MEDICAL PROGRESS
Recent Advances in Understanding the Etiology and Pathogenesis of Pediatric Germ Cell Tumors Christiane H. Mosbech, BSc,* Catherine Rechnitzer, MD, DMSc,w Jesper S. Brok, MD, PhD,w Ewa Rajpert-De Meyts, MD, DMSc,* and Christina E. Hoei-Hansen, MD, DMSc*z
Summary: Pediatric germ cell tumors (GCTs) are rare neoplasms arising predominantly in the gonads and sacrococcygeal, mediastinal, and intracranial localizations. In this article, we review current knowledge of pathogenesis of pediatric GCTs, which differs from adult/adolescent GCTs. One distinctive feature is the absence of a progenitor stage, such as carcinoma in situ or gonadoblastoma, which are seen in adult/adolescent GCTs, except spermatocytic seminoma. The primordial germ cell (PGC) is the suggested origin of all GCTs, with variations in histology reflecting differentiation stage. Expression of pluripotency transcription factors OCT-3/4, NANOG, and AP-2g in germinomas/seminomas/ dysgerminomas is consistent with retaining a germ cell phenotype. Teratomas, in contrast, develop through a pathway of aberrant somatic differentiation of immature germ cells, and the yolk sac tumors and choriocarcinomas result from abnormal extraembryonic differentiation. In pediatric GCTs, origin is suggested at an earlier developmental stage because of predisposing genetic factors, although responsible genes remain largely unknown. Some extragonadal GCTs have been linked to overexpression of the KIT/ KITLG system, allowing for survival of aberrantly migrated ectopic PGCs. Infant gonadal/sacrococcygeal GCTs may be caused by apoptosis-related pathways, consistent with an association with polymorphisms in BAK1. Although recent advances have identified candidate pathways, further effort is needed to answer central questions of pathogenesis of these fascinating tumors. Key Words: germ cell tumor, children, primordial germ cell, pathogenesis
(J Pediatr Hematol Oncol 2014;36:263–270)
G
erm cell tumors (GCT) are a heterogenous group of neoplasms that develop in both sexes at any age. They vary in histology and occur at different anatomic sites. GCTs can be found in the gonads and at extragonadal sites along the midline of the body (Fig. 1). Extragonadal tumors predominate in children younger than 3 years of age, whereas the gonads are the main localization during and after puberty.2 The tumors have been characterized on the basis of histology and site of occurrence, and—more recently—on the pattern of cytogenetic aberrations, gene expression, gene variants, and miRNA expression. Received for publication November 16, 2013; accepted January 15, 2014. From the *Department of Growth and Reproduction; wDepartment of Pediatrics, University Hospital of Rigshospitalet, Copenhagen; and zDepartment of Pediatrics, University Hospital of Hilleroed, Hilleroed, Denmark. Supported by The Danish Cancer Society, The Danish Child Cancer Foundation. The authors declare no conflict of interest. Reprints: Christina E. Hoei-Hansen, MD, DMSc, Department of Pediatrics, University Hospital of Hilleroed, Dyrehavevej 29, DK— 3400 Hilleroed, Denmark (e-mail: [email protected]). Copyright r 2014 by Lippincott Williams & Wilkins
J Pediatr Hematol Oncol
Volume 36, Number 4, May 2014
Each year, 1 of 500 children is diagnosed with cancer worldwide3 with 2.5% being GCTs. Extragonadal tumors are slightly more frequent among girls, but the sex ratio is approximately equal.4 Testicular teratomas make up the majority of GCTs in neonates. The incidence of GCTs peaks first within the first 3 years of age, reflecting the incidence of sacrococcygeal tumors and yolk sac tumors (YST).1 At the onset of puberty, the gonadal tumors of adolescent and adult type account for the second peak in incidence.5 Reports on pediatric GCTs in general differentiate the pediatric population as opposed to the adult population based on age, that is, 15 years, whereas the differences in pathogenesis in the 2 populations most likely are based on the impact of the endocrinological changes occurring in and after puberty. Survival rates of childhood GCTs have markedly improved in the last decades to >70% to 80% after introduction of platinum-based chemotherapy.3 Current chemotherapy protocols take into account age, site, histology, stage, completeness of surgical resection, and treatment response monitored by the tumor markers afetoprotein (AFP), human choriogonadotropin (b-HCG), and appropriate scans.6 In general, gonadal encapsulated tumors can be completely resected, whereas larger gonadal or extragonadal tumors may require neoadjuvant or preoperative chemotherapy. Specific protocols are used for management of intracranial GCTs. Complete resection may be associated with significant morbidity; hence, radiotherapy is part of treatment for some intracranial tumors, but long-term neuropsychological and endocrine sequelae are observed.5,7 With regard to the clinical response to chemotherapeutic regimens, a number of studies have attempted to identify molecular predictive factors, for example, hypermethylation of tumor suppressor genes in some aggressive YSTs,8 or variants in genes coding for transport proteins and receptors associated with ototoxicity of cisplatinum treatment.9 The knowledge about etiology and pathogenesis of GCTs in children is still limited, partly because the malignancy is rare but also because research has focused on GCTs of adults. It has been hypothesized that GCTs originate from primitive germ cells that failed to differentiate further in fetal life and that GCTs in adults originate from a different progenitor and harbor different cytogenetic traits as compared with GCTs in children.10 The developmental stage of origin of the pediatric neoplasms may arise earlier than in adults because of predisposing genetic factors. A question of interest has been whether the histologic phenotype is explained by the differentiation stage of the early germ cells that have undergone neoplastic transformation. In this article, we review the current knowledge concerning the pathogenesis of pediatric GCTs in relation to early development and maturation stages of germ cells. www.jpho-online.com |
263
J Pediatr Hematol Oncol
Mosbech et al
Volume 36, Number 4, May 2014
Tumours” AND “Origin” OR “Etiology” OR “Pathogenesis” OR “Biology” (129 results); “Germ Cell Tumors” OR “Germ Cell Tumours” AND “Cytogenetic” OR “Marker” (74 results). Articles were included based on their relevance to the origin of GCTs in children and the most recent published articles were of highest priority.
DEVELOPMENT AND DIFFERENTIATION OF GERM CELLS DURING EMBRYOGENESIS
FIGURE 1. Relative incidence of GCTs in children below 15 years of age according to site of origin is marked by the size of the circular diagrams: sacrococcygeal region (35%), ovary (25%), testis (20%), central nervous system (CNS; 5%), mediastinum (5%), retroperitoneum (5%), head/neck (3%), and vagina (2%). Relative incidence of histologic subtype is marked for each site of origin: Seminoma/germinoma is marked in red (SE/GER), mature teratoma in turquoise (MT), immature teratoma in yellow (IT), embryonal carcinoma in orange (EC), yolk sac tumor in purple (YST), and mixed GCTs in green (MIX). Modified from Pinkerton et al1 with permission from Elsevier Adaptations are themselves works protected by copyright. So in order to publish this adaptation, authorization must be obtained both from the owner of the copyright in the original work and from the owner of copyright in the translation or adaptation.
METHODS OF THE REVIEW The search for relevant articles was conducted through the PubMed Database using the MeSH function to specify the search criteria. The search was performed in May 2013 by C.H.M. with the following search profile: “Neoplasms, Germ Cell, and Embryonal/Etiology” (Majr) AND “Infant” (MeSH Terms) OR “Child” (MeSH Terms) OR “Adolescent” (MeSH Terms) (9539 results). Subsequent search criteria were added with increasing specificity: “Germ Cell Tumors” OR “Germ Cell Tumours” (374 results); “Germ Cell Tumors” OR “Germ Cell
264 | www.jpho-online.com
Understanding of human embryogenesis and stages of germ cell development is important to recognize the cell of origin of GCTs. A brief summary of germ cell differentiation and maturation is therefore presented here. During the third gestational week formation of the 3 germ layers occurs and the primordial germ cells (PGC) arise from the extraembryonic tissue in the epiblast, where they initially were pluripotent. PGCs can be identified by staining for enzyme alkaline phosphatase or OCT-3/4, a transcription factor characteristic for PGCs and embryonic stem cells.11,12 PGCs do not undergo differentiation in the epiblast, although it has been shown that PGCs undergo methylation, leading to loss of pluripotency. During migration into the gonadal ridge methylation is removed and the cells reenter their pluripotent state.13 At the sixth gestational week PGCs migrate from the yolk sac wall in the midline into the gonadal ridge, aided by folding of the embryo, chemotactic factors, and ameboid movements.12 This migration pattern is controlled by several proteins including KIT/KIT ligand (stem cell factor) signaling pathway and E-cadherin, wherein blocking of E-cadherin may affect PGCs interaction, and possibly their migration, leading to ectopic PGCs.11 A role for microRNAs in regulation of PGC formation and survival has also been suggested.14 Before the sixth gestational week fetal gonads are found in a sexually indifferent stage. Once PGCs colonize the primitive gonads, sex-specific development begins. The coelomic epithelium develops into granulosa cells in females and Sertoli cells in males, mesenchyme into theca and Leydig cells, and PGCs into gonocytes and oogonia/ oocytes, respectively. During normal development the oocytes in the ovaries enter first meiotic division, which specifically is inhibited in the testes until puberty.11 Gonocytes in the testis continue to express pluripotency genes, which are downregulated within the fetal period and early infancy, and thereafter the gonocytes mature to prespermatogonia (also called prospermatogonia). If this process is delayed, for example, in cases of gonadal dysgenesis, the immature gonocytes expressing pluripotency genes may persist after the infantile period. The morphologic subtype of GCTs reflects the differentiation stage of the progenitor cells (summarized in Fig. 2, modified from1,15–17), which in turn, is influenced by interaction of molecular, genetic, and environmental factors.1,18
