize Delta variant compared to other immunosuppressive regimens.

This was also substantially greater than those who were not on immunosuppression. This got worse later on, 5 months postimmunization. Here, all of the people on TNF inhibitors were unable to neutralize Delta variant.

This is in comparison to about half of the people with chronic inflammatory diseases as a whole, so this is a broad swath of immunosuppressive medications and about 17% of the immunocompetent group.

Reassuringly — this is data that we did not present at ACR, it’s now coming out in a manuscript that just got accepted — of the people who were on TNF inhibitors and had no Delta responses but received an additional dose, a third dose of mRNA vaccine, all of them were able to generate antibody responses that were able to neutralize Delta.

This is very encouraging for these people with TNF inhibitor utilization and emphasizes that these people definitely need to get that additional dose in order to restore vaccine responses, particularly to Delta variant in the time where Delta variant is the most predominant strain of SARS-CoV-2 out there.

Basically, in conclusion, most patients with chronic inflammatory diseases mounted good acute humoral responses, albeit at modestly reduced levels. The greatest association of poor antibody responses were found in people who were using B-cell-depleting therapies, mycophenolate, or glucocorticoid users.

The people with TNF inhibitor use were able to mount generally good antibody responses, but the antibodies were not quite of the quality as in other immunosuppressive groups. In other words, they were unable to neutralize Delta, but when they got their additional dose, they were able to neutralize Delta.

The 2 main issues that we still need to understand in the future is, one, of the people who don't get the additional dose with TNF inhibitor use, does this actually increase the risk of them contracting severe COVID-19 in the era of Delta variant?

The reason why I bring this up is that during the era of common variant, work from the COVID-19 Global Rheumatology Alliance demonstrated that TNF inhibitor use prevented hospitalizations compared to methotrexate by about 2-fold. Again, the existing breakthrough data that is out there is all on common variants.

Emerging data, probably that will be coming out later this winter or/and in the spring during the Delta variant, will be very interesting to see whether or not TNF inhibitor use, while it reduces antibody protection from Delta, will it actually cause more severe disease as a result? To be determined in the future.

Of the people who are on B-cell-depleting agents, mycophenolate, prednisone, we know that there are numerous other immune responses present, such as T cell responses.

We are doing a deep dive into the T and B cell responses in these people to see whether or not, A, they're present. B, are they changed, in terms of function compared to those not on immunosuppression? C, to see whether or not we can correlate it to any breakthrough infections and severity of breakthrough infections downstream.

Generally speaking, we've been very fortunate to be able to execute this study. I appreciate the opportunity to present this data to you. I hope you all have a great day. Thank you.