In the largest collected of PML DNA samples, the study identified 4 immune-linked, high effect size, rare variants for use in an iatrogenic PML risk genetic test.
Lawrence Steinman, MD
For the first time, researchers have identified genetic variants that are associated with an increased risk of progressive multifocal leukoencephalopathy (PML), a rare and often lethal brain disorder that develops in a wide range of immunosuppressed groups and can occur as a serious adverse event from exposure to immunosuppressives.1,2
Overall, in individuals taking PML-inducing medications, having at least 1 of the 4 identified genetic variants increased the risk of PML 8.7 times on average, with 1 specific variant increasing it 33-fold. The 4 variants, found in immune system genes with strong biological links, included: C8B, 1-57409459-C-A, rs139498867; LY9 (alias SLAMF3), 1-160769595-AG-A, rs763811636; FCN2, 9-137779251-G-A, rs76267164; STXBP2, 19-7712287-G-C, rs35490401.
"It's critical to be able to identify genetic mutations that greatly increase a person's risk of this devastating infection" Lawrence Steinman, MD, professor of neurology and neurological sciences, pediatrics, and genetics, Stanford University, said in a statement.1 "Preventative screening for these variants should become part of the standard of care. I wish we had more powerful tools like this for other therapies."
Although not affiliated with the study, Steinman and his lab helped develop natalizumab (Tyasbri; Biogen), a therapy for multiple sclerosis (MS) that was temporarily withdrawn from the market because of PML and now carries a Black Box Warning. For context, there are over 3 dozen drugs that include a PML warning in their prescribing information and/or mention PML in their product characteristics.
Previously, the investigators of the study explored the possibility of genetic predisposition to PML in the largest genetic study to date, whole-exome sequencing of 184 PML cases. That study identified 19 rare genetic variants in immune-modulating genes that were more common in this patient population. In the new analysis, an additional 152 cases of PML were included, bringing the total number to 336. Importantly, this study, for the first time, compared 110 drug-exposed PML cases to 718 drug-exposed controls who took PML-linked therapies for at least 2 years.
Variants in genes C8B, FCN2, and STXBP2 were found to be significant (P <.05) when compared with both drug-exposed controls and gnomAD population controls. The LY9 variant was only significant when compared with gnomAD controls, likely a consequence of its low frequency, the authors noted. Notably, the STXBP2 and LY9 variants were absent in drug-exposed controls and demonstrated large effect sizes in relation to PML risk (odds ratio [OR], 33.1 and 19.6, respectively).
Natalizumab, which represented the largest subgroup of PML cases (n = 86) and exposed controls (n = 604), showed comparable results to the full set of drug-exposed cases and controls. The association statistics for 3 of the 4 top variants were slightly improved for the natalizumab subgroup but were less significant for the FCN2 variant.
"Determining genetic susceptibility to PML is an extremely promising method of reducing disease risk. A simple inexpensive test may prove revolutionary in this regard," Joseph Berger, MD, associate chief of the Multiple Sclerosis Division, University of Pennsylvania Perelman School of Medicine, and lead author for the American Academy of Neurology’s PML Guidelines Committee, said in a statement.1
The investigators also conducted a secondary analysis to see if any of the top 4 PML risk variants were associated with MS. Using previously reported MS genome-wide association study data from a large international study containing 32,367 patients with MS and 36,012 health controls, 3 of the 4 top variants in genes—C8B, FCN2, and STXBP2—showed no association with MS. The fourth top variant—LY9 gene—was rare in the general population and had not been reported in the literature to be associated with disease, including MS.
Based on the results of the association analysis in the drug-exposed PML cases, a panel of 4 rare variants in genes with strong immune-linked biology was identified as being potentially useful to identify patients at high risk of PML. Presence of any of these 4 variants was 10.9% in the drug-exposed PML cases vs 1.4% in the drug-exposed controls. Association statistics for the 4-variant panel were strong, and the population attributable fraction, or percentage of drug-induced PML cases that could be avoided with preventative genetic testing was 9.4%.