CLASSIFICATION AND CHARACTERISTICS OF TUMOR SUBTYPES The WHO classification of GCTs distinguishes mature teratomas from immature teratomas, as well as teratomas with malignant transformation. Several other classification systems are based on the original description of endodermal sinus tumors by Gunnar Teilum. He was the first to describe embryonal carcinomas as tumors of totipotential r
2014 Lippincott Williams & Wilkins
J Pediatr Hematol Oncol
Volume 36, Number 4, May 2014
Etiology and Pathogenesis of Pediatric Germ Cell Tumors
FIGURE 2. Marker profile of main histologic subtypes of pediatric GCTs. Staining intensity: + + = expression in the majority of tumor cells; ± = expression in few cells; = no expression. *Carcinoma in situ is the precursor stage in adult testicular GCTs; #gonadoblastoma is associated with dysgenic gonads, no precursor has been found in the pediatric GCTs. AFP indicates a-fetoprotein; b-HCG, b-human choriogonadotropin; CC, choriocarcinoma; CIS, carcinoma in situ; DG, dysgerminoma; Ger, germinoma; IT, immature teratoma; MT, mature teratoma; NA, not analyzed; Sem, seminoma; YST, yolk sac tumor.1,15–17
cells that may give rise to either embryonal neoplasms, with the potential to differentiate into derivatives of all 3 germ layers (teratoma), or extraembryonal neoplasms (YST and choriocarcinoma).15 A classification proposed by Oosterhuis and Looijenga11 separates the GCTs into 5 groups according to the phenotype of the tumor as well as the anatomic localization, originating cell, age group, and epidemiology. Tumors found in neonates and children, which are typically YSTs or benign teratomas, are classified as type I GCTs. Type II GCT’s include germinomas/seminomas/dysgerminomas, embryonal carcinomas, teratomas, choriocarcinomas, and YST in adolescents and young adults, whereas spermatocytic seminomas of older men are distinguished as type III GCTs. Characteristics of the various types are summarized below and schematically depicted in Figure 1. Mature teratomas most frequently occur in the ovary or at extragonadal sites, like the sacrococcygeal region. This is the most common subtype of childhood GCT.2 The teratomas are well differentiated and may contain tissue derived from the 3 germ layers (endodermal, mesodermal, and ectodermal layer).19 Immature teratomas also contain tissue from all 3 germ layers but are poorly differentiated. These teratomas are graded I to III based on the amount of immature neural tissue found in the tumor. High-graded tumors are more likely to have foci of YST. The immature teratomas occur primarily at extragonadal sites and in the ovaries of adolescent girls.10,19 Testicular seminomas and ovarian dysgerminomas most frequently occur in young adults. If they are found in younger children they are typically in association with gonadal dysgenesis.2 They arise from cells that remain in the pluripotent state with morphologic features of undifferentiated germ cells. In extragonadal sites these tumors are named germinomas r
2014 Lippincott Williams & Wilkins
and are the most frequent GCT subtype in the central nervous system, where they are often located in the pineal and suprasellar regions.20 Embryonal carcinomas consist of immature totipotent cells resulting presumably from embryonic reprogramming of germ cells. Tumor cells may exceptionally replicate the structure of an early embryo, forming embryoid bodies, which include germ cell discs and miniature amniotic cavities.15 Embryonal carcinomas often differentiate into or are accompanied by YST and choriocarcinoma, the latter composed of cytotrophoblasts and syncytiotrophoblasts.2,5 YSTs differentiate into structures analogous with the yolk sac. Both choriocarcinomas and YSTs result from an abnormal extraembryonic differentiation.2 Pediatric gonadal tumors, which are not derived from germ cells, are even more rare than GCTs. Granulosa cell tumors, Sertoli cell tumors, and Sertoli-Leydig cell tumors have a sex cord-stromal origin, whereas Leydig cell tumors are derived from androgen-producing interstitial cells.21 These somatic gonadal tumors are not a focus of this review and will not be further discussed.
GCT PROTEIN MARKERS AND GENE EXPRESSION PROFILES Certain proteins secreted by tumors are measurable in serum and can be used in diagnosis and monitoring of treatment of some GCT types. AFP is an a-globulin that has a high serum level at birth. At 8 to 12 months of age the concentration falls to adult levels. YSTs are the main tumor type producing AFP, but AFP can also be secreted at lower levels by some embryonal carcinomas and immature teratomas.5 HCG is a glycoprotein produced by the placenta. To avoid cross-reactivity with other hormones, www.jpho-online.com |
265
Mosbech et al
the b-subunit is typically measured. b-HCG is secreted by choriocarcinomas and at lower levels by germinomas and embryonal carcinoma, indicating the presence of syncytiotrophoblastic giant cells.5 The carbohydrate antigen CA125 is a commonly used marker of ovarian cancer in adults. During gestation, serum levels are high in the fetus, but in children only a few cases of immature teratomas with elevated serum levels of CA-125 have been reported.22 As far as the cellular markers are concerned, only few studies of pediatric GCT gene expression profiles have been published.18,23–26 High expression of several stem cell– related factors is a hallmark of most GCTs that occur in adolescents and adults, which—except spermatocytic seminomas—originate from the precursor stage carcinoma in situ (CIS).27 These proteins have been used as immunohistochemical markers (see examples in Fig. 3). We list here a few markers, which we have previously studied and which are most commonly used in the clinical practice for differential diagnosis of the tumors: KIT, OCT-3/4, NANOG, SOX2, AP-2 g, and GATA4. KIT (c-kit) is a tyrosine kinase receptor for stem cell factor (also called KIT ligand [KITLG]) and one of the factors traditionally described as proto-oncogenes. KIT is expressed in the preinvasive stage of testicular tumors, CIS, and in adult and pediatric germinomas, dysgerminomas, and seminomas.28,29 A prolonged or increased expression of KIT may be linked to the neoplastic transformation of
J Pediatr Hematol Oncol
Volume 36, Number 4, May 2014
germ cells by increasing their survival. KIT/KITLG is important for PGCs migration and homing during embryogenesis, and an abnormally high expression of KIT was suggested as a possible mechanism of survival of ectopic PGCs in the central nervous system.17 OCT-3/4, NANOG, and SOX2 are 3 prototypic embryonic pluripotency-maintaining factors. OCT-3/4 (POU5F1) is a member of the POU family of transcription factors expressed in human embryonic stem cells and germ cells. The protein has a function in regulating and maintaining pluripotency during normal development of the embryo and germ cells. It is detected in CIS and gonadoblastoma, as well as most of GCTs (germinoma/dysgerminoma/seminoma and embryonal carcinoma) except teratomas and extraembryonic-type tumors. In the fetal ovaries OCT-3/4 is silenced at the first meiotic prophase, whereas in the testis downregulation of OCT-3/4 is a gradual process following the differentiation of gonocytes.30,31 NANOG is also a key regulator of self-renewal and pluripotency, and its expression prevents differentiation of embryonic stem cells. Downregulation of NANOG in fetal testes has been seen to occur somewhat earlier than OCT-3/4, suggesting that NANOG acts upstream to OCT-3/4.32 In adult type GCTs, NANOG is expressed in CIS, seminomas, dysgerminomas, and embryonal carcinomas. SOX2 is suggested to act in collaboration with NANOG and OCT-3/4 in promoter
FIGURE 3. Examples of AFP, GATA4, SOX2, and AP-2g expression in pediatric GCTs. A, AFP staining of yolk sac tumor in the ovary of a 15-year-old girl. B, GATA4 staining of granulosa cell tumor in the testis of 1-month-old baby boy. C, SOX2 staining of an immature teratoma in the testis of a 3-month-old baby boy. D, AP-2g staining in a dysgerminoma located in the pelvis of a 13-year-old girl, scalebar 50 mm.
