Drug-induced progressive multifocal leukoencephalopathy: a case/noncase study in the French pharmacovigilance database

Abstract

Progressive multifocal leukoencephalopathy (PML) is an often fatal demyelinating disease of the central nervous system. As effective treatment is unavailable, identification of all drugs that could be associated with PML is essential. The objective of this study was to investigate the putative association of reports of PML and drugs. We used the case/noncase method in the French PharmacoVigilance database (FPVD). Cases were reports of PML in the FPVD between January 2008 and December 2015. Noncases were all other reports during the same period. To assess the association between PML and drug intake, we calculated an adverse drug report odds ratio (ROR) with its 95% confidence interval. We have studied the delay of onset of PML for each drug concerned. Among the 101 cases of PML, 39 drugs were mentioned as suspect. The main therapeutic classes suspected with significant ROR were antineoplastic agents (n = 85), immunosuppressants (n = 67), and corticosteroids. A latent interval from the time of drug initiation to the development of PML is established: the median time to onset was 365 days (123-1095 days). The onset of PML is highly variable and differs among drug classes [from 1 to 96 months (IQR: 39.0-126)]. An association between PML and some immunosuppressant drugs was found as expected, but also with antineoplastic agents and glucocorticoids. An important delay of PML onset after stopping treatment is suspected and should alert prescribers. Prescribers but also patients should be informed about the potential associations with all these drugs. Monitoring could be necessary for many drugs to early detect PML.

What Is the Risk of Progressive Multifocal Leukoencephalopathy in Patients With Ulcerative Colitis or Crohn’s Disease Treated With Vedolizumab?

“…A 2009 study of nationwide US inpatient sample data showed that 20% of cases of PML in rheumatic diseases were associated with the use of only very mild immunosuppression with an antimalarial or low-dose prednisolone, 32 and analysis of French pharmacovigilance reports shows that PML has significantly elevated reported odds ratios (in excess of 4) for both prednisolone and azathioprine. 33 Finally, we should consider what other risks patients might run with alternative therapies. In many cases, the alternatives to vedolizumab will be either immunosuppression or surgery.…”

Drug‐induced hearing loss: a case/non‐case study in the French pharmacovigilance database

“…The ROR was the ratio of the reporting of one specific event to all other events for a given drug compound. This approach, known as the case/non‐case method, measures the disproportionality of combination between a drug and a particular adverse drug report in a pharmacovigilance database . This is the method usually used to generate signals from a pharmacovigilance database.…”

Sensitivity and specificity in signal detection with the reporting odds ratio and the information component

“…In a selection of studies on worsening of myasthenia in antibiotics, 10 antidepressants and neuroleptics, 17 QT prolongation in neuroleptics, 22 transverse myelitis after COVID-19 vaccination, 23 the 30 most frequent ADRs in edoxaban 24 and induction of PML by immunosuppressant drugs 9 there is a large range of expected ADR frequency. The error rates of the drugs in these studies as obtained from simulations with the respective values for P0 and N are shown in Figure 7 together with a simulation for the full range of ADR frequencies with P0 = 0.01 and N = 1 000 000 for expected case numbers below the position of peak n. Figure 3 shows that for n = 3 false positive rates of 0.05 are possible when in fact 0.025 was intended.…”

“…As an example consider the estimation of a probability from an observed frequency for which the simple Wald confidence interval 6 can cover far less than the nominally expected 95% 7,8 if small counts are involved. Counts in pharmacovigilance studies can be very small: Only 0.036% of all ADR reports refer to PML induction, 9 and only 0.02% of all ADR reports refer to worsening of myasthenia. 10 We are not aware of an analysis of the true coverage of the confidence interval of the ROR for such low frequencies.…”