GIRLS WITH IDIOPATHIC premature adrenarche (PA)

Transcription

1 /02/$15.00/0 The Journal of Clinical Endocrinology & Metabolism 87(1): Printed in U.S.A. Copyright 2002 by The Endocrine Society COMMENT Elevated Free IGF-I Levels in Prepubertal Hispanic Girls with Premature Adrenarche: Relationship with Hyperandrogenism and Insulin Sensitivity MIRIAM E. SILFEN, ALEXANDRA M. MANIBO, MICHEL FERIN, DONALD J. MCMAHON, LENORE S. LEVINE, AND SHARON E. OBERFIELD Division of Pediatric Endocrinology (M.E.S., A.M.M., L.S.L., S.E.O.), Department of Obstetrics and Gynecology (M.F.), Columbia University, New York, New York 10032; and Nathan Kline Institute (D.J.M.), Orangeburg, New York Girls with premature adrenarche (PA) (the onset of pubic hair before the age of 8 yr associated with elevated levels of adrenal androgens and no evidence of true puberty or adrenal dysfunction) may be at increased risk for development of polycystic ovarian syndrome (PCOS). Alterations in the IGF system, including elevated free IGF-I, have been demonstrated in PCOS and may be involved in its pathogenesis. Hyperinsulinemia, elevated total IGF-I, and decreased IGF-binding protein-1 (IGFBP-1) have also been reported in PA. Dysregulation of the IGF system may be involved in the pathogenesis of PA and its progression to PCOS. We compared the insulin/igf system in 17 prepubertal girls with PA and nine prepubertal controls. Both groups were predominantly obese. Total and free IGF-I were elevated in Abbreviations: 4-A, 4-Androstenedione; BMI, body mass index; FGIR, fasting glucose to insulin ratio; GTT, glucose tolerance test; IGFBP-1, IGF binding protein-1; IR, insulin resistant; IS, insulin sensitive; OGTT, oral GTT; PA, premature adrenarche; PCOS, polycystic ovarian syndrome; QUICKI, quantitative insulin sensitivity check index. the premature adrenarche group. No differences in basal insulin, insulin area under the curve in response to an oral glucose tolerance test, or IGFBP-1 were noted. These effects persisted when adjusted for adiposity using body mass index-z score. Total and free IGF-I were positively correlated, and IGFBP-1 was negatively correlated with 4-androstenedione, but not with dehydroepiandrosterone sulfate. Free IGF-I trended toward higher levels in the insulin-resistant subgroup, compared with the insulin-sensitive subgroup. These results suggest altered regulation of the insulin/igf system in prepubertal girls with PA and a possible role for free IGF-I in the pathogenesis of the hyperandrogenism of PA as well as its progression to PCOS. (J Clin Endocrinol Metab 87: , 2002) GIRLS WITH IDIOPATHIC premature adrenarche (PA) (the onset of pubic hair before the age of 8 yr associated with elevated levels of adrenal androgens and no evidence of true puberty or adrenal disorders) are at an increased risk for developing polycystic ovarian syndrome (PCOS) (1). PCOS is characterized by elevated androgen levels and anovulation and is associated with complications including insulin resistance, early onset of glucose intolerance and type 2 diabetes mellitus, unfavorable lipid profiles, cardiovascular disease, and infertility (2, 3). Similarly, girls with PA have been shown to have insulin resistance, unfavorable lipid profiles (elevated total cholesterol, LDL, and triglyceride levels, decreased HDL levels, and an increased LDL/HDL ratio), and elevated androgens (4 7). Furthermore, there is a much higher prevalence of type 2 diabetes mellitus and impaired glucose tolerance in first-degree relatives of girls with PA (8). Of note, pubic hair development before 8 yr of age may be more common in black girls compared with white girls, but to our knowledge there are no prevalence data on Hispanic girls (9). Interestingly, work by Dunaif et al. (10) has suggested a predisposition to the development of PCOS in Caribbean-Hispanic women compared with both black and white