266 | www.jpho-online.com
r
2014 Lippincott Williams & Wilkins
J Pediatr Hematol Oncol
Volume 36, Number 4, May 2014
regions of genes that encode 2 types of transcription factors: one that activates expression of those that maintain pluripotency and the one that inhibits expression of those that stimulate differentiation of the germ cells. CIS cells express SOX2 only at the transcriptional level; in addition, GCTs with a germ cell phenotype (seminomas and dysgerminomas) do not express the SOX2 protein. In contrast, embryonal carcinomas and immature teratomas highly express SOX2.33,34 AP-2g (TFAP2C) is a transcription factor regulated by BLIMP1 (PRDM gene family), and both factors are expressed in CIS and seminomas in adult testis, and suggested to have a function in repressing somatic differentiation in germ cells.35,36 In addition to the above-listed pluripotency-related tumor markers, SALL4, a transcription factor involved in the stemness maintenance in various embryonic tissue types and self-renewal in several types of cancers, has been reported as an excellent marker of YST and other GCT components, including even some expression seen in pediatric teratomas.37 Another transcription factor related to embryonic germ cell differentiation is GATA4, which is variably expressed in a subset of CIS, seminomas, dysgerminomas, and YST. Furthermore, it is an excellent marker of granulosa and Sertoli cell tumors. In tumor cells the varying expression of GATA4 in differentiated and nondifferentiated (pluripotent) tumor cells may be explained by cell-specific function in germ cells and somatic cells.38 After many years of studying expression of genes oneby-one at the protein level, the advent of high throughput array methods allowed genome-wide evaluation of entire transcriptome. Several such studies have been performed in adolescent and adult GCTs (reviewed in Alagaratnam et al 201139). To our knowledge, only 1 study evaluated the genome-wide transcriptome of pediatric GCTs, and the data identified some differences in transcript profiles of childhood YST compared with adult counterparts.40 Other studies have focused on selected pathways, especially those involved in early embryonic development and which have also been implicated in cancer. The WNT/b-catenin signaling pathway displays differences in distinct subtypes of GCT, with particularly strong expression of b-catenin in YST and immature teratomas.41 Similarly, multiple components of the TGFb/BMP (bone morphogenetic protein) signaling pathway, including inhibitory SMAD6 and SMAD7 were differentially expressed in childhood germinomas versus YST.42 Profiling of microRNA (miRNA) expression in this pathway revealed miR-155 was most highly expressed in germinomas, suggesting that miR-155 may inhibit BMP signaling in the tumor subtype.42 In contrast, numerous miRNA were significantly higher expressed in YST, including some miRNAs previously identified in this tumor type by global miRNA profiling.43 When the pediatric GCT data were compared with data from adult GCTs from previous work, the RNA and miRNA profiles differed with age (pediatric vs. adult). Germinomas showed an undifferentiated and pluripotent phenotype, overexpressing the embryonal stem cell markers NANOG, OCT-3/4, and UTF1, whereas YSTs displayed extraembryonic differentiation maintaining a proliferative phenotype. This significant difference between miRNA profiles in germinomas and nongerminomas has also been shown in pediatric tumors localized in the central nervous system.44 It opens up for use in future clinical diagnosis and treatment, as the differential miRNA expression (especially r
2014 Lippincott Williams & Wilkins
Etiology and Pathogenesis of Pediatric Germ Cell Tumors
miR-302) is likely to contribute to the relatively aggressive behavior of YST.43 The miR-371-373 is involved in maintaining the pluripotent state, whereas miR-302 is induced during the first stages of in vitro differentiation. The miR302 levels are high in extraembryonic differentiated tumors like YSTs, whereas virtually undetectable in somatically differentiated teratomas. The let-7 family of microRNA tumor suppressor genes is significantly downregulated in pediatric GCTs irrespective of patient’s age, tumor localization, and histology. LIN28 has been shown to downregulate let-7, resulting in expression of a number of oncogenes.45
GENETIC ABERRATIONS, GENE POLYMORPHISMS, AND EPIGENETIC FEATURES The cytogenetic features in GCTs differ depending on age and histology. Virtually, all pediatric teratomas have a normal karyotype.19,46 In contrast, GCTs of adults and adolescents, including teratomas, very consistently have structural abnormalities involving the chromosome 12p sequence; many tumors have an isochromosome 12p, others have large regional gains or amplifications of 12p.11,47 Palmer et al’s48 data showed gain of 12p in more than half of the cases aged 5 to 16 years and in only 4 of 14 cases of YST in children younger than 5 years of age. GCTs without 12p gain more often had a loss on 16p. One of the most frequent aberrations in childhood GCTs is the deletion of 1p, although there is a significant variance among different studies. Zahn et al49 found deletions of 1p in 8/9 pediatric malignant GCTs but not in adult GCTs. Bussey et al47 showed that deletion of chromosome 1p and gain of 1q and 3 were the most frequent abnormalities among the pediatric GCTs from both sexes. Palmer et al48 have shown that loss of 1p, 4q, 6q and gain of 3p is more frequent in YSTs of children as compared with germinomas. Loss of heterozygosity (LOH) of a number of candidate loci has been investigated in pediatric GCTs but very few were identified. An LOH in 5q22 encompassing the ACP gene, which is often mutated in colon neoplasms, was reported in a small series of YSTs and teratomas but mutations have not been identified.50 In recent genome-wide association studies of adult GCTs, susceptibility loci near KITLG, SPRY4, DMRT1, and BAK1 have been identified.51–53 The loci play a role in the survival of PGCs during migration (KIT-KITLG), inhibiting MAP kinase pathway downstream to KIT (SPRY4), promotion of apoptosis (BAK1), and regulation of sexually dimorphic meiotic entry (DMRT1). In their replication study of childhood GCTs (including adolescents), Poynter et al54 suggested that variants in KITLG and SRPY4 are susceptibility alleles only for GCTs of adolescents and young adults, whereas BAK1 may be a susceptibility allele for childhood GCTs, including neonates, young children, and adolescents. Analyses of the imprinting status of GCTs have supported that gonadal and extragonadal GCTs share a common cell of origin, the PGC, but the imprinting varies according to the stage of development of the germ cell.55–57 Schneider et al55 studied DNA methylation of CpG nucleotides in H19, IGF-2, and SNRPN, and the results differed from what was found in adult testicular GCTs. The pediatric GCTs originated from PGCs that showed consistent loss of imprinting of SNRPN and partly loss of imprinting of H19 and IGF-2, suggesting that pediatric www.jpho-online.com |
267
J Pediatr Hematol Oncol
Mosbech et al
Volume 36, Number 4, May 2014
TABLE 1. Molecular Characteristics of Pediatric GCTs
Cytogenetic Aberrations DG SEM GER
i12p (adolescents)
YST
Loss of 1p 4q 6q Gain of 3p 3 1q 2p 13 20q No characteristic alterations
MT IT
IHC Markers
Gene Variants (SNPs)
HCG OCT-3/4 NANOG KIT GATA4 SALL4 AFP GATA4 SALL4
BAK1* (children) KITLG SPRY4 (adolescents)
AFP SOX2
None studied
miRNA Under-expression of miR-200 miR-205 let-7 Over-expression of miR-371-373 miR-142-3/5p miR-146a Under-expression of let-7 Over-expression of miR-371-373 miR-302 miR-122
BAK1* (children) KITLG SPRY4 (adolescents)
No characteristic alterations
For study marked with *there is no distinction of the childhood tumors into histologic subtypes, and DG/SEM/GER and YST are rated collectively.5,23–29,33,37,38,40,42–49,51,54 DG indicates dysgerminoma; GER, germinoma; IHC, immunohistochemistry; IT, immature teratoma; MT, mature teratoma; SEM, seminoma; SNP, single nucleotide polymorphism; YST, yolk sac tumor.
GCTs arise from a different germ cell developmental stage. Molecular characteristics are summarized in Table 1. Extragonadal GCTs develop from a germ cell that has migrated aberrantly during embryogenesis, but it seems that both gonadal and extragonadal GCTs arise from PGCs that have erased their imprinting. The patterns of erasure are different in adults and children; therefore, the development of testicular GCTs in adults occurs at a later stage. Hoffner et al58 have suggested that extragonadal and testicular GCTs in children originate from a mitotic division of a premeiotic germ cell or a pluripotent somatic cell based on centromeric and DNA polymorphism analysis. In contrast, they suggested that ovarian and extragonadal teratomas may arise from germ cells more advanced in their development, because of errors in first meiotic division, second meiotic division, or endoreduplication.
CONCLUDING REMARKS Pediatric GCT origin differs from adult GCTs with regard to patterns of histology, cytogenetics, and absence of a preinvasive stage.18 It is generally accepted that adolescent and adult testicular GCTs, both seminomas and nonseminomas, but not the spermatocytic seminomas, originate from the precursor stage CIS.27 Visfeldt et al59 and Jorgensen et al18 were the first to investigate a possible association between CIS and pediatric testicular GCTs and found no CIS in infantile YSTs and teratomas. In contrast, 5 of 6 cases of adolescent GCTs showed presence of CIS cells, suggesting a common origin with adult GCTs.18 It is important to stress that the presence of CIS markers in the infantile testes does not equal the presence of CIS, because at this age some germ cells may still be at the stage of fetal gonocytes that are morphologically very similar to CIS cells.60,61 The current consensus concerning the pathogenesis of different GCTs stipulates the origin from the same cell type (germ cell) but at different developmental stages. As outlined above, this has been investigated by use of tumor markers, cytogenetic analyses, imprinting status,
268 | www.jpho-online.com
microRNA, and mRNA expression. The pluripotency of the progenitor cell may result in extraembryonically differentiated GCTs with choriocarcinoma and yolk sac elements maintaining a proliferative phenotype. The differentiation of GCTs may be a recapitulation of embryonal somatic differentiation, resulting in mature and immature tissue from all 3 germ cell layers (teratomas). Lastly, the pluripotency state may be maintained in a subset of immature germ cells, with prolonged expression of selfrenewal and pluripotency regulators (eg, OCT-3/4, NANOG, and AP-2g) but without any somatic differentiation, resulting in development of seminomas, dysgerminomas, and germinomas. Exact causative factors involved in initiation of GCTs are not known, but the etiology of these tumors most likely involves complex interplay of genetic mutations, predisposing polymorphisms, and, possibly, also environmental damaging influences. Research has been hampered by the low incidence of GCTs, and multicenter studies to obtain more valid data should be supported.
REFERENCES 1. Pinkerton CR. Malignant germ cell tumours in childhood. Eur J Cancer. 1997;33:895–901. 2. Horton Z, Schlatter M, Schultz S. Pediatric germ cell tumors. Surg Oncol. 2007;16:205–213. 3. Imbach P, Ku¨hne T, Arceci RJ. Pediatric Oncology: A Comprehensive Guide. 2nd ed. Berlin: Springer; 2011. 4. Schneider DT, Calaminus G, Koch S, et al. Epidemiologic analysis of 1442 children and adolescents registered in the German germ cell tumor protocols. Pediatr Blood Cancer. 2004;42:169–175. 5. Gobel U, Calaminus G, Schneider DT, et al. Management of germ cell tumors in children: approaches to cure. Onkologie. 2002;25:14–22. 6. Mann JR, Raafat F, Robinson K, et al. The United Kingdom Children’s Cancer Study Group’s second germ cell tumor study: carboplatin, etoposide, and bleomycin are effective treatment for children with malignant extracranial germ r
2014 Lippincott Williams & Wilkins
J Pediatr Hematol Oncol
7.
8.
9.
10. 11. 12. 13. 14. 15. 16. 17.
18.
19. 20. 21. 22.
23. 24. 25. 26. 27. 28.
