Ruxolitinib is a potent immunosuppressive compound: is it time for anti-infective prophylaxis?

Annkristin Heine, Peter Brossart, Dominik Wolf

To the editor:

Treatment with the Janus-activated kinase (JAK) 1 and 2 inhibitor ruxolitinib decreases constitutional symptoms and spleen size in myelofibrosis. However, accumulating evidence suggests that the drug also exerts substantial immunosuppressive activity.1,2 A very recent report on hepatitis B virus reactivation3 complements the series of severe infections in ruxolitinib-treated patients, among which Cryptococcus neoformans pneumonia,4 toxoplasmosis retinitis,5 disseminated tuberculosis,6 and progressive multifocal leukencephalopathy7 are the most alarming.

The impressive clinical activity of ruxolitinib is predominantly mediated by its profound anti-inflammatory effects, mirrored by a clear reduction of cytokines.8 We have recently shown that ruxolitinib modulates dendritic cell (DC) function resulting in impaired CD4+ and CD8+ T-cell priming both in vitro and in vivo. Notably, we have also demonstrated delayed vaccine-induced hepatic adenoviral clearance in ruxolitinib-treated mice.9 An altered DC and T-cell response as well as imbalances in cytokine production are frequently associated with reactivation or acquisition of infections. As an example, patients receiving tumor necrosis factor α blockers are required to be carefully screened and monitored for symptoms of disseminated tuberculosis infection, and hepatitis B surface antigen carriers must receive either anti–hepatitis B virus treatment or prophylaxis to prevent hepatitis reactivation.

Although clinical evidence is still limited, we assume that ruxolitinib impairs DC and T-cell function, which may lead to impaired cytokine production that results in reduced control of silent infections and, as a consequence, an increased risk of reactivation ending in potentially life-threatening opportunistic infections (eg, progressive multifocal leukencephalopathy). We here propose a provisional screening and monitoring procedure for patients receiving ruxolitinib. We suggest identifying patients who are at risk for developing infections in a way similar to how prophylaxis strategies are determined for patients receiving anti-cytokine or B-cell–depleting therapies in hemato-oncology. Serologic and/or cell-based (ie, QuantiFERON) tests may be used for select hepatitis B surface antigen carriers, patients who have latent tuberculosis, or immunoglobulin Gs that mirror previous contact/infection with varicella zoster virus, herpes simplex virus, cytomegalovirus, Epstein-Barr virus, or Toxoplasma gondii. These patients might benefit from receiving prophylactic therapy with antiviral drugs or antibiotics, according to the suggestions listed in Table 1, or by careful monitoring. In addition, patients need to be closely monitored for their risk of Pneumocystis jirovecii pneumonia or urogenital or bronchopulmonary infections, especially when they develop high-risk World Health Organization (WHO) grade IV neutropenia. Of note, the potential interactions and clinical efficacy of our procedure have not yet been evaluated. In addition, because of its potential harmful effects with respect to the risk of infections, ruxolitinib should be used for patients only within label. It remains of utmost importance to advise patients about the potential risks of using ruxolitinib, including the risk of infections. Finally, future correlative immunologic studies should address whether this immunosuppressive effect is similarly seen with more JAK2-specific inhibitors, especially when use of these drugs is considered before and/or after allogeneic stem cell transplantation, since this might further increase the risk of infection. Conversely, these compounds may exert beneficial effects on alloreactivity (ie, graft-versus-host disease).

View this table:
Table 1

Risk stratification and proposal for an antiviral and antibiotic prophylaxis for patients undergoing ruxolitinib therapy


Contribution: All authors wrote and approved the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Dominik Wolf, University Hospital Bonn, Sigmund-Freud-Strasse 25, 53127 Bonn, Germany; e-mail: dominik.wolf{at}