Introduction

Infertility is a major health problem, and the World Health Organization states that in the western countries infertility affects about 15% of couples trying for a child (Greenhall and Vessey, 1990; Bhasin et al., 1994; Howards, 1995; Dawson and Whitfield, 1996; de Kretser, 1997). In Italy, it is assumed that as many as 60,000 prospective couples each year are affected by this condition and are asking for medical advice in order to obtain proper diagnosis and therapy. Male and female factors are equally contributing to infertility in the couple. Despite the prevalence of infertility, only recent research has focused onto genetic factors accounting both for male and female infertility. It is now clear that genetic abnormalities are present in about 15% of male and 10% of female infertile subjects. They include chromosome aberrations and single gene mutations. Some genetic factors have been outlined and a number of genes related to infertility have been cloned, while other genes have been assigned to specific chromosome regions, but the majority of them still remain undeciphered.Genetic research has expanded in the last years, following the development of in vitro fertilizing techniques, which have made possible reproduction in a wide number infertility conditions. These techniques allow genetic abnormalities to be passed on to the children, the risk of transmission being in many cases quite high. The use of intracytoplasmic sperm injection (ICSI) has raised major concerns about safety in terms of a possible increase of genetic diseases to the offspring, since it bypasses the physiological mechanisms related to fertilization. However, the risk of transmission concerns also the use of standard in vitro fertilization (IVF) and intrauterine insemination (IUI), since also normozoospermic infertile men could have some genetic defect (i.e. 47,XYY males). Natural selection prevents the transmission of mutations causing infertility, while this protective mechanism is overcome by the assisted reproduction techniques (ART). The risk is therefore that genetic defects may persist or even their frequency is increasing in future generations. Thus, the identification of genetic factors in the infertile couple has become mandatory for an appropriate diagnosis, treatment and prognostic assessment. The lack of national or international rules for the genetic approach to the infertile couple, prompted the Italian community of professionals in the field of reproductive medicine to join, exchange experiences and review the literature, and set up guidelines for the genetic diagnosis of male and female infertility. The group of clinical and research experts includes the different disciplines involved in human reproduction (gynaecology, andrology, endocrinology, biology of reproduction, urology, embryology and genetics). They are representatives of twelve national scientific societies, listed at the end of this paper. This Committee worked for five months and in February 2001 the results have been discussed at a convention attended also by external experts from four international societies. The conclusive draft was displayed on Internet for a four month period to allow further comments, based on which the final document was prepared. Therefore, these Guidelines represent a “consensus” document approved by the Italian community operating in human reproduction. The Guidelines have been prepared not to include all the genetic causes of infertility, but only those clinically relevant, both in terms of prevalence in male and female infertility and risk of transmission to offspring. Therefore, genetic syndromes in which infertility is only an accompanying manifestation have not been included, since in these disorders the infertility clinicians are playing only a minor role or even no role. Furthermore, the Guidelines are intended for the management of the infertile couple by the clinicians and have not evaluated technical aspects of specific diagnostic test. Since this field is rapidly expanding, the Guidelines and the diagnostic workup proposed in this paper will be periodically revised. The Committee has suggested some genetic tests as mandatory while others have been considered optional.

MALE INFERTILITY

Knowledge of the molecular mechanisms underlying male infertility is rapidly expanding, and it is likely that many more genes affecting male reproduction will be identified in the next few years. As a result of the most recent researches, many genetic mutations have been recognized as clinically relevant, both in determining male infertility and as risk factors of transmission by ART. These defects include chromosomal aberrations and specific gene mutations. In addition, other gene mutations (i.e. b-subunit of LH, FSH, FSH and LH receptors, other genes involved in early sexual development) appear to have only a minor clinical role and further studies are needed to clarify these aspects. The present Guidelines do not consider these mutations, although they could be included in the future, once scientific evidence will prove their major role. In table 1 the genetic causes of male infertility deserving specific discussion are summarized. Table 2 lists the genetic tests that the Committee has recommended to include in the clinical use. Male infertility has been classified on the basis of the seminal analysis (normo, oligo-, azoospermia) although the commission underlines that infertility and alterations of seminal characteristics are not synonymous. However, this classification is fitting with the clinical practice, since the patients candidate to ART are often classified according to semen analysis results. Other criteria include the hormonal profile, history, associated anomalies, and other findings to be assessed in the management of infertile males. In fact, genetic diagnosis and genetic counselling should always be part of an extensive evaluation of these patients. Therefore, basic clinical analysis, such as search for seminal infections or antisperm antibodies, should precede any genetic analysis. However, genetic evaluation is recommended also when infertility is apparently related to other obvious causes (i.e. varicocele, cryptorchidism), since different causes may coexist.