29.
r
Volume 36, Number 4, May 2014
cell tumors, with acceptable toxicity. J Clin Oncol. 2000;18:3809–3818. Kiltie AE, Gattamaneni HR. Survival and quality of life of paediatric intracranial germ cell tumour patients treated at the Christie Hospital, 1972-1993. Med Pediatr Oncol. 1995; 25:450–456. Jeyapalan JN, Noor DA, Lee SH, et al. Methylator phenotype of malignant germ cell tumours in children identifies strong candidates for chemotherapy resistance. Br J Cancer. 2011;105: 575–585. Pussegoda K, Ross CJ, Visscher H, et al. Replication of TPMT and ABCC3 genetic variants highly associated with cisplatininduced hearing loss in children. Clin Pharmacol Ther. 2013; 94:243–251. Veltman IM, Schepens MT, Looijenga LH, et al. Germ cell tumours in neonates and infants: a distinct subgroup? APMIS. 2003;111:152–160; discussion 160. Oosterhuis JW, Looijenga LH. Testicular germ-cell tumours in a broader perspective. Nat Rev Cancer. 2005;5:210–222. Wylie CC. The biology of primordial germ cells. Eur Urol. 1993;23:62–66. Hajkova P, Erhardt S, Lane N, et al. Epigenetic reprogramming in mouse primordial germ cells. Mech Dev. 2002;117:15–23. Eini R, Dorssers LC, Looijenga LH. Role of stem cell proteins and microRNAs in embryogenesis and germ cell cancer. Int J Dev Biol. 2013;57:319–332. Teilum G. Classification of endodermal sinus tumour (mesoblatoma vitellinum) and so-called “embryonal carcinoma” of the ovary. Acta Pathol Microbiol Scand. 1965;64:407–429. Rescorla FJ. Pediatric germ cell tumors. Semin Pediatr Surg. 2012;21:51–60. Hoei-Hansen CE, Sehested A, Juhler M, et al. New evidence for the origin of intracranial germ cell tumours from primordial germ cells: expression of pluripotency and cell differentiation markers. J Pathol. 2006;209:25–33. Jorgensen N, Muller J, Giwercman A, et al. DNA content and expression of tumour markers in germ cells adjacent to germ cell tumours in childhood: probably a different origin for infantile and adolescent germ cell tumours. J Pathol. 1995;176:269–278. Harms D, Zahn S, Gobel U, et al. Pathology and molecular biology of teratomas in childhood and adolescence. Klin Padiatr. 2006;218:296–302. Echevarria ME, Fangusaro J, Goldman S. Pediatric central nervous system germ cell tumors: a review. Oncologist. 2008;13:690–699. Borer JG, Tan PE, Diamond DA. The spectrum of Sertoli cell tumors in children. Urol Clin North Am. 2000;27:529–541. Lahdenne P, Pitkanen S, Rajantie J, et al. Tumor markers CA 125 and CA 19-9 in cord blood and during infancy: developmental changes and use in pediatric germ cell tumors. Pediatr Res. 1995;38:797–801. Talebagha S, Rizk C, Elawabdeh N, et al. Usefulness of OCT4/ 3 immunostain in pediatric malignant germ cell tumors. Fetal Pediatr Pathol. 2013;32:82–87. Ho DM, Liu HC. Primary intracranial germ cell tumor. Pathologic study of 51 patients. Cancer. 1992;70:1577–1584. Siltanen S, Anttonen M, Heikkila P, et al. Transcription factor GATA-4 is expressed in pediatric yolk sac tumors. Am J Pathol. 1999;155:1823–1829. Perlman EJ, Hawkins EP. Pediatric germ cell tumors: protocol update for pathologists. Pediatr Dev Pathol. 1998;1: 328–335. Skakkebaek NE. Possible carcinoma-in-situ of the testis. Lancet. 1972;2:516–517. Hoei-Hansen CE, Kraggerud SM, Abeler VM, et al. Ovarian dysgerminomas are characterised by frequent KIT mutations and abundant expression of pluripotency markers. Mol Cancer. 2007;6:12. Chou PM, Barquin N, Guinan P, et al. Differential expression of p53, c-kit, and CD34 in prepubertal and postpubertal testicular germ cell tumors. Cancer. 1997;79:2430–2434.
2014 Lippincott Williams & Wilkins
Etiology and Pathogenesis of Pediatric Germ Cell Tumors
30. Looijenga LH, Stoop H, de Leeuw HP, et al. POU5F1 (OCT3/ 4) identifies cells with pluripotent potential in human germ cell tumors. Cancer Res. 2003;63:2244–2250. 31. Rajpert-De Meyts E, Hanstein R, Jorgensen N, et al. Developmental expression of POU5F1 (OCT-3/4) in normal and dysgenetic human gonads. Hum Reprod. 2004;19:1338–1344. 32. Hoei-Hansen CE, Almstrup K, Nielsen JE, et al. Stem cell pluripotency factor NANOG is expressed in human fetal gonocytes, testicular carcinoma in situ and germ cell tumours. Histopathology. 2005;47:48–56. 33. Sonne SB, Perrett RM, Nielsen JE, et al. Analysis of SOX2 expression in developing human testis and germ cell neoplasia. Int J Dev Biol. 2010;54:755–760. 34. Perrett RM, Turnpenny L, Eckert JJ, et al. The early human germ cell lineage does not express SOX2 during in vivo development or upon in vitro culture. Biol Reprod. 2008;78:852–858. 35. Schafer S, Anschlag J, Nettersheim D, et al. The role of BLIMP1 and its putative downstream target TFAP2C in germ cell development and germ cell tumours. Int J Androl. 2011;34:e152–e158. 36. Hoei-Hansen CE, Nielsen JE, Almstrup K, et al. Transcription factor AP-2gamma is a developmentally regulated marker of testicular carcinoma in situ and germ cell tumors. Clin Cancer Res. 2004;10:8521–8530. 37. Cao D, Li J, Guo CC, et al. SALL4 is a novel diagnostic marker for testicular germ cell tumors. Am J Surg Pathol. 2009;33:1065–1077. 38. Salonen J, Rajpert-De Meyts E, Mannisto S, et al. Differential developmental expression of transcription factors GATA-4 and GATA-6, their cofactor FOG-2 and downstream target genes in testicular carcinoma in situ and germ cell tumors. Eur J Endocrinol. 2010;162:625–631. 39. Alagaratnam S, Lind GE, Kraggerud SM, et al. The testicular germ cell tumour transcriptome. Int J Androl. 2011;34:e133–e150. 40. Palmer RD, Barbosa-Morais NL, Gooding EL, et al. Pediatric malignant germ cell tumors show characteristic transcriptome profiles. Cancer Res. 2008;68:4239–4247. 41. Fritsch MK, Schneider DT, Schuster AE, et al. Activation of Wnt/beta-catenin signaling in distinct histologic subtypes of human germ cell tumors. Pediatr Dev Pathol. 2006;9:115–131. 42. Fustino N, Rakheja D, Ateek CS, et al. Bone morphogenetic protein signalling activity distinguishes histological subsets of paediatric germ cell tumours. Int J Androl. 2011;34:e218–e233. 43. Murray MJ, Saini HK, van Dongen S, et al. The two most common histological subtypes of malignant germ cell tumour are distinguished by global microRNA profiles, associated with differential transcription factor expression. Mol Cancer. 2010;9:290. 44. Wang HW, Wu YH, Hsieh JY, et al. Pediatric primary central nervous system germ cell tumors of different prognosis groups show characteristic miRNome traits and chromosome copy number variations. BMC Genomics. 2010;11:132. 45. Murray MJ, Saini HK, Siegler CA, et al. LIN28 expression in malignant germ cell tumors downregulates let-7 and increases oncogene levels. Cancer Res. 2013;73:4872–4884. 46. Schneider DT, Schuster AE, Fritsch MK, et al. Genetic analysis of childhood germ cell tumors with comparative genomic hybridization. Klin Padiatr. 2001;213:204–211. 47. Bussey KJ, Lawce HJ, Olson SB, et al. Chromosome abnormalities of eighty-one pediatric germ cell tumors: sex-, age-, site-, and histopathology-related differences—a Children’s Cancer Group study. Genes Chromosomes Cancer. 1999;25:134–146. 48. Palmer RD, Foster NA, Vowler SL, et al. Malignant germ cell tumours of childhood: new associations of genomic imbalance. Br J Cancer. 2007;96:667–676. 49. Zahn S, Sievers S, Alemazkour K, et al. Imbalances of chromosome arm 1p in pediatric and adult germ cell tumors are caused by true allelic loss: a combined comparative genomic hybridization and microsatellite analysis. Genes Chromosomes Cancer. 2006;45:995–1006.
www.jpho-online.com |
269
Mosbech et al
J Pediatr Hematol Oncol
50. Okpanyi V, Schneider DT, Zahn S, et al. Analysis of the adenomatous polyposis coli (APC) gene in childhood and adolescent germ cell tumors. Pediatr Blood Cancer. 2011;56: 384–391. 51. Rapley EA, Turnbull C, Al Olama AA, et al. A genome-wide association study of testicular germ cell tumor. Nat Genet. 2009;41:807–810. 52. Kanetsky PA, Mitra N, Vardhanabhuti S, et al. Common variation in KITLG and at 5q31.3 predisposes to testicular germ cell cancer. Nat Genet. 2009;41:811–815. 53. Turnbull C, Rapley EA, Seal S, et al. Variants near DMRT1, TERT and ATF7IP are associated with testicular germ cell cancer. Nat Genet. 2010;42:604–607. 54. Poynter JN, Hooten AJ, Frazier AL, et al. Associations between variants in KITLG, SPRY4, BAK1, and DMRT1 and pediatric germ cell tumors. Genes Chromosomes Cancer. 2012;51:266–271. 55. Schneider DT, Schuster AE, Fritsch MK, et al. Multipoint imprinting analysis indicates a common precursor cell for gonadal and nongonadal pediatric germ cell tumors. Cancer Res. 2001;61:7268–7276. 56. Sievers S, Alemazkour K, Zahn S, 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;44:256–264. Bussey KJ, Lawce HJ, Himoe E, et al. SNRPN methylation patterns in germ cell tumors as a reflection of primordial germ cell development. Genes Chromosomes Cancer. 2001;32: 342–352. Hoffner L, Deka R, Chakravarti A, et al. Cytogenetics and origins of pediatric germ cell tumors. Cancer Genet Cytogenet. 1994;74:54–58. Visfeldt J, Jorgensen N, Muller J, et al. Testicular germ cell tumours of childhood in Denmark, 1943-1989: incidence and evaluation of histology using immunohistochemical techniques. J Pathol. 1994;174:39–47. Jorgensen N, Giwercman A, Muller J, et al. Immunohistochemical markers of carcinoma in situ of the testis also expressed in normal infantile germ cells. Histopathology. 1993;22:373–378. Honecker F, Stoop H, de Krijger RR, et al. Pathobiological implications of the expression of markers of testicular carcinoma in situ by fetal germ cells. J Pathol. 2004;203: 849–857.
270 | www.jpho-online.com
57.
58. 59.
60.
