Photo Credit: Frentusha
A recent study led by Philip De Jager, MD, PhD, marked a significant milestone in understanding the clinical heterogeneity of MS outcomes.
Philip De Jager, MD, PhD, presented “Debate: Racial and Ethnic Disease Phenotype Differences are Driven by Genetics: Yes” during ACTRIMS Forum 2024, held February 29 to March 2 in West Palm Beach, FL.
For a study published in Brain—but not necessarily part of Dr. De Jager’s presentation, he and his colleagues sought to overcome the major challenge that heterogeneity in MS clinical presentation causes in identifying genetic variants associated with disease outcomes.
To address this challenge, the researchers utilized prospectively ascertained clinical outcomes data from the largest international multiple sclerosis registry, MSBase. A cohort of deeply phenotyped individuals of European ancestry with relapse-onset MS was assembled. Unbiased genome-wide association study and machine learning approaches were employed to evaluate the genetic contribution to longitudinally defined MS severity phenotypes in 1,813 individuals.
Primary analyses did not reveal any genetic variants of moderate to large effect sizes meeting genome-wide significance thresholds. The strongest signal was associated with rs7289446 (β = -0.4882, P=2.73 × 10-7), located intronic to SEZ6L on chromosome 22. Nonetheless, it was demonstrated that clinical outcomes in relapse-onset MS are linked to multiple genetic loci of small effect sizes.
A machine learning approach incorporating more than 62,000 variants alongside clinical and demographic variables available at MS disease onset was employed, resulting in the prediction of severity with an area under the receiver operator curve of 0.84 (95% CI, 0.79-0.88). The machine learning algorithm achieved a positive predictive value for outcome assignation of 80% and a negative predictive value of 88%, outperforming the machine learning algorithm containing only clinical and demographic variables (area under the receiver operator curve 0.54, 95% CI, 0.48-0.60).
Additionally, sex-stratified analyses identified two genetic loci meeting genome-wide significance thresholds: one in females (rs10967273; βfemale=0.8289, P=3.52 × 10-8) and the other in males (rs698805; βmale = -1.5395, P=4.35 × 10-8), suggesting sex dimorphism in multiple sclerosis severity. Tissue enrichment and pathway analyses revealed an overrepresentation of genes expressed in CNS compartments, particularly in the cerebellum (P=0.023). These genes were associated with mitochondrial function, synaptic plasticity, oligodendroglial biology, cellular senescence, calcium, and G-protein receptor signaling pathways. Six variants with strong evidence for regulating clinical outcomes were identified, with the strongest signal again intronic to SEZ6L (adjusted HR, 0.72; P=4.85 × 10-4).
This study marks a significant milestone in understanding the clinical heterogeneity of MS outcomes, implicating functionally distinct mechanisms in MS risk, wrote the study team in their paper. Importantly, it was demonstrated that machine learning using common single nucleotide variant clusters, along with readily available clinical variables at diagnosis, can enhance prognostic capabilities at diagnosis and, with further validation, has the potential to lead to meaningful changes in clinical practice.
For more information regarding ACTRIMS Forum 2024, visit the event website. Return here often for additional MS-focused highlights, abstracts, features and more from ACTRIMS Forum 2024.