Progressive multifocal leukoencephalopathy (PML) is a potentially fatal neurologic syndrome linked to the John Cunningham virus and to immunosuppression due to disease or therapies, including the α4 integrin antagonist natalizumab. No cases of PML have been reported to date with vedolizumab, a selective antagonist of α4β7 integrin expressed on gut-homing lymphocytes. Here, we examine the hypothetical risk of vedolizumab-associated PML in patients with ulcerative colitis (UC) and Crohn's disease (CD) versus other populations with known levels of risk.Methods:
Calculations were based on the vedolizumab cumulative exposure in UC and CD (clinical trials and post-marketing setting) to 19 May 2016. The expected occurrence of PML in patients receiving vedolizumab was calculated assuming the risk to be similar to natalizumab, the general population and a number of other reference populations where the incidence is elevated and has been quantified (including HIV-free rituximab exposed non-Hodgkin lymphoma [NHL]). PML incidence rates in patients receiving natalizumab and in the reference populations were obtained from the natalizumab United States Prescribing Information and the literature. The probability of ≥1 vedolizumab-associated PML case having occurred annually until 2034 was estimated using the rates in 3 reference populations (general population, HIV-free systemic lupus erythematosus [SLE] and rheumatoid arthritis), the estimated vedolizumab cumulative exposure to May 2016 and the expected vedolizumab use for 2016 to 2034. A sensitivity analysis using a Bayesian approach was conducted for the general population-based analysis.Results:
Estimated cumulative vedolizumab exposure was 54,619 patient-years. In this context, 30.2 (95% CI, 19.4–40.9) PML cases would have been expected if vedolizumab exposure conferred the same risk of PML as that of natalizumab. As no cases of PML have been reported with vedolizumab to date, this suggests that the risk associated with vedolizumab is lower than that of natalizumab. Assuming the same risk as in the reference populations, the expected number of cases with vedolizumab would be between 0.2 (general population, 95% CI, 0.05–0.3) and 131.1 (HIV-free rituximab exposed NHL patients, 95% CI, 54.6–305.9). If the risk of developing PML with vedolizumab were the same as in the general population, the estimated probability of observing the first PML case in vedolizumab-treated patients with UC or CD would be ∼50% by 2018, and 90% by 2022. Using the Bayesian model, the first PML case would occur with a 50% probability by 2019. Assuming a risk similar to that in patients with HIV-free SLE or rheumatoid arthritis, the first PML case would likely occur with ≥90% probability by 2019.Conclusions:
These analyses suggest that any risk of PML with vedolizumab treatment, is significantly lower than that with natalizumab. This would be expected, given vedolizumab's gut-selective mechanism of action. Even if the incidence of PML is no higher in those using vedolizumab than in the background population, it is to be expected that cases will eventually occur. Given the rarity of such cases in the general population however we cannot yet be sure that vedolizumab does not carry any increased risk and on-going active pharmacovigilance will be maintained.