There are substantial differences in the prevalence of prostate cancer (CaP) among populations. However, men whose father or brother have CaP, those with multiple relatives with CaP, and those whose brother had early age-of-onset disease have a higher (2 to 10-fold) risk of developing CaP, regardless of ethnicity, suggesting the existence CaP susceptibility genes.

CaP is a complex, heterogeneous disease, exhibiting a spectrum of clinico-pathologic presentations, and is challenging to study from a genetic standpoint. At the time of diagnosis, parents are usually deceased and offspring too young to be informative. Except, perhaps, for early-onset disease, there are no distinctive features that mark hereditary CaP. Moreover, sporadic CaP may occur by coincidence in hereditary CaP families, confounding genetic analyses. The heterogeneous nature of CaP reflects the multiple genetic and environmental factors important for its etiology. There are likely many genes that can predispose to CaP; however, a permissive environment also may be required.

The search for CaP susceptibility genes has been a struggle. Almost a dozen chromosomal regions have been statistically linked to CaP susceptibility, but replication has proved difficult in independent study populations. To date, only one putative CaP susceptibility gene has been cloned, and its clinical importance has been questioned in subsequent studies. No single genetic locus appears to account for a large proportion of hereditary CaP cases, and the percentage of cases linked to any given locus may depend on the population under study.

Most evidence suggests an autosomal dominant mode of inheritance for hereditary CaP, but there is also evidence for X-linked and recessive forms. Segregation analyses suggest the existence of a highly penetrant gene whose frequency may be about 0.3% in the population and may cause CaP in 63-97% of carriers by age 85 (as compared with only ~5% of non-carriers). Highly penetrant genes may account for 5-10% of CaP cases. The breast cancer susceptibility genes (BRCA1/2) are examples of rare, highly penetrant cancer susceptibility genes. Male carriers of mutations of BRCA1/2 mutations are at increased risk of developing CaP.

Variations in a second type of susceptibility gene causing "familial" (as opposed to "hereditary") CaP are associated with a lower risk but are much more common in the general population and may be responsible for a greater proportion (10-20%) of cases. With these genes, penetrance is variable and environrnental factors may be more important.

Linkage studies of multi-case families have suggested the presence of 3 genes predisposing to Cap on chromosome 1: HPCl (lq24-25), PCAP (lq42.2-43), and CAPE (lp36), as well as genes on chromosomes 2, 12, 15, 16, 20, and X. No susceptibility gene, however, has yet been identified in any of these regions - a fact that underscores the difficulty encountered in mapping genes that contribute to a common complex disease such as CaP.

Recently, the first CaP susceptibility gene, called HPC2/ELAC2, was cloned on chromosome 17. Two polymorphisms of this gene (a serine-to-leucine substitution and an alanine-to-threonine substitution) have been reported to be associated with an increased risk (up to 5-10-fold) of CaP. In particular, a significant increased in the frequency of the threonine allele has been reported in two studies of CaP patients. Two studies have not confirmed an increase in the frequency of the threonine allele in CaP patients (the frequency of the threonine allele has been reported to range from 2.1% to 6.3% in controls and 4.9% to 9.8% in patients with CaP). Furthermore, studies have shown little evidence for excess clustering of the threonine allele within the CaP families, as would be expected if it increased susceptibility, and there no strong evidence for linkage of CaP to the HPC2/ELAC2 region of chromosome 17p 11.2 in most families that are segregating this allele. These results suggest that gennline polymorphisms in HPC2/ELAC2 are unlikely to be a common cause of CaP in the general population.

Polyrnorphisms or mutations of so-called "candidate genes" also can influence the development of CaP. Alleles at these loci may be neither necessary nor sufficient to cause CaP but rather are risk factors in an epidemiological sense. Examples are genes that code for enzymes involved in androgen metabolism, the vitamin D receptor, insulin-like growth factor, or the androgen receptor. For example, studies have found that the length of cytosine-adenine-guanine (CAG) repeats in the androgen receptor gene correlates inversely with transcriptional activity (i.e., effective intensity of androgen effects). Men with shorter CAG repeats may be at higher risk for aggressive CaP.

The mixed results of genetic studies are an indication of the complex nature of CaP, with multiple genetic and environmental factors contributing to its development. Other factors contributing to the lack of consistency across studies are differences in the composition of study populations, the criteria used to select patients for study, and the various statistical methods used.

Newer methods of data analysis, stratifying patients into subsets or performing linkage analysis considering covariates, such as age-of-onset, Gleason grade, male-to-male transmission, number of affected family members, linkage to other putative susceptibility loci, etc., can detect linkage signals missed by traditional methods. These approaches will help account for some of the genetic heterogeneity of CaP and should improve the ability to detect genes responsible for aggressive tumor behavior and other important clinical features of CaP in the future.

The current genome revolution offers the opportunity to determine the molecular foundations of CaP that may be exploited to develop new strategies for risk assessment, early detection, treatment, and even prevention.