1.Karyotype analysis.

It has been known for some 20 years (Chandley, 1979) that the prevalence of constitutional chromosome abnormalities is higher in infertile men, this figure being inversely related to the sperm count. Chromosomal aberrations are also more common in the male partner of couples seeking ICSI, compared to the general population (Peshka et al., 1999; Gekas et al., 2001). Several studies have been carried out, and based on the largest published series it could be estimated that the overall incidence of a chromosomal factor in infertile males is ranging between 2 and 8%, with a mean value of 5% (Chandley, 1979; Retief et al., 1984; Chandley and Hargreave, 1996; Van Assche et al., 1996; Yoshida et al., 1996; Meschede et al., 1998; Tuerlings et al., 1998; Van der Ven et al., 1998; Peshka et al., 1999; Antonelli et al., 2000; Egozcue et al., 2000; Hargreave, 2000; Gekas et al., 2001). This value is increasing to about 15% in azoospermic males, being largely contributed by patients with 47,XXY aneuploidy. Sex chromosomes abnormalities are predominating in azoospermic men, but a wide range of structural autosomal anomalies, including Robertsonian and reciprocal translocations, inversions, duplications and deletions are also found in infertile males. In table 1 chromosome heteromorphisms have been listed, which should be regarded as common variations devoid of any obvious clinical relevance. Sex chromosome abnormalities were found in 3.7%, and autosomal abnormalities in 2.4% of infertile males seeking ICSI programmes in France (Gekas et al., 2001). Sex chromosome anomalies were found in 15.9%, and autosomal anomalies in 2.8% of the azoospermic men. Notably, the corresponding figure in the normozoospermic infertile men was 3.0%, including both sex chromosome aneuploidies (i.e. 47,XYY, mosaicisms) (1.4%) and balanced structural abnormalities (1.6%). Preliminary results on pregnancies conceived through ICSI are suggesting that sex chromosome anomalies may be more frequent that in the naturally occurring pregnancies (Jacobs et al., 1992; In’t Veld et al., 1995; Liebaers et al., 1995a; Van Opstal et al., 1997). In general, children conceived by ICSI have an increased risk of an abnormal karyotype (Liebaers et al., 1995a; 1995b; Westergaard et al., 1999; Loft et al., 1999). Although data are still controversial, a figure of about 3% has been suggested, half of them being transmitted by a chromosomally abnormal prospective father. Based on the frequency of chromosome aberrations in infertile patients with severe testiculopathies, the Committee has recommended to perform karyotype analysis during the diagnostic workup of subjects presenting with azoospermia and severe oligozoospermia. In these subjects the cytogenetic screening is mandatory prior to any ART procedure. Karyotype analysis should be performed in patients candidates to ART (including IUI), also in those cases in which sperm parameters are within the normal ranges or only slightly abnormal. Furthermore, since some karyotype anomalies (for example 47,XYY) may cause male infertility associated with an apparent normozoospermia, this analysis should be performed when no result is reached after one year of sexual intercourse aimed to pregnancy.

2.Microdeletions of the long arm of the Y chromosome.

