Patterns of Inheritance in Osteogenesis Imperfecta - Osteogenesis Imperfecta - page 90

multiple fractures; three of those 79 families (3.8%) had
more than one affected sibling. 17 In the cohort reported
by Byers and colleagues, five of 65 (7.7%) families
with a pregnancy or child with perinatal lethal OI had
more than one affected child. 18 More recently, prospec-
tive studies have further reined and lowered this risk.
Pepin and colleagues found that out of 50 couples
seeking prenatal diagnosis after a previous pregnancy
affected with perinatal lethal OI, only one couple (2%)
had a second affected pregnancy. 2 Pyott and colleagues
conducted the same analysis in a new cohort of 76 cou-
ples, in which again only one couple (1.3%) had a sec-
ond affected pregnancy. 19 Together, these data indicate
that the empiric recurrence risk for perinatal lethal OI is
approximately 1-2%.
It is now known that the majority of perinatal lethal
OI is due to dominant mutations in COL1A1 and
COL1A2 . For example, Bodian and colleagues identi-
ied COL1A1 or COL1A2 mutations in 95% (60/63) of
infants with perinatal lethal OI, whereas the rest had
biallelic mutations in CRTAP or LEPRE1 . 20 The propor-
tion of cases due to recessive forms may be higher in
certain ethnic groups with higher carrier frequencies
or founder mutations, such as LEPRE1 in West Africans
and African Americans 21 and in populations or commu-
nities where inbreeding and consanguineous unions are
more common.
To answer the question of what proportion of recur-
rence is due to recessive inheritance versus paren-
tal mosaicism for a dominant mutation, Pyott and
colleagues studied a group of 37 families with more
than one affected pregnancy born to unaffected par-
ents. 19 Five genes were sequenced: both of the domi-
nant genes ( COL1A1 and COL1A2 ) and the three
recessive genes known at the time ( CRTAP , LEPRE1 and
PPIB ). Twenty-six of 37 (70%) had heterozygous muta-
tions in COL1A1 or COL1A2 , and seven of 37 (19%) had
biallelic mutations in a recessive gene. No mutations
were found in the remaining four families; it is likely
that some, if not all, of those were due to mutations in
the other recessive OI genes not analyzed. This study
demonstrated that the majority of recurrence in peri-
natal lethal OI is due to parental mosaicism for a domi-
nant mutation, rather than due to a recessive mode of
inheritance.
Pyott and colleagues also studied the rate of somatic
mosaicism in parents who had a single child with peri-
natal lethal OI due to COL1A1 or COL1A2 mutations. 19
In six out of 37 families (16%), one parent was found to
be mosaic for the mutation in blood or fibroblasts. Two
out of the six mosaic parents had mild clinical features
of OI, and the rest were phenotypically normal. This
illustrates the importance of testing the parents of a
child with perinatal lethal OI, even if they appear to be
unaffected.

Egg Sperm

Fertilised egg
from which all
body cells arise

*

* *

Kidney Heart Skin Sperm

FIGURE 9.1 Mosaicism. The asterisk (*) represents cells contain-
ing the gene mutation, which spontaneously occurred sometime after
fertilization. In this example showing germline mosaicism, the muta-
tion occurred in a germline precursor cell, thereby eventually consti-
tuting a proportion of all of the gametes (sperm). If the mutation had
occurred earlier during cell differentiation, somatic mosaicism may
result, in that the mutation would be present in a subset of cells in dif-
ferent body tissues. (© Copyright. Centre for Genetics Education for and
on behalf of the Crown in right of the State of New South Wales.)

(see Figure 9.1 ). Mosaicism may be difficult to prove
since the mutation may not be detectable in blood or
other tissues easily sampled, but it is presumed when
an apparently unaffected individual has more than one
child with the same dominant mutation. The recurrence
risk may be as high as 50%, depending on the propor-
tion of parental gametes with the mutation. 2

A UTOSOMAL RECESSIVE O I

Less than 10% of individuals with OI have a recessive
form. To date, at least 12 autosomal recessive OI genes
have been identified: BMP1 , 3 CRTAP , 4 FKBP10 , 5 IFITM5 , 6
LEPRE1 , 7 PLOD2 , 8 PPIB , 9 SERPINF1 , 10 SERPINH1 , 11
SP7 , 12 TMEM38B 13 and WNT1 . 14 It is generally recom-
mended that individuals with a clinical diagnosis of OI
in whom COL1A1 and COL1A2 mutations have been
excluded undergo testing of the recessive genes. 15

PERINATAL LETHAL OI

In the seminal epidemiologic survey by Sillence and
colleagues, two of the 12 pedigrees with a proband with
perinatal lethal OI had more than one affected sibling. 16
It was concluded at the time that perinatal lethal OI
was an autosomal recessive disorder. Subsequent stud-
ies found a lower but still substantial rate of recurrence.
Young and colleagues reported 79 probands born with