Androgen insensitivity syndrome (AIS) is caused by mutations in the androgen receptor gene and is associated with a variety of phenotypes in 46,XY individuals, ranging from phenotypic women [complete form (CAIS)] to men with minor degrees of undervirilization or infertility [partial form (PAIS)]. We studied 32 subjects with male pseudohermaphroditism from 20 families (9 CAIS, 11 PAIS) with the following criteria for AIS: 46,XY karyotype, normal male basal and human chorionic gonadotropin-stimulated levels of serum testosterone and steroid precursors, gynecomastia at puberty, and, in prepubertal patients, a family history suggestive of X-linked inheritance. The entire coding region of the androgen receptor gene was analyzed, and mutations were found in all families with CAIS and in eight of 11 families with PAIS. Fifteen different mutations were identified, including five (S119X, T602P, L768V, I898F, and P904V) that have not been described previously. Detailed clinical and hormonal features were compared with genotype in 25 subjects with AIS and confirmed by mutational analysis. LH hormone levels and the LH x testosterone product were high in all postpubertal subjects with AIS. All subjects with PAIS maintained at postpubertal age the gender identity and social sex that was assigned to them in infancy, in contrast to other forms of pseudohermaphroditism.

Ten male pseudohermaphrodites with 17 beta-hydroxysteroid dehydrogenase 3 (17 beta-HSD3) deficiency were evaluated in 1 clinic with an average follow-up of 10.1 years. The diagnoses were made by demonstrating low to normal serum testosterone levels, high androstenedione levels, and high ratios of serum androstenedione to testosterone in the basal state or after treatment with human chorionic gonadotropin. The molecular features of the underlying mutations were identified in all 7 families. Two additional males in the same families are believed to be affected on the basis of history obtained from family members. All of the 46,XY individuals in these families were registered at birth and raised as females (despite the presence of ambiguous genitalia in all or most), and all virilized after the time of expected puberty due to a rise in serum testosterone to or toward the normal male range. The age at diagnosis varied from 4 to 37 years. Ten individuals were studied by the same psychologist, and change of gender role (social sex) from female to male occurred in 3 subjects and in the 2 presumed affected subjects not studied. The individual with the highest serum testosterone level maintained female sexual identity, and in 2 families some of the affected males changed gender role and others did not. Thus, while androgen action plays a role in the process, additional undefined psychological, social, and/or biologic factors must be determinants of gender identity/role behavior. Management of the 7 individuals who chose to maintain female sex roles included castration, clitoroplasty, vaginal enlargement procedures when appropriate, treatment of hirsutism, cricoid cartilage reduction, and estrogen replacement. Three of the 7 are married (2 twice), 1 is involved in a long-term heterosexual relationship, 1 is engaged to be married, and the other 2 are not married and not believed to be sexually active. The 3 subjects who changed gender role behavior to male underwent hypospadias repair, and 1 was given supplemental testosterone therapy. One of these men is divorced, and the other 2 (aged 29 and 35 years) are unmarried. The diagnosis in 8 of these subjects was made after the time of expected puberty; it is unclear whether the functional and social outcomes would have been different if the diagnosis had been made and therapy begun earlier in life.

Sixteen subjects (from 10 Brazilian families) with male pseudohermaphroditism due to steroid 5alpha-reductase 2 deficiency have been evaluated in 1 clinic. The diagnoses were made on the basis of normal plasma testosterone values, normal or low plasma dihydrotestosterone levels and high testosterone/dihydrotestosterone ratios in the basal state in postpubertal subjects or after treatment with either human chorionic gonadotropin or testosterone in prepubertal subjects. The analysis of the ratios of etiocholanolone to androsterone in urine confirmed the diagnosis in all subjects who were tested, and the molecular basis of the underlying mutations was established in 9 of the families. Fourteen of the individuals were evaluated by the same psychologist. All subjects but 1 were given a female sex assignment at birth. Three of the subjects (1 the sibling of an individual who has undergone female to male social behavior) maintain a female social sex; they have been gonadectomized and treated with exogenous estrogens. Ten of 13 subjects of postpubertal age underwent a change of social sex from female to male, had surgical correction of the hypospadias, and were treated with high-dose testosterone esters by parenteral injection and subsequently with dihydrotestosterone cream. These regimens brought serum dihydrotestosterone levels to the normal male range (or above) but resulted only in limited growth of the prostate and penis and, in some, increase in body and facial hair and enhancement of libido and sexual performance. Treatment of the prepubertal boys with testosterone and/or dihydrotestosterone resulted in a doubling of penis size.

No abstract

Most patients with 46,XY disorders of sex development were satisfied with long-term results of masculinizing genitoplasty, although specific complaints about small penile length, sexual activity and urinary symptoms were frequent. New surgical approaches should be developed to ensure full satisfaction in adulthood among patients with disorders of sex development.

A successful live birth was achieved in a woman with 17-hydroxylase deficiency through IVF, cryopreservation of all embryos, and frozen-thawed embryo transfer after adequate endometrial preparation.

Male gender assignment was more prevalent. Long-term followup revealed adequate pubertal development and sexual activity. Complications involving the urethra developed frequently in male patients.

Stacking methods improve the prediction performance of regression models. A simple way to stack base regressions estimators is by combining them linearly, as done by Breiman [1]. Even though this approach is useful from an interpretative perspective, it often does not lead to high predictive power. We propose the NN-Stacking method (NNS), which generalizes Breiman's method by allowing the linear parameters to vary with input features. This improvement enables NNS to take advantage of the fact that distinct base models often perform better at different regions of the feature space. Our method uses neural networks to estimate the stacking coefficients. We show that while our approach keeps the interpretative features of Breiman's method at a local level, it leads to better predictive power, especially in datasets with large sample sizes.