2.5. Inheritance and risks of recurrence of Prader-Willi syndrome and Angelman syndrome

Most Prader-Willi and Angelman patients have been sporadic, and familial recurrence has been described only in families with structural chromosomal rearrangements, imprinting defects with a deletion in the imprinting centre or, in Angelman syndrome, UBE3A mutations. In the case of IC and UBE3A mutations, the risk for an affected child may be increased even in distant relatives, since the mutation can theoretically be passed silently in the family by the mechanism of inheritance of imprinted diseases.

Based on an estimation derived from cumulative data of many large surveys of Prader-Willi and Angelman cases, the risk of recurrence in families where a proband has a common de novo deletion without chromosomal rearrangement is less than 1% (Horsthemke et al. 1996, Connerton-Moyer et al. 1997, Cassidy et al. 1997b). In contrast, in rare families with small deletions, the recurrence risk might be as high as 50% if the deletion encompassing only the Prader-Willi syndrome chromosome region (PWCR) is inherited from the father or if the mother is a carrier of the deletion encompassing only the Angelman syndrome chromosome region (ASCR) (Hamabe et al. 1991b, Saitoh et al 1992). Other patrilineal (PWCR deletions) and matrilineal (ASCR deletions) relatives may also be at risk to have affected children or grandchildren due to the imprinting inheritance mechanism.

If the deletion identified in a proband is due to unbalanced rearrangement, the risk to the family depends on whether the rearrangement is inherited or de novo (Horsthemke et al. 1996, Stalker & Williams 1998a), being less than 1% in the latter case. Instead, parental transmission of unbalanced rearrangement may lead to 15q11-q13 deletions due to abnormal meiotic segregation and will result in case-specific recurrence risks (Hulten et al. 1991, Smeets et al. 1992). Furthermore, parental translocation may predispose to the formation of de novo deletions with normal chromosomes due to unequal crossing-over between derivative and normal chromosomes 15, as was described in two Prader-Willi patients (Horsthemke et al. 1996).

Based on the lack of recurrence among all known cases of uniparental disomy for chromosome 15, the experience of UPD in other disorders and theoretical consideration regarding the mechanism of UPD, the risk of recurrence in families where the syndromes are due to uniparental disomy without translocation is less than 1% (Stalker & Williams 1998a). In families where the parent of the affected child carries a balanced translocation raising slightly the probability of occurrence of non-disjunction at meiosis, the recurrence risk is case-specific.

Families with an imprinting control centre (IC) deletion have a 50% recurrence risk if one of the parents (father in PWS, mother in AS) carries the deletion (Buiting et al. 1995, Ishikawa et al. 1996, Reis et al. 1994, Teshima et al. 1996, Saitoh et al. 1997, Ming et al. 2000). Other male (in PWS) or female (in AS) relatives who carry a known deletion in IC are also at 50% risk of having an affected child, whereas female (PWS) or male (AS) relatives who carry the mutation are at risk to have affected grandchildren through sons (PWS) and daughters (AS) who inherit the mutation (Fig 8). In contrast, all imprinting defects without IC deletion have been sporadic and thus probably represent a de novo defect in the imprinting process in 15q11-q13 (Bürger et al. 1997, Buiting et al. 2003). Therefore, the risk in such families is less than 1%.

UBE3A mutations, which are detected in approximately 10% of Angelman patients, may be inherited or represent new mutations (Matsuura et al. 1997, Kishino et al. 1997, Fung et al. 1998, Malzac et al. 1998, Tsai et al. 1998, Fang et al. 1999, Ouweland van der et al. 1999, Russo et al. 2000, Lossie et al. 2001). When the proband’s mother has a UBE3A mutation, the risk of recurrence is 50%. Female relatives of the mother who carry the same mutation also have a 50% risk of having a child with AS. Male relatives of the mother who have the defect are not at risk of having affected children, but are at risk of having affected grandchildren through their daughters who have inherited the defect from them (Fig. 8).

Both somatic and gonadal mosaicism in parents may increase the risk of recurrence in families with patients as a result of IC deletion (PWS and AS) (Saitoh et al. 1996, Gilbert et al. 1997, Bielinska et al. 1999) or UBE3A gene mutations (AS) (Malzac et al. 1998). In the case of parental mosaicism, the recurrence risk depends on the ratio of mutant to normal germ cells (Stalker et al. 1998b). The theoretical risk of gonadal mosaicism for the deletion of 15q11-q13 has been estimated to be less than 0.5% (Gardner & Sutherland 1996), but should be taken into consideration in genetic counselling.

The majority of Angelman cases without identifiable molecular aberration have been sporadic, but there are some reports of families with more than one affected sibling, and the theoretical possibility of a 50% recurrence risk cannot be excluded (Stalker & Williams 1998a).

The recurrence risks in families with different genetic mechanisms are summarized in Table 3.