How Immunosuppression May Affect COVID-19 Vaccine Response

Efficacy & Reactogenicity

Although they must be interpreted with caution, the results of early studies shed light on some of the potential concerns at play for patients with rheumatic disease.

Dr. Segev and his group recruited participants for their trials through social media, using observational data from healthcare workers and other individuals who were vaccinated early to obtain data as soon as possible. As reported in a study in the Journal of the American Medical Association, blood samples for 436 transplant recipients were included for analysis after one dose of either the Pfizer-BioNTech or Moderna mRNA vaccines between 14 to 22 days after vaccination.⁴

The team used semiquantitative immunoassays for antibodies against the SARS-CoV-2 spike protein—a test their lab had previously demonstrated correlated well with neutralizing assays.⁵ Such antibodies were detected only in 17% of individuals, with worse responses in older individuals and in those receiving antimetabolite medications, such as mycophenolate mofetil. In contrast, of the immunocompetent individuals tested through the larger Pfizer and Moderna trials, 100% of individuals had such antibodies at roughly this point.

These numbers did improve after a second dose. In a follow-up study of 658 transplant recipients, 54% of participants demonstrated a detectable antibody response at a median of 29 days after their second vaccine dose. Among the patients on antimetabolite medications, 57% had no antibody response after their first or second vaccine dose.⁶

Another study of 242 kidney transplantation patients in France reported similarly dismal results after one dose of the Moderna vaccine. Only 11% demonstrated seroconversion at 28 days after the first dose, with higher response rates in patients who had were taking fewer immunosuppressive agents.⁷

Dr. Calabrese

Dr. Calabrese cautions against over-extrapolating from such studies, noting that patients receiving treatments to prevent organ rejection are often more immunosuppressed than those on therapies for rheumatic disease, due to the type and dosage of their therapies.

In a second, smaller study specifically in patients with rheumatic and musculoskeletal disease, the Johns Hopkins group reported much better antibody responses. In this study, 74% of the 123 patients were found to have antibodies after a first dose. Neither methotrexate nor anti-TNF therapy seemed to impair antibody development; however, patients receiving either rituximab or mycophenolate mofetil were less likely to develop such responses.⁸

From a mechanistic point of view, it makes sense these agents may repair vaccine response. Rituximab is a B cell inhibitor, and as an antimetabolite, mycophenolate mofetil inhibits a rate-limiting enzyme in the synthesis of guanosine nucleotides, a critical pathway for both T and B cell development.⁹

The Johns Hopkins group also reported on safety data from 325 patients with rheumatic and musculoskeletal disease on immunomodulatory therapy, again following their first dose of COVID‑19 mRNA vaccine.¹⁰ Patients reported expected temporary reactions that were usually mild, in line with those in large vaccine trials of immunocompetent individuals. However, the time course of the study did not allow for assessment of disease flares following vaccination.

A study by Deepak et al. reported on vaccine responsiveness two weeks after the second mRNA vaccination dose in 133 patients with various chronic inflammatory diseases, including inflammatory bowel disease, rheumatoid arthritis, lupus and psoriasis.¹¹ The inflammatory disease patients averaged a threefold reduction in antibody titers compared with immunocompetent controls; however, most immunosuppressed participants did develop a robust antibody response.

The study authors also noticed different levels of vaccine responsiveness based on immunotherapy type. B cell depleting agents, such as rituximab, showed the strongest effect, with a 36-fold decrease in total neutralizing antibody titers compared with controls. Glucocorticoids had the next most marked effect, reducing total neutralizing antibody titers 10-fold in a composite analysis of patients on any dosage.¹¹

Janus kinase inhibitors seemed to have a more diminished, though measurable, impact on titers, and antimetabolite therapies seemed to even more modestly reduce vaccine response. However, it should be noted that methotrexate, leflunomide, azathioprine and mycophenolate mofetil were grouped together in this analysis. Moreover, the sample size of patients on specific medications was small, warranting further study before drawing strong conclusions.¹¹

In another small study, researchers examined 26 rheumatic disease patients on immunosuppressants, most commonly TNF inhibitors, IL-17 blockers, conventional disease-modifying anti-rheumatic drugs and/or prednisolone. All patients had SARS-CoV-2 antibodies at one week after the second dose, although IgG titers were lower in patients on immunosuppressants compared to normal controls. No patients in this study were on rituximab or mycophenolate mofetil. No signs of the vaccine triggering disease flare were found in this study, nor have been found in any study to date.¹²

The key question, of course, is not whether antibodies are found in a lab setting but whether patients on various immunosuppressives acquire real-world immunity after their final vaccine dose. Lowered titers compared with controls may still impart immunity if they reach above a certain threshold.

Partly because of the findings from the Johns Hopkins team, the newer version of the guidance document recommends holding mycophenolate mofetil for one week after vaccination in stable patients.³ (In the earlier recommendation, no recommendation to hold was given.)¹

Julie J. Paik, MD, MHS, an assistant professor of medicine in the Division of Rheumatology, Johns Hopkins, is another investigator on the research team. She points out that some of the sickest patients treated by rheuma­tologists are prescribed mycophenolate mofetil (e.g., patients in renal failure from lupus nephritis, patients with scleroderma-associated lung disease and patients with myositis).

Based on these data, Dr. Paik notes, people who take mycophenolate mofetil and get vaccinated may think they have more protection than they really do. “Unfortunately,” she notes, “some of those patients are at highest risk of severe outcomes from COVID-19 if they do get infected. We have to be cautious,” she adds, “because some of these patients, such as those with lupus nephritis, may have just gotten their disease under control, and holding [mycophenolate mofetil] for an extended period might be dangerous.”

Dr. Calabrese emphasizes that clinicians should use the task force guidance to help inform shared decision making between patients and providers. “Not all those recommendations about timing around different therapies are going to be right for every patient, depending on their underlying disease, their activities, their risks, their health beliefs,” she says.

As more data become available from multiple groups currently studying vaccine response in rheumatic patients, the task force will revise this guidance as necessary, as they have already done.

Dr. Segev notes that results from the Johnson and Johnson vaccine—which uses a vector DNA-based platform—may be different, as may results from other vaccines still under development, such as one from Novavax, which uses a protein-based platform. With differing mech­anisms of action, such platforms may be better or worse than the mRNA vaccines for patients on different types of immunosuppressants.

Currently, the ACR task force recommends against routine laboratory testing for SARS-CoV-2 antibodies, because the correlation of protection against SARS-CoV-2 has not been demonstrated. Also, some commercially available assays don’t detect antibody responses to the spike protein thought to be critical to the development of immunity. However, in its guidance document, the task force members noted this recommendation may shift if more data become available about the potential value of antibody testing in select patients.¹

Dr. Segev says that for a patient forced to make decisions about relaxing safety measures, a validated, quantitative antibody test may provide important data, although it must be interpreted in an overall medical context. “This isn’t a clinical recommendation, but personally, if I were on [mycophenolate mofetil] and I got vaccinated, I would want to check my antibodies, probably a month after the second dose,” he says.