women, as well as increased insulin resistance compared with white women. Dysregulation of the insulin/igf system in the pathogenesis of PCOS has been suggested (11 13). In 1995, Zhang et al. (14) speculated that IGF-I might initiate the onset of adrenarche. Both in vivo and in vitro studies support a role for the insulin/igf system in the pathophysiology of PA as well as its progression to PCOS. Clinical studies have demonstrated hyperinsulinemia and decreased levels of IGF -binding protein-1 (IGFBP-1) in girls with PA and women with PCOS, and an inverse relationship between IGFBP-1 and ACTHstimulated adrenal hormone levels in girls with PA (11 13, 15 17). In vitro studies have shown that insulin and IGF-I stimulate ovarian and adrenal steroidogenesis (18 23). Furthermore, IGFBP-1 has been proposed as a marker of insulin sensitivity and has been shown to correlate with insulin sensitivity in early pubertal children and women with PCOS (24 26). Elevated free IGF-I levels have been demonstrated in PCOS but, to our knowledge, have not been reported in PA (12). The primary purpose of the present study was to compare various components of the insulin/igf system, including free IGF-I, between predominantly obese prepubertal girls with PA and prepubertal controls. The secondary aim was to determine the relationship, if any, between the insulin/igf system and androgens and between the IGF system and insulin sensitivity in PA. 398

2 Silfen et al. Comment J Clin Endocrinol Metab, January 2002, 87(1): Subjects Subjects and Methods Twenty-six Hispanic prepubertal girls were enrolled in the study: 17 with PA and 9 controls. Informed consent from a legal guardian of each subject and assent from subjects over the age of 7 yr were obtained before participation in the study. The study was approved by the Institutional Review Board of Columbia-Presbyterian Medical Center (New York, NY). The criteria for entry into the study in the PA group included the appearance of pubic hair before 8 yr of age, adrenal androgens in the Tanner stage 2 range, no breast development on physical examination, and no evidence of an adrenal enzyme defect or other endocrine disorder (27). The criteria for entry in the control group included the absence of breast development and pubic hair on physical examination, and were otherwise the same as for the PA group. Body mass index (BMI) and BMI-Z scores, based on 1 sd reference data developed from National Health and Nutrition Examination Survey I, were calculated for all subjects (28). All subjects were examined for the presence of acanthosis nigricans. Clinical characteristics of the subjects are presented in Table 1. The groups were similar with regard to age, BMI, BMI-Z score, and bone age. The groups were subdivided into insulin-resistant (IR) and insulin-sensitive (IS) subgroups on the basis of the fasting glucose to insulin ratio (FGIR), where FGIR of at least 7 is IS and FGIR of less than 7 is IR, for analysis of the relationship between insulin sensitivity and the IGF system. The FGIR has been validated in prepubertal girls with PA using both the iv glucose tolerance test (GTT) and the oral GTT (OGTT) (7, 29). Insulin sensitivity measures The FGIR was calculated as fasting plasma glucose (G 0 ) divided by fasting serum insulin (I 0 ) levels (30). The quantitative insulin sensitivity check index (QUICKI), a novel method of assessing insulin sensitivity previously validated using the euglycemic insulin clamp, was also calculated for each subject as 1/(logG 0 logi 0 ) (31). Procedures All subjects had basal levels of glucose, insulin, IGFBP-1, IGF-I, free IGF-I, dehydroepiandrosterone sulfate (DHEAS), 4-androstenedione ( 4-A), T, free T, and SHBG measured after an overnight fast. Twenty of the 26 subjects (15 PA and 5 controls) underwent a standard 1.75 g/kg body weight (maximum 75 g) OGTT after an overnight