Introduction: Limited studies have been conducted to identify and validate multiple sclerosis (MS) genetic loci associated with disability progression. Objectives: To predict future worsening of disability in MS using robust and clinically applicable ensemble genetic machine learning models. Aims: To identify MS genetic loci associated with worsening of disability over time, and to develop and validate ensemble genetic machine learning model(s) to identify people with MS (PwMS) at risk of future worsening. Methods: We examined associations of208previously established MS genetic loci with the risk of worsening of disability using penalised Cox models. Using the identified loci, we learned multivariable mixed-effects machine learning ensembles, and developed robust genetic decision rules for predicting worsening events using a training cohort of PwMS (N=202), and validated the obtained predictions in an external cohort (N=67). Results: We found 7 genetic loci (rs7731626: HR=0.92, P=2.4x10- 5;rs12211604: HR =1.16, P=3.2x10-7; rs55858457: HR=0.93, P=3.7x10-7; rs10271373: HR=0.90, P=1.1x10-7; rs11256593: HR=1.13, P=5.1x10-57; rs12588969: HR=1.10, P=2.1x10-10; rs1465697: HR=1.09, P=1.7x10-128) associated with worsening of disability; most of which were located near or tagged to13genomic regions enriched in peptide hormones and steroids biosynthesis pathways by positional and expression quantitative trait loci (eQTL)mapping. The derived ensembles provided a set of genetic decision rules that can be translated to provide additional prognostic values to existing clinical predictions, with the additional benefit of incorporating relevant genetic information into clinical decision making for PwMS. Conclusions: The present study extends our knowledge of MS progression genetics and provides the basis of predicting future disability progression for PwMS.