Since the original report (Tiepolo and Zuffardi, 1976), many studies, both at the cytogenetic and molecular level, have proved that microdeletions of the long arm of the Y chromosome (Yq) are a common cause of male infertility (for references see Foresta et al., 2001). Three different spermatogenesis loci were assigned to this region, which are referred to as “azoospermia factors” (AZFa, b and c). From these regions different candidate genes have been isolated, and shown to be mutated in male infertility: USPY9 and DBY in AZFa (Foresta et al., 2000), RBMY1 in AZFb (Elliott et al., 1997), and DAZ in AZFc (Reijo et al., 1995). However, other Yq genes have been isolated and their contribution to the AZF phenotype is still unknown (Lanh and Page, 1997; Tilford et al., 2001). Microdeletions determine a severe primary testiculopathy resulting in azoospermia or severe oligozoospermia, and they are more frequent in the AZFc locus (about 60%) than in AZFb (about 15%) and AZFa (about 5%). In the other cases a larger deletion involving more than one AZF region is found. The overall prevalence of Yq microdeletions in the infertile males can be estimated in about 10% (Foresta et al., 2001). In selected infertile males this figure is increasing to 15% in subjects with idiopathic severe oligozoospermia and up to 20% in those with idiopathic non-obstructive azoospermia. Individuals presenting with severe testiculopathies associated with other manifest causes of testicular damage, such as varicocele or cryptorchidism, may also have Yq microdeletions (Foresta et al., 1999; Krausz et al., 1999; Moro et al., 2000). Patients with Y chromosome deletions frequently have sperm either in the ejaculate or within the testis, and are therefore suitable candidates to ART. In these cases, the genetic anomaly is invariably transmitted to the male sons, who also receive the paternal disorder (i.e. infertility). However, the actual consequences of this transmission are still unclear, and recent data on sperm aneuploidies in patients with Yq microdeletions at risk of Turner syndrome in the female offspring (Siffroi et al., 2000) raise medical and ethical concerns. The Committee has recommended Yq microdeletion screening during the diagnostic workup of infertile patients with non obstructive azoospermia and severe oligozoospermia, regardless of the presence of other apparent causes of testicular damage. This analysis is not indicated when sperm concentration is higher than 10 million/ml, since only very rare microdeletions have been found in these subjects. Furthermore, all individuals with non obstructive azoospermia and severe oligozoospermia should be analysed for Yq microdeletions prior to any ART procedure.

3.CFTR gene.

Cystic fibrosis is one of the most common autosomal recessive disease among Caucasians, affecting one in 2,500 livebirths, while one in 25 individuals is an asymptomatic heterozygote. The most frequent CFTR gene mutation on chromosome 7q31.1-31.2, is the deletion of a phenylalanine at position 508 (DF508), but more than 800 different mutations have been identified. Congenital bilateral absence of vas deferens (CBAVD) in most cases is regarded as a mild or incomplete form of cystic fibrosis. About 70-80% of these men are heterozygous or compound heterozygotes for a CFTR mutation (Claustres et al., 2000; Quinzii and Castellani, 2000; Attardo et al., 2001; Casals et al., 2001). A distinct mutation related to CBAVD is the “5T allele” (the wild-type allele has 7T or 9T nucleotides in intron 8), causing the skipping of exon 9 and low levels of expression of the CFTR protein (Chu et al., 1993). About 30-40% of patients with CFTR mutations have only one detected mutation, 30-40% are compound heterozygotes with two allelic mutations, and about 20-30% are carrying the 5T allele. Also unilateral absence of vas deferens (CUAVD) has been ascribed to CFTR mutations. Although the prevalence of mutation in this group of subjects is widely ranging in different studies (11 to 75%) (Casals et al., 1995; Mickle et al., 1995; Dork et al., 1997), it is now established that a proportion of CUAVD is due to a defective CFTR protein. Therefore, the clinical manifestation of individuals with CFTR mutations could be both azoospermia and oligozoospermia associated with CBAVD or CUAVD, respectively. Spermatogenesis is normal in these subjects, and therefore they are strong candidate to ICSI using sperm retrieved from testis or epididymis. In these cases the major risk to offspring is fullblown cystic fibrosis, in those couples in which the female partner is heterozygous for a CFTR mutation. The Committee has suggested to perform the screening of CFTR mutations (including the 5T allele) in the infertile individuals with a diagnosis of CBAVD or CUAVD. Whenever the couple is planning a pregnancy by ART, this test should be performed in both the affected male and his partner, and proper genetic counselling provided.

4.KAL1 gene.