fast. Before and 30, 60, 90, and 120 min after the ingestion of oral glucose, blood was sampled for plasma glucose and serum insulin measurements. Assays TABLE 1. Clinical characteristics Premature adrenarche (n 17) Controls (n 9) Age (yr) (range, ) (range, ) BMI (kg/m 2 ) (range, ) (range, ) BMI-Z score (range, ) (range, ) Bone age (yr) a Presence of AN b 11/17 6/9 FH of DM b 14/17 6/9 FH of CV b 14/17 4/9 Data are given as means SD, except where noted. AN, Acanthosis nigricans; FH of DM, family history of adult-onset diabetes mellitus; FH of CV, family history of cardiovascular disease. a Bone ages were performed in only six of the nine controls. b Data are shown here as the proportion of subjects within the group. Insulin (by immunochemiluminometric assay), IGFBP-1, IGF-I, DHEAS, 4-A, T, free T, and SHBG were measured by Esoterix Endocrinology (Calabasas Hills, CA). Plasma glucose levels were measured by the glucose hexokinase method. Free IGF-I was measured in our laboratory using a two-site immunoradiometric assay (Diagnostics Systems Laboratories, Inc., Webster, TX). The intra-assay and interassay coefficients of variation for this assay were 4.3 and 6.4%, respectively. Data analysis Comparison between groups on continuous measures was made using independent t tests with Saitherwaite correction in the event of statistically unequal variances. Analysis of covariance with BMI-Z score entered as a continuous covariate was used to adjust for adiposity. Insulin area under the curve in response to the OGTT (I AUC120 ) was log-transformed (logi AUC120 ) before analysis using analysis of covariance to normalize a positively skewed distribution. Estimation of the direction and strength of the relationships between IGF, insulin resistance, and androgen-related measures was made with simple Pearson correlations. Magnitude of statistical significance of parametric tests is reported as P value rounded up to the next larger significant digit. No adjustment for multiple comparisons was made. All data are reported as means sd. Results Comparison between premature adrenarche and control groups The groups were similar with regard to age, bone age, BMI, and BMI-Z score (Table 1). The average BMI and BMI-Z score (for sex and age) in each group were in the obese range (i.e. 95th percentile for age and sex) (28). Acanthosis nigricans was common in both groups. A positive family history of adult-onset diabetes mellitus and/or cardiovascular disease was elicited frequently in both groups and seemed to be more common in the PA group. Free IGF-I and total IGF-I, but not basal insulin, were significantly higher in the PA group compared with the controls (Table 2). These differences persisted after adjusting for BMI-Z score. IGFBP-1 levels and logi AUC120 were not different in the two groups before or after adjusting for BMI-Z score. Most girls in each group, however, had subnormal IGFBP-1 levels. The ratio of free to total IGF-I was calculated to express the proportion of total IGF-I as free IGF-I. This ratio was significantly higher (P 0.05) in the PA TABLE 2. IGF system and androgens in girls with premature adrenarche and controls Premature adrenarche Controls Insulin ( U/ml) a loginsulin AUC IGFBP-1 (ng/ml) IGF-I (ng/ml) b Free IGF-I (ng/ml) c Androstenedione (ng/dl) DHEAS ( g/dl) d T (ng/dl) d Free T (pg/ml) SHBG (nmol/liter) a Two PA subjects and four controls do not have a value for log- Insulin AUC120 because they did not undergo an OGTT. b Significance at the P 0.01 level for differences between premature adrenarche and control groups, both unadjusted and adjusted for BMI-Z score. c Significance at the P level for differences between premature adrenarche and control groups, both unadjusted and adjusted for BMI-Z score. d Significance at the P 0.05 level between premature adrenarche and control groups, both unadjusted and adjusted for BMI-Z score.

