MCM8 and MCM9 Syndromes

Hypergonadotrophic primary gonadal insufficiency with chromosomal instability, early-onset germ cell tumours and, in MCM9 deficiency, susceptibility to colorectal polyposis, colorectal cancer and gastric cancer

Literature and regulatory information updated: 6 November 2026.

0. Initial Reference Table

1. Introduction

1.1. A syndromic class defined by a DNA repair complex

MCM8- and MCM9-related syndromes are autosomal recessive disorders caused by biallelic loss-of-function variants in two genes encoding the subunits of a single DNA helicase, the MCM8-MCM9 complex [1]. The constant manifestation is primary gonadal insufficiency, that is, hypergonadotrophic hypogonadism: in women it presents as premature ovarian insufficiency (POI) with absent or delayed puberty, primary or secondary amenorrhoea, a hypoplastic uterus and small or streak ovaries; in men, as spermatogenic failure, ranging from non-obstructive azoospermia to severe oligospermia, sometimes with Sertoli cell-only syndrome [1]. Added to this reproductive component is a risk of germ cell tumours in childhood and adolescence in girls and, in MCM9 deficiency, a susceptibility to colorectal polyposis, early-onset CRC and gastric cancer [1, 9]. It is this second, oncological component that distinguishes the class from other monogenic causes of POI and justifies its own approach: a biallelic MCM9 result has reproductive and gastroenterological consequences, and a report describing it only as ovarian dysgenesis is incomplete [1, 9].

1.2. Historical milestones and molecular discovery

The MCM family was characterised as the set of proteins required for the initiation of eukaryotic DNA replication; MCM8 was cloned in 2003 from a hepatitis B virus integration site in a hepatocellular carcinoma, and MCM9 was identified in 2005 by homology with MCM2 and with the Xenopus orthologue [2, 3]. In 2009, two genome-wide association studies, one in 2979 European women and the other in 17,438 women, associated MCM8 polymorphisms at 20p12.3 (rs236114, intronic, and rs16991615, encoding p.Glu341Lys) with age at natural menopause, with p values of 9.71×10⁻¹¹ and 1.2×10⁻²¹, respectively; OMIM records this association as the quantitative locus MENOQ3, included in the POF10 entry [5]. In 2012, Lutzmann and colleagues showed that Mcm8-null mice were sterile in both sexes and that Mcm9-null females lacked oocytes, with genomic instability and defective recruitment of homologous recombination factors in embryonic fibroblasts, according to the OMIM description [2, 3].

Translation to humans came between 2014 and 2015. Wood-Trageser and colleagues described, in two consanguineous Kurdish families, 46,XX girls with primary amenorrhoea and short stature who were homozygous for MCM9 variants (one splicing and one nonsense), with defective repair of chromosome breaks in lymphocytes, and designated the disorder a genomic instability syndrome with hypergonadotrophic hypogonadism and short stature [15]. AlAsiri and colleagues identified, in three Saudi sisters born to first cousins with primary amenorrhoea, hypothyroidism and hypergonadotrophic hypogonadism, the homozygous MCM8 variant p.(Pro149Arg), with defective repair of chromosome breaks in fibroblasts [16]. Tenenbaum-Rakover and colleagues described two consanguineous Arab families in which the homozygous MCM8 variant caused primary amenorrhoea in women and small testes with azoospermia in a man, establishing that the gene is required for gonadal development and maintenance in both sexes [17]. Fauchereau and colleagues confirmed MCM9 in an Arab family with two sisters of normal adult height, broadening the phenotype [18].

The link to cancer emerged soon afterwards. In 2015, Goldberg and colleagues described a multiplex family with early-onset CRC and mixed polyposis associated with hypergonadotrophic hypogonadism, with the homozygous MCM9 variant c.672_673delGGinsC [19]; in 2021, the same group re-evaluated a family with MCM9-related POI and found early CRC and clear cell carcinoma of the cervix in siblings, concluding that, of the nine biallelic carriers published at the time, eight had had early-onset cancer [20]. Golubicki and colleagues proposed, in 2020, MCM8 as a recessive predisposition gene for mismatch repair-deficient CRC with no identified cause (Lynch-like syndrome) [21]. In 2025, Helderman and colleagues published the most extensive characterisation to date: enrichment analyses in the 100,000 Genomes Project and UK Biobank and a series of 54 biallelic carriers (26 MCM8 and 28 MCM9), concluding that MCM9 is associated with polyposis, gastric cancer and early CRC, and that both genes are associated with hypogonadism and early germ cell tumours [9]. The GeneReviews chapter, published in June 2026, systematised the diagnosis, management and surveillance of the class [1], and in April 2026 ClinGen classified the relationship between MCM9 and autosomal recessive gametogenic failure as definitive [8].

1.3. Nomenclature and current classifications

The coexisting designations reflect the history of discovery. OMIM classifies MCM8 deficiency as premature ovarian failure 10 (#612885), with the question mark that denotes a still provisional gene-phenotype relationship, and MCM9 deficiency as ovarian dysgenesis 4 (#616185), integrated respectively into the phenotypic series of premature ovarian failure and of ovarian dysgenesis [5, 6]. Orphanet associates MCM9 with 46,XX ovarian dysgenesis-short stature syndrome (ORPHA:444048) [7]. ClinGen uses the designation "MCM9-related gametogenic failure" [8]. GeneReviews adopts the class designation "MCM8/MCM9-related syndromes", with the names MCM8 deficiency syndrome and MCM9 deficiency syndrome for each entity, and defines the class by the combination of primary gonadal insufficiency and tumour risk [1]. The designation "premature ovarian insufficiency" replaces, in the 2024 European guideline and the 2021 French PNDS, the terms "premature ovarian failure" and "early menopause" [10, 11].

1.4. Definition of the entities within scope

The diagnosis of an MCM8- or MCM9-related syndrome is established in a proband with a compatible phenotype and biallelic pathogenic or likely pathogenic variants in one of the two genes; no consensus clinical criteria have been published [1]. MCM8 deficiency syndrome is defined by primary gonadal insufficiency, delayed puberty and a risk of germ cell tumours; until the publication of GeneReviews no CRC had been described in biallelic carriers and only two had developed polyps, after the initial publication [1]. MCM9 deficiency syndrome shares the gonadal component and adds colorectal polyposis with adenomas, hyperplastic and serrated polyps, CRC and gastric cancer [1]. Heterozygotes are not part of the entities: they may have reduced fertility, but there is no clear evidence of an increased risk of CRC or polyposis [1]. HROB deficiency (a gene encoding a protein that stimulates the MCM8-MCM9 helicase) is treated as a differential diagnosis, with a similar phenotype and still limited data [22].

2. Molecular Genetics and Pathophysiology

2.1. The genes and the proteins

MCM8 is located at 20p12.3, has 19 exons spanning about 44 kb and shares a 5' untranslated region with a CpG island with the GCD10 gene, in head-to-head orientation; it encodes an 840-amino-acid protein with an MCM domain, Walker A and B ATP hydrolysis motifs and a zinc finger motif, whose 250-amino-acid N-terminal segment has no equivalent in the other MCM proteins [2]. Expression is highest in placenta, lung and pancreas, and OMIM records expression in ovarian tissue [2]. MCM9 is located at 6q22.31, has 13 exons and encodes a long 1141-amino-acid protein with zinc finger domains, Walker A and B motifs and the complete domain of the MCM2-7 family [3]. The two proteins form a hexameric ring with DNA helicase activity, and the absence of one destabilises the other: loss of MCM8 or MCM9 in human cells reduces growth and the efficiency of homologous recombination under replication stress [1, 3]. The MANE Select transcripts are NM_032485.6 for MCM8 and NM_017696.3 for MCM9 [4].

2.2. Functions of the MCM8-MCM9 complex

The complex takes part in four processes [1, 23]. In DNA replication, it may support the origin recognition complex and, under certain conditions, promote fork progression when the MCM2-7 replicative helicase is dysfunctional; by recruiting BRCA1 and RAD51, it protects stalled forks from degradation [1]. In meiosis and homologous recombination, it contributes to the formation of crossover products and to the repair of interstrand crosslinks [1]. In mismatch repair (MMR), it may facilitate DNA unwinding after recruitment by hMutS, allowing the recruitment of hMutL [1]. In a 2025 study in men with azoospermia, testicular MCM9 interacted with MSH2 and MLH1 and was expressed mainly in spermatogonia and spermatogonial stem cells [24]. Beyond their joint function, each protein may have independent roles, which may contribute to the phenotypic differences between the two genes [1].

2.3. Disease mechanism

The mechanism is loss of function [1]. Meiosis depends on homologous recombination, and impaired meiosis explains the infertility; deficient homologous recombination and MMR are the proposed basis for the cancer risk [1]. Animal models support this interpretation: in Mcm8-null mice, the testes were apoptotic, with spermatocytes showing persistent DNA damage and unsynapsed chromosomes, arrested in prophase I; ovarian follicles were malformed from early stages and absent in the adult female, and the ovaries of older females developed adenomas and granulosa cell tumours [2]. Mcm9-null females lacked oocytes and the males were subfertile, with an early defect of germ cell proliferation; embryonic fibroblasts from both models accumulated micronuclei and chromosome breaks, failed to recruit Mre11, Rpa and Rad51 to stalled forks and were more sensitive to ionising radiation [2, 3]. In 2026 mouse models lacking helicase activity, this activity was dispensable for meiotic recombination but essential for mitotic homologous recombination and for primordial germ cell development; the two ATPase sites of the complex, asymmetric and regulated by Walker B motif residues, were both required to preserve the ovarian reserve, which links loss of helicase activity to POI through primordial germ cell deficiency (preclinical study) [25].

In patients' cells, chromosomal instability is measurable. In fibroblasts homozygous for MCM8 p.(Pro149Arg), exposure to mitomycin C produced 8 to 10 times more chromosome breaks than in fibroblasts from heterozygous relatives, with complex aberrations in most cells, and heterozygous fibroblasts also had more breaks than wild-type ones; the mutant protein had lower affinity for single-stranded DNA and was not recruited to damage sites [2, 16]. In lymphocytes from patients with biallelic MCM9, the number of breaks was higher than in heterozygous relatives and controls, and the protein truncated by loss of exon 9 was not recruited to damage sites [3, 15]. This chromosomal fragility test has been used in research and is not validated as a diagnostic test [16, 17].

In carriers' tumours, the molecular signature is not the one expected of a classical repair defect. In the 2025 series, most tumours were diploid; the tumour mutational burden ranged from less than 1 to 360.42 mutations per megabase (median 27.83); the clock-like signatures SBS1 and SBS5 dominated in all tumours with matched germline sequencing; the MMR-deficiency signature SBS26 appeared in one tumour and explained 13% of mutations; and the homologous recombination deficiency signatures SBS3 and ID6 each appeared in one tumour, both without matched germline sequencing [9]. In The Cancer Genome Atlas pan-cancer atlas, somatic MCM8 and MCM9 variants appeared in tumours with ultraviolet radiation, APOBEC, MMR-deficiency and POLE-deficiency signatures, suggesting that somatic damage to the complex may be secondary to other genomic instability pathways [1, 9]. It is therefore not established whether carriers' tumours differ molecularly from sporadic ones [1].

2.4. Inheritance pattern

Inheritance is autosomal recessive for both entities [1, 5, 6]. Two observations require discussion of heterozygotes. In the 2025 series, 29% (14/49) of monoallelic MCM8 carriers and 22% (10/45) of MCM9 carriers had hypogonadism, figures higher than the population prevalence of POI, but in a context of ascertainment bias through fertility studies [9]. Guo and colleagues found three heterozygous MCM9 variants with reduced repair capacity in vitro in 192 Chinese women with sporadic POI and raised the hypothesis of a dosage effect, but without excluding other genes [26, 27]. GeneReviews concludes that it is not clear that heterozygotes have an increased risk of CRC or polyposis [1]. The accepted model is therefore recessive; the hypothesis of a heterozygous effect remains unproven.

2.5. Variant spectrum and nomenclature

GeneReviews identified no correlation between variant type and phenotype: missense and predicted loss-of-function variants produce overlapping presentations [1]. Variants cluster in two domains: the N-terminal DNA-binding domain (55% of MCM8 variants and 20% of MCM9 variants in biallelic carriers in the 2025 series) and the AAA+ core, essential for helicase activity (45% and 60%, respectively) [1, 9]. Some variants are recurrent: MCM8 c.482A>C p.(His161Pro), classified as a variant of uncertain significance (VUS), was shared by six biallelic carriers from two families with hypogonadism, and MCM9 c.394C>T p.(Arg132Ter), pathogenic, by seven biallelic carriers from four unrelated families [9]. To date, no large intragenic deletions or duplications have been described [1].

The literature uses different transcripts. The MCM8 variants of the Tenenbaum-Rakover family were described on NM_001281520.1, and that of Mishra and colleagues on NM_001282717.2; the MCM9 variants of Fauchereau on NM_017696.2 [2, 3, 28]. The following table presents the reference variants normalised in VariantValidator to the MANE transcripts (GRCh38), with the original nomenclature where it differs [4].

Other published variants, described here as in the original articles, are MCM9 c.672_673delGGinsC p.(Glu225Lysfs*4), homozygous in a family with mixed polyposis and CRC [19], MCM9 c.1651C>T p.(Gln551*), homozygous in a woman with POI [30], MCM8 c.998delG p.(Gly333Glufs*50) in two siblings of different sexes [31] and the pair MCM8 p.(Lys118Glufs*5) and p.(Ile138Met) in trans in a man with Lynch-like CRC [21].

2.6. Penetrance and expressivity

In pedigrees, penetrance appears almost complete: nearly all biallelic carriers had at least one of the described phenotypes [1]. Two biallelic MCM8 carriers without a phenotype have been described, siblings of patients, without prolonged follow-up that would allow late manifestations to be excluded [1]. Expressivity varies in stature (short in the Wood-Trageser families, normal in the Fauchereau family and in the 2024 Belgian adolescent), in the degree of gonadal dysgenesis (primary amenorrhoea with non-visible ovaries or secondary amenorrhoea after spontaneous menarche) and in the male phenotype (azoospermia with normal puberty) [6, 27, 29]. Cancer penetrance has no validated estimate, because the series were identified through fertility studies, the patients were young and few underwent colonoscopy [1, 9].

2.7. Gene-to-gene correlations, allelic disorders and modifiers

The gene-to-gene distinction is the main clinically useful correlation. In the 100,000 Genomes Project, biallelic MCM9 carriers had significant enrichment of colonic polyps, rectal polyps and gastric cancer, whereas MCM8 carriers had no enrichment of any phenotype; in the clinical series, polyposis, CRC and gastric cancer occurred only in MCM9 carriers [9]. Germ cell tumours occurred in both entities [9]. No other disorders associated with biallelic germline variants in the two genes are known [1]. Common MCM8 polymorphisms modulate age at natural menopause in the population (MENOQ3), a quantitative effect distinct from the monogenic disease [5]. In one participant in the 2025 series, a carrier of MCM8 or MCM9 variants with three polyps analysed, biallelic VUS were also found in HROB, a gene encoding a protein that supports the function of the complex [9]. The 2024 Belgian adolescent also had a heterozygous pathogenic FSHR variant, probably irrelevant to the phenotype because FSHR-related POI is recessive [27].

3. Clinical Spectrum and Manifestations by Organ System

3.1. Data sources and their limits

The clinical description rests on three sources of different nature. The first comprises the families published between 2014 and 2023, almost all identified through infertility or amenorrhoea, many of them consanguineous, of Kurdish, Saudi, Arab, Turkish, Tunisian or Jewish origin [1, 27]. The second is the aggregated series of Helderman and colleagues (2025), which brought together the cases in the literature up to August 2023 and unpublished cases obtained through ERN GENTURIS, and which included 26 biallelic MCM8 carriers (15 with pathogenic or likely pathogenic variants and 11 with VUS with a CADD (Combined Annotation-Dependent Depletion) score above 20 and a gnomAD frequency below 1%) and 28 MCM9 carriers (22 and 6, respectively) [9]. The third comprises the population analyses of the 100,000 Genomes Project and of 200,000 UK Biobank exomes, based on diagnostic codes and in silico variant filtering [9]. GeneReviews cautions that the high frequency of hypogonadism reflects recruitment through fertility studies, that the frequency of cancer and polyposis may be underestimated because the patients were young, without longitudinal follow-up and rarely underwent colonoscopy, and that the data do not apply to heterozygotes or to carriers of VUS that do not pass the pathogenicity filters [1].

3.2. Frequency of manifestations

Source: GeneReviews table 2, with data from the series of Helderman and colleagues [1, 9].

3.3. Female reproductive system

Primary gonadal insufficiency is the most constant manifestation: gonadal dysfunction in 88% (23/26) of biallelic MCM8 carriers and 93% (26/28) of MCM9 carriers, typically recognised between 10 and 30 years of age [1, 9]. In women, the presentation ranges from complete absence of puberty to delayed or incomplete puberty, primary amenorrhoea, or secondary amenorrhoea and oligomenorrhoea after spontaneous menarche [1]. On pelvic ultrasound, a hypoplastic uterus and small or streak ovaries are frequent, especially before oestrogen exposure [1]. In the 2025 series, among women with POI, 14/20 (70%) MCM8 carriers had undetectable or small ovaries and 13 (65%) an infantile or absent uterus; among MCM9 carriers, 14/23 (61%) had invisible or small ovaries and 12/23 (52%) an infantile or absent uterus [9]. In the founder families, the girls had primary amenorrhoea with a 46,XX karyotype, low oestradiol and raised gonadotrophins (FSH, follicle-stimulating hormone, and LH, luteinising hormone), ovaries not identified on ultrasound and an infantile uterus; in the Tenenbaum-Rakover family, the proband was completely prepubertal at 15 years and at 22 years, on HRT, had only breast buds, a 6 mm uterus and a rudimentary vagina on magnetic resonance imaging [5, 6]. Presentations with residual ovaries do exist: in one of the Fauchereau sisters the small uterus and ovaries were visible, and in a Chinese family with compound heterozygous MCM8 variants POI had paediatric onset with delayed puberty and growth [6, 32]. Intermittent ovarian function and spontaneous pregnancy may occur in POI in general, and GeneReviews recommends bearing them in mind in this class as well [1].

3.4. Male reproductive system

In men, the typical presentation is infertility due to spermatogenic failure, ranging from azoospermia to severe oligospermia or oligoasthenospermia, with semen parameters that fluctuate with age and with intercurrent factors such as smoking, obesity, stress and heat exposure [1]. Sertoli cell-only syndrome, with complete absence of germ cells on testicular biopsy, has been described with MCM8 and MCM9 [1, 29, 31]. Male puberty may be normal: the brother in the Belgian family homozygous for MCM9 p.(Arg132Ter) had normal pubertal development and hormones and non-obstructive azoospermia with Sertoli cell-only histology [29]; the 46,XY brother in the Tenenbaum-Rakover family had normal pubic hair and penis, 3 mL testes, azoospermia and raised basal and stimulated gonadotrophins [5]. In two Chinese men with homozygous MCM9 loss-of-function variants, MCM9 was absent in the testis and microsurgical testicular sperm extraction (micro-TESE) retrieved no sperm [24]. GeneReviews stresses that the presence of sperm, even with normal parameters on a single assessment, does not exclude the diagnosis when the hormonal or family picture is suggestive [1]. Ascertainment of men in the series is limited (four per gene in the GeneReviews table) [1].

3.5. Growth, bone and other endocrine manifestations

Short stature occurs mainly in patients with delayed puberty, and final height depends on age at onset and adherence to hormone therapy [1]. In the two Kurdish families of Wood-Trageser the patients had short stature and low weight, and combined oestrogen-progestogen therapy produced regular menses, appropriate bone age and breast and pubic development, with gains in height and weight in the 16-year-old girl whose bone age was 11 years [6]. In the Fauchereau family, adult height was normal despite delayed bone age [6]. In the 2025 series, seven biallelic MCM9 carriers and one MCM8 carrier, all with hypogonadism, had osteoporosis or delayed bone age [9]. Thyroid dysfunction has been described in three biallelic MCM8 carriers: the three Saudi sisters had hypothyroidism that responded to thyroxine, and in one patient with hypothyroidism diagnosed in childhood thyroid function was normal at assessment, but anti-thyroid peroxidase antibodies persisted [1, 5].

3.6. Germ cell tumours

Germ cell tumours have been described only in women: two biallelic MCM8 carriers and one MCM9 carrier, diagnosed between 11 and 15 years of age [1, 9]. Two were endodermal sinus tumours arising in dysgerminomas which had in turn arisen from gonadoblastomas, and the third was a germ cell tumour of unspecified origin; GeneReviews describes them as endodermal sinus tumours, dysgerminomas and gonadal tumours in the setting of marked gonadal dysgenesis [1, 9]. In the review by Goldberg and colleagues, a biallelic MCM9 c.1473dupT carrier from another series had a germ cell tumour at 14 years and POI at 20 [20]. OMIM records that Mcm8-null female mice developed ovarian adenomas and granulosa cell tumours with age [2].

3.7. Digestive tract: polyposis, colorectal cancer and gastric cancer

Colorectal polyposis is observed mainly in MCM9 deficiency [1]. In the 2025 series, 6/28 (21%) biallelic MCM9 carriers had polyposis, typically with more than 20 polyps, of adenomatous, hyperplastic and serrated histology; age at identification of polyps ranged from 19 to 68 years [1, 9]. CRC was the most frequent neoplasm in MCM9 deficiency, in 6/28 (21%): three carriers of likely pathogenic variants had CRC between 30 and 40 years and three VUS carriers between 40 and 60 years [9]. In the Goldberg families, a biallelic MCM9 p.(Glu495Ter) carrier had CRC at 31 years and POI at 15; in the 2015 family, one sister had more than 40 polyps and CRC at 34 years, with a microsatellite-stable tumour, and another more than 20 polyps and two CRCs at 37 years, both with POI before 20 years [19, 20]. In two siblings with biallelic MCM9 variants in the Golubicki series, CRC occurred at 45 years (with microsatellite instability) and at 55 years (stable) [20].

Gastric cancer was described in one biallelic MCM9 carrier in the clinical series, and the population analysis supports the association: in the 100,000 Genomes Project, the odds ratio (OR) for biallelic MCM9 variants was 27.03 (95% confidence interval, 95% CI, 2.93–248.5; p=0.004) for gastric cancer, 6.51 (1.24–34.11; p=0.03) for colonic polyps and 8.40 (1.28–55.35; p=0.03) for rectal polyps, adjusted for age, sex and ethnicity; the OR for CRC (1.49; 0.12–18.11) was not significant [9]. The width of these intervals reflects the small number of cases (two damaging alleles in gastric cancer cases and four to six in polyp cases) [9]. No enrichment was observed in UK Biobank [9].

