Genetic factors and the founder effect explain familial MS in Sardinia M.G. Marrosu, MD; M. Lai, MD; E. Cocco, MD; V. Loi, MD; G. Spinicci, MD; M.P. Pischedda, MD; S. Massole, MD; G. Marrosu, MD; and P. Contu, MD Abstract—Objective: To estimate the presence of familial aggregation and determine the contribution of genetic factors to familial clustering of MS in patients coming from Sardinia, a Mediterranean island considered a genetically homogeneous, isolated area having high disease incidence and prevalence. Methods: Recurrence risk in siblings of 901 Sardinian patients and factors influencing risk (patient and sibling sex, patient age at onset, sibling birth cohort, and presence of affected relatives other than siblings) were examined. The presence of distant familial relationships among patients was evaluated by tracing the extended pedigrees of all patients with MS born in one Sardinian village. Results: Twenty-three brothers and 36 sisters of the 2,971 siblings were affected with MS. Recurrence risk was greater in siblings of index patients with onset age less than 30 years (p  0.01, increased risk 2.33 times) and having a relative with MS other than a sibling or parent (p  0.01, increased risk 2.90 times). Pedigree analysis of patients from the village of L. showed that all 11 patients descended from 3 pairs of ancestors, whereas no cases occurred in the remaining 2,346 inhabitants. In descendants from the 3 couples, MS prevalence was dramatically greater than the regional average and 1.5 times greater than that observed in siblings of affected cases. Conclusions: Data from this study indicate that MS familial aggregation in Sardinians is influenced by genetic factors and that founder effect and the isolation of Sardinia can be considered causes of the enrichment of “etiologic” MS genes. NEUROLOGY 2002;58:283–288 The etiology of MS is believed to be complex and multivariant, the result of both environmental and multiple low and moderate effect genetic factors. 1 Although studies showing that MS recurrence risk in twins, 2,3 full siblings of patients, 4-8 and offspring of conjugal MS individuals 9 is greater than in the gen- eral population did not provide unequivocal evidence that genetic factors are preponderant in determining familial aggregation of the disease, this fact has been clearly demonstrated by studies on “adoptees” 4 and half siblings. 5 Familial clustering of the disease is defined by  s —that is, the ratio between lifetime risk for a sibling of an MS patient divided by risk in the general population. 10-12 Determination of  s is critical in establishing the feasibility of finding genes in- volved in the disease, making it possible to detect linkage and providing the sample size required for such analysis: the smaller the  s , the more difficult it is to find the responsible gene(s). 10-12  s is higher in MS populations of Scandinavian descent—for exam- ple, people from the British Isles and Canadians, 6,7 — whereas it is lower in other European MS populations. 8 These variations suggest the existence of heterogeneity in predisposing gene stock, which probably will be explained in the future by defining population-specific frequencies of susceptibility and protective alleles at various disease loci. Reviewing recent epidemiologic studies on MS, it has been sug- gested that genetic susceptibility to the disease had been disseminated by the Vikings. 13 Expressing this idea as a genetic model, liability to MS spreads from ancestors: the greater the distance from allele-muta- tion– carrying ancestors, the less the chance of find- ing genes involved in the disease and vice versa. In principle, a suitable population for a study of the relative contribution of genetic factors to disease oc- currence is represented by a genetically homoge- neous population in which all individuals tend to be subjected to similar environmental factors. The pop- ulation of Sardinia, a large (24,089 km 2 ) Italian is- land, meets both these requirements. In fact, the Sardinian population is an ancient, genetically iso- lated and highly homogeneous one, 14 with a distribu- tion of alleles at multiple loci different from other European populations, 15 and a high incidence and prevalence of MS. 16,17 Therefore, evidence for a causal role of alleles and polymorphisms in disease can be obtained by comparison of associations in Sar- dinians with those in other populations. Although in patients in northern Europe and of northern Euro- pean descent the disease is associated with the HLA- DRB1*1501 (DR2) allele, 18 in Sardinia, MS has been associated with the HLA-DRB1*0301 (DR3) and -DRB1*0405 (DR4) alleles, 19 which are part of two extended HLA-DRB1,DQA1,DQB1 haplotypes, namely -DRB1*0301,DQA1*0501,DQB1*0201 and -DRB1*0405,DQA1*0501,DQB1*0301, 20 both spe- cific to the Sardinian population in their high fre- From the Multiple Sclerosis Center (Drs. M. Marrosu, Lai, Cocco, Loi, Spinicci, Pischedda, Massole, and G. Marrosu), Department of Neuroscience, Ospedale Binaghi, Cagliari, Italy; and Department of Public Health (Dr. Contu), University of Cagliari, Italy. Received January 31, 2001. Accepted in final form October 7, 2001. Address correspondence and reprint requests to Dr. M. Marrosu, Centro Sclerosi Multipla, Ospedale Binaghi, Via Is Guadazzonis, 2, 09126 Cagliari, Italy; e-mail: gmarrosu@unica.it Copyright © 2002 by AAN Enterprises, Inc. 283