Genetic Variant Panel Identified for Progressive Multifocal Leukoencephalopathy

A 4 genetic variant risk panel for progressive multifocal leukoencephalopathy (PML) was identified, according to a study published in the journal Frontiers in Neurology.

Among individuals who are immunosuppressed, a reactivation of JC virus (JCV) can cause PML, which is an aggressive demyelinating disorder with high neurological disability and mortality risk. Although both immunosuppression and JCV infection are necessary components for developing PML, individuals with both are not always diagnosed with the disease.

For the study, researchers assessed whether specific genetic variants may predict PML risk by collecting genetic samples from patients with PML and matched non-PML control individuals from 9 institutions. A total of 336 individuals with PML, 110 of whom were drug-exposed, and 718 drug-exposed matched control individuals were included in this study. Data from the gnomAD population were also used as a general population comparator cohort.

Among individuals with PML, 85% had definite PML and 15% probable, 156 had HIV, 94 had multiple sclerosis (MS), and 41 had blood cancers (BCs). A total of 184 individuals with PML were men; 281 were of European ancestry, and 55 of African ancestry. For control individuals, all had MS; 645 were of European and 73 were of African ancestry.

The 19 PML risk variants that were previously published were tested among drug-exposed cases.

A total of 4 of the variants (syntaxin binding protein 2 [STXBP2], lymphocyte antigen 9 [LY9; also known as SLAMF3], complement C8 beta chain [C8B], ficolin 2 [FCN2]) associated with PML compared with gnomAD control individuals (odds ratio [OR] range, 4.3-63.4; all P ≤.0373) and 3 (STXBP2, C8B, FCN2) compared with drug-exposed control individuals (OR range, 5.7-33.1; all P ≤.0175).

These 4 variants were observed to be rare in the general population, with allele frequencies ranging from 0.000072 (LY9) to 0.004331 (C8B). Three of the 4 variants were missense mutations and the fourth (LY9) was a loss of function frameshift mutation. All 4 variants had combined annotation dependent depletion (CADD) scores >20 indicating a high likelihood of deleterious consequences.

To evaluate whether these 4 variants may be as useful as a diagnostic panel for PML, its utility was evaluated among drug-exposed cases and control individuals. Individuals with PML who were drug-exposed carried these variants at a frequency of 10.9% compared with 1.4% among control individuals (P =3.50×10^-6). This 4-variant panel had a sensitivity of 10.9%, specificity of 98.6%, positive predictive value of 19.5%, and negative predictive value of 97.3%.

Among all cases, 3 of the 4 variants were observed in both individuals with European and African ancestry and all 4 variants were observed in HIV, MS, BC, and other PML cases.

The results of this study may be biased as some cases were not definite PML cases.

The study authors concluded that as there is a lack of treatment for PML and its life-threatening nature, “it would seem unethical not to test individuals considering immunosuppressive therapies with PML risk for our top four variants, and advising those with a positive result to consider an alternative therapy or treatment strategy.”

Disclosure: Multiple authors declared affiliations with industry. Please refer to the original article for a full list of disclosures.