3 400 J Clin Endocrinol Metab, January 2002, 87(1): Silfen et al. Comment compared with the control group. As expected, the PA group had elevated DHEAS levels compared with controls. Normal low levels of T were seen in both groups, though significantly higher T levels were observed in the PA group. No difference was noted in insulin sensitivity, as measured by FGIR and QUICKI, between the PA and control groups before or after adjustment for BMI-Z score. Comparison between IR and IS subjects Six of the 17 PA subjects and 4 of the 9 controls were classified as IR according to their FGIR. The IR girls in each group had significantly higher BMI and BMI-Z scores, and in the PA IR subgroup also had higher free T and 4-A and lower SHBG levels. As shown in Fig. 1, there was a tendency toward higher total and free IGF-I and lower IGFBP-1 levels in the IR, compared with IS, girls with premature adrenarche. In the control subjects, the IR subgroup tended toward lower IGFBP-1 levels, but no difference was seen in total or free IGF-I levels. Correlations within IGF system in PA Free IGF-I levels correlated significantly with total IGF-I levels in the PA group (r 0.71; P 0.001), but not in the control group. In the PA group, logi AUC120, but not basal insulin, inversely correlated significantly with IGFBP-1 (r 0.78; P 0.005). Free IGF-I was inversely correlated with IGFBP-1 (r 0.58; P 0.02), and total IGF-I tended to correlate inversely with IGFBP-1 (r 0.44; P 0.08). Basal insulin did not correlate with total or free IGF-I. Correlations between IGF system and androgens in PA 4-A, but not DHEAS, correlated with total and free IGF-I (r 0.50 and 0.50, respectively; P 0.05), correlated inversely with IGFBP-1 (r 0.52; P 0.05), and did not correlate with basal insulin or logi AUC120 in the PA group. 4-A and DHEAS were not correlated with BMI-Z score. Correlations between IGF system and insulin sensitivity in PA In the PA group, logi AUC120 (r 0.78; P 0.005), but not basal insulin or any measures of insulin sensitivity, correlated significantly with IGFBP-1. No measure of insulin sensitivity correlated with total or free IGF-I levels. A correlation was noted in the PA group between QUICKI and SHBG (r 0.51; P 0.05), and between logi AUC120 and SHBG (r 0.63; P 0.02). We observed a strong correlation of both total (r 0.81; P 0.05) and free IGF-I (r 0.85; P 0.02) with 4-A and of free IGF-I with DHEAS (r 0.79; P 0.05) in the PA IR but not the PA IS group. PA IR subjects exhibited a trend toward lower IGFBP-1 levels than PA IS subjects and a trend toward higher total and free IGF-I (Fig. 2). FIG. 1. Relationship between total IGF-I, free IGF-I, and IGFBP-1 with insulin resistance status by FGIR in girls with PA and controls (CO). FIG. 2. Schematic representation of hypothesis relating dysregulation of the IGF system to the development of premature adrenarche and PCOS.

4 Silfen et al. Comment J Clin Endocrinol Metab, January 2002, 87(1): Discussion Although idiopathic PA is the most common cause of the premature development of sexual hair in girls, the pathogenesis of PA has not yet been clearly defined. Recent observations have suggested that PA may be pathological and may predispose some of these girls to the development of PCOS and its complications (1, 4, 5, 8). Previous studies suggest a possible etiological role for the IGF system in both PA and PCOS (11 13, 15, 17). Fig. 2 is a schematic representation of the hypothesis relating the IGF system with PA and PCOS. Although hyperandrogenism in PCOS has been attributed in part to LH and estrogen, this seems very unlikely in prepubertal girls with PA (32). The present study confirms that there are alterations in the IGF system in girls with PA. In prepubertal girls with PA, we have found increased total and free IGF-I levels compared with controls, independent of BMI-Z score, as well as a correlation between total and free IGF-I and 4-A, but not DHEAS. Only one earlier report by Ibañez et al. (17) has examined the IGF system in girls with PA compared with a control group, similarly finding elevated total IGF-I as well as decreased IGFBP-1. The same study