Kallmann syndrome affects one in 10,000 males, and consists of congenital, isolated, idiopathic hypogonadotropic hypogonadism (HH) associated with anosmia. Anosmia is due to agenesis of the olfactory lobes, and HH is due to deficiency of GnRH. Isolated HH without anosmia can occur as an isolated symptoms or in association with a wide–range of somatic abnormalities, resulting from different genetic mutations (Seminara et al., 2000). Three different forms of Kallmann syndrome do exist, inherited as X-linked, autosomal dominant or recessive traits. The X-linked form causes 10-15% of Kallmann syndrome cases (Seminara et al., 2000; Oliveira et al., 2001). The X-linked gene, KAL1, encodes a protein (anosmin) with a key role in the migration of GnRH neurons and olfactory nerves to the hypothalamus. Additional clinical features in these subjects include cryptorchidism, unilateral renal agenesis, cleft palate, and colour blindness. Although Kallmann syndrome is rare and mutation analysis of KAL1 gene is not easily available, the Committee has recommended to include the KAL1 gene screening in all azoospermic men with HH and anosmia. This decision was supported by evidence that hormonal treatment may restore fertility and, in turn, risk of transmitting the mutation to offspring.

5.Androgen receptor gene.

Androgen insensitivity syndrome is an X-linked recessive disorder due to a defect in the androgen receptor (AR) gene located in Xq11-12. Affected individuals may show variable phenotypes, ranging from fullblown female appearance, to overt genital ambiguity, or even infertile male (Quigley et al., 1995). Over 300 different mutations have been reported (http//www.mcgill.ca/androgendb/), the majority of which representing point mutations leading to aminoacid substitutions. Infertile males carrying AR gene mutations show azoospermia or severe oligozoospermia, either as an isolated feature or in association with other anomalies resulting from defective sensibility to androgens (such as cryptorchidism, hypospadia, gynaecomastia, poor virilization). They also exhibit distinct hormonal profile, with increased LH levels and normal-high testosterone plasma concentrations. The product of LH x testosterone (expressed in U x nmol/l2), also referred to as androgen sensitivity index (ASI), can be useful for detecting patients at risk for AR gene mutation. In fact, it has been suggested that the higher the ASI, the more likely an abnormality within the AR gene (Hiort et al., 2000). The frequency of point mutations in the AR gene in infertile subjects with azoospermia or severe oligozoospermia is estimated in 2-3% (Hiort et al., 2000). Since identical mutations in AR gene may be associated with different phenotypes, the consequences in children born by use of ART cannot be easily predictable and the couple should be informed about the possibility of a worsening of the clinical features in offspring inheriting the disease-gene. Large expansion of the trinucleotide (CAG) repeat in exon 1 (more than 40) causes spinal bulbar muscular atrophy (SBMA or Kennedy disease), which is characterized by progressive muscle weakness and atrophy, and is associated with low virilization, infertility, and testicular atrophy. It is still unclear if minor CAG repeat expansions result in isolated defective spermatogenesis (Mifsud et al., 2001, Patrizio et al., 2001). Only few studies have reported AR gene mutation screening in infertile patients without other congenital anomalies. Therefore, the Committee has recommended to perform this test in azoospermic and severely oligozoospermic men with high ASI, although this is not a mandatory rule. Furthermore, in these cases the test should be performed as a second step, following karyotype analysis. Obviously, when other clinical manifestations of androgen insensitivity are present, the AR gene should be screened for mutations. Contrasting data have been reported concerning the role of the CAG triplet expansion in male infertility. Therefore, at present this analysis should not be performed in the clinical practice.

6.5 a-reductase-2 gene (SRD5A2).

This enzyme converts testosterone to 5a-dihydrotestosterone in specific androgen-dependent target tissues (Fratianni and Imperato-McGinley, 1994). Deficiency of 5 a-reductase-2 causes male pseudohermaphroditism usually presenting with pseudovaginal perineoscrotal hypospadia at birth. Virilization occurs at puberty with phallic growth and testicular descent. Adult subjects may present with azoospermia or severe oligozoospermia sometimes associated with undescended testes and hypospadia (Cai et al., 1994; Katz et al., 1997). Due to underdevelopment of prostate and seminal vesicles, seminal analysis frequently shows low volume of ejaculate and highly viscous semen. Pregnancy by IUI has been reported in patients with 5 a-reductase deficiency (Katz et al., 1997). Given the low incidence of 5 a-reductase deficiency in infertile males, the Committee has recommended this analysis only in selected subjects presenting clinical features fitting with this disorder.

7.Aneuploidy analysis in spermatozoa.

