For the rheumatologist, one of the greatest challenges since the start of the coronavirus disease 2019 (COVID-19) pandemic has been determining which of our patients are at greatest risk for severe COVID-19, and who should be triaged for aggressive outpatient management with preexposure prophylaxis (PrEP), additional vaccine doses, and treatment with antivirals and/or monoclonal antibody products when infected. At present, although the US federal government declared an end to the public health emergency for COVID-19 in May 2023, SARS-CoV-2 is still with us. The pandemic has now morphed into 2 epidemics: one of immunocompetent vaccinated persons, and one of the immunocompromised. The question remains: Which immunocompromised patients are at greatest risk?

Confusion surrounding who is “high risk” and an optimal candidate for PrEP products, vaccine boosters, and antiviral treatments stems largely from lumping “immunocompromised persons” into one bucket. Three percent of the US population is classified as being immunocompromised, yet we are well aware that this is a highly heterogeneous group, encompassing a highly diverse range of immunosuppressed states that range from well-controlled patients with HIV with normal or near-normal immune function, to highly vulnerable bone marrow transplant recipients. Even within the realm of rheumatologic patients, there is a high degree of diversity with regard to their immunosuppressive state and vulnerability to serious infectious complications based on variables such as age, comorbidities, and especially immunosuppressive regimens. Even within the category of biologic and targeted therapies, there is marked heterogeneity, as a 50-year-old patient on monotherapy with an interleukin (IL)-23 inhibitor would be viewed by most clinicians to be in a different COVID-19 risk category than a 40-year-old on a B cell–depleting drug.

Centers for Disease Control and Prevention (CDC) guidance on who are the moderately-to-severely immunocompromised is far from clear, particularly in regard to the effects of immunotherapeutics.1 This guidance document is both vague and inaccurate and (of greatest concern) has remained unchanged throughout the pandemic. Thus, it does not reflect advances in our understanding of the immunocompromised state relevant to COVID-19 outcomes despite a growing body of evidence assessing risks of various biologic therapeutics. The COVID-19 Global Rheumatology Alliance was our first source of information on COVID-19 risk associations with various drugs and demonstrated that tumor necrosis factor inhibitor (TNFi)-treated patients fare relatively well with COVID-19, with some studies demonstrating more favorable outcomes associated with this drug class.2 Other studies have confirmed these observations as well.3,4 This lack of communication and specific guidance from the CDC led to uncertainty as to who was an optimal candidate for PrEP and antivirals, continues to leave practitioners wondering which of their patients truly need these interventions, and has also led to overprescribing to persons who are unlikely to benefit.

The availability of tixagevimab/cilgavimab for COVID-19 PrEP for our most vulnerable patients was a high point of the pandemic and a potential game changer for our most severely immunocompromised patients. It was granted emergency use authorization in December 2021 for the prevention of COVID-19 in patients who were unlikely to have responded to the vaccines. Early on, doses were limited and thus it was even more important to triage based on risk status. We have since learned that B cell–depleted patients are in the highest risk category for severe COVID-19 and likely stand to benefit most from PrEP due to low rates of seroconversion after vaccination.5 Unfortunately, given its targeting of highly specific epitopes within the spike protein, tixagevimab/cilgavimab neutralization resistance rapidly emerged with the introduction of Omicron subvariants and was removed from the market in January 2023. Although no PrEP product is currently active in our armamentarium, multiple products are in the pipeline, including AZD3152, a single monoclonal antibody developed against more current variants of concern, studied in the SUPERNOVA trial, which has closed enrollment.6 The fact that patients have been left without a viable PrEP product for nearly a year is directly attributable to the prolonged approval process required to make new monoclonal antibodies, which, thus far at least, incorporate virtually the same constant region. The process is highly time consuming and resource intensive, requiring repeated randomized clinical trials for each new iteration, in contrast to the rapid approval of a succession of new vaccines based largely on ex vivo neutralization data and preclinical modeling. At the minimum, this approval process should be critically appraised and candidly debated for it has left the most vulnerable among our patients unprotected.