also found hyperinsulinemia, but not insulin resistance, at all Tanner stages and decreased IGFBP-1 in Tanner stage 5 girls with a history of premature adrenarche. The dissimilar results with regard to hyperinsulinemia may reflect differences in study population. Our PA and control groups, both of whom were comparatively more obese than those of Ibañez, had higher insulin levels after oral glucose load. Elevation of free IGF-I has not been reported previously in this population, but has been reported in women with PCOS (12). The free form of IGF-I is considered to possess the most potent biological activity at target tissues (33). We found a tendency toward higher total and free IGF-I and lower IGFBP-1 levels in the PA IR, compared with the PA IS girls, suggesting that the IGF system may be more important in the subgroup of girls with insulin resistance. The inverse relationship between logi AUC120 and IGFBP-1 lends further support for insulin suppression of hepatic IGFBP-1 production. This relationship was not seen with fasting insulin levels most likely because, as suggested by Le Stunff and Bougneres (34), stimulated hyperinsulinemia may precede the development of fasting hyperinsulinemia in children. Unexpectedly, our data did not reveal hyperinsulinemia or decreased levels of IGFBP-1 in the girls with PA compared with the controls when adjusted for degree of obesity as measured by BMI-Z score. We speculate that the very low levels of IGFBP-1 in many of the subjects may have limited these analyses and obscured a difference between the groups. Our study is in agreement with the findings in other populations, including PCOS, that have revealed an inverse correlation between IGFBP-1 and free IGF-I and further supports a role for IGFBP-1 in the regulation of bioavailable IGF-I (12, 35). However, there may be other factors in addition to IGFBP-1 regulating free IGF-I concentrations. The basis for the higher free IGF-I levels in the PA girls is not clear from the present study. Because there is a higher free to total IGF-I ratio and the percentage increase in free IGF-I levels (150%) is greater than for total IGF-I levels (62%) in the PA compared with the control group, it seems unlikely that the difference in free IGF-I is attributable solely to the total IGF-I level. The relationship between IGF-I and IGFBP-1 is, in all probability, complex and has not been fully elucidated. A single previous study by Vuguin et al. (15) has assessed the relationship between the IGF system and adrenal steroids in prepubertal girls with PA, finding a significant positive correlation of total IGF-I, and a negative correlation of IGFBP-1, with ACTH-stimulated androgen levels. The positive relationship between total and free IGF-I and 4-A in the present study further supports the hypothesis that the IGF system may influence adrenal androgen production and may contribute to the pathogenesis of the adrenal hyperandrogenism seen in PA. However, our study is not conclusive in this regard because levels of DHEAS, an adrenal androgen more commonly found to be elevated in PA, were not found to be related to the components of the IGF system studied. This might be due to the fact that 4-A levels exhibit a circadian rhythm with higher levels in the early morning due to ACTH secretion, whereas DHEAS levels do not (36). It is possible that ACTH-stimulated androgens (and not basal levels alone), as demonstrated by Vuguin et al. (15), may be necessary to determine this association. Alternatively, it may be that PA is a heterogeneous disorder with a subset of girls whose adrenal hyperandrogenism is induced by the IGF system. This is supported by our finding of a strong correlation of total and free IGF-I with 4-A in the PA IR but not the PA IS group. These results, however, should be interpreted cautiously because there are intrinsic problems with conclusions based solely on correlational analyses. It is also important to note that our subjects and controls had mean BMI and BMI-Z scores in the obese range, and therefore our findings may not be reproducible in a lean group of girls with PA. There have been conflicting