The idiopathic forms of premature menopause and early menopause show the same genetic pattern
First Department of Obstetrics and Gynaecology, University of Milan
Via della Commenda 12, 20122 Milan
Obstetrics and Gynaecology Unit, Department of Clinical and Biological Sciences, University of Insubria
V.le Borri 57, 20121 Varese
Department of Clinical and Biological Sciences, University of Insubria, Ospedale di Circolo
V.le Borri 57, 20121 Varese
Department of Biology and Genetics for Medical Sciences, University of Milan
Via Viotti 3/5, 20133 Milan
5To whom correspondence should be addressed Genetic factors may influence the timing of menopause. Premature ovarian failure (POF) has recently been identified as a genetic entity, but no genetic data are available on early menopause (EM). We investigated 36 patients with EM (age of menopause between 40 and 45 years of age) using cytogenetic and pedigree analysis. In 30 patients of this study the EM was idiopathic and 15 subjects (50%) had a familial condition of EM or POF. Pedigree analysis revealed a dominant pattern of inheritance of EM through maternal or paternal relatives. Our data reveal that POF and EM patients show the same genetic features and we postulate that these conditions may be a variable expression of the same genetic disease.
Over the past few decades, the timing of menopause has
remained unchanged: the median age at menopause is currently
around 50 years in Western industrialized society (Ginsburg,
1991; Sowers and La Pietra, 1995). Few surveys have reported
the distribution of age at menopause in the general population
(Coulam et al., 1986). Recently it was reported that the mean
age at menopause is over 45 years in 8 8 % of women, under
45 in 9.7% (early menopauseEM), and under 40 years
(premature ovarian failurePOF) in only 1.9% (Torgerson
et al., 1997).
In a large Italian study, Vegetti et al. (1998) showed that in
one-third of the idiopathic POF patients this condition was an
inheritable entity with a dominant maternal and/or paternal
transmission characterized by incomplete penetrance. In addi
tion three sporadic POF patients in that study shared an Xq
interstitial deletion spanning from q22.3 to q26.3 (one case)
and from q22-23 to qter (two cases). This finding confirms
that a critical region of the X chromosome plays an important
causative role in POF pathogenesis (Powell etal, 1994). Other
authors reported familial and sporadic POF patients carrying
Xq abnormalities and in particular large deletions spanning
from q22-25 to qter and deletions in Xq26-27 (Krauss et al.,
1987; Veneman etal., 1991; Tharapel etal., 1993; Ishizuka
et al., 1997). Moreover recent and detailed cytogenetic studies
have identified the region q21 on the X chromosome (Sala
et al., 1997) as being critical and the Diaphanous gene mapping
in Xq21.3 has very recently been proposed as a candidate
gene for POF (Bione et al., 1998).
On the contrary, very few data are reported in the literature
on the EM condition. During a case-control study (Cramer
et al., 1995), an association was observed between family
history and EM, suggesting the presence of genetic factors as
a cause of EM. In a study on patients with POF, EM, and
menopause over the age of 45 (Torgerson et al., 1997), it was
reported that a significant relationship existed between the
menopausal age of mothers and daughters. A brilliant twin
study (Snieder et al., 1998) demonstrated that genetic factors
are determinants of age at menopause. Nevertheless, little is
known of the genetic factors involved in EM.
Because the menopause occurred between the ages of 40
and 45 in several female relatives of the POF index cases in
the Italian study previously described (Vegetti et al., 1998),
the authors decided to examine the families of patients with
The aim of this study therefore, was to investigate the
presence of genetic factors that may influence the timing of
menopause in EM patients.
Materials and methods
Between May 1996 and December 1998, we observed 81 POF patients
(reported in Vegetti et al., 1998) and 36 patients who entered
menopause between the ages of 40 and 45 years. The patients were
recruited by the Reproductive Endocrinology Services, Department
of Obstetrics and Gynecology in Milan and Varese. Both centres
followed a pre-established protocol for patient selection. For the
purpose of this study, secondary hypergonadotrophic amenorrhoea
was defined as cessation of menses for a duration of at least 6 months
between the ages of 40 and 45 and follicle stimulating hormone
(FSH) concentrations > 4 0 IU/l on at least two occasions. Patients
with amenorrhoea due to known causes or associated with autoimmune
diseases were excluded from the study.
All patients underwent the following clinical assessments: complete
medical history, complete gynaecological history including age at
menarche and previous menses, complete obstetric history with
previous pregnancy outcome, history of cigarette-smoking, alcohol
consumption, and diet, clinical gynaecological examination, ultra
sound pelvic evaluation and serum gonadotrophin assessment.
For karyotyping, peripheral blood lymphocyte cultures were set up
conventionally and incubated for 72 h. In order to obtain high
resolution chromosomal spreads ( > 5 0 0 bands), 1 0 - 3 mmol/l
methotrexate (MTX; Sigma, Italy) was added to cultures for 17 h and
replaced, after washing, with 1.2 X 1 0 - 4 mmol/l thymidine (Sigma, St
Louis, MO, USA) for an additional 4 h. Colcemid (Boehringer
Mannheim, Mannheim, Germany) at a final concentration of 0.1 |ig/
ml was added for 10 min to the culture. Chromosome spreads and
QFQ banding were prepared according to the standard methods. The
reconstructed karyotypes followed the guidelines expressed in the
International System for Chromosome Nomenclature 1995 (ISCN,
1995). For each karyotype a minimum of 20 cells was analysed and
an additional 50 cells were assessed to exclude sex chromosome
Family history was reviewed during genetic counselling and family
members were traced back three generations. To avoid recall bias,
each family history was carefully reviewed with the index patients
and all affected members, when possible, were visited in person.