Normal males produce a variable proportion of spermatozoa (ranging between 1 and 15%) with visible chromosome aberrations, most of them being of the structural type (about 90%) (Rosenbush, 1995). Some evidence suggests that a number of clinical conditions related to male infertility are associated with increased numbers of hypo- or hyperaploid spermatozoa. In infertile patients with severe testiculopathies numerical sperm anomalies are increased as a consequence of mitotic and meiotic failures (Bernardini et al., 2000; Vegetti et al., 2000; Calogero et al., 2001). Furthermore, chemotherapy and radiotherapy have a clastogenic effect persisting over a period of time (up to six months) after discontinuing these treatments (Robbins et al., 1997; De Palma et al., 2000). Fluorescent in situ hybridization (FISH), which is the technique of choice for assessing the sperm aneuploidies, has some limitations. In fact, it is unable to detect structural abnormalities. In addition, only a limited number of spermatozoa are evaluated and standardization of the hybridization procedure is needed. At present there is not agreement about the clinical application of FISH, and the Committee has recommended that this test be not performed as a routine analysis in infertile males. Once that more robust scientific data become available, FISH analysis could be indicated in primary severe testiculopathies and after chemo-radiotherapy. At least sex chromosomes and chromosome 21 should be included in this analysis.

8.Other conditions.

Other genetic anomalies have been evaluated by the Committee, but have not been included as tests for routine assessment of the infertile subjects, either because they are extremely rare in the setting of an infertility centre, or because scientific data are still limited. They include mutations in genes essential for gonadal development and testicular function (deficiency of steroidogenic enzymes, mutations in the genes for b-subunit of LH and FSH, and LH and FSH receptors), and pleiotropic genes whose mutation alter also spermatogenesis (miotonic dystrophy, DAX1). Furthermore, a number of genetic disorders may affect spermatogenesis, but clinical manifestations are so complex that infertility represents a minor feature (table 3). Furthermore, some forms of male infertility characterized by uniform and permanent morphological and structural alterations of the entire sperm population (such as round headed sperm, miniacrosome etc.) are probably the results of hitherto unknown genetic mutations. This aspect should be kept in mind in subjects candidate to ART.

FEMALE INFERTILITY

Following the progress of molecular biology, many studies have been undertaken in recent years to clarify the genetic mechanisms underlying female reproductive disorders. Apart from chromosomal abnormalities, alterations in female sexual maturation or reproductive function may be caused by gene defects at various levels of the hypothalamic-pituitary-ovarian axis or in gonadal and adrenal steroid biosynthesis or reception (Fauser and Hsulh, 1995). Moreover, the possibility of single or multiple gene defects in common clinical conditions, such as polycistic ovary syndrome (PCOS) or premature ovarian failure (POF), have been recently described. Technical progresses in ART have stimulated research in this field. As a consequence, some clinical aspects specifically related to these techniques, such as poor response to ovarian stimulation, have been investigated as genetically determined. Finally, a causal relationship has been established between many chromosomal abnormalities and the spontaneous early fetal demise (Plachot, 1997). In table 4 the genetic causes of female infertility have been summarized. In table 5 the genetic tests that the Committee has recommended to include in the clinical management of infertile woman are outlined. Similarly to male infertility, also female infertility has been classified primarily on clinical criteria, especially abnormalities of the ovarian function resulting in abnormal menstrual cycle. Even if recurrent fetal loss is not actually a cause of primary infertility, it has been included, given its clinical relevance and frequency of genetic causes in this condition.

1.Karyotype analysis.