61.
r
Volume 36, Number 4, May 2014
2014 Lippincott Williams & Wilkins
Recent Advances in Understanding the Etiology and Pathogenesis of Pediatric Germ Cell Tumors Christiane H. Mosbech, BSc,* Catherine Rechnitzer, MD, DMSc,w Jesper S. Brok, MD, PhD,w Ewa Rajpert-De Meyts, MD, DMSc,* and Christina E. Hoei-Hansen, MD, DMSc*z
Summary: Pediatric germ cell tumors (GCTs) are rare neoplasms arising predominantly in the gonads and sacrococcygeal, mediastinal, and intracranial localizations. In this article, we review current knowledge of pathogenesis of pediatric GCTs, which differs from adult/adolescent GCTs. One distinctive feature is the absence of a progenitor stage, such as carcinoma in situ or gonadoblastoma, which are seen in adult/adolescent GCTs, except spermatocytic seminoma. The primordial germ cell (PGC) is the suggested origin of all GCTs, with variations in histology reflecting differentiation stage. Expression of pluripotency transcription factors OCT-3/4, NANOG, and AP-2g in germinomas/seminomas/ dysgerminomas is consistent with retaining a germ cell phenotype. Teratomas, in contrast, develop through a pathway of aberrant somatic differentiation of immature germ cells, and the yolk sac tumors and choriocarcinomas result from abnormal extraembryonic differentiation. In pediatric GCTs, origin is suggested at an earlier developmental stage because of predisposing genetic factors, although responsible genes remain largely unknown. Some extragonadal GCTs have been linked to overexpression of the KIT/ KITLG system, allowing for survival of aberrantly migrated ectopic PGCs. Infant gonadal/sacrococcygeal GCTs may be caused by apoptosis-related pathways, consistent with an association with polymorphisms in BAK1. Although recent advances have identified candidate pathways, further effort is needed to answer central questions of pathogenesis of these fascinating tumors. Key Words: germ cell tumor, children, primordial germ cell, pathogenesis
(J Pediatr Hematol Oncol 2014;36:263–270)
G
erm cell tumors (GCT) are a heterogenous group of neoplasms that develop in both sexes at any age. They vary in histology and occur at different anatomic sites. GCTs can be found in the gonads and at extragonadal sites along the midline of the body (Fig. 1). Extragonadal tumors predominate in children younger than 3 years of age, whereas the gonads are the main localization during and after puberty.2 The tumors have been characterized on the basis of histology and site of occurrence, and—more recently—on the pattern of cytogenetic aberrations, gene expression, gene variants, and miRNA expression. Received for publication November 16, 2013; accepted January 15, 2014. From the *Department of Growth and Reproduction; wDepartment of Pediatrics, University Hospital of Rigshospitalet, Copenhagen; and zDepartment of Pediatrics, University Hospital of Hilleroed, Hilleroed, Denmark. Supported by The Danish Cancer Society, The Danish Child Cancer Foundation. The authors declare no conflict of interest. Reprints: Christina E. Hoei-Hansen, MD, DMSc, Department of Pediatrics, University Hospital of Hilleroed, Dyrehavevej 29, DK— 3400 Hilleroed, Denmark (e-mail: [email protected]). Copyright r 2014 by Lippincott Williams & Wilkins
J Pediatr Hematol Oncol
Volume 36, Number 4, May 2014
Each year, 1 of 500 children is diagnosed with cancer worldwide3 with 2.5% being GCTs. Extragonadal tumors are slightly more frequent among girls, but the sex ratio is approximately equal.4 Testicular teratomas make up the majority of GCTs in neonates. The incidence of GCTs peaks first within the first 3 years of age, reflecting the incidence of sacrococcygeal tumors and yolk sac tumors (YST).1 At the onset of puberty, the gonadal tumors of adolescent and adult type account for the second peak in incidence.5 Reports on pediatric GCTs in general differentiate the pediatric population as opposed to the adult population based on age, that is, 15 years, whereas the differences in pathogenesis in the 2 populations most likely are based on the impact of the endocrinological changes occurring in and after puberty. Survival rates of childhood GCTs have markedly improved in the last decades to >70% to 80% after introduction of platinum-based chemotherapy.3 Current chemotherapy protocols take into account age, site, histology, stage, completeness of surgical resection, and treatment response monitored by the tumor markers afetoprotein (AFP), human choriogonadotropin (b-HCG), and appropriate scans.6 In general, gonadal encapsulated tumors can be completely resected, whereas larger gonadal or extragonadal tumors may require neoadjuvant or preoperative chemotherapy. Specific protocols are used for management of intracranial GCTs. Complete resection may be associated with significant morbidity; hence, radiotherapy is part of treatment for some intracranial tumors, but long-term neuropsychological and endocrine sequelae are observed.5,7 With regard to the clinical response to chemotherapeutic regimens, a number of studies have attempted to identify molecular predictive factors, for example, hypermethylation of tumor suppressor genes in some aggressive YSTs,8 or variants in genes coding for transport proteins and receptors associated with ototoxicity of cisplatinum treatment.9 The knowledge about etiology and pathogenesis of GCTs in children is still limited, partly because the malignancy is rare but also because research has focused on GCTs of adults. It has been hypothesized that GCTs originate from primitive germ cells that failed to differentiate further in fetal life and that GCTs in adults originate from a different progenitor and harbor different cytogenetic traits as compared with GCTs in children.10 The developmental stage of origin of the pediatric neoplasms may arise earlier than in adults because of predisposing genetic factors. A question of interest has been whether the histologic phenotype is explained by the differentiation stage of the early germ cells that have undergone neoplastic transformation. In this article, we review the current knowledge concerning the pathogenesis of pediatric GCTs in relation to early development and maturation stages of germ cells. www.jpho-online.com |
263
J Pediatr Hematol Oncol
Mosbech et al
Volume 36, Number 4, May 2014
Tumours” AND “Origin” OR “Etiology” OR “Pathogenesis” OR “Biology” (129 results); “Germ Cell Tumors” OR “Germ Cell Tumours” AND “Cytogenetic” OR “Marker” (74 results). Articles were included based on their relevance to the origin of GCTs in children and the most recent published articles were of highest priority.
DEVELOPMENT AND DIFFERENTIATION OF GERM CELLS DURING EMBRYOGENESIS
FIGURE 1. Relative incidence of GCTs in children below 15 years of age according to site of origin is marked by the size of the circular diagrams: sacrococcygeal region (35%), ovary (25%), testis (20%), central nervous system (CNS; 5%), mediastinum (5%), retroperitoneum (5%), head/neck (3%), and vagina (2%). Relative incidence of histologic subtype is marked for each site of origin: Seminoma/germinoma is marked in red (SE/GER), mature teratoma in turquoise (MT), immature teratoma in yellow (IT), embryonal carcinoma in orange (EC), yolk sac tumor in purple (YST), and mixed GCTs in green (MIX). Modified from Pinkerton et al1 with permission from Elsevier Adaptations are themselves works protected by copyright. So in order to publish this adaptation, authorization must be obtained both from the owner of the copyright in the original work and from the owner of copyright in the translation or adaptation.
METHODS OF THE REVIEW The search for relevant articles was conducted through the PubMed Database using the MeSH function to specify the search criteria. The search was performed in May 2013 by C.H.M. with the following search profile: “Neoplasms, Germ Cell, and Embryonal/Etiology” (Majr) AND “Infant” (MeSH Terms) OR “Child” (MeSH Terms) OR “Adolescent” (MeSH Terms) (9539 results). Subsequent search criteria were added with increasing specificity: “Germ Cell Tumors” OR “Germ Cell Tumours” (374 results); “Germ Cell Tumors” OR “Germ Cell
264 | www.jpho-online.com