reports in the literature as to the effect of obesity on the IGF system, with both normal and increased free IGF-I in childhood obesity being reported (37, 38). Given that IGFBP-1 may be a marker of insulin sensitivity, we investigated whether free IGF-I might also predict insulin sensitivity (39). Vuguin et al. (15) have noted an association between total IGF-I and insulin sensitivity as assessed by the frequently sampled iv GTT. We found no relationship between free IGF-I and FGIR, QUICKI, or logi AUC120. However, more precise methods of assessing insulin sensitivity, such as the euglycemic insulin clamp, might be required before concluding that free IGF-I is not useful in predicting insulin sensitivity. We did not detect a difference in insulin sensitivity between our prepubertal PA and control girls, in agreement with a previous study by Ibañez et al. (17), and suggest that it may be more a function of obesity in our study population. We chose as our measures of insulin sensitivity FGIR, because it has been validated against the iv GTT and OGTT in children with PA, and QUICKI because it has been validated, in adults, against the gold standard euglycemic clamp (7, 29, 31). Again, a controlled study using the euglycemic clamp technique may be necessary to clarify this issue. Further study of the IGF system, particularly free IGF-I, in girls with PA as they mature will be important in determining whether free IGF-I predicts a subset at risk for the development of PCOS or ovarian hyperandrogenism. Ibañez et

5 402 J Clin Endocrinol Metab, January 2002, 87(1): Silfen et al. Comment al. (17) found that the elevated total IGF-I in prepubertal girls with PA did not persist during puberty, whereas decreased IGFBP-1 levels were seen at Tanner stage 5 as well as prepubertally. As Miller (40) has suggested, increased IGF-I may be a trigger for the commencement of adrenarche, which may not persist after its onset. Additionally, these findings need to be confirmed in other racial and ethnic groups, because racial differences in the IGF system have been reported (41, 42). The role of free IGF-I in promoting linear growth in children with premature adrenarche calls for investigation as well. In conclusion, we have examined prepubertal girls with PA to further evaluate the hypothesis that the IGF system, chiefly free IGF-I, plays a role in its pathogenesis. We have shown that free IGF-I is elevated in our population of predominantly obese Hispanic prepubertal girls with PA. Our findings corroborate previous reports of both elevated total IGF-I in premature adrenarche and a relationship between adrenal androgens and the IGF system. This relationship seems to be more prominent in an IR subset of girls with PA. Acknowledgments We thank members of the Pediatric Endocrine Division at Columbia- Presbyterian Medical Center (New York, NY) for their kind referral of subjects, as well as the subjects for agreeing to participate. We also thank Allison Murphy, R.N., and the nurses and staff of the Pediatric General Clinical Research Center (GCRC) for their outstanding help. We acknowledge and thank Genentech, Inc., Eli Lilly & Co., and Pharmacia & Upjohn, Inc. for grant support. We gratefully acknowledge Esoterix Endocrinology and the GCRC laboratory for performing laboratory measurements. Received May 3, Accepted September 28, Address all correspondence and requests for reprints to: Sharon E. Oberfield, M.D., 630 West 168th Street, PH-5E-522, New York, New York seo8@columbia.edu. This work was supported in part by grants from the NIH (RR00645), Genentech, Inc., Eli Lilly & Company, and Pharmacia & Upjohn. This work was presented in part at the Lawson Wilkins Pediatric Endocrine Society Meeting in Montreal, Canada, July References 1. Ibañez L, Potau N, Virdis R, Zampolli M, Terzi C, Gussinye M, Carrascosa A, Vicens-Calvet E 1993 Postpubertal outcome in girls diagnosed of premature pubarche during childhood: increased frequency of functional ovarian hyperandrogenism. J Clin Endrocrinol Metab 76: Dunaif A, Graf M, Mandeli J, Laumas V, Dobrjansky A 1987 Characterization of groups of hyperandrogenic women