Turner syndrome is the most common chromosome abnormality in infertile women, but also a variety of structural autosomal aberrations may be found with relatively high frequency. Contradictory data have been published concerning the frequency of chromosomal abnormalities in female infertility (Meschede et al., 1998; Badovinac et al., 2000), which, can be estimated in about 5% (Gekas et al., 2001). In the female partners of couples undergoing the ICSI procedure, 2.8% has numerical sex chromosome abnormalities and 2.1% structural autosomal abnormalities (Gekas et al., 2001). The phenotype of women affected by sex chromosomal aberrations is highly variable, in terms of external and internal genitalia and physical features. For example, females with Turner syndrome and other syndromes with short stature and ovarian dysgenesis may show a range of dysmorphisms and defects. However, a feature shared by all these chromosome imbalances is primary ovarian dysfunction (hypergonadotropic) with primary or secondary amenorrhoea (including POF) or oligomenorrhoea. For example, about 30% of primary amenorrhoeas are caused by Turner syndrome. In addition, autosomal structural abnormalities may cause recurrent fetal loss (Goddijn et al., 2000). The Committee has recommend karyotype analysis during the diagnostic workup of infertile women presenting with primary ovarian dysfunction or recurrent fetal loss. This screening is mandatory in women candidates to ART. As for the males, karyotype analysis should be included as diagnostic test, when no result is achieved after one year of sexual intercourse aimed to pregnancy. This is suggested by evidence of abnormal karyotypes (such as 47,XXX) in women without apparent causes of infertility (Gekas et al., 2001). The Committee suggests caution in establishing karyotype-phenotype correlations in infertile women presenting with low level sex chromosomes mosaicism and careful control of the laboratory results. In fact, the eventual confirmation of this correlation should also imply a higher risk of recurrent fetal loss, perinatal death, and sex chromosome aneuploidies in offspring.

2.Fragile X syndrome (FRAXA).

This is the most common cause of mental retardation in males and it is caused by the expansion of the CGG trinucleotide repeat in exon 1 of the FMR1 gene located in Xq27.3 (Jacobs, 1991; Verkerk et al., 1991). In the general population less than 50 repeats are found, while more than 200 (full mutation) are causing mental retardation. A premutation is defined as 50-200 repeats and can be associated with POF in otherwise normal women (Schwatrz et al., 1994; Fryns, 1986; Conway et al., 1995; 1998; Allingham-Hawkins et al., 1999; Davis et al., 2000). In fact, 15-25% of premutated women are affected by POF and 6.5% of women with POF carry a FRAXA premutation (Sherman, 2000a; 2000b). Patients who will develop a POF frequently show a period of oligomenorrhoea with a progressive increase in gonadotropins (Murray et al., 2000). Furthermore, premutation has been shown to be associated with low response to ovarian stimulation during in vitro ART (Ferraretti et al., 2000). Premutation is predisposing to further expansion of the repeat in the germ line, and therefore premutated women are at high risk of delivering male children with mental retardation. The Committee has recommended to include this test during the diagnostic workup of women with oligomenorrhoea or amenorrhoea caused by primary ovarian dysfunction (including POF), especially when ART is considered. Furthermore, premutation screening should be performed also when no obvious abnormality is present but poor response to ovarian stimulation in previous IVF or ICSI cycles was noticed.

3.KAL1 gene.

Kallmann syndrome due to mutation in the KAL1 gene is exceedingly rare in women, resulting in primary amenorrhoea with hypergonadotropinism and anosmia. A higher incidence of uterine malformations has been also reported (Brandenberger et al., 1994). Heterozygous females have no discernible abnormalities (Oliveira et al., 2001). Kallmann syndrome should be suspected, and KAL1 gene analysis performed, only in infertile women with HH Autosomal recessive and dominant causes of HH may be present, but the genes responsible for these conditions are to date unknown (Oliveira et al., 2001).

4.CFTR gene.

Mutations in the CFTR gene have not been unequivocally associated with reduced female infertility. However, women affected by cystic fibrosis are subfertile and have a higher risk of complicated pregnancy (Cohen et al., 1980). The prevalence of the CFTR mutation in the general population (1:25), and the consequence high probability of a CFTR mutation to occur in the male partners of ICSI couples, recommend this test in the female partners candidate to ART.

5.Other conditions.

Similarly to the males, other genetic conditions may determine female infertility. However, they are very rare and of little clinical relevance for specialists in the field of reproduction. Notable examples include androgen receptor gene mutations causing complete androgen insensitivity syndrome, syndromes in which infertility is a minor manifestation (table 6), and syndromes in which infertility represents the major phenotype (mutations in FSH and LH receptors, FSH gene, GnRH receptor). It is worth of note that karyotypically normal women produce a variable percentage (more than 20%) of chromosomally abnormal oocytes. This percentage is increasing with advanced maternal age, as a consequence of altered crossing-over and increased non-disjunction. Genetic counselling is therefore recommended when ART is performed in older women. (traduzione dell'Autore)

Carlo Foresta

Assistant Professor of Internal Medicin University of Padova. Director of the Centre for Cryopreservation of Male Gametes University of Padova.