Paternal age effect in fibrodysplasia ossificans progressiva

Journal of Medical Genetics, 1979, 16, 147-148
Paternal age effect in fibrodysplasia ossificans
progressival
JOHN G. ROGERS AND GARY A. CHASE2
From the Division of Medical Genetics, Johns Hopkins University School of Medicine, and
Johns Hopkins Hospital, Baltimore, Maryland, USA
SUMMARY Analysis by the method of Smith (1972) of birth order and parental age data collected
from 38 of 42 patients with fibrodysplasia ossificans progressiva shows a significant paternal age
effect. This finding among the sporadically occurring cases would support the proposition that this
condition usually arises as a new dominant mutation.
Tunte et al. (1967) suggested that there was a paternal
age effect in fibrodysplasia ossificans progressiva. We
undertook a survey of patients with fibrodysplasia
ossificans progressiva (FOP) to assess their clinical
status. This study has been reported in detail elsewhere
(Rogers and Geho, 1978). The data collected
included complete information on parental age and
birth order in 38 of 42 patients. This short communication
is to report the confirmation of a
paternal age effect using the method of Smith (1972).
Methods and results
The ages of the parents at the time of birth of the
affected child and the birth order are shown in
Table 1. The birth order was based on the number of
children born to the mother, regardless of sire. The
birth order of FOP patients is shown in Table 2, in
comparison with their 'expected birth order' calculated
from the 1960 US Census figures, which are
derived in the same manner. Of the 42 patients, 38
provided complete data for analysis, by the method
of Smith (1972), to determine the presence of direct
maternal age, paternal age, and birth order defects.
This analysis attempts, through conventional regression
techniques, to estimate the effects of maternal
age, paternal age, and birth order, after having
accounted for linear relationships among these
effects. The direct estimates of these effects reach
statistical significance when they are greater than
twice the standard error of the method.
The estimate of direct maternal age effect dm was
0-77 ± 2-96 (2 SE). The estimate of direct paternal
'Supported by NIH Grant No. 5 TOI GM 00795 and by
Proctor and Gamble.
2Investigator, Howard Hughes Medical Institute.
Received for publication 3 July 1978
age effect df was 2 87 ± 2' 82. The estimate of direct
birth order effect db was -0 37 + 0 62.
Table 1 Age ofparents and birth order
Patient Father's age at birth of Mother's age at birth Birth
no affected child of affected child order
1 27-75 27 1
2 32-08 27-67 2
3 - - 4
4 33-67 32-75 6
5 23 21-67 1
6 36 22-58 1
7 33-33 25.83 1
8 44-50 39*50 5
9 32-67 31-50 5
10 26-92 23-50 1
11 27-92 28-75 3
12 20-50 19-83 1
13 36-75 28-58 1
14 31-08 32-83 1
15 21-17 17-75 1
16 47-83 36-83 3
17 23 75 16-50 1
18 20-42 17-83 4
19 27-50 21-33 3
20 30*75 28-08 4
21 45-75 36-75 5
22 37.25 25-83 2
23 33 30-50 4
24 - - 2
25 45 40-50 3
26 29-42 24-92 1
27 - 24-42 2
28 34 27-17 2
29 37-75 26-67 3
30 39-75 39*33 6
31 29-58 25-42 1
32 25 42 23-67 4
33 42-92 35-42 4
34 28-17 27-92 2
35 22-17 19-42 1
36 39-45 37-75 3
37 42-75 35-83 5
38 43-33 40-58 4
39 28- 58 28-33 1
40 39-08 32-75 1
41 34-42 29-83 1
42 - 24-50 2
147
Downloaded from hwmaint.jmg.bmj.com on 1 February 2010
Rogers and Chase
Table 2 Birth order
Observed Fxpected
1st born 16 10-53
2nd born 6 9-98
3rd born 6 7.73
4th born 6 4-80
5th born 4 2.59
6th born 2 1-37
Discussion
Birth order and parental age are liable to change with
time. Comparing our population, born over many
years, to a single year in the US Census data may
introduce bias. The year 1960 was chosen as the
median year of birth of our sample. However, it is
likely that all three parameters, birth order, maternal
age, and paternal age will change with time.
The results of direct estimate of paternal age
effect shows that a significant effect remained after
adjustment by the method of Smith (1972). The
estimate df was more than twice its standard error,
but no significant effect was detected either for birth
order or maternal age. These data are set out in
comparison with the data of Tiinte et al. (1967) and
US 1960 Census data in Table 3.
In the sample of Tunte et al. (1967), the mean age
Table 3 Mean and standard deviation ofpaternal age,
maternal age, and birth order
Paternal age Maternal Birth
age order
Present study 32-9±7-6 (SD) 29*3 ±6 2-6±1*7
US Population (1960) 29-8+6-9 28-5±6-8 2-8±1-7
Data of Tunte et al.
(1967) 37-2±7-2 (SD) 31-7±6 3-2±2
of fathers is considerably older than in our sample
and the differences in mean ages of both parents are
greater. In addition, the mean paternal age shows a
greater departure from the reference population
mean than maternal age or birth order. They used a
mixed reference population from Denmark, Norway,
England, and Japan.
Advanced paternal age has been shown as a factor
in a number of dominant conditions arising as new
mutants (Jones et al., 1975). The confirmation of a
paternal age effect in FOP would support the proposal
that it usually occurs as a new mutation. This
feature would account for the lack of affected sibs.
Low reproductive fitness accounts for the lack of
reports of offspring.
We should like to thank Dr Blair W. Geho for his
help in collecting these data, and Mrs Susan Vane
for help in collating data.
References
Jones, K. L., Smith, D. W., Harvey, M. A. S., Hall, B. D.,
and Quan, L. (1975). Older paternal age and fresh gene
mutation. Data on additional disorders. Journal of
Pediatrics, 86, 84-88.
Rogers, J. G., and Geho, W. B. (1979). Fibrodysplasia
ossificans progressiva-a report of forty two cases. Journal
of Bone and Joint Surgery. (In the press.)
Smith, C. A. B. (1972). Note on the estimation of parental
age effects. Annals of Human Genetics, 35, 337-342.
Tunte, W., Becker, P. E., and Knorre, G. V. (1967). Zur
genetik der myositis ossificans progressiva. Humangenetik,
4, 320-351.
Requests for reprints to Dr John G. Rogers, Genetics
Research Unit, Royal Children's Hospital Research
Foundation, Parkville, Victoria 3052, Australia.