Thirty-six patients with secondary hypergonadotrophic
amenorrhoea between the ages of 40 and 45 years were
identified. After screening, six patients were excluded from
the study because of EM phenotype-related clinical conditions.
Of the six excluded patients, four were withdrawn because of
previous ovarian surgery, one because of congenital
hypothyroidism and one because of an Xq deletion [karyotype:
46,XX, del(q21-q25)], revealed by chromosome analysis. A
total of 30 patients was defined as idiopathic and analysed.
The mean age at menopause was 42.3 1.2 years (range
41-44), the mean FSH concentration was 67.1 19.8 IU/l
(41.2-105), the mean age at menarche was 12.6 1.6 years
(9-15) and the mean duration of menstrual cyclicity was
29.3 2.3 years (22-34). Patients had a mean body weight of
60.0 7.4 kg (45-75) and a mean height of 159.1 5.2 cm
(150-170). Twenty-four patients had had at least one previous
conception and 20 of them delivered at least one child. Four
patients had had a previous conception that had ended in mis
carriage. Of the 30 patients, 10 were regular consumers of alcohol
and 13 were smokers (more than five cigarettes per day). Medical
history showed a previous dysthyroidism in two patients and
two cases of cancer (breast cancer and chondrosarcoma) treated
only with surgery.
The mean maternal age at menopause was 45.8 5.5
(36-54) and pedigree analysis showed that 15 out of 30
index cases (50%) had a familial condition of EM. In addition,
some relatives with POF were also recorded. Six of the families
had relatives showing EM only, and nine had relatives showing
both EM and POF, as summarized in Table I.
Pedigree analysis revealed a dominant pattern of inheritance
of EM through maternal (12 families) or paternal (three
families) relatives. Figure 1 shows three pedigree examples
with maternal (family 1) and paternal (family 2) transmission
of EM. Family 3, shown in the same figure, reveals EM and
POF conditions in the same pedigree.
In all but one family, individuals showing EM or POF
appeared to be phenotypically normal on physical examination.
One EM index case showed congenital deafness, and so did
the mother and one sister but they were of physiological age
EM = early menopause.
POF = premature ovarian failure.
NA = not available, patient deceased before menopause.
aThe number of EM and POF relatives indicates the number of females in
the same family of the index case showing EM or POF features.
Physiological age at menopause is considered to be between
45 and 55 years. Much evidence has shown that menopausal age
is influenced by environmental and genetic factors (Ginsburg,
1991). The hypothesis that a strong genetic factor may deter
mine the occurrence of POF has been suggested in the literature
(van Starup and Sele, 1973; Coulam et al., 1983; Mattison
et al., 1984; Snieder etal., 1998). Two recent epidemiological
studies have suggested that EM may have genetic determinants
(Cramer etal., 1995; Torgerson etal., 1997).
In a previous study (Vegetti et al., 1998) a group of POF
patients showed E M relatives. This finding prompted us to
examine the EM patients more carefully. Fifty per cent of the
EM cases examined in the present study had a familial
condition of EM and pedigree analysis suggested a dominant
X-linked or an autosomal pattern of inheritance. Both patterns
shared an obvious sex-limited transmission of the disorder.
In this study we found the presence of a familial pattern of
EM, as well as the concomitant condition of POF and EM in
the same family. This latter finding strongly suggests that there
is a genetic relationship between EM and POF.
Because EM patients have genetic features similar to those
observed in patients with POF, this may suggest that E M is
Genetics of early menopause
due to the same genetic causes as POF. This hypothesis is in
agreement with preliminary observations (Cramer et al., 1995;
Torgerson etal., 1997) from which a correlation was postulated
between family history and the occurrence of early menopause.
In a population of 344 women with menopause before the age
of 46 years, it was found that 37.5% of cases were familial
(POF and EM patients were analysed together) (Cramer et al.,
1995). These workers obtained an odds ratio of 6.1 (95% CI
4.0-9.3) for the disease associated with family history of the
same condition. It has been found, in a population of patients
with premature and physiological menopause, that age at
menopause depended on the maternal age at menopause
(Torgerson et al., 1997).
The present study is the first that examines and characterizes
families showing the recurrence of EM. The high percentage
of familial cases (50%) found in this study is probably due to
the small sample size. Nevertheless, all the affected relatives
were visited in person in order to avoid recall bias.
From a clinical point of view, we did not find any difference
in clinical and gynaecological history between POF (Vegetti
et al., 1998) and EM patients in this study. As expected,
patients with EM showed a more favourable reproductive
performance due to the later onset of menopause compared to
POF patients: 80% of EM patients (24 out of 30) experienced
at least one previous pregnancy compared with the 56.3%
already reported (Vegetti et al., 1998) in POF patients. The
percentage of patients who experienced at least one miscarriage
is similar: 2 5 % in EM patients and 24.6% in POF patients.
In conclusion, this study suggests that idiopathic POF and
EM, differing only in age of menopause onset, may represent
a variable expression ofthe same genetic disease. The different
age of menopause onset in these patients may be explained
by genetic heterogeneity and/or by different environmental
factors. Otherwise these results suggest that the timing of
menopause is controlled by different genes probably located
on the X chromosome. Genetic mutations (chromosome dele
tions or molecular mutations) of one or more genes may
explain the premature or early onset of menopause. A linkage
analysis in POF and E M families is currently in progress and
preliminary results on haplotype-sharing are in agreement with
an X-linked mode of inheritance.
This work was supported by grants from I.S.S., 93/G/T28. The
authors would like to thank Yvonne Pomposo for her assistance in
the preparation of the manuscript.