with acanthosis nigricans, impaired glucose tolerance, and/or hyperinsulinemia. J Clin Endocrinol Metab 65: Dunaif A, Segal KR, Futterweit W, Dobrjansky A 1989 Profound peripheral insulin resistance, independent of obesity, in polycystic ovary syndrome. Diabetes 38: Oppenheimer E, Linder B, DiMartino-Nardi J 1995 Decreased insulin sensitivity in prepubertal girls with premature adrenarche and acanthosis nigricans. J Clin Endocrinol Metab 80: Ibañez L, Potau N, Chacon P, Pascual C, Carrascosa A 1998 Hyperinsulinaemia, dyslipaemia and cardiovascular risk in girls with a history of premature pubarche. Diabetologia 41: Ibañez L, Potau N, Francois I, de Zegher F 1998 Precocious pubarche, hyperinsulinism and ovarian hyperandrogenism in girls: relation to reduced fetal growth. J Clin Endocrinol Metab 83: DiMartino-Nardi J 1999 Premature adrenarche: findings in prepubertal African-American and Caribbean-Hispanic girls. Acta Paediatr Suppl 433: Ibañez L, Castell C, Tresserras R, Potau N 1999 Increased prevalence of type 2 diabetes mellitus and impaired glucose tolerance in first-degree relatives of girls with a history of precocious pubarche. Clin Endocrinol (Oxf) 51: Herman-Giddens ME, Slora EJ, Wasserman RC, Bourdony CJ, Bhapkar MV, Koch GG, Hasemeier CM 1997 Secondary sexual characteristics and menses in young girls seen n office practice: a study from the Pediatric Research in Office Settings network. Pediatrics 99: Dunaif A, Sorbara L, Delson R, Green G 1993 Ethnicity and polycystic ovary syndrome are associated with independent and additive decreases in insulin action in Caribbean-Hispanic women. Diabetes 42: Suikkari AM, Ruutianen K, Erkkola R, Seppala M 1989 Low levels of low molecular weight insulin-like growth factor-binding protein in patients with polycystic ovarian disease. Hum Reprod 4: Thierry van Dessel HJ, Lee PD, Faessen G, Fauser BC, Giudice LC 1999 Elevated serum levels of free insulin-like growth factor I in polycystic ovary syndrome. J Clin Endocrinol Metab 84: Homburg R, Pariente C, Lunenfeld B, Jacobs HS 1992 The role of insulin-like growth factor-i and IGF binding protein-i in the pathogenesis of polycystic ovary syndrome. Hum Reprod 7: Zhang LH, Rodriguez H, Ohno S, Miller WL 1995 Serine phosphorylation of human P450c17 increases 17,20-lyase activity: implications for adrenarche and the polycystic ovary syndrome. Proc Natl Acad Sci USA 92: Vuguin P, Linder B, Rosenfeld FG, Saenger P, DiMartino-Nardi J 1999 The roles of insulin sensitivity, insulin-like growth factor I, and IGF-binding protein-1 and 3 in the hyperandrogenism of African-American and Caribbean Hispanic girls with premature adrenarche. J Clin Endocrinol Metab 84: Morales AJ, Laughlin GA, Butzow T, Maheshwari H, Baumann G, Yen SS 1996 Insulin, somatotropic, and luteinizing hormone axes in lean and obese women with polycystic ovary syndrome: common and distinct features. J Clin Endocrinol Metab 81: Ibañez L, Potau N, Zampolli M, Rique S, Saenger P, Carrascosa A 1997 Hyperinsulinemia and decreased insulin-like growth factor-binding protein-1 are common features in prepubertal and pubertal girls with a history of premature pubarche. J Clin Endocrinol Metab 82: Poretsky L, Cataldo NA, Rosenwaks Z, Giudice LC 1999 The insulin-related ovarian regulatory system in health and disease. Endocr Rev 20: Kristiansen SB, Endoh A, Casson PR, Buster JE, Hornsby PJ 1997 Induction of steroidogenic enzyme genes by insulin and IGF-I in cultured adult human adrenocortical cells. Steroids 62: Mesiano S, Katz SL, Lee JY, Jaffe RB 1997 Insulin-like growth factors augment steroid production and expression of steroidogenic enzymes in human fetal adrenal cortical cells: implications for adrenal androgen regulation. J Clin Endocrinol Metab 82: L Allemand D, Penhoat A, Lebrethon M-C, Ardevol R, Baehr V, Oelkers W, Saez JM 1996 Insulin-like growth factors enhance steroidogenic enzyme and corticotropin steroidogenic responsiveness in cultured human adrenocortical cells. J Clin Endocrinol Metab 81: Barbieri RL, Makris A, Randall RW, Daniels G, Kistner RW, Ryan KJ 1986 Insulin stimulates androgen accumulation in incubations of ovarian