In this issue of The Journal of Rheumatology, Kawano et al report outcomes of breakthrough COVID-19 in patients with systemic rheumatic disease after receipt of tixagevimab/cilgavimab and are the first to report this experience over nearly the entire approval period of the product, collecting cases through November 2022.7 Out of 444 patients, almost one-fifth experienced breakthrough COVID-19 over 2637.6 person-months of follow-up with mostly favorable outcomes. Only 7 patients were hospitalized and 1 died. Nearly half of the cohort had rheumatoid arthritis (RA), followed by systemic lupus erythematosus (14.9%) and antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV; 11.7%); almost half of the cohort received B cell–depleting therapy. Eighty percent of the patients received additional treatment for COVID-19 infection, such as oral antivirals. In terms of risk factors, higher baseline antispike antibodies as well as older age were associated with decreased risk of breakthrough, with the latter possibly being explained by behavioral differences. Interestingly, treatment with B cell–depleting therapy was not associated with a higher risk of breakthrough. Similarly, in a retrospective cohort study of 400 rituximab (RTX)-treated patients in the UK, Yusof et al reported that whereas one-third of the cohort experienced breakthrough COVID-19, only a small proportion of patients experienced severe COVID-19.8

These findings in the current Kawano et al study7 and the recent Yusof et al8 study are in stark contrast to those of numerous other studies throughout the pandemic that demonstrated a strong association between severe COVID-19 and B cell–depleting therapies, throughout all periods of the pandemic.5 The reasons for these varying outcomes in terms of risk of severe COVID-19 in patients with B cell depletion remain unclear. One hypothesis centers on diagnosis plus treatment, rather than solely on treatment. The studies showing more favorable COVID-19 outcomes in the setting of B cell depletion included a higher proportion of patients with RA, and those demonstrating worse outcomes included significantly more patients with AAV. In the current study, nearly half of the cohort in Kawano et al were patients with RA, with only 12% AAV, yet half of the cohort received RTX. A natural question is why should an underlying diagnosis influence outcomes of SARS-CoV-2 infection when the background biologic is RTX in both conditions? Beyond the obvious differences in underlying diagnoses, comorbidities, and implicit difference in concomitant cotherapies, the major determinant may be the likelihood of generating vaccine responsiveness.

The recently published MELODY study9 is a landmark investigation examining the rates of seropositivity after at least 3 COVID-19 vaccines across different immunocompromised states including transplantation, cancer, and immune-mediated inflammatory diseases during the Omicron era, which appears to have shed needed light on factors that may belie some of these observed differences. Overall, MELODY found that having small-vessel vasculitis was the leading associated disorder among all conditions for having a nonresponse to 3 or more vaccines (49%).9 Given that Kawano et al7 and Yusof et al8 both examined a predominantly RA population receiving B cell–depleting agents, with only a minority with underlying small-vessel vasculitis, the diagnosis may be an important contributing factor to the welcome but surprisingly favorable outcomes of COVID-19 in these investigations.

Additional evidence supporting differing outcomes in patients on B cell–depleting therapies for various diagnoses after serious infections beyond COVID-19 is not new and comes from the study of other infections. Using Pneumocystis jirovecii pneumonia (PJP) as an example, the incidence of PJP is significantly higher in RTX-treated patients with AAV compared to patients receiving RTX for other indications such as RA.10 Use of concomitant glucocorticoids likely accounts for some of the discrepancies; however, there are other differences, some known and some less understood.

Regardless, there is no question that patients with AAV represent a different risk class in terms of serious infection, and having fewer patients with AAV in the study by Kawano et al likely explains the more favorable COVID-19 outcomes. Another explanation could be the high usage of COVID-19–specific treatments, such as the oral antiviral nirmatrelvir/ritonavir, so we are unable to tease out the role of tixagevimab/cilgavimab alone in reducing severe outcomes. The use of antivirals to treat COVID-19 is increasingly considered standard of care in this patient population,11 especially for patients receiving B cell–depleting therapy. Thus, increasing use of these drugs in those at both high as well as low risk from immunosuppression combined with decreasing pathogenicity of more recent SARS-CoV-2 variants may explain in part why patients on B cell–depleting therapy did not fare worse than the other groups. For now, currently available oral antivirals are heavily understudied in our immune-mediated inflammatory disease populations and ongoing investigations such as this are needed to inform us of both benefits and risks of aggressive management of COVID-19 in our patients.

CC has received speaker fees from Sanofi and AstraZeneca, and consulting fees from Sanofi, AstraZeneca, Pfizer, and Lilly.

See COVID-19 after tixagevimab/cilgavimab, page 305