Testicular embryonic differentiation is due to a cascade of gene activations in the bipotential gonadal ridge. Sertoli cells differentiate by the sequential expression of Sry and SOX9 that prompts their clustering with recently arrived germ cells to form testis cords, the precursors of postnatal seminiferous tubules. Alterations in this process may lead to testicular dysgenesis, the phenotypic expression of Disorders of Sex Development (DSD), congenital anomalies that involve impaired development of chromosomal, gonadal or anatomic sex.
Testicular dysgenesis comprises a wide spectrum of anomalies ranging from a well-differentiated testis with mild dysgenesis to a completely undifferentiated gonadal tissue. Dysgenetic gonads carrying Y chromosome material are at increased risk of developing germ cell cancer, which derives from pre-invasive stages first described by Skakkebæk as Carcinoma in Situ (CIS) of the testis (also known as Intratubular Germ Cell Neoplasia (ITGCN). CIS originates in malignant transformation of fetal gonocytes that have failed to differentiate to postnatal spermatogonia. The diagnosis of CIS-ITGCN is based on characteristic histologic and immunohistochemical criteria. The existence of abnormal infantile germ cells and development of CIS in boys with DSD and cryptorchidism was first reported by Müller et al, but the validity of its presence in children did not get universal acceptance.
We evaluated testicular dysgenesis by quantifying cytological germ cell abnormalities, incidence of immunohistochemical markers of undifferentiation-multipotency (OCT-3/4, PLAP) and quantification of germ cell DNA (ploidy) in testicular biopsies or gonadectomies from 43 patients with DSD and testicular dysgenesis. Main histopathologic parameters studied were thickness of the tunica albuginea, cortico-medullar differentiation, cortical branching of seminiferous cords, penetration of the tunica albuginea by seminiferous cords, presence of cortical ovarian-like stroma and widening of the interstitial space.
All patients consulted for ambiguous genitalia with variable degrees of masculinization (phallus between a hypertrophic clitoris and a small penis, complete, or almost, labial fusion and perineo-scrotal urethral opening). Bilateral dysgenetic testes were present in 21 patients (BT-DSD), one dysgenetic testis was combined with one streak gonad in another 13 patients (CT-DSD) and the other 9 had variable arrays of dysgenetic testes and ovaries (OT-DSD). Our main purpose was to analyze the relationship between the degrees of objectively- characterized testicular dysgenesis with various forms of germ cell proliferations and tumors to define the phenotype of an infantile form of CIS and differentiate it from non-specific germ cell proliferations and delayed maturation of testicular germ cells.
We have previously demonstrated that testicular dysgenesis mimics the first stages of embryogenesis and represents a form of immaturity, as if the gonad had failed to complete full embryonic maturation (Chemes et al, 2003). Incidence of cortico-medullar differentiation, dichotomic bifurcation of seminiferous cords and their penetration of a thin albuginea are probably the result of arrested testicular development, an important component of the of testicular dysgenesis syndrome.
The replacement of fetal gonocytes by postnatal spermatogonia and the downregulation of OCT 3/4 expression in the first postnatal months may fail in testicular dysgenesis resulting in delayed germ cell maturation, a risk factor for germ cell malignancies at or after puberty. The distinction of CIS from delayed maturation may be quite difficult even to experienced pathologists because neoplastic changes in the prepubertal testis do not conform to the classic adult CIS phenotype. A careful histopathologic study permits a proper distinction between these two conditions.
Microscopic features suggestive of malignant potential paralleled the incidence of CIS + TGCT in the three groups of patients. Indeed, cytologic abnormalities, OCT 3/4 positivity and germ cell aneuploidy were highest in BT-DSD testes (that had maximal numbers of CIS and TGCT), intermediate in OT-DSD and lowest in CT-DSD testes that had only 1 CIS.
Adult CIS shows germ cell-depleted small seminiferous tubules containing big basal atypical germ cells with increased mitosis and aneuploid DNA content. Sertoli cell nuclei are displaced to a second row. Here we show that infantile CIS is characterized by solid seminiferous cords with abundant basal and central atypical gonocyte-like germ cells, most of them positive for OCT 3/4, infrequent mitosis and hypertriploid to hypotetraploid DNA content. These seminiferous cords gather in areas of lobulillar distribution separated from normal testicular parenchyma by thin connective tissue septa. As documented in the present report, the neoplastic potential of infantile CIS is confirmed by its evolution to invasive germ cell cancer after puberty. During late infancy, there is an evolution from the infantile to the adult CIS phenotype. Infantile CIS is a “dormant” form of intratubular malignancy with unpredictable and sometimes protracted progression. With present methods it is difficult to anticipate the moment when CIS will evolve to invasive cancer. Adult CIS is also dormant because seminoma develops predominantly in the 3rd to the 5th decades although these patients should have harbored CIS at least since puberty.
The intensity of testicular dysgenesis was inversely related to the occurrence of CIS as shown by the highest numbers of CIS+TGCT in BT-DSD patients that had milder dysgenesis. Conversely, maximal dysgenesis in CT-DSD coincided with lower incidence of CIS. This dissociation between dysgenesis and neoplastic risk is an intriguing finding that was previously noted in dysgenetic gonads of adult infertile patients. The development of CIS derives from the failure of gonocyte differentiation beyond its original multipotency probably due to a disrupted interaction between germ and somatic cells in the fetal testis.
The incidence of CIS in our population of ovotesticular DSD is higher than previously thought. However, the reported low CIS risk may be biased because until recently most ovotestes were resected during infancy, before the possible appearance of TGCT. Interestingly, current trends to preserve testicular tissue in cases of ovotestes with male sex assignment have coincided with increased rates of seminoma.
The data here presented clearly delineate the infantile CIS phenotype in testicular dysgenesis and demonstrates that the use of markers of germ cell multipotency together with quantification of germ cell atypia and ploidy can satisfactorily identify infantile CIS with high risk of malignant evolution and set it apart from delayed germ cell maturation with lower neoplastic potential. However, it seems reasonable to remember that, from the standpoint of the pathologist, the diagnosis of infantile CIS should primarily rest in a careful histopathologic examination of cytological signs of malignancy and their topographic distribution, and be confirmed by specific immunohistochemical markers and (in most cases) aneuploidy.
Centro de Investigaciones Endocrinológicas Dr. César Bergadá (CEDIE), CONICET, Buenos Aires, Argentina.
Abstract: Is a CIS phenotype apparent in children with Disorders of Sex Development? The infantile precursor of testicular germ cell cancer, or how milder testicular dysgenesis means higher risk of malignancy.