stroma obtained from women with hyperandrogenism. J Clin Endocrinol Metab 62: Nestler JE, Jakubowicz DJ, Falcon de Vargas A, Brik C, Quintero N, Medina F 1998 Insulin stimulates testosterone biosynthesis by human thecal cells from women with polycystic ovary syndrome by activating its own receptor and using inositolglycan mediators as the signal transduction system. J Clin Endocrinol Metab 83: Saitoh H, Kamoda T, Nakahara S, Hirano T, Matsui A 1999 Insulin-like growth factor binding protein-1 as a predictor of glucose-stimulated hyperinsulinemia in prepubertal obese children. Eur J Endocrinol 140: Travers SH, Labarta JI, Gargosky SE, Rosenfeld RG, Jeffers BW, Eckel RH 1998 Insulin-like growth factor binding protein-1 levels are strongly associated with insulin sensitivity and obesity in early pubertal children. J Clin Endocrinol Metab 83: Morris DV, Falcone T 1996 The relationship between insulin sensitivity and insulin-like growth factor-binding protein-1. Gynecol Endocrinol 10: Oberfield SE, Mayes DM, Levine LS 1990 Adrenal steroidogenic function in a Black and Hispanic population with precocious pubarche. J Clin Endocrinol Metab 70: Must A, Dallal GE, Dietz WH 1991 Reference data for obesity: 85th and 95th percentiles of body mass index (wt/ht 2 ) a correction. Am J Clin Nutr 54: Silfen ME, Manibo AM, McMahon DJ, Levine LS, Murphy AR, Oberfield SE 2001 Comparison of simple measures of insulin sensitivity in young girls with premature adrenarche: the fasting glucose the insulin ratio may be a simple and useful measure. J Clin Endocrinol Metab 86: Legro RS, Finegood D, Dunaif A 1998 A fasting glucose to insulin ratio is a useful measure of insulin sensitivity in women with polycystic ovary syndrome. J Clin Endocrinol Metab 83: Katz A, Nambi SS, Mather K, Baron A, Follmann DA, Sullivan G, Quon MJ 2000 Quantitative insulin sensitivity check index: a simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab 85: Carmina E, Gonzalez F, Vidali A, Stanczyk FZ, Ferin M, Lobo RA 1999 The contributions of oestrogen and growth factors to increased adrenal androgen secretion in polycystic ovary syndrome. Hum Reprod 14:

6 Silfen et al. Comment J Clin Endocrinol Metab, January 2002, 87(1): Brismar K, Holl K 1993 Clinical application of IGFBP-1 and its regulation. Growth Regul 3: Le Stunff C, Bougnères P 1994 Early changes in postprandial insulin secretion, not in insulin sensitivity, characterize juvenile obesity. Diabetes 44: Lee PD, Durham SK, Martinez V, Vasconez O, Powell DR, Guevara-Aguirre J 1997 Kinetics of insulin-like growth factor and IGF-binding protein responses to a single dose of growth hormone. J Clin Endocrinol Metab 82: Pang S, Lerner AJ, Stoner E, Levine LS, Oberfield SE, Engel I, New MI 1985 Late-onset adrenal steroid 3 -hydroxysteroid dehydrogenase deficiency. I. A cause of hirsutism in pubertal and postpubertal women. J Clin Endocrinol Metab 60: Argente J, Caballo N, Barrios V, Pozo J, Munoz MT, Chowen JA, Hernandez M 1997 Multiple endocrine abnormalities of the growth hormone and insulinlike growth factor axis in prepubertal children with exogenous obesity: effect of short- and long-term weight reduction. J Clin Endocrinol Metab 82: Park MJ, Kim HS, Kang JH, Kim DH, Chung CY 1999 Serum levels of insulin-like growth factor-1, free IGF-1, IGF binding protein-1, IGFBP-3 and insulin in obese children. J Pediatr Endocrinol Metab 12: Travers SH, Labarta JI, Gargosky SE, Rosenfeld RG, Jeffers BW, Eckel RH 1998 Insulin-like growth factor binding protein-1 levels are strongly associated with insulin sensitivity and obesity in early pubertal children. J Clin Endocrinol Metab 83: Miller WL 1999 The molecular basis of premature adrenarche: an hypothesis. Acta Paediatr Suppl 433: Wong WW, Copeland KC, Hergenroeder AC, Hill RB, Stuff JE, Ellis KJ 1999 Serum concentrations of insulin, insulin-like growth factor-1, and insulin-like growth factor binding proteins are different between white and African American girls. J Pediatr 135: Yanovski JA, Sobik KN, Nguyen TT, Sebring NG 2000 Insulin-like growth factors and bone mineral density in African American and White girls. J Pediatr 137: