Current knowledge of the genetic basis of PBC is at best incomplete and at worst poor. Studies so far may be used as a guide to the pitfalls that await unwary investigators and also in deciding where to look and which genes or systems are most likely to yield informative results. The Human Genome Project has revealed a vast array of polymorphism that is too much to contemplate even with the best of current techniques. The crucial processes are the selection of candidates and study design. The strong genetic associations so far in PBC are with chromosomes 6p21.3 and 2q and include; HLA DRBI*08 haplotypes, CTLA4* G and IL1RN-IL1B haplotypes, CASP8, and nramp1. Many of the latter should be considered with caution until confirmed in independent series. Other associations with MBL, APOE and VDR remain to be confirmed. There are also several informative negatives, MMP3 and IL10 for example. It is unlikely that the only genes that influence disease susceptibility and progression in PBC are immunoregulatory genes concerned with T cell immunity. Recent studies indicate a new era for immunogenetics, when genes encoding all immune active proteins may be considered as candidates. One should not concentrate solely on the immune response as recent investigations of mannose binding lectin and apolipoprotein-E testify. One is only just beginning to understand the genetic basis of complex diseases like PBC. The key issues for future investigators are: defining the mechanisms where by self tolerance is broken, defining the mechanisms that determine the rate of disease progression, and identifying genetic markers to predict progression and malignancy. Assessing the genetic basis of variability in disease progression. The significant variation in rate of progression of PBC has led to the hypothesis that genes, in addition to contributing to disease susceptibility, may also determine the rate of disease progression. Several of the studies mentioned earlier have suggested associations between alleles at polymorphic loci and rate of progression . All studies performed to date, however, have been retrospective in nature. One problem inherent in such studies is that of definition of disease progression. One simple definition, that of histological progression to Scheuer stage IV disease , requires liver biopsy. The need to perform repeat biopsies raises ethical problems in cases where there is no other clinical indication. Studies of histologic progression in patients in the control arm of therapeutic trials represent one scenario where repeat biopsy would be indicated. However, the typical time course of such trials is 2 years, insufficient for meaningful assessment of disease progression and natural history, particularly in PBC where there is marked heterogeneity and, as a result, tissue sampling error. Yet, alternative systems for assessing disease progression, such as the Mayo prognostic score, lack sensitivity in any scenario other than existing advanced disease. Outlook for the future. Clinical observations support a significant genetic component to disease susceptibility. Elucidating predisposing genetic associations will markedly assist in understanding the pathophysiology of disease. Investigations to date have been restricted to various community-based case-control association studies, with well-recognized limitations. In future SNP maps and haplotype maps from the Human Genome Project will be available. Studies will require the collection of several well-characterized patients. To meet the required statistical power this will necessitate collaboration on a national and international scale. It is essential that these studies address the relationship between genes and disease progression. The possibility of identifying, in the early stages of disease, patients who are at elevated risk for more rapid progression, would have obvious clinical benefit in terms of patient management and therapy.