Understanding of human embryogenesis and stages of germ cell development is important to recognize the cell of origin of GCTs. A brief summary of germ cell differentiation and maturation is therefore presented here. During the third gestational week formation of the 3 germ layers occurs and the primordial germ cells (PGC) arise from the extraembryonic tissue in the epiblast, where they initially were pluripotent. PGCs can be identified by staining for enzyme alkaline phosphatase or OCT-3/4, a transcription factor characteristic for PGCs and embryonic stem cells.11,12 PGCs do not undergo differentiation in the epiblast, although it has been shown that PGCs undergo methylation, leading to loss of pluripotency. During migration into the gonadal ridge methylation is removed and the cells reenter their pluripotent state.13 At the sixth gestational week PGCs migrate from the yolk sac wall in the midline into the gonadal ridge, aided by folding of the embryo, chemotactic factors, and ameboid movements.12 This migration pattern is controlled by several proteins including KIT/KIT ligand (stem cell factor) signaling pathway and E-cadherin, wherein blocking of E-cadherin may affect PGCs interaction, and possibly their migration, leading to ectopic PGCs.11 A role for microRNAs in regulation of PGC formation and survival has also been suggested.14 Before the sixth gestational week fetal gonads are found in a sexually indifferent stage. Once PGCs colonize the primitive gonads, sex-specific development begins. The coelomic epithelium develops into granulosa cells in females and Sertoli cells in males, mesenchyme into theca and Leydig cells, and PGCs into gonocytes and oogonia/ oocytes, respectively. During normal development the oocytes in the ovaries enter first meiotic division, which specifically is inhibited in the testes until puberty.11 Gonocytes in the testis continue to express pluripotency genes, which are downregulated within the fetal period and early infancy, and thereafter the gonocytes mature to prespermatogonia (also called prospermatogonia). If this process is delayed, for example, in cases of gonadal dysgenesis, the immature gonocytes expressing pluripotency genes may persist after the infantile period. The morphologic subtype of GCTs reflects the differentiation stage of the progenitor cells (summarized in Fig. 2, modified from1,15–17), which in turn, is influenced by interaction of molecular, genetic, and environmental factors.1,18
CLASSIFICATION AND CHARACTERISTICS OF TUMOR SUBTYPES The WHO classification of GCTs distinguishes mature teratomas from immature teratomas, as well as teratomas with malignant transformation. Several other classification systems are based on the original description of endodermal sinus tumors by Gunnar Teilum. He was the first to describe embryonal carcinomas as tumors of totipotential r
2014 Lippincott Williams & Wilkins
J Pediatr Hematol Oncol
Volume 36, Number 4, May 2014
Etiology and Pathogenesis of Pediatric Germ Cell Tumors
FIGURE 2. Marker profile of main histologic subtypes of pediatric GCTs. Staining intensity: + + = expression in the majority of tumor cells; ± = expression in few cells; = no expression. *Carcinoma in situ is the precursor stage in adult testicular GCTs; #gonadoblastoma is associated with dysgenic gonads, no precursor has been found in the pediatric GCTs. AFP indicates a-fetoprotein; b-HCG, b-human choriogonadotropin; CC, choriocarcinoma; CIS, carcinoma in situ; DG, dysgerminoma; Ger, germinoma; IT, immature teratoma; MT, mature teratoma; NA, not analyzed; Sem, seminoma; YST, yolk sac tumor.1,15–17
cells that may give rise to either embryonal neoplasms, with the potential to differentiate into derivatives of all 3 germ layers (teratoma), or extraembryonal neoplasms (YST and choriocarcinoma).15 A classification proposed by Oosterhuis and Looijenga11 separates the GCTs into 5 groups according to the phenotype of the tumor as well as the anatomic localization, originating cell, age group, and epidemiology. Tumors found in neonates and children, which are typically YSTs or benign teratomas, are classified as type I GCTs. Type II GCT’s include germinomas/seminomas/dysgerminomas, embryonal carcinomas, teratomas, choriocarcinomas, and YST in adolescents and young adults, whereas spermatocytic seminomas of older men are distinguished as type III GCTs. Characteristics of the various types are summarized below and schematically depicted in Figure 1. Mature teratomas most frequently occur in the ovary or at extragonadal sites, like the sacrococcygeal region. This is the most common subtype of childhood GCT.2 The teratomas are well differentiated and may contain tissue derived from the 3 germ layers (endodermal, mesodermal, and ectodermal layer).19 Immature teratomas also contain tissue from all 3 germ layers but are poorly differentiated. These teratomas are graded I to III based on the amount of immature neural tissue found in the tumor. High-graded tumors are more likely to have foci of YST. The immature teratomas occur primarily at extragonadal sites and in the ovaries of adolescent girls.10,19 Testicular seminomas and ovarian dysgerminomas most frequently occur in young adults. If they are found in younger children they are typically in association with gonadal dysgenesis.2 They arise from cells that remain in the pluripotent state with morphologic features of undifferentiated germ cells. In extragonadal sites these tumors are named germinomas r
2014 Lippincott Williams & Wilkins
and are the most frequent GCT subtype in the central nervous system, where they are often located in the pineal and suprasellar regions.20 Embryonal carcinomas consist of immature totipotent cells resulting presumably from embryonic reprogramming of germ cells. Tumor cells may exceptionally replicate the structure of an early embryo, forming embryoid bodies, which include germ cell discs and miniature amniotic cavities.15 Embryonal carcinomas often differentiate into or are accompanied by YST and choriocarcinoma, the latter composed of cytotrophoblasts and syncytiotrophoblasts.2,5 YSTs differentiate into structures analogous with the yolk sac. Both choriocarcinomas and YSTs result from an abnormal extraembryonic differentiation.2 Pediatric gonadal tumors, which are not derived from germ cells, are even more rare than GCTs. Granulosa cell tumors, Sertoli cell tumors, and Sertoli-Leydig cell tumors have a sex cord-stromal origin, whereas Leydig cell tumors are derived from androgen-producing interstitial cells.21 These somatic gonadal tumors are not a focus of this review and will not be further discussed.
GCT PROTEIN MARKERS AND GENE EXPRESSION PROFILES Certain proteins secreted by tumors are measurable in serum and can be used in diagnosis and monitoring of treatment of some GCT types. AFP is an a-globulin that has a high serum level at birth. At 8 to 12 months of age the concentration falls to adult levels. YSTs are the main tumor type producing AFP, but AFP can also be secreted at lower levels by some embryonal carcinomas and immature teratomas.5 HCG is a glycoprotein produced by the placenta. To avoid cross-reactivity with other hormones, www.jpho-online.com |
265
Mosbech et al
the b-subunit is typically measured. b-HCG is secreted by choriocarcinomas and at lower levels by germinomas and embryonal carcinoma, indicating the presence of syncytiotrophoblastic giant cells.5 The carbohydrate antigen CA125 is a commonly used marker of ovarian cancer in adults. During gestation, serum levels are high in the fetus, but in children only a few cases of immature teratomas with elevated serum levels of CA-125 have been reported.22 As far as the cellular markers are concerned, only few studies of pediatric GCT gene expression profiles have been published.18,23–26 High expression of several stem cell– related factors is a hallmark of most GCTs that occur in adolescents and adults, which—except spermatocytic seminomas—originate from the precursor stage carcinoma in situ (CIS).27 These proteins have been used as immunohistochemical markers (see examples in Fig. 3). We list here a few markers, which we have previously studied and which are most commonly used in the clinical practice for differential diagnosis of the tumors: KIT, OCT-3/4, NANOG, SOX2, AP-2 g, and GATA4. KIT (c-kit) is a tyrosine kinase receptor for stem cell factor (also called KIT ligand [KITLG]) and one of the factors traditionally described as proto-oncogenes. KIT is expressed in the preinvasive stage of testicular tumors, CIS, and in adult and pediatric germinomas, dysgerminomas, and seminomas.28,29 A prolonged or increased expression of KIT may be linked to the neoplastic transformation of
J Pediatr Hematol Oncol
Volume 36, Number 4, May 2014
germ cells by increasing their survival. KIT/KITLG is important for PGCs migration and homing during embryogenesis, and an abnormally high expression of KIT was suggested as a possible mechanism of survival of ectopic PGCs in the central nervous system.17 OCT-3/4, NANOG, and SOX2 are 3 prototypic embryonic pluripotency-maintaining factors. OCT-3/4 (POU5F1) is a member of the POU family of transcription factors expressed in human embryonic stem cells and germ cells. The protein has a function in regulating and maintaining pluripotency during normal development of the embryo and germ cells. It is detected in CIS and gonadoblastoma, as well as most of GCTs (germinoma/dysgerminoma/seminoma and embryonal carcinoma) except teratomas and extraembryonic-type tumors. In the fetal ovaries OCT-3/4 is silenced at the first meiotic prophase, whereas in the testis downregulation of OCT-3/4 is a gradual process following the differentiation of gonocytes.30,31 NANOG is also a key regulator of self-renewal and pluripotency, and its expression prevents differentiation of embryonic stem cells. Downregulation of NANOG in fetal testes has been seen to occur somewhat earlier than OCT-3/4, suggesting that NANOG acts upstream to OCT-3/4.32 In adult type GCTs, NANOG is expressed in CIS, seminomas, dysgerminomas, and embryonal carcinomas. SOX2 is suggested to act in collaboration with NANOG and OCT-3/4 in promoter
FIGURE 3. Examples of AFP, GATA4, SOX2, and AP-2g expression in pediatric GCTs. A, AFP staining of yolk sac tumor in the ovary of a 15-year-old girl. B, GATA4 staining of granulosa cell tumor in the testis of 1-month-old baby boy. C, SOX2 staining of an immature teratoma in the testis of a 3-month-old baby boy. D, AP-2g staining in a dysgerminoma located in the pelvis of a 13-year-old girl, scalebar 50 mm.
