Gonadal tumours and DSD

Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

Contents lists available at ScienceDirect

Best Practice & Research Clinical Endocrinology & Metabolism journal homepage: www.elsevier.com/locate/beem

9

Gonadal tumours and DSD Leendert H.J. Looijenga, PhD a, *, Remko Hersmus, BSc, PhD student a, Bertie H.C.G.M. de Leeuw, PhD a, Hans Stoop, BSc, PhD student a, Martine Cools, MD, PhD b, J. Wolter Oosterhuis, MD, PhD a, Stenvert L.S. Drop, MD, PhD c, Katja P. Wolffenbuttel, MD d a

Department of Pathology, Erasmus MC-University Medical Center Rotterdam, Josephine Nefkens Institute, Daniel den Hoed Cancer Center, Rotterdam, The Netherlands b Department of Pediatric Endocrinology, University Hospital Ghent, Belgium c Department of Pediatric Endocrinology, Erasmus MC-University Medical Center Rotterdam, Sophia, Rotterdam, The Netherlands d Department of Pediatric Urology, Erasmus MC-University Medical Center Rotterdam, Sophia, Rotterdam, The Netherlands

Keywords: disorders of sex development (DSD) undervirilisation gonadal dysgenesis germ cell tumours (GCTs) seminoma non-seminoma carcinoma in situ (CIS) gonadoblastoma (GB) maturation delay testicularisation OCT3/4 testis-specific protein on the Y chromosome (TSPY) stem cell factor (SCF)

Disorders of sex development (DSD), previously referred to as intersex, has been recognised as one of the main risk factors for development of type II germ cell tumours (GCTs), that is, seminomas/dysgerminomas and non-seminomas (e.g., embryonal carcinoma, yolk sac tumour, choriocarcinoma and teratoma). Within the testis, this type of cancer is the most frequent malignancy in adolescent and young adult Caucasian males. Although these males are not known to have dysgenetic gonads, the similarities in the resulting tumours suggest a common aetiological mechanism(s), –genetically, environmentally or a combination of both. Within the group of DSD patients, being in fact congenital conditions, the risk of malignant transformation of germ cells is highly heterogeneous, depending on a number of parameters, some of which have only recently been identified. Understanding of these recent insights will stimulate further research, with the final aim to develop an informative clinical decision tree for DSD patients, which includes optimal (early) diagnosis without overtreatment, such as prophylactic gonadectomy in the case of a low tumour risk. Ó 2009 Elsevier Ltd. All rights reserved.

* Corresponding author. Department of Pathology, Erasmus MC-University Medical Center Rotterdam, Josephine Nefkens Institute, Building Be, Room 430b, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands. Tel.: þ31 10 70 44329; Fax: þ31 10 70 44365. E-mail address: [email protected] (L.H.J. Looijenga). 1521-690X/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.beem.2009.10.002

292

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

Human germ cell tumours (GCTs) comprise a heterogeneous group of neoplasms with different pathogenesis and clinical behaviour.1 Recognition of the existence of the various subtypes is both scientifically and clinically relevant. The subtypes are characterised by a set of defined parameters, most of them related either directly or indirectly to their different origin, that is, maturation stage of the germ cell, and the associated pathobiology. This knowledge is, on the one hand, informative to investigate the pathogenesis of these types of tumours, and, on the other hand, of value to identify individuals who are at increased risk to develop such a malignancy in comparison with the general population. In fact, recent insights can be used for further development of individualised treatment protocols with the aim of optimal clinical management, both related to early diagnosis of the tumour and efficient treatment, with emphasis on prevention of both under- and overtreatment (either (prophylactic) gonadectomy, irradiation and/or chemotherapy). Proper decision making is of utmost importance because of the general young age of the patients at clinical diagnosis and the curability of the disease. This article specifically focusses on patients with disorders of sex development (DSD), of which recent observations indicate that heterogeneity exists between the various subgroups of patients regarding the risk for malignant transformation of germ cells, leading to the so-called type II GCTs. Application of the knowledge on predictive parameters will allow a better pre-selection of patients into tumour-risk groups (i.e., high, intermediate and low), and subsequent application of the most appropriate treatment modalities. It must be borne in mind that gonadectomy at an early age will preclude spontaneous puberty and fertility and necessitate lifelong hormonal supplementation. To facilitate an evidence-based decision regarding the various possible treatment approaches, understanding the biology of normal germ cells, gonadal development and the pathobiology of the type II GCTs is of great importance. Therefore, in this article, the first two sections describe many relevant aspects of normal embryonic germ cell development, both pre-gonadal and gonadal. Subsequently, the type II GCTs are introduced, including their precursor lesions, followed by a description of the parameters related to tumour risk. In this context, the most appropriate markers for (early) diagnosis are discussed. Finally, suggestions for future studies are made with the aim to further improve assessment of tumour risk in DSD with minimal burden to the patients. Physiological pre-gonadal germ cell development During human intra-uterine development, embryonic germ cells can be identified at weeks 5–6 gestational age, referred to as primordial germ cells (PGCs).2 These germ cells are unique and characterised by several markers, such as alkaline phosphatase (AP), VASA, c-KIT and OCT3/4 (also known as POU5F1). PGCs are progenitors of the germ cell lineage, resulting in spermatozoa in males and oocytes in females in later life, capable of transmitting genetic information to the next generation.3 To fulfil this special task, they have unique characteristics (recently discussed elsewhere4 and references cited therein). PGCs begin migrating from the proximal epiblast through the hindgut and mesentery to the genital ridge (Fig. 1), for which the stem cell factor (SCF) – c-KIT pathway – is crucial.5 PGCs express the receptor, while the SCF functions as a chemo-attractant as well as survival factor.6–8. Disturbances in the function of the c-KIT pathway, depending on the ligand stem SCF, results in various anomalies, including sub-infertility or infertility. Recently, Sox17 has been identified to be crucial for proper migration of PGCs as well.9 Physiological gonadal germ cell development On reaching the genital ridge, PGCs are called gonocytes (Fig. 1) independent of the chromosomal constitution. Their original bi-parental pattern of genomic imprinting is completely erased. This epigenetic modification, demonstrated by the bi-allelic expression of the imprinted genes (see Ref. 10 and 11 for review), is required to allow proper development of the gender-specific germ cell lineage. The fate of the gonocytes is determined by the microenvironment, referred to as gonadal sex, (i.e., development of either testis or ovary; see Fig. 2). In this article, the term ‘testicularisation’ will henceforth be used for the active process of testis formation, in the broadest context. Development of gonadal sex is initiated by the chromosomal sex, that is, XY (male) versus XX (female) of the gonadal stromal compartment, which is established during fertilisation. Testis development, in fact, depends on

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

293

Fig. 1. Representative H & E stained (left panels) whole mount sections of a human male embryo at developmental stage week 10. Parallel OCT3/4 stained slide is shown in the right panels. Note the presence of OCT3/4 positive cells in the immature testis (black boxes; indicated by arrowheads), being gonocytes, as well during migration (red box).

expression of a single transcription factor in stromal gonadal cells, known as SRY (Ref. 12, for review; Fig. 2, right panel). If sufficient SRY protein is present, both spacially and temporally, it induces expression of the transcription factor SOX9, finally resulting in formation of Sertoli cells13, eventually leading to the male phenotypic sex later in life (related to functionality of Leydig cells, being the source of testosterone, amongst others) (see Ref. 14 for review). If no functional SRY is present, stromal cells will follow the female pathway and become granulosa cells (see Fig. 2, left panel). Like in the male, the female developmental pathway is also dependent on a number of gene products, including FOXL2 (Ref. 15, for review). Based on the formation of either the male or female microenvironment (specifically, presence of Sertoli or granulosa cells), the gonocytes will mature to either pre-spermatogonia or oogonia. As part of this process, they will gradually lose expression of the embryonic markers (i.e., AP, OCT3/4).16–21 In the testis, they are substituted by others, such as MAGE-4A.19,20 However, the embryonic markers can still be present during the first months of the first postnatal year in males, specifically in the germ cells located within the luminal space of the seminiferous tubules, indicating an immature status. Moreover, c-KIT is still detectable at a low level in human spermatogonia (see Ref. 22, for review). Further development of germ cells is beyond the scope of this article and will not be discussed here. Human GCTs: Introduction Traditionally, GCTs are classified based on their histological appearance, as judged by the pathologist.23–26 A more pathobiologically based classification system has been proposed1, which provides an

294

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

Fig. 2. Schematic representation of normal gonadal development, including maturation of germ cells (upper panel) and pathogenesis of type II GCTs (lower panel). Note the initial presence of a bipotential gonad (formed under control of a number of gene products, including WT1, DMRT1, and SF1), to which the primordial germ cells (PGCs) migrate from the epiblast (yolk sac). The PGCs (now referred to as gonocytes) are characterised by a series of markers, including OCT3/4, NANOG, c–KIT etc., as well as their demethylated DNA and erased pattern of genomic imprinting (GI). Depending on the chromosomal constitution (i.e., presence of a Y chromosome, specifically the SRY gene, right panel of the figure), SOX9 will be formed, resulting in generation of Sertoli cells. This results in maturation of the gonocytes to (pre-)spermatogonia, which lose the embryonic markers, and express others, like MAGE4A and TSPY, and at the same time migrate from the centre to the basal lamina of the tubules. This continuing process of testis formation is referred to as testicularisation, resulting eventually to the generation of mature spermatozoa with a male pattern of GI after puberty. Based on the embryonic formation of Leydig cells, virilisation will occur, as a result of synthesis of testosterone. In the absence of a Y chromosome, under control of other proteins (like DAX1 and FOXL2), granulosa cells will be formed (being the female counterpart of Sertoli cells) (left panel of the figure). However, maturation of gonocytes to oogonia (which will go into meiosis I arrest) is autonomous, and therefore independent of the presence of granulosa cells. However, this is again associated with loss of the embryonic markers. Theca cells will be formed, being the female counterpart of Leydig cells, forming oestrogen and progesteron (Pr). The timing of male and female germ cell maturation compared to birth (B) and puberty (P) is different. The type II GCTs (lower panel of the figure) can only be formed in case of retention of PGC/gonocyte-like cells, in which delayed maturation is a risk factor. This can be due to various reasons, both genetic as well as environmental, both in the context of Disorders of Sex Development (DSD: hypovirilisation and gonadal dysgenesis) and Testicular Dysgenesis Syndrome (TDS: cryptorchidism, subfertility, and genetic predisposition). This can result in the formation of either gonadoblastoma (GB) in case of low level of testicularisation and carcinoma in situ (CIS) of the testis. These precursor lesions will only be formed when part of the Y chromosome is presence in the germ cells, being the Gonadoblastoma on the Y chromosome region (GBY), of which TSPY is the most interesting candidate so far. In addition, the presence of stem cell factor (SCF) is characteristic for these precursor lesion, as it is absent in germ cells showing delayed maturation. The GB and CIS lesions will progress to an invasive type II GCTs, either with a seminomatous of non-seminomatous histology. In the testis this occurs after puberty, while in patients with DSD this can happen at early age. In this transition to invasiveness, gain of the short arm of chromosome 12 is always found. The histological composition can be heterogeneous, but a seminomatous histology (seminoma in the testis and dysgerminoma in the ovary/dysgenetic gonad) is predominantly found when the gonad is located intra-abdominally.

insight into the pathogenesis of the various tumour types. The site of presentation of the primary tumour, age of the patient at diagnosis, histological composition and chromosomal constitution are informative parameters. Based on these criteria, five categories (I–V) of GCTs are identified.1 Because only the type II GCTs are of relevance in the context of DSD, only this type is discussed here. The other types of GCTs are discussed elsewhere.1,27–29

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

295

Type II GCTs and precursor lesions Human type II GCTs are different from other solid cancers of adults in several aspects, both regarding pathobiology and clinical behaviour.1 This is likely related to their embryonic origin, either a PGC or gonocyte. Histologically and clinically, these tumours are dichotomised into seminomatous and non-seminomatous types. The seminomatous tumours, referred to as seminoma of the testis, dysgerminoma of the ovary and dysgenetic gonad and germinoma of the brain, morphologically and immunohistochemically mimic the PGCs/gonocytes. Within the group of non-seminomatous tumours, many differentiated variants can be identified, being teratoma (somatic differentiation), yolk sac tumour and choriocarcinoma (extra-embryonic differentiation), including the germ cell lineage and their stem cell component, embryonal carcinoma. Tumours containing both seminomatous and nonseminomatous components are referred to as combined tumour, according to the British classification30, and as non-seminoma in the World Health Organization (WHO) classification.31 It is proposed that the origin of type II GCTs also explains their overall sensitivity to DNA-damaging agents (i.e., irradiation and cisplatin-based chemotherapy)32, supported by the fact that this is influenced by the histological composition of the tumour.33 The well-accepted precursor of type II GCTs of the adult testis is the so-called carcinoma in situ of the testis (CIS)34, also referred to as intratubular germ cell neoplasia unclassified (IGCNU)31 and testicular intra-epithelial neoplasia (TIN).35 A representative example is shown in Fig. 3. Although better and more accurate terms (i.e., CIS is not a carcinoma) can be applied, they have not been accepted to date. Therefore, the term CIS/IGCNU will be used here. In contrast to the situation in the testis, the precursor lesions for the ovarian (without dysgenesis), mediastinal and intracranial type II GCTs have not been identified so far, although they are expected to be similar to a certain extent.36,37 The CIS/ IGCNU counterpart of the dysgenetic gonads (with a low level of virilisation) is known as

Fig. 3. Representative parallel slides of a testicular parenchyma sample of a patient with a type II GCT stained with H & E, as well as various markers specific for normal germ cells (SSX and VASA), as well as CIS/IGCNU (PLAP, c-KIT, and OCT3/4) (left panel of the figure). All images represent both CIS/IGCNU (left upper panel) and spermatogenesis (lower right panel). For comparison, OCT3/4 staining of a testicular biopsy of a patient with delayed maturation of germ cells is included (right upper panel), showing positively stained germ cells in the luminal position within the seminiferous tubules. These cells are not CIS/IGCNU cells. Note that the germ cell positioned under the tight junction of the Sertoli cells, on the basal lamina is negative for OCT3/4 (indicated by an arrow).

296

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

gonadoblastoma (GB)38, of which a representative example is shown in Fig. 4. This lesion also contains germ cells showing the same characteristics as CIS/IGCNU cells, but accompanied by stromal cells that have not differentiated into Sertoli cells39–41, as reviewed elsewhere.42–44 The precursor of GB is known as undifferentiated gonadal tissue (UGT)41 (see below). The consistent bi-allelic expression of imprinted genes in invasive type II GCTs, as well as CIS/IGCNU, is in agreement with the origin from an erased embryonic germ cell.45–50 In addition, high-throughput expression profiling shows that CIS/ IGCNU cells show strong overlap with embryonic stem cells regarding expression profile.51 The hypothesised initiation of this tumour during embryogenesis is in line with the epidemiological observation of the dip in the incidence of this type of cancer in males conceived during World War II52,53, as well as other risk factors (discussed elsewhere54 and below). The alternative model is that type II GCTs originate from a pachytene spermatocyte (in the testis).55 Possibly the most convincing argument against this latter model is the fact that patients with various forms of DSD, most of whom will never develop proper spermatogenesis, not even spermatogonia, have an increased risk for this type of cancer. Therefore, it can be concluded that the cell of origin of type II GCTs is a germ cell blocked in the PGC/gonocyte stage. This also explains why similar tumours can be found in the ovary, as well as at extragonadal sites, like in patients with Klinefelter syndrome (see below). Risk factors, integrated in the testicular dysgenesis syndrome (TDS) Besides the aforementioned risk factor of DSD, several other risk factors have been identified for type II GCTs, predominantly of the testis, being cryptorchidism, in(sub)fertility, familial predisposition,

Fig. 4. Representative images of a gonadoblastoma (upper panel), including macroscopical appearance, characterised by the presence of calcifications also demonstrated by ultrasound analysis (lower left image). The histological representation (showing the presence of both germ cells and granulosa like cells) is shown based on H & E staining, as well as immunohistochemical detection of PLAP and OCT3/4. In addition, the similarities between CIS/IGCNU and GB are illustrated using staining for OCT3/4 and TSPY (lower left panel of the figure). Note that the neoplastic germ cells are positive for both markers. While the Sertoli cells in the testis are SOX9 positive, the granulosa cells in GB are FOXL2 positive. This is also found in case of a combined presence of both CIS/IGCNU and GB in a single gonad (lower right panel of the figure), even within a single tubule-like structure.

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

297

birth weight and birth order.56–63 These again support the model that the initiating step in the pathogenesis of this cancer occurs during embryonal development. In addition, it has been shown that an early age of orchiopexy indeed reduces the risk for a testicular type II GCT.64,65 This is likely related to the still ongoing maturation of PGCs/gonocytes to pre-spermatogonia (see above). Therefore, surgical intervention in DSD patients might also reduce the risk for tumour development in case of a cryptorchid testis. Till most recently, no gene or genes involved in familial type II GCTs had been identified yet, in spite of the linkage to the long arm of chromosome X.66,67 However, highly interesting data have recently been reported, showing a role of variants of the SCF.68,69 The associated variants are more frequently present in the Caucasian than in the blacks and Asian populations, well in parallel with the epidemiological distribution of this type of cancer. This observation is of specific relevance because of the finding that immunohistochemical detection of SCF has diagnostic value for identification of the precursor lesions of these tumours, especially in DSD patients (see below). The complexity of the role of genetic and possibly environmental factors (see below) is well illustrated by the observation that immigrants from Finland to Sweden, who have a lower initial risk for type II GCTs of the testis, obtain the risk of the Swedish population at the second generation.70 This demonstrates a significant effect of the environment as well on the incidence during a limited period of time. In this context, it is relevant that a recent meta-analysis demonstrated that both low and high birth weight increase the risk for type II testicular GCTs.71 In addition, male trisomy 21 patients have an increased risk72, and a delayed maturation of germ cells has been identified, which has been thought to be related to the pathogenesis of this cancer (see below). Interestingly, most, if not all, identified risk factors have in common that they in one way or the other negatively affect the maturation of embryonic germ cells. These factors have been brought together into the TDS.73–77 This model integrates various elements, in which the final outcome will have a negative effect on testicular function, including sub(in)fertility, cryptorchidism and/or an increased risk for development of a type II GCT. In this model the role of the supportive elements, that is, the Leydig and Sertoli cells, is crucial. Understanding of the mechanisms involved will also help to further elucidate the processes related to the development of malignancy in patients with DSD. Type II GCTs and DSD The group of developmental anomalies, referred to as DSD, previously referred to as intersex, is defined as conditions of incomplete or disordered genital or gonadal development leading to a discordance between genetic sex (i.e., determined by the chromosomal constitution, of the X and Y chromosomes), gonadal sex (the testicular or ovarian development of the gonad) and phenotypic sex (the physical appearance of the individual). Recently, a revised classification system has been proposed, with the aim to reduce uncertainties on description.78,79 The classification system is primarily based on the karyotype of the patient, resulting in three entities: (1) sex chromosomal DSD, (2) 46,XY DSD and (3) 46,XX DSD. The first group includes all patients with numerical sex chromosomal anomalies, such as 47,XXY (Klinefelter syndrome and variants), 45,X (Turner syndrome and variants), 45,X/46,XY (mixed gonadal dysgenesis) and 46,XX/46,XY (chimaerism). The 46,XY DSD group include all patients with (A) disorders of gonadal (testicular) development, including (1) complete and partial gonadal dysgenesis (due to mutation of SRY, SOX9, SF1, WT1, etc), (2) ovotesticular DSD (Fig. 5) and (3) testis regression; (B) disorders in androgen synthesis or action, including (1) disorders of androgen synthesis, (2) action and (C) others. The 46,XX DSD group includes all patients with (A) disorders of gonadal (ovarian) development, including (1) gonadal dysgenesis, (2) ovotesticular DSD and (3) testicular DSD (i.e., SRY presence, duplication SOX9 and mutation RSPO1); (B) androgen excess, including (1) foetal, (2) foetaplacental, (3) maternal and (C) others. Patients with the same phenotypical characteristics can be part of a different category due to the karyotype used for the primary sub-classification. In addition, patients might also switch from category, for example, based on additional information gathered, such as presence of mosaicisms. This might be a limitation in some cases. However, the consensus report must be interpreted as a living document, which at least allows a better description of the various types of disorders than the previous nomenclatures. It, therefore, allows a more straightforward sub-classification and correlative analyses, including investigation of the risk for malignant transformation of

298

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

Fig. 5. Representative H & E stained slide of an ovotestis, showing testicular development on the right side, and ovarian development on the left.

germ cells. In summary, only the DSD patients with hypovirilisation or gonadal dysgenesis may have an increased risk for development of type II GCTs, although in a highly heterogeneous pattern. This is found to be related to several specific parameters, as discussed below. With a single exception, to be presented first, all these patients are at risk for gonadal GCTs. Klinefelter syndrome; occurrence of mediastinal type II GCTs Klinefelter syndrome patients (47,XXY), a variant of sex chromosomal DSD, have no increased risk of type II GCTs of the testis, but rather of the mediastinum.80–85 The absence of gonadal (testicular) GCTs in these patients is likely due to induction of apoptosis of germ cells related to an improper microenvironment.86 The resulting pituitary/gonadal overstimulation may play a role in the formation of the mediastinal tumour. Indications of apoptosis of gonadal germ cells have also been reported in the ovary of Turner syndrome patients (45,XO), another variant of sex chromosomal DSD87, as well as patients with complete androgen insensitivity, a variant of 46,XY DSD88 (see below). Parameters related to gonadal type II GCT risk in DSD Introduction DSD patients with hypervirilisation (various forms of 46,XX DSD with androgen excess in the new classification) (see above) do not have an increased risk for GCTs compared with the general population. However, this is completely different for DSD patients with either hypovirilisation or gonadal dysgenesis. These can in fact be part of all three categories in the new classification: sex chromosomal DSD, 46,XY DSD and 46,XX DSD. Several reviews on this topic have been published recently (see Ref. 42–44,89, for review). Therefore, only the most important issues are summarised here. Furthermore, based on additional studies, not included in reviews so far, new information is presented here, including a number of case reports, showing the impact of knowledge related to DSD and tumour risk. Gonadal localisation The anatomical position of the gonad is an informative predictive parameter for development of gonadal type II GCTs. This is in line with the fact that cryptorchidism is one of the strongest risk factors for type II GCTs of the testis in the general (Caucasian) population.90–93 Indeed, within DSD patients,

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

299

the risk for tumour development is higher within the same patient category when the gonad is abdominal rather than scrotal (see Ref. 42–44,89, for review). This is only the case if the other requirements for malignant transformation of germ cells are fulfilled, including the presence of the socalled GBY region (see below). Interestingly, it has been demonstrated that a seminomatous histology of the tumour is more frequently found in intra-abdominal testes than in testes localised in the scrotum.94 This likely also explains the preferential occurrence of dysgerminomas in the ovary as well as dysgenetic gonad, which are abdominally located.95–99 In view of these observations, it comes as no surprise that seminoma of the testis and dysgerminoma of the dysgenetic gonad/ovary are similar regarding morphology, immunohistochemical characteristics, chromosomal constitution as well as gene expression profile (Fig. 6). Precursor lesion depending on gonadal virilisation status As indicated above, the precursor of type II gonadal GCTs can be either CIS/IGCNU or GB, or a combination of both (Ref. 100 and unpublished observations). This is in fact found to be related to the level of testicularisation of the gonad: CIS in the testis and GB in the dysgenetic (undifferentiated) gonad, or a mixture of both. This can be well demonstrated by the use of immunohistochemistry for SOX9 (read-out of SRY function and Sertoli cell differentiation) and FOXL2 (granulosa cell differentiation).101 Representative examples are shown in Fig. 4 of a combined presence of CIS/IGCNU and GB within a single gonad, even within a single tubule-like structure. These results indicate that CIS/ IGCNU and GB are part of a histological continuum, as originally proposed39, in which the development into either Sertoli – or granulosa – cell direction determines the histological context of the pre-malignant cells.44 The precursor lesion of GB has been identified, being UGT41, allowing a better histological description of the gonadal tissue available, either a biopsy or gonadectomy specimen. Again, PGC/gonocyte-like pre-malignant germ cells are present in these UGT lesions, being OCT3/4 (AP and c-KIT) positive. In addition, the stromal cells show expression of predominantly FOXL2 compared to SOX9.

Fig. 6. Representative illustrations of immunohistochemistry for OCT3/4 on a testicular seminoma (left upper panel) and a dysgerminoma of a dysgenetic gonad (left lower panel). Illustration of the presence of numerical chromosomal changes of the matched tumors as investigated using array CGH (middle panels). Gain is represented by measure points above the zero line, and loss below. Note the presence of additional copies of the short arm of chromosome 12 in both tumours (indicated by a red arrow). Expression profiling of all the genes within the genome is performed using Affymetrix arrays (right panel), showing that the seminomas (SE) and dysgerminomas (DG) can not be separated from each other. In other words, regarding morphology, immunohistochemical characteristics, chromosomal constitution, as well as expression profile, seminomas of the testis and dysgerminomas of the ovary/ dysgenetic gonads, must be considered as the same type of cancer.

300

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

Involvement of the Y chromosome The risk of development of a type II GCT in DSD patients is directly related to presence of a specific part of the Y chromosome, known as the gonadoblastoma region of the Y chromosome (GBY).102 This area maps around the centromeric region and excludes the SRY gene as candidate. This is indeed supported by the clinical observation that patients with a translocation of the SRY gene to an X chromosome or an autosome, resulting in 46,XX males, have no increased risk for this type of cancer. Although SRY is not the gene of interest in this context, knowledge of its function is relevant (see below). The recent report on the predictive value of the presence of SRY for GB in 45,X (Turner) patients is likely related to the fact that all the patients are mosaic and also possess the GBY region.103 The first downstream target of SRY is the transcription factor SOX9, which in the testis is Sertoli cell specific (see above), initiating the process of testicularisation (Fig. 2). A suboptimal presence of SOX9 is related to the risk for type II GCTs, as recently reported.104 Because this case well illustrates the value of knowledge of the pathogenesis of type II GCTs, in particular, the possible histological presentations and the need of a specialised multidisciplinary team, this is summarised below. The full information is published elsewhere.104 This is followed by the role of the Y chromosome, that is, TSPY. Case report I A 26-year-old phenotypical female patient with a history of treatment-resistant irregular menstrual cycles, after menarche at the age of 14 years, presented with abdominal pain on the right side, initially diagnosed as irritable bowel syndrome. Echographic examination demonstrated an enlarged ovary, suspected to be a dermoid cyst, treated by salpingo-ophorectomy. Histological examination revealed an exceptionally large GB, positive for OCT3/4, TSPY and SCF, and with FOXL2-positive supporting cells, in the absence of SOX9. In addition, a small dysgerminoma and non-dysgerminoma component were identified. The presence of a GB, staining positive for TSPY, initiated discussion about the karyotype of the patient. Fluorescent in situ hybridisation on the tumour tissue and karyotyping of peripheral lymphocytes demonstrated the presence of an XY chromosomal constitution. After extensive discussion with the patient, the remaining left gonad was surgically removed, presenting as a streak, morphologically, as well as functionally. Detailed histological examination showed a small GB, confirmed by immunohistochemistry. Direct sequence analysis of the SRY gene showed a single nucleotide change at position 209 (G to T), resulting in a missense mutation (tryptophan (W) to leucine (L) amino acid change) at position 70 in the SRY protein, being an unknown mutation so far. This mutation is located within the N-terminal NLS sequence of SRY, and indeed, the mutant W70L showed a significant reduction of nuclear accumulation, most likely explaining the female phenotype of this patient. In conclusion, this case illustrates that a proper pathological classification can be the trigger for a full clinical description of a so-far undiagnosed, DSD patient, at high risk for tumour development. This interference resulted in prevention of development of an invasive cancer, likely to be treated by chemotherapy. It well demonstrates that knowledge about the mechanisms involved in normal gonadal development is relevant in the context of understanding the pathobiology of type II GCTs, especially in DSD patients. This patient is retrospectively, therefore, classified as 46,XY DSD, previously as Swyer syndrome, known for their high risk for type II GCTs. Similar cases have been reported recently.105–108 The first patient presented at an age of 16 years with primary amenorrhoea, undervirilisation, a hypoplastic uterus and a left adrenal mass. The right gonad showed a GB, while the left contained a GB, a dysgerminoma and a yolk sac tumour component. Genetic mutation was not investigated. The second case is of specific interest, because it refers to two half-sisters, both developing a GB (and an invasive tumour), although no SRY mutation or chromosomal mosaicism was identified. Interestingly, the Y chromosome was lost from the invasive tumour component (see below). Involvement of the Y chromosome; TSPY as candidate gene In contrast to the link between ovarian differentiation and FOXL2 and testicular differentiation and SOX9, the correlation between the presence of the Y chromosome and testicular development is less obvious.109 In fact, no correlation between the amount of Y chromosomes and testis development has

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

301

been identified in patients with sex chromosomal mosaicisms, for which no explanation is available so far. Several candidate genes map within the GBY region, of which TSPY is one of the most interesting ones. It stands for Testis-Specific Protein on the Y chromosome, which is in fact a multicopy gene.110 It has similarities to the DEK/CAN family of proteins, and it interacts with cyclin B1, thereby supposed to be involved in cell-cycle regulation. Various splice variants have been reported, which indeed can be present in type II GCTs.111 Protein expression analysis demonstrate that the corresponding protein is present in spermatogonia during normal development.40,112,113 The level of protein is increased in CIS/ IGCNU and GB (Fig. 4), for which the mechanistic basis is unknown so far.40,100,111,114–116 In fact, the increased level of this protein is used as supportive parameter to distinguish a malignant germ cell from a germ cell showing delayed maturation (see below). Upon invasive growth, expression of the gene is mostly lost, associated with subsequent absence of the protein, although the Y chromosome can still be retained. Therefore, this is due to the down-regulation of expression. Transfection expression analysis demonstrated that induction of TSPY in human cells lacking this protein results in an increase in proliferation, both in vitro and in vivo. In fact, the cells show a shorter G2 phase of the cell cycle.117 Interestingly, a subsequent study shows that a number of the up-regulated genes in the TSPY transfected cells map to the short arm of chromosome 12, always found to be over-represented in invasive type II GCTs.118 In fact, a correlation between the level of TSPY and expression of these genes, including KRAS2 and NANOG, was only found in the precursor lesion CIS/IGCNU, and not in the invasive tumours.112 So far, GB was not investigated in this context. This observation fits well with the downregulation of TSPY upon progression of the tumour towards invasiveness. To illustrate a number of issues, a second case report is presented.

Case report II This case has not been published so far. A phenotypically male patient was initially diagnosed as Denys Drash, a known risk factor for type II GCTs.42–44 However, based on careful retrospective clinical data analysis, the patient was suspected to be a Frasier syndrome patient, specifically based on the type of kidney anomalies. Indeed, a Frasier-specific mutation affecting the splice site of intron 9 of the Wilms’ tumour suppressor gene 1 (WT1) gene was identified, being IVS9 þ 4C > T.119 This leads to mainly formation of the KTS WT1 protein variant, resulting in less stabilisation of the SRY mRNA and subsequent protein. Both the Denys Drash and Frasier syndrome patients are recognised as 46,XY DSD variants with a high risk for malignant transformation of germ cells, leading to the type II GCTs. The patient was diagnosed previously with hypospadias and an inguinal hernia, for which he was operated at the age of 2. No vas deferens was identified at that time, and the right gonad was lost during surgery. No information was available about the left gonad. At the age of 5, the patient was operated for a hydrocoele, after which he showed nephritis and a progressive glomerulosclerose, for which he underwent kidney transplantation. At the age of 13, he was operated for an utriculuscyst. The left gonad was originally intra-abdominally localised, but placed at a scrotal position at early age. The patient was on testosterone substitution therapy and requested testicular prostheses. The atrophic gonad showed microlithiasis (microcalcifications on ultrasound examination) and was removed at the age of 21 years. Histological examination demonstrated extensive CIS (i.e., OCT3/4, AP and c-KIT positive), and possibly GB, as well as a pre-existing small invasive seminoma (see Fig. 7). Because of the stage of the disease (i.e., small invasive tumour, localised to the testis, stage I), surveillance was warranted. Timely gonadectomy prevented the seminoma from spreading to lymph nodes, necessitating further treatment, either irradiation or chemotherapy. This would have been highly undesirable in view of the immune suppressed state of the patient because of his kidney transplantation. An additional case of a Frasier syndrome patient has been reported recently120, demonstrating the same issue. Interesting is the fact that both cases show a male phenotype, with CIS/IGCNU as precursor lesion, identified by the various markers, amongst others, OCT3/4, as well as SCF. More frequently, Frasier syndrome is diagnosed in female individuals, who are likely to develop GB instead of CIS.121 Therefore, individuals with 46,XY gonadal dysgenesis are at high risk, irrespectively of their (male of female) phenotype.

302

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

Fig. 7. Macroscopy (left panels) and microscopy (right lower panels) of the gonadectomy specimen (being testis) of the Frasier patient at the age of 21 years. The intron 9 specific mutation (þ4C > T) is demonstrated by sequencing, and shown in the upper right panel. The presence of neoplastic germ cells (detected by OCT3/4 and direct alkaline phosphatase staining on frozen tissue, representing enzymatic reactivity of the presence of PLAP) demonstrated the diagnosis of CIS/IGCNU and seminoma.

Optimal diagnostic markers for type II GCTs and precursors in DSD OCT3/4 has been reported as a highly informative diagnostic marker for embryonal carcinoma, seminoma (dysgerminoma and germinoma) as well as the precursor lesions CIS/IGCNU and GB.18,40,122– 126 In fact, this marker is also informative in detecting CIS/IGCNU cells in the semen of high-risk males.127 However, there are (at least) two exceptions in which OCT3/4, as well as all other embryonic germ cell-specific markers, including c-KIT and PLAP, are not sufficiently informative. That is in the case of tissue obtained during the first year of life (see above), and in the case of gonads showing germ cell maturation delay, in spite of proper development of the stromal component. In these conditions, OCT3/ 4, as well as of all the other embryonic germ cell markers, can be demonstrated immunohistochemically in germ cells which have not undergone malignant transformation.88 Based on morphology, as well as additional criteria, supportive arguments can be obtained to diagnose or rule out malignancy. These criteria are not easy to apply in routine pathology, and they are not without any restriction.88 For this purpose, availability of a more informative marker would be of great clinical diagnostic value in these patients. A possible marker fulfilling these criteria is SCF, being the ligand of cKIT. It is crucial for proper migration and survival of PGCs (see above). Two variants of SCF can be generated by Sertoli cells; one is membrane bound and is highly effective in supporting survival of PGCs128,129 while the soluble form activates Leydig cells present in the stromal compartment of the testis. Under normal physiological conditions, both embryonic and adult, no SCF can be identified in human gonads by immunohistochemistry using a specific antibody.22 However, it is consistently present in the testes with CIS/IGCNU (and GB), but not in case of maturation delay. Upon invasive growth of the tumour cells, SCF, like c-KIT130, is predominantly down-regulated, although it can be present heterogeneously in various histological elements. It could be demonstrated that SCF has a significant additional value to detect the earliest malignant changes in germ cells. This suggests the

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

303

presence of an autocrine loop, which is in line with the observations in cell lines131, both biologically and diagnostically relevant. It suggests that during the early stages in the pathogenesis of type II GCTs, a switch occurs between a paracrine to an autocrine loop of the SCF and c-KIT pathway. Upon development of an invasive tumour, either seminoma or non-seminoma, SCF and c-KIT may lose significance and not be under selective pressure anymore. In this context, amplification of c-KIT and activating mutations, the latter in preferentially bilateral testicular type II GCTs, are of interest.132–134 In addition, the finding of c-KIT mutations in dysgerminomas identified in patients without DSD is of specific interest (Ref. 135 and unpublished observations). Hormonal factors in DSD The mechanistic basis of the increased risk in the various conditions remains to be elucidated, but the possible role of oestrogen and anti-androgen functions, being the basis of the TDS model (see above), is worthy of investigating in more detail. This hypothesis is supported by multiple observations. The higher level of testosterone in blacks might indeed explain the lower incidence of this type of cancer.136 This is supposedly related to the role of testosterone during embryonal development in pushing the PGCs/gonocytes along their maturation pathway to spermatogonia, which thereby lose their embryonic characteristics, and therefore their capacity to form CIS/IGCNU and GB (see above). The higher risk of the first child in birth order is in line with a role of a higher level of oestrogen exposure at early embryonal developmental age.137 Although type II GCTs are rather specific for the Caucasian population, the Maori are an interesting exception.138 Males of this ethnic group show a similar incidence as Caucasians, possibly again related to an increased level of oestrogen. Several studies also indicate that polymorphisms in enzymes, which increase the level of oestrogen, are related to a higher risk of type II GCTs.139 Moreover, the differences between Denmark and Finland are associated with different exposures to chemicals that have an oestrogen or anti-androgen activity.140,141 A counterargument on the role of increased oestrogen is that during early intra-uterine development, the level of oestrogen is high; however, a critical window could be relevant is this context. Of specific interest is that an animal model for disrupted testicular development, used as model for endocrine disruptors in the generation of TDS, indicates such a window.142 That at least a certain amount of testosterone is needed for the precursor lesion to progress to an invasive disease, at least in patients with hypovirilisation, is supported by multiple findings. Patients with complete androgen insensitivity syndrome (CAIS) have a significantly lower risk than the patients with the partial form of this disorder (PAIS).42,88,143,144 Most likely this is related to the induction of apoptosis of germ cells in the testis of CAIS patients, as observed in Klinefelter syndrome patients (see above). Moreover, complete absence or very low level of testosterone also diminishes the risk of a type II GCT. This is well illustrated by patients with hypogonadotropic hypogonadism, who often have cryptorchid testis, but no TGCT is reported so far. Towards development of a clinical decision tree Based on the different levels of information described, a comprehensive model for the pathogenesis of type II GCTs can be proposed (Fig. 2). Although not all (relevant) information is included, it allows further development of a clinical decision tree for patients with DSD. In fact, it might also be applicable to patients without DSD, being at risk for a (testicular) type II GCTs, similar to those with TDS. The proposed scheme is presented in Fig. 8. It is a modified version of the proposal published earlier42,89, although it includes most recent insights into the pathobiology of the type II GCTs. In addition, it links the gonadal morphology to the clinical phenotype145, which will allow comparative studies in time. As stated above, only patients with hypovirilisation or gonadal dysgenesis are at risk for cancer development. This is only the case when the so-called GBY region is present, resulting in expression of TSPY in the embryonic germ cells, which are surviving in immature niches, not in line with the expected maturation of the gonad expected based on age of the patient. In addition, positive immunohistochemical staining for SCF must be present to confirm that the germ cells are indeed transformed, and not only delayed in their maturation.

304

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

Fig. 8. Schematic representation of the interrelationship between clinic and patho-biology. The patient presents in the clinic with either a form of Disorder of Sex Development (DSD) or Testicular Dysgenesis Syndrome (TDS). The patients can be described using the External Masculinization Score (EMS). The main question (in the context of this review) is related to the risk of malignant transformation, leading to an invasive type II GCT. This is likely related to various parameters, including environment as well as genetic/chromosomal anomalies. This might interfere with the physiological maturation of primordial germ cells (PGCs)/gonocytes, to either oogonia in the ovary and prespermatogonia in the testis. This is related to the level of testicularisation of the gonad (absence is female and full presence is male). Four parameters are required to allow malignant transformation, of which only the presence of the GBY region (including the TSPY gene) is not functionally related to the level of testicularisation. These include the presence of immature gonadal tissue and presence of embryonic germ cells (characterised by OCT3/4 amongst others) (due to hypotesticularisation), as well as the presence of the c-KIT ligand, being the SCF. This specific constitution is the prerequisite for the formation of the precursor lesion, either GB or CIS/IGCNU, depending on the level of testicularisation of the gonad, directly related to the presence of either FOXL2 and SOX9. Depending on this knowledge, clinical intervention can be performed, varying between no action at all (in case of absence of hypovirilisation and gonadal dysgenesis), surveillance, irradiation (in case of proven presence of CIS/IGCNU), or even prophylactic gonadectomy, if the patient fulfils all criteria for the high risk group for cancer development (see text for further explanation).

Opportunities for further research; research agenda In spite of the wealth of recent and novel information on the pathogenesis of type II GCTs in DSD patients, there are still a (large) number of unsolved issues, which are of relevance in clinical management of these patients, as well as understanding the pathobiology of the cancer. These relate to relatively simple issues, like primary handling of biopsy samples (proper surgical intervention and fixation), to more complex issues. This, for example, relates to possible heterogeneity of the gonadal histology, which will mainly remain unknown based on a single biopsy. This is specifically of relevance in the context of possible under-diagnosis in case when it is decided for a conservative (surgical) approach, that is, in which the gonad(s) is left in situ. This will require a proper follow-up protocol, in which high-resolution visualisation techniques (ultrasound and magnetic resonance imaging (MRI)) might find a place, as well as the application of possible serum markers. In addition, functional tests for the presence of testicular tissue (e.g., hCG, inhibin or anti-Mu¨llerian hormone test) might be informative, which could also be developed for ovarian tissue (FSH stimulation test). Because of the rarity of the various forms of DSD, international collaboration is required to obtain sufficient numbers of cases

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

305

to study the actual risk for malignant transformation in the subtypes. Therefore, protocols for histological description (besides the primary handling, as mentioned above) of the biopsy samples must become available, agreed upon by specialised pathologist, including the use of informative markers to characterise the maturation stage of the germ cells. For this specific purpose, OCT3/4, SCF, TSPY, FOXL2 and SOX9 seem to be an informative set so-far. Possible mosaicism must be considered leading to putative under-diagnosis of the presence of the GBY region in peripheral blood or (genital) skin fibroblasts. To answer the various questions, it is highly recommended that a multidisciplinary database will become available, including a multidisciplinary set up. In fact, such an initiative is taken with the Euro-DSD framework (http://www.eurodsd.eu/). The requirements are demanding, in time, energy, regulations, as well as financial resources, but will result in improved diagnosis and follow-up, and overall care of patients with DSD. In addition, it will also result in better understanding of the mechanisms of the different levels of generation the different sexes, using the variations seen within the natural population. In summary, DSD patients with hypervirilisation are not at risk for GCTs, while the risk is highly variable within the group of patients with hypovirilisation and gonadal dysgenesis. Recognition hereof resulted in stratification into a high-, intermediate-, low (and unknown)-risk category. Within these subgroups, tumour risk is associated with a number of criteria (most likely interrelated), amongst others: the nature of the genetic anomaly, presence of Y-chromosomal material (i.e., GBY region), as well as gonad-related characteristics, including the anatomical position (e.g., scrotal, inguinal and abdominal) and status of maturation (i.e., undifferentiated, female, male or a mixture of those). In general, it can be stated that the tumour risk decreases with a more accomplished maturation state of the gonad (ovary, testis, or even combined, i.e., ovotestis). Conversely, the risk for tumour development increases with (non-physiological) retention of immature structures and/or characteristics, here referred to as inhibited ‘testicularisation’. This allows embryonic germ cells (positive for various markers, including OCT3/4, c-KIT and PLAP) to survive in these niches, while they would normally mature into oogonia (female) or pre-spermatogonia (male). During normal maturation they will lose expression of the embryonic markers and thereby their capacity for malignant transformation. Embryonic germ cells might also disappear in time via cell death, for example, in CAIS patients. This might explain the significant lower tumour risk in CAIS than in PAIS patients. Besides the microenvironment, presence of part of the Y chromosome is of relevance for malignant transformation. This relates to the so-called GBY (gonadoblastoma on the Y chromosome) region, for which TSPY is one of the preferred candidate genes. Co-expression of OCT3/4 and TSPY defines the (embryonic) germ cells at risk. This probably fulfils the pathogenetic requirements for suppression of apoptosis and induction of proliferation. Recognition of the pre-invasive lesions of type II GCTs, carcinoma in situ of the testis (CIS) and gonadoblastoma (GB) of the dysgenetic gonad with a low level of differentiation, efficiently identified by immunohistochemical staining for OCT3/4 and TSPY, combined with SOX9 (Sertoli cell marker) and FOXL2 (granulosa cell marker) is informative to diagnose DSD patients based on gonadal biopsies. Immunohistochemistry for the stem cell factor (SCF: c-KIT ligand) allows distinction between premalignant germ cells from germ cells showing (only) delayed maturation. This is a further step in the stratification of DSD patients for their actual risk for development of a type II GCT. Acknowledgements Thanks to all co-workers who made the analyses on the pathogenesis of germ cell tumours possible, these include technicians, students, PhD students, as well as post docs. The patients participating in the various studies and the collaborating clinicians (pathologists, urologists, medical oncologists, etc.) are all greatly acknowledged for their support. In addition, the various organisations financially supporting the different studies are thanked (Dutch Cancer Society, Grant EMCR 2006–3607; Erasmus MC Translational research; Flanders Research Foundation). References *1. Oosterhuis J & Looijenga L. Testicular germ-cell tumours in a broader perspective. Nature Reviews. Cancer 2005 Mar; 5(3): 210–222.

306

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

2. McLaren A. Primordial germ cells in the mouse. Developmental Biology 2003 Oct 1; 262(1): 1–15. 3. Donovan PJ. The germ cell–the mother of all stem cells. The International Journal of Developmental Biology 1998; 42(7): 1043–1050. 4. van de Geijn GJ, Hersmus R & Looijenga LH. Recent developments in testicular germ cell tumor research. Birth Defects Research. Part C, Embryo Today: Reviews 2009 Mar; 87(1): 96–113. 5. Donovan PJ. Growth factor regulation of mouse primordial germ cell development. Current Topics in Developmental Biology 1994; 29: 189–225. 6. Godin I, Deed R, Cooke J et al. Effects of the steel gene product on mouse primordial germ cells in culture. Nature 1991; 352: 807–809. 7. Wylie CC. The biology of primordial germ cells. European Urology 1993; 23: 62–67. 8. Runyan C, Schaible K, Molyneaux K et al. Steel factor controls midline cell death of primordial germ cells and is essential for their normal proliferation and migration. Development (Cambridge, England) 2006 Dec; 133(24): 4861–4869. 9. Hara K, Kanai-Azuma M, Uemura M et al. Evidence for crucial role of hindgut expansion in directing proper migration of primordial germ cells in mouse early embryogenesis. Developmental Biology 2009 Jun 15; 330(2): 427–439. 10. Szabo PE & Mann JR. Biallelic expression of imprinted genes in the mouse germ line: implications for erasure, establishment, and mechanisms of genomic imprinting. Genes & Development 1995; 9: 1857–1868. 11. Surani MA. Reprogramming of genome function through epigenetic inheritance. Nature 2001 Nov 1; 414(6859): 122–128. 12. Wilhelm D, Palmer S & Koopman P. Sex determination and gonadal development in mammals. Physiological Reviews 2007 Jan; 87(1): 1–28. 13. Polanco JC & Koopman P. Sry and the hesitant beginnings of male development. Developmental Biology 2007 Feb 1; 302(1): 13–24. 14. Wilhelm D & Koopman P. The makings of maleness: towards an integrated view of male sexual development. Nature Reviews. Genetics 2006 Aug; 7(8): 620–631. 15. Ottolenghi C, Uda M, Crisponi L et al. Determination and stability of sex. BioEssays: News and Reviews in Molecular, Cellular and Developmental Biology 2007 Jan; 29(1): 15–25. 16. Rajpert-De Meyts E, Jorgensen N, Brondum-Nielsen K et al. Developmental arrest of germ cells in the pathogenesis of germ cell neoplasia. APMIS: acta pathologica, microbiologica, et immunologica Scandinavica 1998; 106(1): 198–204. discussion -6. 17. Jørgensen N, Rajpert-De Meyts E, Graem N et al. Expression of immunohistochemical markers for testicular carcinoma in situ by normal fetal germ cells. Laboratory Investigation; a Journal of Technical Methods and Pathology 1995; 72: 223–231. 18. Honecker F, Stoop H, De Krijger R et al. Pathobiological implications of the expression of markers of testicular carcinoma in situ by foetal germ cells. The Journal of Pathology 2004; 203: 849–857. 19. Gaskell TL, Esnal A, Robinson LL et al. Immunohistochemical profiling of germ cells within the human fetal testis: identification of three subpopulations. Biology of Reproduction 2004 Dec; 71(6): 2012–2021. 20. Gashaw I, Dushaj O, Behr R et al. Novel germ cell markers characterize testicular seminoma and fetal testis. Molecular Human Reproduction 2007 Oct; 13(10): 721–727. 21. Kerr CL, Hill CM, Blumenthal PD et al. Expression of pluripotent stem cell markers in the human fetal testis. Stem Cells 2008 Feb; 26(2): 412–421. 22. Stoop H, Honecker F, van de Geijn G et al. Stem cell factor as a novel diagnostic marker for early malignant germ cells. The Journal of Pathology 2008; 216: 43–54. 23. Scully RE. Germ cell tumors. In Scully RE (ed.). Tumors of the ovary and maldeveloped gonads. 1st ed. Washington, DC: Armed Forces of Pathology, 1978, pp. 226–286. 24. Mostofi FK & Sesterhenn IA. Pathology of germ cell tumors of testes. Progress in Clinical and Biological Research 1985; 203: 1–34. 25. Mostofi FK, Sesterhenn IA & Davis CJJ. Immunopathology of germ cell tumors of the testis. Seminars in Diagnostic Pathology 1987; 4: 320–341. 26. Donohue JP. The pathology of germ cell tumors of the testis. In Libertino JA (ed.). Testis tumors (International Perspectives in Urology). 7th ed. Baltimore\London: Williams&Wilkins, 1990, pp. 23–54. 27. Looijenga LH & Oosterhuis JW. Pathogenesis of testicular germ cell tumours. Reviews of Reproduction 1999 May; 4(2): 90–100. 28. Looijenga LH, de Munnik H & Oosterhuis JW. A molecular model for the development of germ cell cancer. International Journal of Cancer 1999 Dec 10; 83(6): 809–814. 29. Oosterhuis JW & Looijenga LH. Current views on the pathogenesis of testicular germ cell tumours and perspectives for future research: highlights of the 5th Copenhagen Workshop on Carcinoma in situ and Cancer of the Testis. APMIS: acta pathologica, microbiologica, et immunologica Scandinavica 2003 Jan; 111(1): 280–289. 30. Pugh RCB. Combined tumours. In Pugh RCB (ed.). Pathology of the testis. Oxford: Blackwell, 1976, pp. 245–258. 31. Woodward PJ, Heidenreich A, Looijenga LHJ et al. Testicular germ cell tumors. In Eble JN, Sauter G, Epstein JI & Sesterhann IA (eds.). World Health Organization classification of tumours pathology and genetics of the urinary system and male genital organs. Lyon: IARC Press, 2004, pp. 217–278. 32. Hong Y & Stambrook PJ. Restoration of an absent G1 arrest and protection from apoptosis in embryonic stem cells after ionizing radiation. Proc Natl Acad Sci U S A 2004 Oct 5; 101(40): 14443–14448. *33. Masters JR & Koberle B. Curing metastatic cancer: lessons from testicular germ-cell tumours. Nature Reviews. Cancer 2003 Jul; 3(7): 517–525. *34. Skakkebæk NE. Possible carcinoma-in-situ of the testis. Lancet 1972: 516–517. 35. Loy V & Dieckmann KP. Carcinoma in situ of the testis: intratubular germ cell neoplasia or testicular intraepithelial neoplasia? Human Pathology 1990; 21: 457–458. 36. Looijenga LH, Stoop H, Hersmus R et al. Genomic and expression profiling of human spermatocytic seminomas: pathogenetic implications. International Journal of Andrology 2007 Aug; 30(4): 328–335. discussion 35–6. 37. Oosterhuis JW, Stoop H, Honecker F et al. Why human extragonadal germ cell tumors occur in the midline of the body; old concepts, new perspectives. International Journal of Andrology 2007; 30: 256–263. 38. Scully RE. Gonadoblastoma/A review of 74 cases. Cancer 1970; 25: 1340–1356.

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

307

39. Jørgensen N, Muller J, Jaubert F et al. Heterogeneity of gonadoblastoma germ cells: similarities with immature germ cells, spermatogonia and testicular carcinoma in situ cells. Histopathology 1997; 30(2): 177–186. 40. Kersemaekers AM, Honecker F, Cools M et al. Identification of germ cells at risk for neoplastic transformation in gonadoblastomas: an immunohistochemical study for OCT3/4 and TSPY. Human Pathology 2005; 36: 512–521. 41. Cools M, Stoop H, Kersemaekers AM et al. Gonadoblastoma arising in undifferentiated gonadal tissue within dysgenetic gonads. The Journal of Clinical Endocrinology and Metabolism 2006 Jun; 91(6): 2404–2413. *42. Cools M, Drop SL, Wolffenbuttel KP et al. Germ cell tumors in the intersex gonad: old paths, new directions, moving frontiers. Endocrine Reviews 2006 May 30; 27: 468–484. 43. Looijenga LH, Hersmus R, Oosterhuis JW et al. Tumor risk in disorders of sex development (DSD). Best Practice & Research. Clinical Endocrinology & Metabolism 2007 Sep; 21(3): 480–495. 44. Hersmus R, de Leeuw BHCGM, Wolffenbuttel KP et al. New insights into type II Germ Cell Tumor pathogenesis based on the studies of patients with various forms of disorders of sex development (DSD). Molecular and Cellular Endocrinology 2008; 291: 1–10. *45. Van Gurp RJ, Oosterhuis JW, Kalscheuer V et al. Human testicular germ cell tumors show biallelic expression of the H19 and IGF2 gene. Journal of the National Cancer Institute 1994; 86: 1070–1075. 46. Rachmilewitz J, Elkin M, Looijenga LHJ et al. Characterization of the imprinted IPW gene: allelic expression in normal and tumorigenic human tissues. Oncogene 1996; 13: 1687–1692. 47. Verkerk AJ, Ariel I, Dekker MC et al. Unique expression patterns of H19 in human testicular cancers of different etiology. Oncogene 1997; 14(1): 95–107. 48. Looijenga LH, Verkerk AJ, Dekker MC et al. Genomic imprinting in testicular germ cell tumours. APMIS: acta pathologica, microbiologica, et immunologica Scandinavica 1998; 106(1): 187–195. discussion 96–7. 49. Sievers S, Alemazkour K, Zahn D et al. IGF2/H19 imprinting analysis of human germ cell tumors (GCTs) using the methylation-sensitive single-nucleotide primer extension method reflects the origin of GCTs in different stages of primordial germ cell development. Genes, Chromosomes & Cancer 2005; 44(3): 256–264. 50. Kawakami T, Zhang C, Okada Y et al. Erasure of methylation imprint at the promoter and CTCF-binding site upstream of H19 in human testicular germ cell tumors of adolescents indicate their fetal germ cell origin. Oncogene 2006 Jun 1; 25(23): 3225–3236. 51. Almstrup K, Hoei-Hansen CE, Nielsen JE et al. Genome-wide gene expression profiling of testicular carcinoma in situ progression into overt tumours. British Journal of Cancer 2005 May 23; 92(10): 1934–1941. 52. MØller H. Clues to the aetiology of testicular germ cell tumours from descriptive epidemiology. European Urology 1993; 23: 8–15. 53. Moller H & Skakkebæk NE. Risk of testicular cancer and cryptorchidism in relation to socio-economical status and related factors: case-control studies in Denmark. International Journal of Cancer 1996; 66: 287–293. 54. Krausz C & Looijenga LHJ. Genetic aspects of testicular germ cell tumors. Cell Cycle 2008; 7: 3519–3524. 55. Chaganti RS & Bosl GJ. Germ cell tumors: unraveling a biological paradox. Laboratory Investigation; a Journal of Technical Methods and Pathology 1995; 73: 593–595. 56. Moller H. Clues to the aetiology of testicular germ cell tumours from descriptive epidemiology. European Urology 1993; 23(1): 8–13. discussion 4–5. 57. Skakkebaek NE, Rajpert-De Meyts E, Jorgensen N et al. Germ cell cancer and disorders of spermatogenesis: an environmental connection? APMIS: acta pathologica, microbiologica, et immunologica Scandinavica 1998 Jan; 106(1): 3– 11. discussion 2. 58. Jacobsen R, Bostofte E, Engholm G et al. Risk of testicular cancer in men with abnormal semen characteristics: cohort study. BMJ (Clinical Research Ed.) 2000; 321(7264): 789–792. 59. McGlynn KA, Devesa SS, Sigurdson AJ et al. Trends in the incidence of testicular germ cell tumors in the United States. Cancer 2003 Jan 1; 97(1): 63–70. 60. Pamenter B, De Bono JS, Brown IL et al. Bilateral testicular cancer: a preventable problem? Experience from a large cancer centre. BJU International 2003 Jul; 92(1): 43–46. 61. Raman JD, Nobert CF & Goldstein M. Increased incidence of testicular cancer in men presenting with infertility and abnormal semen analysis. The Journal of Urology 2005 Nov; 174(5): 1819–1822. discussion 22. 62. Sonke GS, Chang S, Strom SS et al. Prenatal and perinatal risk factors and testicular cancer: a hospital-based casecontrol study. Oncology Research 2007; 16(8): 383–387. 63. Cook MB, Graubard BI, Rubertone MV et al. Perinatal factors and the risk of testicular germ cell tumors. International Journal of Cancer 2008 Mar 6; 122: 2600–2606. 64. Walsh TJ, Dall’era MA, Croughan MS et al. Prepubertal orchiopexy for cryptorchidism may be associated with lower risk of testicular cancer. The Journal of Urology 2007 Oct; 178(4): 1440–1446. 65. Pettersson A, Richiardi L, Nordenskjold A et al. Age at surgery for undescended testis and risk of testicular cancer. The New England Journal of Medicine 2007 May 3; 356(18): 1835–1841. 66. Rapley EA, Crockford GP, Teare D et al. Localization to Xq27 of a susceptibility gene for testicular germ-cell tumours. Nature Genetics 2000; 24(2): 197–200. 67. Holzik MF, Rapley EA, Hoekstra HJ et al. Genetic predisposition to testicular germ-cell tumours. The Lancet Oncology 2004 Jun; 5(6): 363–371. *68. Rapley EA, Turnbull C, Al Olama AA et al. A genome-wide association study of testicular germ cell tumor. Nature Genetics 2009 May 31. *69. Kanetsky PA, Mitra N, Vardhanabhuti S et al. Common variation in KITLG and at 5q31.3 predisposes to testicular germ cell cancer. Nature Genetics 2009 May 31. 70. Hemminki K, Li X & Czene K. Cancer risks in first-generation immigrants to Sweden. International Journal of Cancer 2002 May 10; 99(2): 218–228. 71. Michos A, Xue F & Michels KB. Birth weight and the risk of testicular cancer: a meta-analysis. International Journal of Cancer 2007 Sep 1; 121(5): 1123–1131.

308

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

72. Cools M, Honecker F, Stoop H et al. Maturation delay of germ cells in trisomy 21 fetuses results in increase risk for the development of testicular germ cell tumors. Human Pathology 2006; 37: 101–111. 73. Skakkebæk NE, Rajpert-De Meyts E & Main KM. Testicular dysgenesis syndrome: an increasingly common developmental disorder with environmental aspects. Human Reproduction (Oxford, England) 2001; 16(5): 972–978. 74. Fisher JS, Macpherson S, Marchetti N et al. Human ‘testicular dysgenesis syndrome’: a possible model using in-utero exposure of the rat to dibutyl phthalate. Human Reproduction (Oxford, England) 2003 Jul; 18(7): 1383–1394. 75. Skakkebaek NE. Testicular dysgenesis syndrome. Hormone Research 2003; 60(Suppl. 3): 49. 76. Rajpert-De Meyts E. Developmental model for the pathogenesis of testicular carcinoma in situ: genetic and environmental aspects. Human Reproduction Update 2006 May–Jun; 12(3): 303–323. 77. Sonne SB, Kristensen DM, Novotny GW et al. Testicular dysgenesis syndrome and the origin of carcinoma in situ testis. International Journal of Andrology 2008 Jan 16; 31: 275–287. 78. Hughes IA, Houk C, Ahmed SF et al. Consensus statement on management of intersex disorders. Archives of Disease in Childhood 2006 Apr 19. *79. Hughes IA. Disorders of sex development: a new definition and classification. Best Practice & Research. Clinical Endocrinology & Metabolism 2008 Feb; 22(1): 119–134. 80. Isurugi K, Imao S, Hirose K et al. Seminoma in Klinefelter’s syndrome with 47, XXY, 15sþ karyotype. Cancer 1977; 39(5): 2041–2047. 81. Lee MW & Stephens RL. Klinefelter’s syndrome and extragonadal germ cell tumors. Cancer 1987; 60(5): 1053–1055. 82. Nichols CR, Heerema NA, Palmer C et al. Klinefelter’s syndrome associated with mediastinal germ cell neoplasms. Journal of Clinical Oncology: Official Journal of the American Society of Clinical Oncology 1987; 5(8): 1290–1294. 83. Hasle H, Jacobsen BB, Asschenfeldt P et al. Mediastinal germ cell tumour associated with Klinefelter syndrome. A report of case and review of the literature. European Journal of Pediatrics 1992; 151: 735–739. 84. Hasle H, Mellemgaard A, Nielsen J et al. Cancer incidence in men with Klinefelter syndrome. British Journal of Cancer 1995; 71: 416–420. 85. Volkl TM, Langer T, Aigner T et al. Klinefelter syndrome and mediastinal germ cell tumors. American Journal of Medical Genetics. Part A 2006 Mar 1; 140(5): 471–481. 86. Aksglaede L, Wikstrom AM, Rajpert-De Meyts E et al. Natural history of seminiferous tubule degeneration in Klinefelter syndrome. Human Reproduction Update 2006 Jan–Feb; 12(1): 39–48. 87. Modi DN, Sane S & Bhartiya D. Accelerated germ cell apoptosis in sex chromosome aneuploid fetal human gonads. Molecular Human Reproduction 2003 Apr; 9(4): 219–225. 88. Cools M, van Aerde K, Kersemaekers AM et al. Morphological and immunohistochemical differences between gonadal maturation delay and early germ cell neoplasia in patients with undervirilization syndromes. The Journal of Clinical Endocrinology and Metabolism 2005 Sep; 90(9): 5295–5303. 89. Cools M, Looijenga LH, Wolffenbuttel KP et al. Disorders of sex development: update on the genetic background, terminology and risk for the development of germ cell tumors. World Journal of Pediatrics: WJP 2009; 5(2): 93–102. 90. Batata MA, Whitmore Jr. WF, Chu FCH et al. Cryptorchidism and testicular cancer. The Journal of Urology 1980; 124: 382–387. 91. Muller J, Skakkebaek NE, Nielsen OH et al. Cryptorchidism and testis cancer. Cancer 1984; 54: 629–634. 92. Giwercman A, Grindsted J, Hansen B et al. Testicular cancer risk in boys with maldescended testis: a cohort study. The Journal of Urology 1987; 138(5): 1214–1216. 93. Abratt RP, Reddi VB & Sarembock LA. Testicular cancer and cryptorchidism. British Journal of Urology 1992; 70: 656–659. 94. Ogunbiyi JO, Shittu OB, Aghadiuno PU et al. Seminoma arising in cryptorchid testes in Nigerian males. East African Medical Journal 1996; 73(2): 129–132. 95. Susnerwala SS, Pande SC, Shrivastava SK et al. Dysgerminoma of the ovary: review of 27 cases. Journal of Surgical Oncology 1991 Jan; 46(1): 43–47. 96. Dietl J, Horny HP, Ruck P et al. Dysgerminoma of the ovary. An immunohistochemical study of tumor- infiltrating lymphoreticular cells and tumor cells. Cancer 1993; 71(8): 2562–2568. 97. Chow SN, Yang JH, Lin YH et al. Malignant ovarian germ cell tumors. International Journal of Gynecologyand Obstetrics 1996; 53: 151–158. 98. Cusido MT, Jorda B, Gonzalez J et al. Ovarian germ cell tumors. European Journal of Gynaecological Oncology 1998; 19(2): 130–134. 99. Tewari K, Cappuccini F, Disaia PJ et al. Malignant germ cell tumors of the ovary. Obstetrics and Gynecology 2000 Jan; 95(1): 128–133. 100. Li Y, Vilain E & Conte F. Rajpert-De Meyts E, Lau YF. Testis-specific protein Y-encoded gene is expressed in early and late stages of gonadoblastoma and testicular carcinoma in situ. Urologic Oncology 2007 Mar-Apr; 25(2): 141–146. 101. Hersmus R, Kalfa N, De Leeuw B et al. FOXL2 and SOX9 as parameters of female and male gonadal differentiation in patients with various forms of disorders of sex development (DSD). The Journal of Pathology 2008; 215: 31–38. 102. Page DC. Hypothesis: a Y-chromosomal gene causes gonadoblastoma in dysgenetic gonads. Development (Cambridge, England) 1987; 101(Suppl): 151–155. 103. Bianco B, Lipay M, Guedes A et al. SRY gene increases the risk of developing gonadoblastoma and/or nontumoral gonadal lesions in Turner syndrome. International Journal of Gynecological Pathology 2009 Jan 30. 104. Hersmus R, de Leeuw BH, Stoop H et al. A novel SRY missense mutation affecting nuclear import in a 46, XY female patient with bilateral gonadoblastoma. European Journal of Human Genetics : EJHG 2009 Jun 10. 105. Ng SB, Yong MH, Knight LA et al. Gonadoblastoma-associated mixed germ cell tumour in 46, XY complete gonadal dysgenesis (Swyer syndrome): analysis of Y chromosomal genotype and OCT3/4 and TSPY expression profile. Histopathology 2008 Apr; 52(5): 644–646. 106. Beaulieu Bergeron M, Bouron-Dal Soglio D, Maietta A et al. Co-existence of a choriocarcinoma and a gonadoblastoma in the gonad of a 46, XY female: A SNP array analysis. Pediatric and Developmental Pathology 2009 May 8: 1. 107. Kini U, Bantwal G, Ayyar V et al. Bilateral gonadoblastomas with a left sided dysgerminoma in a true hermaphrodite (disorder of sexual differentiation) with 46, XY karyotype. The Journal of the Association of Physicians of India 2008 Jul; 56: 549–551.

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

309

108. Simon RA, Laughlin TS, Nuccie B et al. 46 XY phenotypic female adolescent with bilateral gonadal tumors consisting of five different components. International Journal of Gynecological Pathology 2008 Jul; 27(3): 407–411. 109. Cools M, Boter M, van Gurp R et al. Impact of the Y-containing cell line on histological differentiation patterns in dysgenetic gonads. Clinical Endocrinology 2007 Aug; 67(2): 184–192. 110. Vogel T & Schmidtke J. Structure and function of TSPY, the Y-chromosome gene coding for the ‘‘testis-specific protein’’. Cytogenetics and Cell Genetics 1998; 80(1–4): 209–213. *111. Lau YF. Gonadoblastoma, testicular and prostate cancers, and the TSPY gene. American Journal of Human Genetics 1999; 64(4): 921–927. 112. Li Y, Tabatabai ZL, Lee TL et al. The Y-encoded TSPY protein: a significant marker potentially plays a role in the pathogenesis of testicular germ cell tumors. Human Pathology 2007 Oct; 38(10): 1470–1481. 113. Lau YF, Lau HW & Komuves LG. Expression pattern of a gonadoblastoma candidate gene suggests a role of the Y chromosome in prostate cancer. Cytogenetic and Genome Research 2003; 101(3–4): 250–260. 114. Schnieders F, Dork T, Arnemann J et al. Testis-specific protein, Y-encoded (TSPY) expression in testicular tissues. Human Molecular Genetics 1996; 5(11): 1801–1807. 115. Hildenbrand R, Schroder W, Brude E et al. Detection of TSPY protein in a unilateral microscopic gonadoblastoma of a Turner mosaic patient with a Y-derived marker chromosome. The Journal of Pathology 1999; 189(4): 623–626. 116. Delbridge ML, Longepied G, Depetris D et al. TSPY, the candidate gonadoblastoma gene on the human Y chromosome, has a widely expressed homologue on the X - implications for Y chromosome evolution. Chromosome Research 2004; 12(4): 345–356. 117. Oram SW, Liu XX, Lee TL et al. TSPY potentiates cell proliferation and tumorigenesis by promoting cell cycle progression in HeLa and NIH3T3 cells. BMC Cancer 2006; 6: 154. 118. Looijenga LHJ, Zafarana G, Grygalewicz B et al. Role of gain of 12p in germ cell tumour development. APMIS: acta pathologica, microbiologica, et immunologica Scandinavica 2003; 111: 161–173. 119. Hammes A, Guo JK, Lutsch G et al. Two splice variants of the Wilms’ tumor 1 gene have distinct functions during sex determination and nephron formation. Cell 2001; 106(3): 319–329. 120. Schumacher V, Gueler B, Looijenga LH et al. Characteristics of testicular dysgenesis syndrome and decreased expression of SRY and SOX9 in Frasier syndrome. Molecular Reproduction and Development 2008 Feb 12. 121. Andrade JG, Guaragna MS, Soardi FC et al. Clinical and genetic findings of five patients with WT1-related disorders. Arquivos brasileiros de endocrinologia e metabologia 2008 Nov; 52(8): 1236–1243. *122. Looijenga LHJ, Stoop H, De Leeuw PJC et al. POU5F1 (OCT3/4) identifies cells with pluripotent potential in human germ cell tumors. Cancer Research 2003; 63: 2244–2250. 123. de Jong J, Stoop H, Dohle GR et al. Diagnostic value of OCT3/4 for pre-invasive and invasive testicular germ cell tumours. The Journal of Pathology 2005 Jun; 206(2): 242–249. 124. De Jong J & Looijenga LHJ. Stem cell marker OCT3/4 in tumor biology and germ cell tumor diagnostics: history and future. OCT3/4 in oncogenesis. Critical Reviews in Oncogenesis 2006; 12(3–4): 171–203. 125. Cheng L, Sung MT, Cossu-Rocca P et al. OCT4: biological functions and clinical applications as a marker of germ cell neoplasia. The Journal of Pathology 2007 Jan; 211(1): 1–9. 126. van Casteren NJ, Boellaard WP, Dohle GR et al. Heterogeneous distribution of ITGCNU in an adult testis: consequences for biopsy-based diagnosis. International Journal of Surgical Pathology 2008 Jan; 16(1): 21–24. 127. van Casteren NJ, Stoop H, Dohle GR et al. noninvasive detection of testicular carcinoma in situ in semen using OCT3/4. European Urology 2008 Jul; 54(1): 153–160. 128. Lev S, Blechman JM, Givol D et al. Steel factor and c-kit protooncogene: genetic lessons in signal transduction. Critical Reviews in Oncogenesis 1994; 5(2–3): 141–168. 129. Yan W, Kero J, Huhtaniemi I et al. Stem cell factor functions as a survival factor for mature Leydig cells and a growth factor for precursor Leydig cells after ethylene dimethane sulfonate treatment: implication of a role of the stem cell factor/c-Kit system in Leydig cell development. Developmental Biology 2000 Nov 1; 227(1): 169–182. 130. Biermann K, Klingmuller D, Koch A et al. Diagnostic value of markers M2A, OCT3/4, AP-2gamma, PLAP and c-KIT in the detection of extragonadal seminomas. Histopathology 2006 Sep; 49(3): 290–297. 131. Goddard NC, McIntyre A, Summersgill B et al. KIT and RAS signalling pathways in testicular germ cell tumours: new data and a review of the literature. International Journal of Andrology 2007 Aug; 30(4): 337–348. discussion 49. 132. McIntire KR, Summersgill B, Grygalewicz B et al. Amplification and overexpression of the KIT gene is associated with progression in the seminoma subtype of testicular germ cell tumors of adolescents and adults. Cancer Research 2005; 65: 8085–8089. 133. Looijenga LHJ, De Leeuw PJC, Van Oorschot M et al. Stem cell factor receptor (c-KIT) codon 816 mutations predict development of bilateral testicular germ cell tumors. Cancer Research 2003; 63: 7674–7678. 134. Biermann K, Goke F, Nettersheim D et al. c-KIT is frequently mutated in bilateral germ cell tumours and down-regulated during progression from intratubular germ cell neoplasia to seminoma. The Journal of Pathology 2007 Nov; 213(3): 311–318. 135. Hoei-Hansen CE, Kraggerud SM, Abeler VM et al. Ovarian dysgerminomas are characterised by frequent KIT mutations and abundant expression of pluripotency markers. Molecular Cancer 2007; 6: 12. 136. Henderson BE, Bernstein L, Ross RK et al. The early in utero oestrogen and testosterone environment of blacks and whites: potential effects on male offspring. British Journal of Cancer 1988; 57(2): 216–218. 137. Weir HK, Marrett LD, Kreiger N et al. Pre-natal and peri-natal exposures and risk of testicular germ-cell cancer. International Journal of Cancer 2000; 87(3): 438–443. 138. Wilkinson TJ, Colls BM & Schluter PJ. Increased incidence of germ cell testicular cancer in New Zealand Maoris. British Journal of Cancer 1992; 65(5): 769–771. 139. Starr JR, Chen C, Doody DR et al. Risk of testicular germ cell cancer in relation to variation in maternal and offspring cytochrome p450 genes involved in catechol estrogen metabolism. Cancer Epidemiology, Biomarkers & Prevention 2005 Sep; 14(9): 2183–2190.

310

L.H.J. Looijenga et al. / Best Practice & Research Clinical Endocrinology & Metabolism 24 (2010) 291–310

140. Toppari J, Larsen JC, Christiansen P et al. Male reproductive health and environmental xenoestrogens. Environmental Health Perspectives 1996; 104(Suppl. 4): 741–803. 141. Rajpert-De Meyts E, Toppari J, Hoi-Hansen CE et al. Testicular neoplasia in childhood and adolescence. Endocrine Development 2003; 5: 110–123. 142. Welsh M, Saunders PT, Fisken M et al. Identification in rats of a programming window for reproductive tract masculinization, disruption of which leads to hypospadias and cryptorchidism. The Journal of Clinical Investigation 2008 Apr; 118(4): 1479–1490. 143. Hannema SE, Scott IS, Rajpert-De Meyts E et al. Testicular development in the complete androgen insensitivity syndrome. The Journal of Pathology 2006 Mar; 208(4): 518–527. 144. Cheikhelard A, Morel Y, Thibaud E et al. Long-term followup and comparison between genotype and phenotype in 29 cases of complete androgen insensitivity syndrome. The Journal of Urology 2008 Oct; 180(4): 1496–1501. 145. Ahmed SF, Khwaja O & Hughes IA. The role of a clinical score in the assessment of ambiguous genitalia. BJU International 2000 Jan; 85(1): 120–124.