In MCM8 deficiency, no CRC was described in the clinical series, and in the 100,000 Genomes Project the 51 biallelic carriers had no enrichment of CRC, polyposis or other neoplasms (OR for colonic polyps 1.20; 95% CI 0.20–7.11) [1, 9]. Two biallelic MCM8 carriers in the series developed polyps after publication: a sessile serrated lesion and a tubular adenoma with low-grade dysplasia at 26 years and, in her 35-year-old sister, six tubular adenomas with low-grade dysplasia and three hyperplastic polyps [1]. The proposal of MCM8 as a recessive CRC gene rests mainly on a man with stage IIIB mucinous left-sided colon cancer at 40 years, with microsatellite instability and loss of MLH1/PMS2 without somatic MLH1 methylation, fertility problems and two MCM8 variants in trans [21]. The frequency of polyposis may be underestimated in both entities, because many carriers were young or had never undergone colonoscopy [1].

3.8. Other neoplasms and isolated findings

In the clinical series, one biallelic MCM9 carrier had human papillomavirus-unrelated clear cell carcinoma of the cervix, another melanoma, and one biallelic MCM8 carrier breast cancer [9]. The clear cell carcinoma of the cervix was diagnosed at 37 years in a biallelic MCM9 p.(Glu495Ter) carrier with POI since the age of 15 [20]. In the 100,000 Genomes Project, breast cancer, melanoma, endometrial cancer and epilepsy occurred in biallelic carriers without significant enrichment [9]. Intellectual disability, temporal lobe epilepsy, conductive hearing loss and unilateral renal agenesis (in a consanguineous patient whose brother also had a 22q11 microdeletion), facial naevi and shield chest, and pilomatricomas have been described in isolated individuals; GeneReviews considers it uncertain whether they are part of the phenotype [1, 5].

3.9. Heterozygotes

Heterozygotes may have reduced fertility: in the 2025 series, 29% (14/49) of monoallelic MCM8 carriers and 22% (10/45) of MCM9 carriers had hypogonadism [1, 9]. CRC and polyposis were rarely described (two CRCs and two cases of polyposis among MCM8 heterozygotes; six CRCs and seven cases of polyposis among MCM9 heterozygotes, three with both), at ages above 40 years, and GeneReviews considers that there is no clear evidence of increased risk [1]. In the Goldberg family, the heterozygous parents had three polyps each at 66–68 years and a heterozygous brother had CRC at 35 years with a microsatellite-stable tumour and oligoteratoasthenozoospermia [20]. In the Hartwig Medical Foundation metastatic cohort, four monoallelic MCM8 carriers and three MCM9 carriers had CRC, with no biallelic carriers [9].

3.10. Natural history and treated disease

The available data describe disease that was mostly treated on the endocrine side and not monitored on the oncological side. In the founder families, oestrogen-progestogen therapy induced regular menses, breast development and gain in height [5, 6]. There are no studies comparing oncological outcomes with and without surveillance, nor data on the effect of HRT on bone density or cardiovascular risk specific to this class; these effects are extrapolated from POI in general, in which absence of HRT is associated with reduced life expectancy, mainly due to cardiovascular disease [10]. The typical chronology, built from the 2025 series, is: gonadal insufficiency between 10 and 30 years; germ cell tumours between 11 and 15 years; polyps from 19 years; and CRC between 30 and 60 years in MCM9 deficiency [1, 9].

4. Clinical Recognition and Warning Signs

4.1. When to suspect

GeneReviews proposes as suggestive findings, for both entities, primary gonadal insufficiency (primary or secondary amenorrhoea, oligomenorrhoea, a hypoplastic uterus and small or streak ovaries on ultrasound), impaired spermatogenesis (non-obstructive azoospermia, oligospermia, Sertoli cell-only syndrome), delayed or incomplete puberty, especially in girls, and germ cell tumours before 20 years of age in girls [1]. For MCM9 deficiency, it adds a history of 20 or more adenomas, hyperplastic or serrated polyps at any age, or of 10 to 20 adenomas in an individual with another suggestive finding, CRC diagnosed before 50 years and gastric cancer before 50 years [1]. A family history compatible with recessive inheritance, with affected siblings or parental consanguinity, strengthens the suspicion, but its absence does not exclude the diagnosis [1]. The authors of the 2025 series recommend considering biallelic MCM9 variants in individuals and families with unexplained polyposis, gastric cancer, germ cell tumours or early CRC, especially with recessive inheritance or known hypogonadism, and biallelic MCM8 variants in unexplained germ cell tumours, and consider it prudent to think also of MCM8 in unexplained CRC or polyposis while data remain limited [9].

4.2. By age and clinical context

In the adolescent girl, the typical context is delayed puberty: absence of breast development at 13 years or of menarche at 15 years, which may be self-limiting but may also be the first sign of POI [27]. The combination of raised gonadotrophins, low oestradiol, undetectable anti-Müllerian hormone (AMH) and a hypotrophic uterus with atrophic ovaries, with negative autoimmunity, a normal karyotype and absent FMR1 premutation, was the path that led to the diagnosis in the 2024 Belgian adolescent [27]. In the adult woman, suspicion arises with secondary amenorrhoea or infertility with POI of unidentified cause, particularly when there are affected sisters [1, 10]. In men, the context is non-obstructive azoospermia, especially with Sertoli cell-only syndrome, sisters with POI or consanguinity [1, 29, 31]. In gastroenterology and oncology, suspicion should arise with mixed or adenomatous polyposis without a variant in APC, MUTYH or the other polyposis genes, with CRC or gastric cancer before 50 years, and especially when the same person or a sister has POI [1, 9]. In MMR-deficient CRC without a germline cause in the Lynch genes and without somatic MLH1 methylation, MCM8 has been proposed as a candidate gene [21].

4.3. Clues from history and examination

The history should include a three-generation pedigree with the ages at menarche and menopause of female relatives, consanguinity, male infertility in siblings and a history of polyps and digestive cancer [1, 33]. On examination, height, Tanner stage and testicular volume guide the assessment; the absence of dysmorphic features does not rule out the diagnosis, and the dysmorphic findings described are rare and of uncertain significance [1, 27]. Turner features should be sought, because sex chromosome aneuploidy is the most frequent genetic cause of hypergonadotrophic hypogonadism [1, 33].

4.4. Warning and emergency signs

The most relevant warning sign is an adnexal mass, abdominal pain or increasing abdominal girth in a girl with gonadal dysgenesis due to MCM8 or MCM9, because the germ cell tumours described arose between 11 and 15 years of age [1]. In biallelic MCM9 carriers, rectal bleeding, anaemia, change in bowel habit or passage of mucus should prompt colonoscopy regardless of age, in line with what the European polyposis guideline recommends for adenomatous polyposis syndromes [12]. There are no metabolic emergencies in this class.

4.5. Newborn screening

There is no newborn screening for these syndromes. Presymptomatic diagnosis is made by targeted testing of the siblings of an index case [1].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Criteria

There are no consensus clinical criteria [1]. The diagnosis requires a compatible phenotype and biallelic pathogenic or likely pathogenic variants in MCM8 or MCM9, according to the criteria of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP); the identification of two VUS, or of one pathogenic variant and one VUS, neither establishes nor excludes the diagnosis [1]. The POI phenotypic component is defined by the 2024 ESHRE/ASRM/CREWHIRL/IMS guideline: disturbance of the menstrual cycle (spontaneous amenorrhoea or irregular cycles) for at least 4 months and FSH (follicle-stimulating hormone) above 25 IU/L, a single measurement being sufficient with a characteristic clinical picture and repetition after 4 to 6 weeks being reserved for diagnostic doubt; low oestradiol confirms hypo-oestrogenism but is not used for diagnosis, and AMH may help when FSH is inconclusive [10]. This is a change from the 2015/2016 recommendations, which required two FSH measurements [10]. Hormonal contraception may mask amenorrhoea and lower FSH and may need to be stopped before confirmation; pregnancy must be excluded [10].

5.2. Aetiological assessment in order

The recommended sequence for a woman with non-iatrogenic POI is: karyotype (strong recommendation, ⊕⊕◯◯), FMR1 premutation testing (strong recommendation, ⊕⊕◯◯) and, where available and after genetic counselling, next-generation sequencing (NGS) for other POI genes (conditional recommendation, ⊕⊕◯◯), with no age restriction on access [10]. The 2021 French PNDS proposes the same order, with the NGS panel after a normal karyotype and negative FMR1 premutation, and stresses that DNA repair and meiosis genes constitute the most important causal family, with a very unfavourable fertility prognosis and tumour implications, so that testing for them should be performed by a multidisciplinary team experienced in presymptomatic diagnosis [11]. GeneReviews asks for aneuploidies (Turner, Klinefelter) and the FMR1 premutation to be excluded early [1]. Anti-21-hydroxylase antibody testing is indicated in POI of unknown cause and thyroid-stimulating hormone (TSH) at diagnosis, repeated every 5 years; anti-ovarian antibodies are not useful for the diagnosis of autoimmune POI and anti-thyroid peroxidase antibodies are not tested routinely [10].

For molecular testing, GeneReviews proposes two options [1]. In the first, when the phenotype suggests the syndrome, concurrent sequencing of MCM8 and MCM9 or a multigene panel that includes them is performed, followed by targeted deletion and duplication analysis if only one variant or none is found. In the second, for atypical phenotypes, exome sequencing, the most widely used, or genome sequencing is performed; as most described variants lie in the coding region, they should be detected by the exome [1]. POI panels may include both genes, such as the 34-gene panel used in the 2024 Belgian adolescent [27]. In digestive phenotypes, the 2024 European guideline recommends a multigene germline panel from 20 cumulative adenomas, with a threshold of 10 adenomas if the diagnosis is made before 60 years, if there is a family history of polyposis or CRC or extracolonic manifestations, and in all digestive cancers before 50 years; this guideline does not include MCM8 or MCM9 in its gene list, and the authors of the 2025 series advocate their inclusion in polyposis, early CRC and gastric cancer panels [9, 12].

5.3. Diagnostic yield

The contribution of the class to POI is small and varies with population and phenotype. GeneReviews attributes about 50% of described cases to each gene [1]. In an international series, MCM8 was sequenced in 155 women with POI and MCM9 in 151: 3/155 (2%) had possibly damaging heterozygous MCM8 variants, 7/151 (5%) MCM9 variants, and one woman was homozygous for MCM9 p.(Gln551*) [30]. In a series of 109 women with idiopathic POI, Wood-Trageser found no causal MCM9 variants [3]. In 149 British women with POI of onset before 25 years, 11/17 families (64.7%) with familial POI had candidate variants, including homozygous MCM9 variants, and biallelic variants were identified in MCM9 (c.1217C>T p.A406V and c.724G>A p.G242R) and MCM8 (c.698delC) [34]. In 63 Russian adolescents with 46,XX POI before 18 years, the exome identified causal variants in 11 (17.5%) [35], and among 20 Chinese women with complete 46,XX gonadal dysgenesis one had a homozygous MCM9 variant [36]. In Brazil, the MCM8 variant rs138761187 (c.1954-1G>A) was tested in 100 women with POI and 100 fertile controls from Santo André, São Paulo, with low frequency and no difference between groups; the study tested a single variant and does not assess the contribution of the gene [37]. In CRC and polyposis, no biallelic carrier was found in 44 families with serrated polyposis, in 24 patients from 16 families meeting the Amsterdam criteria without polyposis or in 632 metastatic CRCs [9].

5.4. Limitations of the tests

Sequencing detects all variants described so far, but this proportion refers to the known spectrum and is not equivalent to clinical sensitivity [1]. No large intragenic deletions or duplications have been described; a result with a single variant should nonetheless lead to copy number analysis, and apparent homozygosity in a patient without consanguinity may correspond to an undetected deletion on the other allele or to uniparental isodisomy [1]. POI panels do not detect structural rearrangements, repeats, deep intronic variants or promoter variants [27]. The FMR1 premutation requires its own test [10]. Many missense variants are classified as VUS, and the 2025 series included VUS carriers selected by in silico criteria, which limits estimates of phenotype frequency [9]. A negative polyposis or CRC panel excludes the class only if it includes both genes [9, 12].

5.5. Report content, VUS and segregation

The report should state the gene, the transcript with version (NM_032485.6 for MCM8 and NM_017696.3 for MCM9), the variant in Human Genome Variation Society (HGVS) nomenclature, the zygosity and the ACMG/AMP classification, and should mention the oncological and not only the reproductive implications when the gene is MCM9 [1, 4]. The phase (cis or trans) of the variants should be confirmed by parental testing, as in the Belgian adolescent, in whom the frameshift variant was paternal and the nonsense variant maternal [1, 27]. If only one parent is a carrier and paternity and maternity are confirmed, one of the variants may be de novo in the proband or result from parental post-zygotic mosaicism [1]. Segregation in affected and unaffected siblings contributes to VUS reclassification [1].

5.6. Practical algorithm

5.7. Prenatal and preimplantation diagnosis

Once the two pathogenic variants have been identified in the family, prenatal diagnosis and preimplantation genetic testing are possible [1]. The appropriate time to discuss the risk and these options is before pregnancy [1]. There are no prenatal ultrasound markers for the class.

6. Differential Diagnosis

6.1. Comparative table

6.2. Main differential diagnoses

The first differential is the common cause of hypergonadotrophic hypogonadism. Sex chromosome aneuploidy and the FMR1 premutation are the most frequent genetic causes and must be excluded before any panel, because the panel and the exome do not detect the premutation [1, 10]. In sporadic idiopathic POI, about 2% of women have an FMR1 premutation, and in familial POI about 14% [33]. Interpreting low-grade 45,X/46,XX mosaicism is difficult because X loss increases with age, from 1.5–2.5% of cells in prepubertal girls to 4.5–5% above 75 years [33]. In primary amenorrhoea, gonadal dysgenesis accounts for 48.5% of cases, followed by uterovaginal agenesis (16.2%), GnRH deficiency (8.3%) and constitutional delay (6.0%), which places MCM8/MCM9 syndromes within the first group [42].

The second differential comprises the other DNA repair genes that cause POI with tumour risk. Fanconi anaemia, Bloom syndrome and Werner syndrome share the pathway and the chromosomal instability, but have their own somatic manifestations: bone marrow failure and radial ray malformations in Fanconi anaemia, severe growth restriction and facial erythema in Bloom syndrome, premature ageing in Werner syndrome [1, 38, 39]. Fanconi anaemia may occur in children without dysmorphic features (25% are structurally normal), and the median age at haematological onset is 7 years [38]. Neither bone marrow failure nor radial ray malformations have been described in MCM8/MCM9 syndromes, and sensitivity to crosslinks is a research finding [1, 16]. HROB, which encodes a protein that stimulates the MCM8-MCM9 helicase, produces an almost overlapping picture: in a 2026 family, two individuals with biallelic HROB variants had hypergonadotrophic hypogonadism and colonic polyposis, and the authors recommend including the gene in hypergonadotrophic hypogonadism panels, especially with polyposis [22].

The third differential is that of the polyposis syndromes. Patients with MCM9 deficiency usually have more than 20 polyps of mixed adenomatous, hyperplastic and serrated histology, which brings them close to hereditary mixed polyposis, serrated polyposis and the attenuated forms of APC and MUTYH [1, 19]. In the general population, up to 30% of people have one or two adenomas by 70 years, but only 1/1000 has more than two, and 20–30% of cases with fewer than 100 adenomas remain without a molecular diagnosis, which defines the group in which MCM9 should be sought [40]. About 25% of people with more than 9 adenomas and no dominant history have MUTYH-associated polyposis, which is also recessive, but without gonadal dysfunction [40]. MUTYH-associated polyposis may produce CRC with microsatellite instability through somatic second hits in the MMR genes, mimicking Lynch syndrome; the same reasoning applies to the Lynch-like cases in which MCM8 has been proposed [21, 41]. The most useful clinical distinction is the combination of polyposis or early digestive cancer with hypogonadism in the same person or in a sister, which does not occur in the other polyposis syndromes [1].

7. Treatments and Risk Management Strategies

7.1. Principles and evidence base

There is no treatment that corrects the MCM8-MCM9 complex defect or restores gonadal function, and there are no class-specific clinical guidelines; the GeneReviews recommendations are based on the authors' personal experience [1]. Management is organised on three fronts: treatment of gonadal insufficiency and its sequelae, which follows the general guidelines for POI and male infertility; treatment of tumours, which follows standard oncological practice; and cancer surveillance, specific to the class and the entity [1]. Care should be multidisciplinary, with paediatric or reproductive endocrinology, gynaecology, andrology or urology, gastroenterology, oncology and clinical genetics, and there should be early referral to reproductive medicine [1]. ESHRE recommends that the diagnosis of POI be communicated sensitively, with personalised information, shared decision-making, continuity of care and referral to support groups and mental health care [10].

7.2. Baseline assessment after diagnosis

GeneReviews recommends, if not already performed, an assessment by reproductive or paediatric endocrinology with pubertal staging, sexual history and enquiry about symptoms of sex steroid deficiency (vasomotor or genitourinary); in women, menstrual history, FSH, LH, oestradiol and pelvic ultrasound; in men, testosterone and semen analysis after puberty, with sperm cryopreservation if sperm are present [1]. In people with gonadal insufficiency, it recommends bone densitometry by dual-energy X-ray absorptiometry (DXA), assessment of calcium and vitamin D intake, blood pressure, weight and body mass index (BMI), smoking status, lipid profile and diabetes screening (glycated haemoglobin, HbA1c, or blood glucose) [1]. For tumours, it recommends considering pelvic ultrasound in girls from 10 years, colonoscopy in biallelic MCM9 carriers from 25 years and upper gastrointestinal endoscopy in biallelic MCM9 carriers from 30 years; it adds that larger cohorts are needed to define the ideal starting age [1]. The surveillance table of the same chapter proposes upper endoscopy combined with colonoscopy, and the summary refers to both every 2 years from 25 years, so there is an internal five-year discrepancy regarding the start of upper endoscopy [1]. Genetic counselling completes the assessment [1].

7.3. Pubertal induction

Pubertal induction is indicated in girls without spontaneous pubertal development or with incomplete development [1]. ESHRE recommends inducing or completing puberty with oestradiol, starting with a low dose at 11 years and increasing gradually over 2 to 3 years (strong recommendation, ⊕⊕◯◯); in late diagnoses, without concern about growth, it allows a modified regimen (conditional); the transdermal route may be preferred because it produces more physiological concentrations, without conclusive evidence on the optimal route; the combined oral contraceptive should not be used for induction (strong); and cyclical progestogens are started after about 2 years of oestrogen or when bleeding occurs [10]. The French PNDS recommends 17β-oestradiol, transdermal by default, from the physiological age of pubertal onset, between 11 and 13 years, without delaying induction beyond 12 years; it describes starting doses of about 8 to 10% of the adult dose, or 5 µg/kg/day with increases every 6 months up to the adult dose over 2 to 3 years, using divisible 25 µg patches, and recommends a growth chart and bone age before starting [11].

Neither ESHRE nor the PNDS gives a year-by-year stepwise regimen for POI. By extrapolation from the most studied analogous condition, Turner syndrome, the 2024 international guideline proposes for girls with raised FSH starting between 11 and 12 years with 7 µg/24 h of transdermal oestradiol (half a 14 µg patch or a quarter of a 25 µg patch) or 0.25 mg/day orally, with increases every 6 to 12 months guided by breast stage and serum oestradiol, reaching the adult dose in the 4th year [43]. This extrapolation is reasonable because the physiological aim is the same (replacing absent ovarian production in a girl with growth potential), but it has not been studied in MCM8/MCM9 syndromes.

In girls with short stature, the low starting doses are maintained so as not to compromise the pubertal growth spurt, according to the PNDS [11]. In the 2024 Belgian adolescent, with a bone age of 12 years at 16, induction was performed with oral oestradiol valerate 0.5 mg/day, increased to 2 mg/day, until breast stage 4 and a mature uterus on ultrasound [27]. Oestradiol patches are not indicated in paediatric patients in the US label, which recommends monitoring bone maturation if oestrogen is given before completion of growth; patch fractionation and induction doses are therefore off-label use supported by the guidelines [11, 43, 44].

7.4. Hormone replacement therapy in the adolescent and adult woman

HRT is recommended to treat the symptoms of hypo-oestrogenism (strong, ⊕⊕◯◯), and the absence of HRT is associated with a higher risk of cardiovascular events and mortality, so it is recommended that it be continued until the usual age of menopause (strong) [10]. GeneReviews adopts the same recommendation, citing the opinion of the American College of Obstetricians and Gynecologists [1]. The French PNDS advises continuing HRT at least until 51 years, the average age of natural menopause, and the Oxford reference book on clinical genetics indicates at least until 50 years [11, 33]. On reaching the usual age of menopause, continuation should be based on a personalised assessment of risks and benefits [10]. ESHRE states that there is no evidence that HRT increases the risk of breast cancer in women with POI compared with women of the same age without POI (conditional, ⊕⊕◯◯) [10].

Doses differ between sources, and the guideline dose is the reference. ESHRE suggests, to optimise bone mineral density, a daily HRT dose of no less than 2 mg of oral oestradiol or 100 µg of transdermal oestradiol, or equivalent (conditional, ⊕◯◯◯), and, if the combined oral contraceptive is used, a continuous or extended regimen to avoid the oestrogen-free interval (strong) [10]. The French PNDS indicates a dose producing serum oestradiol of about 100 pg/mL, between 50 and 100 µg/day transdermally, 2 to 3 pumps of gel per day or 2 to 4 mg of oral oestradiol, and notes that doses of about 100 µg transdermally generally achieve physiological concentrations [11]. The Turner guideline gives an adult range of 50–200 µg/24 h transdermally, 1.5–3 mg/day of gel or 2–4 mg/day orally, with the approximate equivalence of 50/100 µg transdermal = 2 mg oral = 20 µg of ethinylestradiol [43]. The ESHRE minimum dose (100 µg transdermal) coincides with the upper limit of the PNDS.

The labels are designed for physiological menopause and serve for comparison. In Portugal, the SmPC of Dermestril® (oestradiol, transdermal systems of 25, 50 and 100 µg/24 h applied twice weekly) indicates HRT for symptoms of oestrogen deficiency in women at least 6 months after their last menstrual period and prevention of osteoporosis with the 50 and 100 µg strengths, usually starting with 25 µg, and requires a progestogen for at least 12 to 14 days per cycle in women with a uterus [45]. The SmPC of Estrofem® (oestradiol 1 and 2 mg) indicates one tablet daily without interruption for postmenopausal symptoms, at the lowest effective dose [46]. In the United States, the transdermal oestradiol label includes the indication of hypo-oestrogenism due to hypogonadism, castration or primary ovarian failure, starting at 0.0375 mg/day twice weekly for vasomotor symptoms and 0.025 mg/day for osteoporosis prevention [44]. The guideline doses for POI (≥100 µg/24 h) are above the starting doses of these labels and, in adolescents, outside the approved indications. In Brazil, Estradot® (oestradiol hemihydrate, Sandoz) and Utrogestan® (progesterone, Besins), among others, have active registration with ANVISA [47].

A progestogen is combined with oestrogen in all women with a uterus to prevent endometrial hyperplasia and cancer (strong, ⊕⊕◯◯), and its dose should increase when higher oestrogen doses are used [10]. The sequential regimen produces regular withdrawal bleeds; if the woman does not want them, the progestogen can be given continuously [11]. In a woman with intermittent ovarian function who wishes to conceive naturally, ESHRE recommends a sequential regimen, which does not alter the probability of natural conception [10]. HRT is not contraceptive: ESHRE recommends contraception for women with non-surgical POI who wish to avoid pregnancy (strong), and GeneReviews repeats the warning [1, 10]. On HRT, absence of bleeding after the progesterone phase should prompt a pregnancy test [11]. Transdermal testosterone may be considered for hypoactive sexual desire when other causes have been excluded (conditional in non-iatrogenic POI), with no long-term safety data [10].

Adverse effects and contraindications follow those of the SmPC. With Dermestril®, breast discomfort was reported by 20–35% of patients and application site reactions, mainly mild erythema with or without pruritus, by 10–25%; headache, abdominal pain, nausea, depressed mood, rash and pruritus are common [45]. Contraindications include confirmed, suspected or previous breast cancer, oestrogen-dependent malignant tumours, undiagnosed genital bleeding, untreated endometrial hyperplasia and previous or current venous thromboembolism or known thrombophilia; before elective surgery with prolonged immobilisation, the SmPC recommends stopping HRT 4 to 6 weeks beforehand [45]. Oral progesterone may cause drowsiness and dizziness and should be stopped in the event of jaundice, a significant rise in blood pressure or new-onset migraine-type headache [48]. ESHRE does not consider migraine a contraindication to HRT and recommends the transdermal route in migraine with aura [10].

7.5. Bone, cardiovascular system and metabolism

POI is associated with changes in bone microarchitecture and reduced mineral density (strong, ⊕⊕◯◯) [10]. ESHRE recommends DXA at diagnosis in all women, where available (strong); if density is normal and adequate HRT is adhered to, it considers repeating DXA before 5 years to be of little value; and, with osteoporosis or low density, it recommends reassessment every 1 to 3 years according to risk factors [10]. A decrease in density should lead to a review of HRT and of the factors contributing to bone loss [10]. Calcium and vitamin D supplementation may be needed in women with vitamin D deficiency or insufficient calcium intake [10]. In the 2025 series, eight biallelic carriers with hypogonadism had osteoporosis or delayed bone age [9]. Women with POI should be informed of the increased risk of coronary heart disease, heart failure and stroke and of the favourable cardiometabolic effect of oestrogens, and encouraged to adopt a healthy lifestyle (no smoking, a healthy diet, physical activity and appropriate weight); the lipid profile and diabetes screening are performed at diagnosis and thereafter according to risk [10].

7.6. Men: hypogonadism and infertility

In men, spermatogenic failure is managed in andrology or urology [1]. If sperm are present in the ejaculate, cryopreservation should be considered; if not, surgical sperm retrieval with assisted reproductive techniques may be considered when appropriate [1]. Prognostic information is scarce: in the two Chinese men with Sertoli cell-only syndrome and homozygous MCM9 loss-of-function variants, micro-TESE failed, and the authors propose that these variants might serve as a biomarker of micro-TESE failure, which has not been validated [24]. In men who wish to have current or future fertility, testosterone monotherapy should be avoided because it may suppress spermatogenesis, according to the American Urological Association and American Society for Reproductive Medicine guideline cited in GeneReviews [1, 50]. In men with androgenic hypogonadism, testosterone replacement follows the general standards for male hypogonadism, which the sources consulted for this class do not specify.

7.7. Female fertility

Women with POI have greatly reduced fertility, but intermittent ovarian function and spontaneous pregnancy are possible [1]. Most of the patients described had severe gonadal dysfunction, and natural conception is unlikely [1]. Fertility preservation is possible only if there is gonadal reserve; in the 2024 Belgian adolescent, oocyte cryopreservation was not offered because AMH was undetectable [1, 27]. Oocyte donation is an established option (strong, ⊕⊕◯◯) [10]. When the donor is a sister, ESHRE recommends informing the woman that the genetic risk is shared and that the cancellation rate of stimulation cycles is higher (strong); in a recessive syndrome with affected sisters, the donor sister should be tested for the familial variants before donation [1, 10]. Pregnancies from oocyte donation are high risk and should be monitored in an appropriate obstetric unit (strong), and the woman should be investigated for the aetiology of POI before donation (strong) [10]. In in vitro fertilisation (IVF) with donated oocytes, the recipient uterus is prepared with its own HRT regimen [33]. The SmPC of Utrogestan® includes, by the vaginal route, progesterone replacement in women without ovarian function in oocyte donation programmes [48].

7.8. Germ cell tumours

Germ cell tumours are treated according to standard oncological practice [1]. Prophylactic gonadectomy is not part of the GeneReviews recommendations, and the sources consulted do not evaluate this option in MCM8/MCM9 syndromes; the proposed strategy is pelvic ultrasound surveillance, to be considered annually from 10 to 20 years in girls [1]. The authors of the 2025 series base the start at 10 years on the age of 11 to 15 years observed in the three cases [9]. The presence of tumours arising in gonadoblastomas in the setting of marked gonadal dysgenesis justifies investigating without delay any adnexal mass in a girl with the diagnosis [1, 9]. ESHRE recommends avoiding HRT in women with hormone-dependent ovarian or uterine tumours, including granulosa cell tumours and sex cord-stromal tumours (strong, ⊕⊕⊕◯), and considers the effect of HRT on recurrence of non-epithelial ovarian cancers uncertain, advising a personalised decision that takes account of the tumour's hormone receptors [10]; this recommendation is relevant for survivors of germ cell tumours who need HRT.

7.9. Polyposis, colorectal cancer and gastric cancer

In MCM9 deficiency, polyps are resected during colonoscopy; when the size and density of polyps can no longer be controlled endoscopically, subtotal colectomy or proctocolectomy should be performed, according to the characteristics and location of the polyps [1]. CRC and gastric cancer are treated according to standard practice [1]. The 2024 European guideline for MUTYH-associated polyposis, also recessive and with an attenuated polyposis phenotype, serves as a model: preference for endoscopic resection of adenomas when the burden allows; a surgical decision in a multidisciplinary meeting, considering the polyp burden in the colon and rectum, age, comorbidities, patient preference and adherence to surveillance; colectomy with ileorectal anastomosis as the first option and proctocolectomy with an ileal pouch if dense rectal polyposis cannot be controlled; and prophylactic surgery not recommended in carriers without polyps or cancer [12]. Applying this model to MCM9 deficiency is an extrapolation, because there are no data on surgical outcomes in this class. GeneReviews recommends, in biallelic MCM9 carriers, testing for and eradication of Helicobacter pylori combined with upper gastrointestinal endoscopy, as a gastric cancer risk-reduction measure [1]. There are no eradication studies in patients with MCM9 deficiency; regimens, doses and confirmation of eradication follow the general guidelines for adults, the 2022 Maastricht VI/Florence European consensus and the 2024 American College of Gastroenterology (ACG) guideline, the latter developed using the GRADE method for adults aged 18 years or over [51, 52]. The indication to treat is compatible with both sources: the Maastricht VI consensus considers that eradication reduces the risk of gastric cancer at any adult age, with greater benefit before severe atrophic gastritis, and classifies hereditary gastric cancer as a distinct entity in which the role of infection remains to be clarified (grade D2); the ACG guideline includes among the indications for testing and treatment first-degree relatives of patients with gastric cancer, people with hereditary syndromes with an increased risk of gastric cancer, by shared decision, and adults living with a person with a positive non-serological test [51, 52].

The choice of regimen depends on clarithromycin resistance. According to the Maastricht VI consensus, without individual susceptibility testing and in regions with clarithromycin resistance above 15% or unknown, first-line treatment is bismuth quadruple therapy (proton pump inhibitor, PPI, bismuth, tetracycline and metronidazole) for 14 days, unless 10-day regimens with proven local efficacy are available (grades B1 and D2); failing that, non-bismuth concomitant quadruple therapy (PPI, amoxicillin, clarithromycin and a nitroimidazole given simultaneously) for 14 days is the alternative, discouraged when dual resistance to clarithromycin and metronidazole exceeds 15%; clarithromycin triple therapy is used only where resistance is low and local efficacy is at least 90%, also for 14 days; clarithromycin susceptibility testing, molecular or by culture, is recommended before any clarithromycin-containing regimen [51]. In the European Hp-EuReg registry, single-capsule bismuth quadruple therapy for 10 days achieved eradication of 94.6% in first line (modified intention-to-treat analysis, 2100 patients, mainly from Spain, Italy and Portugal) and of 92% in the overall set of 5068 patients [51]. The ACG guideline recommends "optimised" bismuth quadruple therapy for 14 days as first line (strong recommendation, moderate quality) and suggests as empirical alternatives rifabutin triple therapy and potassium-competitive acid blocker (P-CAB) plus amoxicillin dual therapy, both for 14 days (conditional recommendations); P-CAB clarithromycin triple therapy is suggested instead of PPI clarithromycin triple therapy, but reserved for strains with proven susceptibility, and concomitant therapy is not suggested instead of bismuth quadruple therapy [52]. In confirmed penicillin allergy, both sources indicate bismuth quadruple therapy; the ACG also proposes referral to allergy and immunology to confirm the allergy and consider desensitisation, because less than 1% of the population has true IgE-mediated allergy [51, 52].

The doses are taken from tables 5 and 6 of the ACG guideline; the Maastricht VI consensus does not set doses per regimen, but defines high-dose PPI as 40 mg of omeprazole (double dose) or equivalent twice daily, which increases the efficacy of triple therapy with no demonstrated benefit in quadruple regimens, indicates 500 mg/day of levofloxacin and reports that amoxicillin with a PPI in four daily doses achieved more eradication than lower dosing (87% versus 73%, meta-analysis of 15 randomised trials) [51, 52]. The ACG advises against substituting doxycycline for tetracycline and using metronidazole below 1.5 g/day, because both modifications reduce eradication; bismuth subsalicylate is contraindicated in salicylate allergy; levofloxacin carries a warning of tendinitis and tendon rupture; and a non-optimised bismuth quadruple therapy (less than 10 days, metronidazole below 1.5 g/day, doxycycline or insufficient PPI) is the only failed regimen that can be repeated, after optimisation [52]. After a first failure, the Maastricht VI consensus avoids repeating regimens already used: after bismuth quadruple therapy, fluoroquinolone quadruple or triple therapy or high-dose dual therapy; after clarithromycin triple therapy or concomitant therapy, bismuth quadruple therapy or a fluoroquinolone regimen; and, in third line, rifabutin, high-dose dual therapy or bismuth with other antibiotics when fluoroquinolone resistance is high [51]. The ACG guideline suggests optimised bismuth quadruple therapy in patients who have not yet received it and rifabutin triple therapy in those who have, with rifabutin preferably 50 mg 3 times daily (Talicia®) or 150 mg twice daily [52]. Eradication is confirmed in all treated patients by urea breath test, monoclonal stool antigen or biopsy-based test, at least 4 weeks after the end of treatment, without antibiotics or bismuth in the preceding 4 weeks and with the PPI stopped 14 days beforehand (Maastricht VI, grade A1); H2 antagonists and antacids do not interfere, and serology is not useful to confirm eradication [51, 52].

7.10. Agents and exposures to consider

There are no drugs specifically contraindicated in this class [1]. Cells lacking MCM8 or MCM9 are highly sensitive to agents that induce interstrand crosslinks, such as mitomycin C, and fibroblasts from deficient mice are more sensitive to ionising radiation [3, 21]. There are no clinical data on the toxicity of chemotherapy with alkylating agents, platinum agents or radiotherapy in biallelic carriers, so these laboratory observations justify only that the oncology team be aware of the diagnosis when planning treatment of a tumour; they do not support dose changes. Smoking, obesity, stress and heat may impair semen parameters and should be avoided in men trying to preserve fertility [1]. ESHRE advises against compounded so-called "bioidentical" hormone preparations, owing to a lack of efficacy and safety data [10].

7.11. Pregnancy, perioperative care and transition

Pregnancy, whether spontaneous or achieved by assisted reproduction, is monitored according to the usual obstetric indications and the woman's comorbidities [1]; pregnancies from oocyte donation are high risk [10]. Oral progesterone is restricted, in the SmPC of Utrogestan®, to non-obstetric indications, and use in pregnancy is limited to the first trimester and the vaginal route [48]. Perioperatively, for elective surgery with prolonged immobilisation, the SmPC of Dermestril® recommends stopping HRT 4 to 6 weeks beforehand and resuming it only with full mobilisation [45]. The transition from paediatric to adult endocrinology is a time of risk for HRT discontinuation (in Turner syndrome, up to 37% of women under 51 years stop HRT prematurely) [43]; ESHRE stresses that adherence to HRT is necessary to minimise long-term risks and requires prolonged follow-up (strong) [10]. At transition, the cancer surveillance plan (pelvic ultrasound up to 20 years, colonoscopy from 25 years in MCM9 deficiency) should be explicitly handed over to the adult team [1].

7.12. Regulatory status

There are no medicines approved specifically for MCM8/MCM9 syndromes, and the sources consulted do not mention orphan drug designations or regulatory applications [1]. The drugs used are oestrogens and progestogens for general use, authorised through national procedures.

7.13. Trials and developments in the last 12 months

A search of ClinicalTrials.gov (6 November 2026) for "MCM8" or "MCM9" found no registered studies, and GeneReviews indicates that there may be no trials for this class [1]. There are therefore no interventional trials to tabulate; the following table summarises the clinical and population studies supporting the recommendations, with their design and limitations.

These studies do not allow cumulative risks to be estimated, and the wide confidence intervals of the odds ratios reflect very small absolute numbers; a high OR should not be read as a high absolute risk. In the last 12 months, the milestones were: the definitive classification of MCM9 in ClinGen (April 2026) [8]; the description of HROB deficiency with hypogonadism and polyposis (March 2026) [22]; the preclinical mouse study demonstrating the requirement for helicase activity for primordial germ cell development (May 2026) [25]; the functional characterisation of MCM8 variants in two paediatric patients (June 2026) [32]; and the publication of GeneReviews (June 2026) [1]. None of these studies is a therapeutic trial.

7.14. Gene and targeted therapies

There is no gene therapy, genome editing or targeted pharmacological therapy in clinical development for MCM8/MCM9 syndromes, and the ClinicalTrials.gov search identified no studies [1]. The available research is preclinical and mechanistic: the 2026 mouse models indicate that ovarian loss begins during primordial germ cell development, which places the biological window for any correction before birth and reduces the plausibility of restoring fertility once dysgenesis is established [25]. No therapeutic strategy based on this observation has been tested in humans.

7.15. Recommendations from the sources and discrepancies

The discrepancy most relevant to practice is the oestradiol dose: ESHRE sets as the minimum for bone the value that the PNDS gives as the upper limit. For H. pylori eradication, the European consensus accepts non-bismuth concomitant quadruple therapy when bismuth is unavailable and high-dose dual therapy as rescue, whereas the North American guideline does not suggest concomitant therapy instead of bismuth quadruple therapy and makes no recommendation on high-dose dual therapy as rescue; the North American guideline also includes rifabutin and vonoprazan regimens approved in the USA [51, 52]. The third is the age at starting upper endoscopy in MCM9 deficiency, with an internal discrepancy within GeneReviews itself; the choice of starting upper endoscopy together with the first colonoscopy at 25 years is consistent with the surveillance table and with the proposal of the series authors, and that of starting at 30 years with the baseline assessment table [1, 9].

8. Reproductive Options and Family Counselling

8.1. Recurrence risks

MCM8/MCM9 syndromes are transmitted in an autosomal recessive manner [1]. The parents of a proband are presumed heterozygous, and molecular testing should confirm that both are carriers, so that the recurrence risk is reliable [1]. If both are heterozygous, each sibling of the proband has, at conception, a 25% probability of inheriting both variants and being affected, 50% of being heterozygous and 25% of inheriting neither [1]. If the variant is found in only one parent, with maternity and paternity confirmed, the other may have arisen de novo in the proband or in a parent with post-zygotic mosaicism [1]. If the proband appears homozygous, a deletion of one or more exons not detected by sequencing, which mimics homozygosity, or uniparental isodisomy of the carrier parent's chromosome should be considered [1]. The children of a patient are obligate heterozygotes, but natural conception is unlikely, because most of the patients described have severe gonadal dysfunction [1]. Each sibling of the proband's parents has a 50% probability of being heterozygous [1]. In consanguineous families, the risk extends to other branches: in the second Tenenbaum-Rakover family, after the diagnosis in three sisters, two paternal cousins aged 30 and 28 years were also diagnosed with primary hypergonadotrophic hypogonadism [5].

8.2. Testing of relatives and carriers

Heterozygosity testing in at-risk relatives requires prior identification of the familial variants [1]. GeneReviews recommends clarifying the genetic status of asymptomatic older and younger siblings, to identify as early as possible those who may benefit from treatment of POI and delayed puberty, reproductive planning and cancer surveillance [1]. ESHRE recommends offering counselling and testing to relatives of women with an identified genetic cause of POI and informing the sisters and daughters of women with non-iatrogenic POI of their increased risk, of the signs and symptoms of POI and that there are no established methods to predict or prevent it, some of whom may consider family planning and fertility preservation [10]. In recessive polyposis syndromes, the European guideline recommends always offering testing to siblings [12]. In heterozygotes, GeneReviews does not recommend cancer surveillance beyond population screening, and the authors of the 2025 series consider that the data do not justify surveillance in these carriers [1, 9]. Testing the partner of a patient or carrier allows the risk to offspring to be calculated; the frequency of loss-of-function variants in the population is low (1.4 and 2.5 per 100,000 individuals in gnomAD for MCM8 and MCM9), but rare missense variants are much more frequent and mostly of unknown significance [1].

8.3. Germline mosaicism

The sources consulted do not describe germline mosaicism in this class. In a recessive model, the relevant parental mosaicism is that of a post-zygotic de novo variant in a parent, which GeneReviews cites as an explanation when only one parent is a carrier [1].

8.4. Reproductive options

The ideal time to determine genetic risk and discuss prenatal and preimplantation diagnosis is before pregnancy [1]. For heterozygous couples with an affected child, the options include prenatal diagnosis, preimplantation genetic testing, the use of donor gametes and acceptance of the risk; views on the use of prenatal and preimplantation diagnosis may differ among professionals and within families, and the decision is personal [1]. For the affected person themselves, reproductive options depend on gonadal reserve: fertility preservation while reserve still exists, oocyte donation, IVF and intracytoplasmic sperm injection (ICSI) with sperm from the ejaculate or retrieved surgically [1, 10]. In oocyte donation by a sister, the genetic risk is shared and the sister should be tested [1, 10]. A woman with POI should know that spontaneous conception is possible and that HRT is not contraceptive [1].

8.5. Counselling of minors and young adults

Testing of minor siblings has direct benefit, because it allows timely pubertal induction and, in girls, ultrasound surveillance from 10 years [1]. GeneReviews recommends offering genetic counselling, including discussion of the risks to offspring and of reproductive options, to young adults who are affected, heterozygous or at risk of being so [1]. The French PNDS stresses that the identification of variants in DNA repair and meiosis genes should be carried out in a multidisciplinary team experienced in presymptomatic diagnosis, because of the tumour implications, and in compliance with applicable legislation [11].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Frequency

The prevalence of MCM8/MCM9 syndromes is unknown [1]. Orphanet assigns 46,XX ovarian dysgenesis-short stature syndrome a prevalence below 1/1,000,000 [7]. Allele frequency data measure the presence of variants, not disease. In gnomAD v2.1.1 (141,456 unrelated individuals, accessed May 2023), predicted loss-of-function variants in MCM8 occurred in 1.4 per 100,000 individuals, with a maximum of 5.5 per 100,000 in Africans and African Americans, and those in MCM9 in 2.5 per 100,000, with a maximum of 5.7 per 100,000 in Finns; missense variants occurred in 462.4 and 1173.3 per 100,000, without it being known how many are pathogenic [1, 9]. In the gnomAD co-occurrence tool, 23 individuals (0.02%) were classified as biallelic carriers of missense or more severe variants in MCM8 and 22 (0.02%) in MCM9, one of whom was homozygous for loss of function [9]. This value of 0.02% is a genotype frequency that includes variants of unknown effect and should not be read as the prevalence of manifest disease [1, 9]. POI in general is common: the prevalence of non-iatrogenic POI ranges from about 1% in older studies to 3.5% in recent ones, and POI and early menopause combined reach 12.2% [10]; MCM8/MCM9 syndromes are a small fraction of this whole.

9.2. Populations described

The first families were consanguineous, of Kurdish (MCM9), Saudi and Arab (MCM8) and Arab (MCM9) origin, and until 2024 homozygous variants had been found almost exclusively in Arab, Turkish or Jewish families [15–18, 27]. Non-consanguineous European cases include the Belgian adolescent with compound heterozygous MCM9 variants and a Belgian family homozygous for MCM9 p.(Arg132Ter) [27, 29]. The 2025 series included unpublished cases from several European ERN GENTURIS centres and the analysis of UK participants [9]. There are also series from China, the United Kingdom and Russia [26, 34, 35]. The MCM8 variant c.1954-1G>A (rs138761187) was observed in 1/8599 European American exomes and in 1/4545 genomes of the 1000 Genomes Project [2]. In Brazil, a study from Santo André (São Paulo) assessed this variant in 100 women with POI and 100 fertile controls and found a low frequency, with no difference between groups [37]. The sources consulted do not include data from Portugal. The absence of data in the sources does not demonstrate the absence of patients, and the reliance on consanguinity in the first descriptions reflects identification by homozygosity mapping.

9.3. Special populations

In children, early diagnosis allows pubertal induction with preserved growth potential and surveillance for germ cell tumours [1, 43]. In men, ascertainment in the series is limited and the phenotype may be confined to azoospermia with normal puberty [1, 29]. In women pregnant through oocyte donation, the obstetric risk is high [10]. In consanguineous families, screening of cousins and other branches has a yield [5].

9.4. Reference networks and patient associations

ERN GENTURIS (genetic tumour risk syndromes) covers the oncological component and was the network through which the 2025 series identified unpublished cases [9, 14]. Endo-ERN covers rare endocrine conditions, including disorders of puberty and gonadal function [13]. GeneReviews lists as resources the Colorectal Cancer Alliance, the National Cancer Institute and the ReproductiveFacts portal of the American Society for Reproductive Medicine [1]. The Daisy Network, in the United Kingdom, supports women with POI [33]. ESHRE recommends referring women with POI to support groups [10].

10. Illustrative Clinical Case

10.1. Adolescent with delayed puberty and compound heterozygous MCM9 variants

Nauwynck and colleagues described, in 2024, a 16-year-old girl from a white, non-consanguineous Belgian family, referred to a paediatric endocrinology clinic for delayed puberty [27]. She had no pubic hair, breast development or vaginal discharge, and reported no galactorrhoea, hirsutism, hot flushes, mood changes, weight change or cyclical abdominal pain; her personal history was unremarkable and she was taking no medication [27]. Both parents had had constitutional delay of puberty (maternal menarche at 14 years; the father had received intramuscular testosterone), her sister had started breast development at 13.5 years, and there was no history of autoimmune or endocrine disease, early menopause or fragile X syndrome [27]. On examination, she weighed 46.5 kg (Z-score −1.5), measured 157.9 cm (Z-score −1.3), had a blood pressure of 119/66 mmHg, Tanner stage I breast and pubic development, a normal thyroid and no dysmorphic features [27].

Investigations showed FSH of 81.1 IU/L and LH of 29.4 IU/L, oestradiol below 25 ng/L, AMH below 0.03 µg/L and inhibin B below 10 ng/L, with normal prolactin, TSH and free T4; repeat testing confirmed hypergonadotrophic hypogonadism [27]. Anti-ovarian, anti-21-hydroxylase, anti-thyroid peroxidase and anti-parietal cell antibodies were negative [27]. Bone age was 12 years at a chronological age of 16 years, and pelvic magnetic resonance imaging showed a hypotrophic uterus and atrophic ovaries [27]. Array comparative genomic hybridisation and FMR1 repeat analysis were normal [27].

A 34-gene POI panel identified two MCM9 variants: NM_017696.3:c.2237_2238dup p.(Phe747IlefsTer78), frameshift, and NM_017696.3:c.820C>T p.(Gln274Ter), nonsense, both rare in gnomAD v4.1.0 and classified as likely pathogenic (PVS1 and PM2) [27]. Segregation showed that the first was paternal and the second maternal, confirming compound heterozygosity; a heterozygous maternal pathogenic FSHR variant (NM_000145.4:c.1222G>T p.(Asp408Tyr)) was also found, considered irrelevant to the phenotype because FSHR-related POI is recessive [27]. The pedigree and electropherograms are shown in figure 1 of the article (https://pmc.ncbi.nlm.nih.gov/articles/PMC11558971/figure/fig1/) [27].

Pubertal induction was carried out with oral oestradiol valerate, 0.5 mg daily, increased progressively to 2 mg daily, until Tanner breast stage 4 and a mature uterus on ultrasound [27]. The patient was referred to reproductive medicine and, after counselling, oocyte cryopreservation was not proposed because AMH was very low [27]. Bone mineral density was not assessed, and the team planned to start surveillance for CRC and cervical cancer from 18 years, with advice from a specialist at Leiden University Medical Center, in the Netherlands [27].

10.2. Critical reading of the case

The case brings together the elements that should lead to the diagnosis: delayed puberty that is not constitutional (raised gonadotrophins, undetectable AMH), negative autoimmunity, normal karyotype and FMR1, and uterine and ovarian dysgenesis on imaging [27]. It also shows that the disease exists outside consanguineous families and that stature may be normal [27]. Some approaches differ from current recommendations. ESHRE recommends oestradiol for induction, with the transdermal route as the preferable option and a low starting dose, and the Turner regimen starts with 0.25 mg orally [10, 43]; the regimen described started with 0.5 mg orally in a 16-year-old adolescent with a bone age of 12 years, which corresponds to the modified regimen allowed in late diagnoses [10]. DXA, recommended at diagnosis by ESHRE and GeneReviews, was not performed [1, 10]. The surveillance plan from 18 years predates GeneReviews, which proposes colonoscopy from 25 years and annual pelvic ultrasound from 10 to 20 years for germ cell tumours [1]. The reference to cervical surveillance is a decision of the team, which the later sources do not include in their recommendations.

10.3. Family with early colorectal cancer

The family re-evaluated by Goldberg and colleagues in 2021 illustrates the oncological component [20]. In a family with POI due to MCM9 c.1483G>T p.(Glu495Ter), a homozygous sister with POI since the age of 15 had CRC at 31 years, and another, also homozygous and with POI since 15, had clear cell carcinoma of the cervix at 37 years; a heterozygous brother had CRC at 35 years, with a microsatellite-stable tumour, and oligoteratoasthenozoospermia, and the heterozygous parents had three polyps each at 66–68 years [20]. The pedigree is shown in figure 1 of the article (https://pmc.ncbi.nlm.nih.gov/articles/PMC8460657/figure/Fig1/) [20]. The family shows that POI may precede cancer by 16 to 22 years, which gives the reproductive clinic an opportunity to initiate surveillance, and that the interpretation of cancer in heterozygotes remains uncertain [1, 20].

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis

The reproductive prognosis is unfavourable: gonadal dysfunction occurs in 88% of biallelic MCM8 carriers and 93% of MCM9 carriers and is usually severe, and natural conception is unlikely [1, 9]. The French PNDS classifies DNA repair and meiosis genes as the group with the worst fertility prognosis among the genetic causes of POI [11]. In men with Sertoli cell-only syndrome, surgical sperm retrieval was unsuccessful in the two cases with a published study [24]. The prognosis of the sequelae of hypo-oestrogenism depends on HRT: in POI in general, absence of HRT is associated with reduced life expectancy, mainly due to cardiovascular disease (strong, ⊕⊕◯◯), and bone density is reduced [10]. Final height depends on the age at onset and adherence to hormone therapy [1].

The oncological prognosis has no cumulative risk estimates. The available data are frequencies in biased series (germ cell tumours in 3/54 biallelic carriers, at 11–15 years; CRC in 6/28 biallelic MCM9 carriers, at 30–60 years) and population odds ratios with wide intervals for polyps and gastric cancer in MCM9 deficiency [9]. In the 2021 review, eight of nine biallelic MCM9 carriers published at the time had had early-onset cancer, which mainly reflects the way the families were identified [20]. In MCM8 deficiency, there is no evidence of increased CRC risk in the population, but follow-up is short [9]. There are no data on mortality or on the effect of surveillance on survival.

11.2. Surveillance table

The oncological intervals are proposals to be considered in shared decision-making; the GeneReviews authors indicate that larger cohorts and longitudinal studies are needed to define them [1]. By comparison, the European guideline for MUTYH-associated polyposis recommends colonoscopy from 18 years, every 1 or 2 years according to polyp burden, and upper endoscopy from 35 years, at an interval of no more than 3 years [12]; the later starting age proposed for MCM9 deficiency is based on the observed onset of CRC between 30 and 60 years [9]. The 2010 British recommendations reproduced in the Oxford handbook provided, for MUTYH-associated polyposis, colonoscopy from 25 years and upper endoscopy from 30 years, with biennial colonoscopy or colectomy and gastroduodenoscopy every 3 to 5 years, which shows how these thresholds have evolved as data have accumulated [56].

11.3. Follow-up and coordination

Follow-up should be coordinated by a team that integrates the reproductive and oncological components, and the surveillance plan should accompany the patient through the transition to adult care [1]. At-risk siblings should be tested early and, if affected, enter the same plan [1]. New data, such as the inclusion of MCM8 and MCM9 in polyposis and cancer panels, may change the recommendations, and the plan should be revised when cumulative risks are published [9].

12. Summary and Final Considerations

MCM8/MCM9 syndromes are autosomal recessive disorders caused by loss of function of a DNA helicase involved in replication, meiotic and mitotic homologous recombination and mismatch repair [1]. The constant manifestation is hypergonadotrophic hypogonadism, which in women presents as POI with delayed or absent puberty, primary or secondary amenorrhoea and uterine and ovarian dysgenesis, and in men as non-obstructive azoospermia, sometimes with Sertoli cell-only syndrome and normal puberty [1, 29]. The two entities share a risk of germ cell tumours in childhood and adolescence in girls, and MCM9 deficiency is associated with colorectal polyposis of mixed histology, CRC between 30 and 60 years and gastric cancer; in MCM8 deficiency, the digestive risk has not been demonstrated [1, 9].

The diagnosis requires biallelic pathogenic or likely pathogenic variants and is made by panel or exome after aneuploidies and the FMR1 premutation have been excluded; sequencing detects all described variants, but VUS are frequent and do not establish the diagnosis [1, 10]. Treatment is that of POI and male infertility, with pubertal induction from 11 years, HRT with oestradiol at doses of at least 100 µg transdermally or 2 mg orally according to ESHRE (50–100 µg according to the French PNDS) and a progestogen in women with a uterus until the usual age of menopause, andrological management and avoidance of testosterone monotherapy in men who wish to have children, and early referral to reproductive medicine [1, 10, 11]. The proposed cancer surveillance, based on expert opinion, includes annual pelvic ultrasound from 10 to 20 years in girls and, in MCM9 deficiency, colonoscopy every 2 years from 25 years with upper gastrointestinal endoscopy and H. pylori testing [1].

Three limitations constrain practice. The series were identified through infertility, which overestimates hypogonadism and underestimates cancer; there are no cumulative cancer risks or data on the benefit of surveillance; and the role of heterozygotes, who may have reduced fertility, remains to be clarified [1, 9]. Including MCM8, MCM9 and HROB in panels for POI, male infertility, polyposis and early digestive cancer is the condition for these gaps to narrow [9, 22].

13. References

1. Helderman NC, Ben Yahia S, Rajkovic A, Goldberg Y, Nielsen M. MCM8/MCM9-Related Syndromes. 2026 Jun 16. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK623157/
2. Online Mendelian Inheritance in Man (OMIM). *608187 Minichromosome maintenance complex component 8; MCM8. Baltimore: Johns Hopkins University; last edited 1 Jul 2016. Available from: https://www.omim.org/entry/608187
3. Online Mendelian Inheritance in Man (OMIM). *610098 Minichromosome maintenance complex component 9; MCM9. Baltimore: Johns Hopkins University; last edited 27 May 2025. Available from: https://www.omim.org/entry/610098
4. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348. GRCh38 REST query on 6 Nov 2026: https://rest.variantvalidator.org
5. Online Mendelian Inheritance in Man (OMIM). #612885 Premature ovarian failure 10; POF10. Baltimore: Johns Hopkins University; last edited 12 Dec 2023. Available from: https://www.omim.org/entry/612885
6. Online Mendelian Inheritance in Man (OMIM). #616185 Ovarian dysgenesis 4; ODG4. Baltimore: Johns Hopkins University; last edited 9 Jun 2025. Available from: https://www.omim.org/entry/616185
7. Orphanet. 46,XX ovarian dysgenesis-short stature syndrome (ORPHA:444048). Available from: https://www.orpha.net/en/disease/detail/444048
8. ClinGen. Gene-disease validity: MCM9 – MCM9-related gametogenic failure (MONDO:1060226), autosomal recessive, Definitive. Male Infertility Gene Curation Expert Panel; 1 Apr 2026. Available from: https://search.clinicalgenome.org/kb/gene-validity/CGGV:assertion_50750869-1d3c-4833-bac8-045122b16610-2026-04-01T160000.000Z
9. Helderman NC, Yang T, Palles C, Terlouw D, Mei H, Vorderman RHP, et al. Clinical syndromes linked to biallelic germline variants in MCM8 and MCM9. HGG Adv. 2025;6(4):100480. doi:10.1016/j.xhgg.2025.100480
10. ESHRE, ASRM, CREWHIRL and IMS Guideline Group on POI; Panay N, Anderson RA, Bennie A, Cedars M, Davies M, Ee C, et al. Evidence-based guideline: premature ovarian insufficiency. Hum Reprod Open. 2024;2024(4):hoae065. doi:10.1093/hropen/hoae065
11. Haute Autorité de Santé; Centre de Référence des Maladies Endocriniennes Rares de la Croissance et du Développement (CRMERCD). Protocole National de Diagnostic et de Soins (PNDS): Insuffisance ovarienne prématurée. Saint-Denis: HAS; mars 2021 (published 6 May 2021). Available from: https://www.has-sante.fr/jcms/p_3264990
12. Zaffaroni G, Mannucci A, Koskenvuo L, de Lacy B, Maffioli A, Bisseling T, et al. Updated European guidelines for clinical management of familial adenomatous polyposis (FAP), MUTYH-associated polyposis (MAP), gastric adenocarcinoma, proximal polyposis of the stomach (GAPPS) and other rare adenomatous polyposis syndromes: a joint EHTG-ESCP revision. Br J Surg. 2024;111(5):znae070. doi:10.1093/bjs/znae070
13. European Reference Network on Rare Endocrine Conditions (Endo-ERN). Available from: https://endo-ern.eu
14. European Reference Network on Genetic Tumour Risk Syndromes (ERN GENTURIS). Available from: https://www.genturis.eu
15. Wood-Trageser MA, Gurbuz F, Yatsenko SA, Jeffries EP, Kotan LD, Surti U, et al. MCM9 mutations are associated with ovarian failure, short stature, and chromosomal instability. Am J Hum Genet. 2014;95(6):754-762. doi:10.1016/j.ajhg.2014.11.002
16. AlAsiri S, Basit S, Wood-Trageser MA, Yatsenko SA, Jeffries EP, Surti U, et al. Exome sequencing reveals MCM8 mutation underlies ovarian failure and chromosomal instability. J Clin Invest. 2015;125(1):258-262. doi:10.1172/JCI78473
17. Tenenbaum-Rakover Y, Weinberg-Shukron A, Renbaum P, Lobel O, Eideh H, Gulsuner S, et al. Minichromosome maintenance complex component 8 (MCM8) gene mutations result in primary gonadal failure. J Med Genet. 2015;52(6):391-399. doi:10.1136/jmedgenet-2014-102921
18. Fauchereau F, Shalev S, Chervinsky E, Beck-Fruchter R, Legois B, Fellous M, et al. A non-sense MCM9 mutation in a familial case of primary ovarian insufficiency. Clin Genet. 2016;89(5):603-607. doi:10.1111/cge.12736
19. Goldberg Y, Halpern N, Hubert A, Adler SN, Cohen S, Plesser-Duvdevani M, et al. Mutated MCM9 is associated with predisposition to hereditary mixed polyposis and colorectal cancer in addition to primary ovarian failure. Cancer Genet. 2015;208(12):621-624. doi:10.1016/j.cancergen.2015.10.001
20. Goldberg Y, Aleme O, Peled-Perets L, Castellvi-Bel S, Nielsen M, Shalev SA. MCM9 is associated with germline predisposition to early-onset cancer—clinical evidence. NPJ Genom Med. 2021;6(1):78. doi:10.1038/s41525-021-00242-4
21. Golubicki M, Bonjoch L, Acuña-Ochoa JG, Díaz-Gay M, Muñoz J, Cuatrecasas M, et al. Germline biallelic Mcm8 variants are associated with early-onset Lynch-like syndrome. JCI Insight. 2020;5(18):e140698. doi:10.1172/jci.insight.140698
22. Helderman NC, Tops CM, Legebeke J, Yang T, Gay MD, Terlouw D, et al. Biallelic germline inactivation of HROB causes primary gonadal insufficiency and is potentially associated with colonic polyposis predisposition. Am J Med Genet A. 2026;200(8):1773-1782. doi:10.1002/ajmg.a.70128
23. Helderman NC, Terlouw D, Bonjoch L, Golubicki M, Antelo M, Morreau H, et al. Molecular functions of MCM8 and MCM9 and their associated pathologies. iScience. 2023;26(6):106737. doi:10.1016/j.isci.2023.106737
24. Sha X, Zhang X, Geng H, Li Y, Xia X, Li G, et al. MCM9 deficiency impairs DNA damage repair during spermatogenesis, leading to Sertoli cell-only syndrome in humans. Cell Death Discov. 2025;11(1):292. doi:10.1038/s41420-025-02581-y
25. Jiao X, Li Z, Tang Z, Xu J, Lu LY, Liu Y. MCM8-9 helicase activity protects primordial germ cell development to prevent premature ovarian insufficiency. Proc Natl Acad Sci U S A. 2026;123(19):e2535910123. doi:10.1073/pnas.2535910123
26. Guo T, Zheng Y, Li G, Zhao S, Ma J, Qin Y. Novel pathogenic mutations in minichromosome maintenance complex component 9 (MCM9) responsible for premature ovarian insufficiency. Fertil Steril. 2020;113(4):845-852. doi:10.1016/j.fertnstert.2019.11.015
27. Nauwynck E, De Vos M, Gheldof A, Dequeker BJ, Van Der Kelen A, Hes F, et al. MCM9 compound heterozygosity in an adolescent with premature ovarian insufficiency. Endocrinol Diabetes Metab Case Rep. 2024;2024(4):24-0012. doi:10.1530/EDM-24-0012
28. Mishra R, Kumar N, Bargali A, Maich G. Understanding the novel MCM8 gene mutation: primary ovarian insufficiency and uterine hypoplasia in siblings. BMJ Case Rep. 2024;17(7):e259433. doi:10.1136/bcr-2023-259433
29. Potorac I, Laterre M, Malaise O, Nechifor V, Fasquelle C, Colleye O, et al. The role of MCM9 in the etiology of Sertoli cell-only syndrome and premature ovarian insufficiency. J Clin Med. 2023;12(3):990. doi:10.3390/jcm12030990
30. Desai S, Wood-Trageser M, Matic J, Chipkin J, Jiang H, Bachelot A, et al. MCM8 and MCM9 nucleotide variants in women with primary ovarian insufficiency. J Clin Endocrinol Metab. 2017;102(2):576-582. doi:10.1210/jc.2016-2565
31. Dong J, Zou Z, Wang W, Chen L, Ma R, Ge X, et al. A novel homozygous frameshift mutation in MCM8 causes primary gonadal dysgenesis in both genders. J Assist Reprod Genet. 2025;42(4):1197-1204. doi:10.1007/s10815-025-03443-3
32. Wang F, Zang S, Li P, Yin X. MCM8 variants in two patients with primary ovarian insufficiency: clinical findings and in vitro defective DNA repair revealed by an MCM8 double-mutant construct. Front Endocrinol (Lausanne). 2026;17:1807887. doi:10.3389/fendo.2026.1807887
33. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 778–779.
34. McGlacken-Byrne SM, Suntharalingham JP, Ishida M, Buonocore F, Del Valle I, Cameron-Pimblett A, et al. A tiered approach to exome sequencing analysis in early-onset primary ovarian insufficiency. J Clin Endocrinol Metab. 2025;110(11):3142-3154. doi:10.1210/clinem/dgaf124
35. Tsabai P, Kumykova Z, Averkova V, Pavlova N, Maslennikov D, Bolshakova A, et al. Novel variants associated with premature ovarian insufficiency in Russian adolescents. Front Endocrinol (Lausanne). 2025;16:1687148. doi:10.3389/fendo.2025.1687148
36. Ding L, Deng S, Zhang P, Zhang D, Tian Q. Identification of novel variants and candidate genes in women with 46,XX complete gonadal dysgenesis. Reprod Biol Endocrinol. 2024;22(1):140. doi:10.1186/s12958-024-01309-4
37. Rabello BXC, Rocha LAD, Barbosa CP, Bianco B, Christofolini DM. Investigation of an MCM8 gene variant in women with premature ovarian insufficiency. Einstein (Sao Paulo). 2024;22:eAO0712. doi:10.31744/einstein_journal/2024AO0712
38. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 452–453.
39. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 122–123.
40. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 568–573.
41. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 556–561.
42. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 722–723.
43. Gravholt CH, Andersen NH, Christin-Maitre S, Davis SM, Duijnhouwer A, Gawlik A, et al. Clinical practice guidelines for the care of girls and women with Turner syndrome. Eur J Endocrinol. 2024;190(6):G53-G151. doi:10.1093/ejendo/lvae050
44. U.S. Food and Drug Administration. Estradiol transdermal system (Amneal Pharmaceuticals) — prescribing information, sections 1, 2 and 8.4; rev. May 2024. Available from: https://api.fda.gov/drug/label.json
45. INFARMED. Dermestril 25, 50 e 100 microgramas/24 h sistema transdérmico (estradiol). Resumo das Características do Medicamento. Lisboa: INFARMED, I.P.; accessed 2 November 2026. Available from: https://extranet.infarmed.pt/INFOMED-fo/
46. INFARMED. Estrofem 1 mg e 2 mg comprimidos revestidos por película (estradiol). Resumo das Características do Medicamento (approved 25-09-2024). Lisboa: INFARMED, I.P.; accessed 6 November 2026. Available from: https://extranet.infarmed.pt/INFOMED-fo/
47. Agência Nacional de Vigilância Sanitária (ANVISA). Dados abertos: medicamentos registados (DADOS_ABERTOS_MEDICAMENTOS.csv), filtered for estradiol and progesterone with active registration. Brasília: ANVISA; accessed 6 November 2026. Available from: https://dados.anvisa.gov.br/dados/DADOS_ABERTOS_MEDICAMENTOS.csv
48. INFARMED. Utrogestan 100 mg cápsulas moles (progesterona). Resumo das Características do Medicamento. Lisboa: INFARMED, I.P.; accessed 6 November 2026. Available from: https://extranet.infarmed.pt/INFOMED-fo/
49. U.S. Food and Drug Administration. PROMETRIUM (progesterone, USP) capsules — prescribing information; rev. Jul 2026. Available from: https://api.fda.gov/drug/label.json?search=openfda.brand_name:%22PROMETRIUM%22
50. Schlegel PN, Sigman M, Collura B, De Jonge CJ, Eisenberg ML, Lamb DJ, et al. Diagnosis and treatment of infertility in men: AUA/ASRM guideline part II. Fertil Steril. 2021;115(1):62-69. doi:10.1016/j.fertnstert.2020.11.016
51. Malfertheiner P, Megraud F, Rokkas T, Gisbert JP, Liou JM, Schulz C, et al. Management of Helicobacter pylori infection: the Maastricht VI/Florence consensus report. Gut. 2022;71(9):1724-1762. doi:10.1136/gutjnl-2022-327745. PMID: 35944925
52. Chey WD, Howden CW, Moss SF, Morgan DR, Greer KB, Grover S, et al. ACG Clinical Guideline: Treatment of Helicobacter pylori Infection. Am J Gastroenterol. 2024;119(9):1730-1753. doi:10.14309/ajg.0000000000002968. PMID: 39626064
53. H2-Pharma, LLC. PYLERA (bismuth subcitrate potassium, metronidazole, and tetracycline hydrochloride) capsules: prescribing information. NDA050786. DailyMed setid 5cdacbf5-67be-42eb-bd54-8acc32ea1abc; version of 2026 Jul 1 [cited 2026 Nov 7]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=5cdacbf5-67be-42eb-bd54-8acc32ea1abc
54. INFARMED. Pylera 140 mg + 125 mg + 125 mg cápsulas: resumo das características do medicamento. Approved 2022 Dec 28. Infomed [cited 2026 Nov 7]. Available from: https://extranet.infarmed.pt/INFOMED-fo/
55. Casper Pharma LLC. HELIDAC Therapy (bismuth subsalicylate chewable tablets, metronidazole tablets, tetracycline hydrochloride capsules): prescribing information. NDA050719, supplement 026, 2025 Mar 31. Drugs@FDA [cited 2026 Nov 7]. Available from: https://www.accessdata.fda.gov/drugsatfda_docs/label/2025/050719s026lbl.pdf
56. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 857–862.