266 | www.jpho-online.com
r
2014 Lippincott Williams & Wilkins
J Pediatr Hematol Oncol
Volume 36, Number 4, May 2014
regions of genes that encode 2 types of transcription factors: one that activates expression of those that maintain pluripotency and the one that inhibits expression of those that stimulate differentiation of the germ cells. CIS cells express SOX2 only at the transcriptional level; in addition, GCTs with a germ cell phenotype (seminomas and dysgerminomas) do not express the SOX2 protein. In contrast, embryonal carcinomas and immature teratomas highly express SOX2.33,34 AP-2g (TFAP2C) is a transcription factor regulated by BLIMP1 (PRDM gene family), and both factors are expressed in CIS and seminomas in adult testis, and suggested to have a function in repressing somatic differentiation in germ cells.35,36 In addition to the above-listed pluripotency-related tumor markers, SALL4, a transcription factor involved in the stemness maintenance in various embryonic tissue types and self-renewal in several types of cancers, has been reported as an excellent marker of YST and other GCT components, including even some expression seen in pediatric teratomas.37 Another transcription factor related to embryonic germ cell differentiation is GATA4, which is variably expressed in a subset of CIS, seminomas, dysgerminomas, and YST. Furthermore, it is an excellent marker of granulosa and Sertoli cell tumors. In tumor cells the varying expression of GATA4 in differentiated and nondifferentiated (pluripotent) tumor cells may be explained by cell-specific function in germ cells and somatic cells.38 After many years of studying expression of genes oneby-one at the protein level, the advent of high throughput array methods allowed genome-wide evaluation of entire transcriptome. Several such studies have been performed in adolescent and adult GCTs (reviewed in Alagaratnam et al 201139). To our knowledge, only 1 study evaluated the genome-wide transcriptome of pediatric GCTs, and the data identified some differences in transcript profiles of childhood YST compared with adult counterparts.40 Other studies have focused on selected pathways, especially those involved in early embryonic development and which have also been implicated in cancer. The WNT/b-catenin signaling pathway displays differences in distinct subtypes of GCT, with particularly strong expression of b-catenin in YST and immature teratomas.41 Similarly, multiple components of the TGFb/BMP (bone morphogenetic protein) signaling pathway, including inhibitory SMAD6 and SMAD7 were differentially expressed in childhood germinomas versus YST.42 Profiling of microRNA (miRNA) expression in this pathway revealed miR-155 was most highly expressed in germinomas, suggesting that miR-155 may inhibit BMP signaling in the tumor subtype.42 In contrast, numerous miRNA were significantly higher expressed in YST, including some miRNAs previously identified in this tumor type by global miRNA profiling.43 When the pediatric GCT data were compared with data from adult GCTs from previous work, the RNA and miRNA profiles differed with age (pediatric vs. adult). Germinomas showed an undifferentiated and pluripotent phenotype, overexpressing the embryonal stem cell markers NANOG, OCT-3/4, and UTF1, whereas YSTs displayed extraembryonic differentiation maintaining a proliferative phenotype. This significant difference between miRNA profiles in germinomas and nongerminomas has also been shown in pediatric tumors localized in the central nervous system.44 It opens up for use in future clinical diagnosis and treatment, as the differential miRNA expression (especially r
2014 Lippincott Williams & Wilkins
Etiology and Pathogenesis of Pediatric Germ Cell Tumors
miR-302) is likely to contribute to the relatively aggressive behavior of YST.43 The miR-371-373 is involved in maintaining the pluripotent state, whereas miR-302 is induced during the first stages of in vitro differentiation. The miR302 levels are high in extraembryonic differentiated tumors like YSTs, whereas virtually undetectable in somatically differentiated teratomas. The let-7 family of microRNA tumor suppressor genes is significantly downregulated in pediatric GCTs irrespective of patient’s age, tumor localization, and histology. LIN28 has been shown to downregulate let-7, resulting in expression of a number of oncogenes.45
GENETIC ABERRATIONS, GENE POLYMORPHISMS, AND EPIGENETIC FEATURES The cytogenetic features in GCTs differ depending on age and histology. Virtually, all pediatric teratomas have a normal karyotype.19,46 In contrast, GCTs of adults and adolescents, including teratomas, very consistently have structural abnormalities involving the chromosome 12p sequence; many tumors have an isochromosome 12p, others have large regional gains or amplifications of 12p.11,47 Palmer et al’s48 data showed gain of 12p in more than half of the cases aged 5 to 16 years and in only 4 of 14 cases of YST in children younger than 5 years of age. GCTs without 12p gain more often had a loss on 16p. One of the most frequent aberrations in childhood GCTs is the deletion of 1p, although there is a significant variance among different studies. Zahn et al49 found deletions of 1p in 8/9 pediatric malignant GCTs but not in adult GCTs. Bussey et al47 showed that deletion of chromosome 1p and gain of 1q and 3 were the most frequent abnormalities among the pediatric GCTs from both sexes. Palmer et al48 have shown that loss of 1p, 4q, 6q and gain of 3p is more frequent in YSTs of children as compared with germinomas. Loss of heterozygosity (LOH) of a number of candidate loci has been investigated in pediatric GCTs but very few were identified. An LOH in 5q22 encompassing the ACP gene, which is often mutated in colon neoplasms, was reported in a small series of YSTs and teratomas but mutations have not been identified.50 In recent genome-wide association studies of adult GCTs, susceptibility loci near KITLG, SPRY4, DMRT1, and BAK1 have been identified.51–53 The loci play a role in the survival of PGCs during migration (KIT-KITLG), inhibiting MAP kinase pathway downstream to KIT (SPRY4), promotion of apoptosis (BAK1), and regulation of sexually dimorphic meiotic entry (DMRT1). In their replication study of childhood GCTs (including adolescents), Poynter et al54 suggested that variants in KITLG and SRPY4 are susceptibility alleles only for GCTs of adolescents and young adults, whereas BAK1 may be a susceptibility allele for childhood GCTs, including neonates, young children, and adolescents. Analyses of the imprinting status of GCTs have supported that gonadal and extragonadal GCTs share a common cell of origin, the PGC, but the imprinting varies according to the stage of development of the germ cell.55–57 Schneider et al55 studied DNA methylation of CpG nucleotides in H19, IGF-2, and SNRPN, and the results differed from what was found in adult testicular GCTs. The pediatric GCTs originated from PGCs that showed consistent loss of imprinting of SNRPN and partly loss of imprinting of H19 and IGF-2, suggesting that pediatric www.jpho-online.com |
267
J Pediatr Hematol Oncol
Mosbech et al
Volume 36, Number 4, May 2014
TABLE 1. Molecular Characteristics of Pediatric GCTs
Cytogenetic Aberrations DG SEM GER
i12p (adolescents)
YST
Loss of 1p 4q 6q Gain of 3p 3 1q 2p 13 20q No characteristic alterations
MT IT
IHC Markers
Gene Variants (SNPs)
HCG OCT-3/4 NANOG KIT GATA4 SALL4 AFP GATA4 SALL4
BAK1* (children) KITLG SPRY4 (adolescents)
AFP SOX2
None studied
miRNA Under-expression of miR-200 miR-205 let-7 Over-expression of miR-371-373 miR-142-3/5p miR-146a Under-expression of let-7 Over-expression of miR-371-373 miR-302 miR-122
BAK1* (children) KITLG SPRY4 (adolescents)
No characteristic alterations
For study marked with *there is no distinction of the childhood tumors into histologic subtypes, and DG/SEM/GER and YST are rated collectively.5,23–29,33,37,38,40,42–49,51,54 DG indicates dysgerminoma; GER, germinoma; IHC, immunohistochemistry; IT, immature teratoma; MT, mature teratoma; SEM, seminoma; SNP, single nucleotide polymorphism; YST, yolk sac tumor.
GCTs arise from a different germ cell developmental stage. Molecular characteristics are summarized in Table 1. Extragonadal GCTs develop from a germ cell that has migrated aberrantly during embryogenesis, but it seems that both gonadal and extragonadal GCTs arise from PGCs that have erased their imprinting. The patterns of erasure are different in adults and children; therefore, the development of testicular GCTs in adults occurs at a later stage. Hoffner et al58 have suggested that extragonadal and testicular GCTs in children originate from a mitotic division of a premeiotic germ cell or a pluripotent somatic cell based on centromeric and DNA polymorphism analysis. In contrast, they suggested that ovarian and extragonadal teratomas may arise from germ cells more advanced in their development, because of errors in first meiotic division, second meiotic division, or endoreduplication.
CONCLUDING REMARKS Pediatric GCT origin differs from adult GCTs with regard to patterns of histology, cytogenetics, and absence of a preinvasive stage.18 It is generally accepted that adolescent and adult testicular GCTs, both seminomas and nonseminomas, but not the spermatocytic seminomas, originate from the precursor stage CIS.27 Visfeldt et al59 and Jorgensen et al18 were the first to investigate a possible association between CIS and pediatric testicular GCTs and found no CIS in infantile YSTs and teratomas. In contrast, 5 of 6 cases of adolescent GCTs showed presence of CIS cells, suggesting a common origin with adult GCTs.18 It is important to stress that the presence of CIS markers in the infantile testes does not equal the presence of CIS, because at this age some germ cells may still be at the stage of fetal gonocytes that are morphologically very similar to CIS cells.60,61 The current consensus concerning the pathogenesis of different GCTs stipulates the origin from the same cell type (germ cell) but at different developmental stages. As outlined above, this has been investigated by use of tumor markers, cytogenetic analyses, imprinting status,
268 | www.jpho-online.com
microRNA, and mRNA expression. The pluripotency of the progenitor cell may result in extraembryonically differentiated GCTs with choriocarcinoma and yolk sac elements maintaining a proliferative phenotype. The differentiation of GCTs may be a recapitulation of embryonal somatic differentiation, resulting in mature and immature tissue from all 3 germ cell layers (teratomas). Lastly, the pluripotency state may be maintained in a subset of immature germ cells, with prolonged expression of selfrenewal and pluripotency regulators (eg, OCT-3/4, NANOG, and AP-2g) but without any somatic differentiation, resulting in development of seminomas, dysgerminomas, and germinomas. Exact causative factors involved in initiation of GCTs are not known, but the etiology of these tumors most likely involves complex interplay of genetic mutations, predisposing polymorphisms, and, possibly, also environmental damaging influences. Research has been hampered by the low incidence of GCTs, and multicenter studies to obtain more valid data should be supported.
REFERENCES 1. Pinkerton CR. Malignant germ cell tumours in childhood. Eur J Cancer. 1997;33:895–901. 2. Horton Z, Schlatter M, Schultz S. Pediatric germ cell tumors. Surg Oncol. 2007;16:205–213. 3. Imbach P, Ku¨hne T, Arceci RJ. Pediatric Oncology: A Comprehensive Guide. 2nd ed. Berlin: Springer; 2011. 4. Schneider DT, Calaminus G, Koch S, et al. Epidemiologic analysis of 1442 children and adolescents registered in the German germ cell tumor protocols. Pediatr Blood Cancer. 2004;42:169–175. 5. Gobel U, Calaminus G, Schneider DT, et al. Management of germ cell tumors in children: approaches to cure. Onkologie. 2002;25:14–22. 6. Mann JR, Raafat F, Robinson K, et al. The United Kingdom Children’s Cancer Study Group’s second germ cell tumor study: carboplatin, etoposide, and bleomycin are effective treatment for children with malignant extracranial germ r
2014 Lippincott Williams & Wilkins
J Pediatr Hematol Oncol
7.
8.
9.
10. 11. 12. 13. 14. 15. 16. 17.
18.
19. 20. 21. 22.
23. 24. 25. 26. 27. 28.
