Tissue Antigens ISSN 0001-2815 HLA-DRB1 associations with disease susceptibility and clinical course in Australians with multiple sclerosis J. Stankovich 1 , H. Butzkueven 2,3,4 , M. Marriott 2,4 , C. Chapman 4 , N. Tubridy 4,5 , B. D. Tait 6 , M. D. Varney 6 , B. V. Taylor 1 , S. J. Foote 1 , The ANZgene Consortium*, T. J. Kilpatrick 2,3,4 & J. P. Rubio 2,3 1 The Menzies Research Institute, Hobart, Tasmania, Australia 2 The Howard Florey Institute, University of Melbourne, Melbourne, Victoria, Australia 3 The Centre for Neuroscience, University of Melbourne, Melbourne, Victoria, Australia 4 The Royal Melbourne Hospital, Melbourne, Victoria, Australia 5 St Vincent’s Healthcare Group, Dublin, Ireland 6 The Australian Red Cross Blood Service, Melbourne, Victoria, Australia Key words Australians; clinical course; human leucocyte antigen-DRB1; multiple sclerosis; susceptibility Correspondence Dr Justin Rubio The Neurogenetics Laboratory The Howard Florey Institute The University of Melbourne Victoria 3010 Australia Tel: 161 3 8344 6386 Fax: 161 3 9348 1707 e-mail: justin.rubio@florey.edu.au Received 11 November 2008; revised 10 March 2009; accepted 16 March 2009 doi: 10.1111/j.1399-0039.2009.01262.x Abstract Human leucocyte antigen (HLA)-DRB1*1501 and other class II alleles influence susceptibility to multiple sclerosis (MS), but their contribution if any to the clinical course of MS remains uncertain. Here, we have investigated DRB1 alleles in a large sample of 1230 Australian MS cases, with some enrichment for subjects with primary progressive (PPMS) disease (n ¼ 246) and 1210 healthy controls. Using logistic regression, we found that DRB1*1501 was strongly associated with risk (P ¼ 7  10 245 ), as expected, and after adjusting for DRB1*1501, a predisposing effect was also observed for DRB1*03 (P ¼ 5  10 27 ). Individuals homozygous for either DRB1*15 or DRB1*03 were considerably more at risk of MS than heterozygotes and non-carriers. Both the DRB1*04 and the DRB1*01/DRB1*15 genotype combination, respectively, protected against PPMS in comparison to subjects with relapsing disease. Together, these data provide further evidence of heterogeneity at the DRB1 locus and confirm the importance of HLA variants in the phenotypic expression of MS. Introduction Multiple sclerosis (MS) is the most common (1:1000) demyelinating disease of the central nervous system, and it is both clinically heterogeneous and genetically complex. Key features of the MS phenotype include the occurrence of relapses (episodic exacerbations of symptoms that can occur over days or weeks), remissions (periods in between relapses of relatively normal function than can last for weeks to years) and progression (steady decline in neuro- logical function with accumulating disability). MS can therefore be subdivided on the basis of clinical phenotyping into (a) relapsing–remitting (RRMS) or secondary pro- gressive (SPMS) disease, the latter comprising patients with progressive disease with a history of relapses, and (b) pri- mary progressive (PPMS) disease in whom relapses have never occurred and whose disabilities steadily accumulate with time. Around 5%–15% of all MS patients have PPMS, and currently available immunotherapies are ineffective in this subgroup. For clarity, we will group RRMS and SPMS subjects together and refer to them as ‘relapsing MS’ from here on. Relatively little is known about the genetic factors that predispose to PPMS or, conversely, those that increase the risk of relapses; however, family-based studies have pro- vided evidence that these loci exist (1–3). In this regard, we *ANZgene Consortium members: D. R. Booth, S. Broadley, J.M. Greer, L. R. Grifffiths, R. N. Heard, J. Lechner-Scott, M. J. Pender, R. J. Scott, and G. J. Stewart. ª 2009 John Wiley & Sons A/S  Tissue Antigens 74, 17–21 17