29.
r
Volume 36, Number 4, May 2014
cell tumors, with acceptable toxicity. J Clin Oncol. 2000;18:3809–3818. Kiltie AE, Gattamaneni HR. Survival and quality of life of paediatric intracranial germ cell tumour patients treated at the Christie Hospital, 1972-1993. Med Pediatr Oncol. 1995; 25:450–456. Jeyapalan JN, Noor DA, Lee SH, et al. Methylator phenotype of malignant germ cell tumours in children identifies strong candidates for chemotherapy resistance. Br J Cancer. 2011;105: 575–585. Pussegoda K, Ross CJ, Visscher H, et al. Replication of TPMT and ABCC3 genetic variants highly associated with cisplatininduced hearing loss in children. Clin Pharmacol Ther. 2013; 94:243–251. Veltman IM, Schepens MT, Looijenga LH, et al. Germ cell tumours in neonates and infants: a distinct subgroup? APMIS. 2003;111:152–160; discussion 160. Oosterhuis JW, Looijenga LH. Testicular germ-cell tumours in a broader perspective. Nat Rev Cancer. 2005;5:210–222. Wylie CC. The biology of primordial germ cells. Eur Urol. 1993;23:62–66. Hajkova P, Erhardt S, Lane N, et al. Epigenetic reprogramming in mouse primordial germ cells. Mech Dev. 2002;117:15–23. Eini R, Dorssers LC, Looijenga LH. Role of stem cell proteins and microRNAs in embryogenesis and germ cell cancer. Int J Dev Biol. 2013;57:319–332. Teilum G. Classification of endodermal sinus tumour (mesoblatoma vitellinum) and so-called “embryonal carcinoma” of the ovary. Acta Pathol Microbiol Scand. 1965;64:407–429. Rescorla FJ. Pediatric germ cell tumors. Semin Pediatr Surg. 2012;21:51–60. Hoei-Hansen CE, Sehested A, Juhler M, et al. New evidence for the origin of intracranial germ cell tumours from primordial germ cells: expression of pluripotency and cell differentiation markers. J Pathol. 2006;209:25–33. Jorgensen N, Muller J, Giwercman A, et al. DNA content and expression of tumour markers in germ cells adjacent to germ cell tumours in childhood: probably a different origin for infantile and adolescent germ cell tumours. J Pathol. 1995;176:269–278. Harms D, Zahn S, Gobel U, et al. Pathology and molecular biology of teratomas in childhood and adolescence. Klin Padiatr. 2006;218:296–302. Echevarria ME, Fangusaro J, Goldman S. Pediatric central nervous system germ cell tumors: a review. Oncologist. 2008;13:690–699. Borer JG, Tan PE, Diamond DA. The spectrum of Sertoli cell tumors in children. Urol Clin North Am. 2000;27:529–541. Lahdenne P, Pitkanen S, Rajantie J, et al. Tumor markers CA 125 and CA 19-9 in cord blood and during infancy: developmental changes and use in pediatric germ cell tumors. Pediatr Res. 1995;38:797–801. Talebagha S, Rizk C, Elawabdeh N, et al. Usefulness of OCT4/ 3 immunostain in pediatric malignant germ cell tumors. Fetal Pediatr Pathol. 2013;32:82–87. Ho DM, Liu HC. Primary intracranial germ cell tumor. Pathologic study of 51 patients. Cancer. 1992;70:1577–1584. Siltanen S, Anttonen M, Heikkila P, et al. Transcription factor GATA-4 is expressed in pediatric yolk sac tumors. Am J Pathol. 1999;155:1823–1829. Perlman EJ, Hawkins EP. Pediatric germ cell tumors: protocol update for pathologists. Pediatr Dev Pathol. 1998;1: 328–335. Skakkebaek NE. Possible carcinoma-in-situ of the testis. Lancet. 1972;2:516–517. Hoei-Hansen CE, Kraggerud SM, Abeler VM, et al. Ovarian dysgerminomas are characterised by frequent KIT mutations and abundant expression of pluripotency markers. Mol Cancer. 2007;6:12. Chou PM, Barquin N, Guinan P, et al. Differential expression of p53, c-kit, and CD34 in prepubertal and postpubertal testicular germ cell tumors. Cancer. 1997;79:2430–2434.
2014 Lippincott Williams & Wilkins
Etiology and Pathogenesis of Pediatric Germ Cell Tumors
30. Looijenga LH, Stoop H, de Leeuw HP, et al. POU5F1 (OCT3/ 4) identifies cells with pluripotent potential in human germ cell tumors. Cancer Res. 2003;63:2244–2250. 31. Rajpert-De Meyts E, Hanstein R, Jorgensen N, et al. Developmental expression of POU5F1 (OCT-3/4) in normal and dysgenetic human gonads. Hum Reprod. 2004;19:1338–1344. 32. Hoei-Hansen CE, Almstrup K, Nielsen JE, et al. Stem cell pluripotency factor NANOG is expressed in human fetal gonocytes, testicular carcinoma in situ and germ cell tumours. Histopathology. 2005;47:48–56. 33. Sonne SB, Perrett RM, Nielsen JE, et al. Analysis of SOX2 expression in developing human testis and germ cell neoplasia. Int J Dev Biol. 2010;54:755–760. 34. Perrett RM, Turnpenny L, Eckert JJ, et al. The early human germ cell lineage does not express SOX2 during in vivo development or upon in vitro culture. Biol Reprod. 2008;78:852–858. 35. Schafer S, Anschlag J, Nettersheim D, et al. The role of BLIMP1 and its putative downstream target TFAP2C in germ cell development and germ cell tumours. Int J Androl. 2011;34:e152–e158. 36. Hoei-Hansen CE, Nielsen JE, Almstrup K, et al. Transcription factor AP-2gamma is a developmentally regulated marker of testicular carcinoma in situ and germ cell tumors. Clin Cancer Res. 2004;10:8521–8530. 37. Cao D, Li J, Guo CC, et al. SALL4 is a novel diagnostic marker for testicular germ cell tumors. Am J Surg Pathol. 2009;33:1065–1077. 38. Salonen J, Rajpert-De Meyts E, Mannisto S, et al. Differential developmental expression of transcription factors GATA-4 and GATA-6, their cofactor FOG-2 and downstream target genes in testicular carcinoma in situ and germ cell tumors. Eur J Endocrinol. 2010;162:625–631. 39. Alagaratnam S, Lind GE, Kraggerud SM, et al. The testicular germ cell tumour transcriptome. Int J Androl. 2011;34:e133–e150. 40. Palmer RD, Barbosa-Morais NL, Gooding EL, et al. Pediatric malignant germ cell tumors show characteristic transcriptome profiles. Cancer Res. 2008;68:4239–4247. 41. Fritsch MK, Schneider DT, Schuster AE, et al. Activation of Wnt/beta-catenin signaling in distinct histologic subtypes of human germ cell tumors. Pediatr Dev Pathol. 2006;9:115–131. 42. Fustino N, Rakheja D, Ateek CS, et al. Bone morphogenetic protein signalling activity distinguishes histological subsets of paediatric germ cell tumours. Int J Androl. 2011;34:e218–e233. 43. Murray MJ, Saini HK, van Dongen S, et al. The two most common histological subtypes of malignant germ cell tumour are distinguished by global microRNA profiles, associated with differential transcription factor expression. Mol Cancer. 2010;9:290. 44. Wang HW, Wu YH, Hsieh JY, et al. Pediatric primary central nervous system germ cell tumors of different prognosis groups show characteristic miRNome traits and chromosome copy number variations. BMC Genomics. 2010;11:132. 45. Murray MJ, Saini HK, Siegler CA, et al. LIN28 expression in malignant germ cell tumors downregulates let-7 and increases oncogene levels. Cancer Res. 2013;73:4872–4884. 46. Schneider DT, Schuster AE, Fritsch MK, et al. Genetic analysis of childhood germ cell tumors with comparative genomic hybridization. Klin Padiatr. 2001;213:204–211. 47. Bussey KJ, Lawce HJ, Olson SB, et al. Chromosome abnormalities of eighty-one pediatric germ cell tumors: sex-, age-, site-, and histopathology-related differences—a Children’s Cancer Group study. Genes Chromosomes Cancer. 1999;25:134–146. 48. Palmer RD, Foster NA, Vowler SL, et al. Malignant germ cell tumours of childhood: new associations of genomic imbalance. Br J Cancer. 2007;96:667–676. 49. Zahn S, Sievers S, Alemazkour K, et al. Imbalances of chromosome arm 1p in pediatric and adult germ cell tumors are caused by true allelic loss: a combined comparative genomic hybridization and microsatellite analysis. Genes Chromosomes Cancer. 2006;45:995–1006.
www.jpho-online.com |
269
Mosbech et al
J Pediatr Hematol Oncol
50. Okpanyi V, Schneider DT, Zahn S, et al. Analysis of the adenomatous polyposis coli (APC) gene in childhood and adolescent germ cell tumors. Pediatr Blood Cancer. 2011;56: 384–391. 51. Rapley EA, Turnbull C, Al Olama AA, et al. A genome-wide association study of testicular germ cell tumor. Nat Genet. 2009;41:807–810. 52. Kanetsky PA, Mitra N, Vardhanabhuti S, et al. Common variation in KITLG and at 5q31.3 predisposes to testicular germ cell cancer. Nat Genet. 2009;41:811–815. 53. Turnbull C, Rapley EA, Seal S, et al. Variants near DMRT1, TERT and ATF7IP are associated with testicular germ cell cancer. Nat Genet. 2010;42:604–607. 54. Poynter JN, Hooten AJ, Frazier AL, et al. Associations between variants in KITLG, SPRY4, BAK1, and DMRT1 and pediatric germ cell tumors. Genes Chromosomes Cancer. 2012;51:266–271. 55. Schneider DT, Schuster AE, Fritsch MK, et al. Multipoint imprinting analysis indicates a common precursor cell for gonadal and nongonadal pediatric germ cell tumors. Cancer Res. 2001;61:7268–7276. 56. Sievers S, Alemazkour K, Zahn S, 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;44:256–264. Bussey KJ, Lawce HJ, Himoe E, et al. SNRPN methylation patterns in germ cell tumors as a reflection of primordial germ cell development. Genes Chromosomes Cancer. 2001;32: 342–352. Hoffner L, Deka R, Chakravarti A, et al. Cytogenetics and origins of pediatric germ cell tumors. Cancer Genet Cytogenet. 1994;74:54–58. Visfeldt J, Jorgensen N, Muller J, et al. Testicular germ cell tumours of childhood in Denmark, 1943-1989: incidence and evaluation of histology using immunohistochemical techniques. J Pathol. 1994;174:39–47. Jorgensen N, Giwercman A, Muller J, et al. Immunohistochemical markers of carcinoma in situ of the testis also expressed in normal infantile germ cells. Histopathology. 1993;22:373–378. Honecker F, Stoop H, de Krijger RR, et al. Pathobiological implications of the expression of markers of testicular carcinoma in situ by fetal germ cells. J Pathol. 2004;203: 849–857.
270 | www.jpho-online.com
57.
58. 59.
60.
61.
r
Volume 36, Number 4, May 2014
2014 Lippincott Williams & Wilkins
