|Year : 2013 | Volume
| Issue : 3 | Page : 76-83
Immunohistochemical study of estrogen and androgen receptors in skin tags
Arwa M Hassan1, Laila M Mohammad1, Ahmad M Nassar1, Mona A Abd-Elazeem2, Marwa M Abaza3
1 Dermatology and Venereology Department, Faculty of Medicine, Tanta University, Egypt
2 Pathology Department, Faculty of Medicine, Tanta University, Egypt
3 Dermatology and Venereology Department, Damanhour Hospital, Egypt
|Date of Web Publication||23-Jan-2018|
Arwa M Hassan
14 Moatassem Street, Tanta
Source of Support: None, Conflict of Interest: None
Skin tags (ST) are the most common fibro-epithelial skin tumors. They are tumors of the middle-aged and elderly people. However, in women, the age of 50 is suggested to be the turning point of life at which the development of skin tags comes to a stop. A major event that occurs around this period of life is menopause/andropause, after menopause, estrogen receptors amounts decrease significantly. Estrogen receptors (ER) and androgen receptors (AR) are members of the steroid receptor gene superfamily. Estrogen receptors positivity has been demonstrated in most skin tumors by both histochemical and biochemical assays.
This study aimed to assess the role of estrogen and androgen receptors in the development of skin tags by assessment of their immunohistochemical expression. Twenty five patients were included in this study. Snip excision of skin tags was done for each patient, punch biopsy from adjacent sites from skin tags have been taken as perilesional skin biopsy only for twelve patients. Histopathology was performed to confirm the diagnosis. Immuno-histochemical staining was done to detect ER and AR for skin tags, perilesional skin, and control normal skin.
Immunohistochemical staining for ER was positive in 64% of skin tag cases and AR were positive in 60% of them. There was a statistically significant difference in ER and AR staining pattern between skin tags, perilesional skin and normal skin. It could be concluded that ST correlated positively with ER and AR immunohistochemical staining which may implicate a role for them in the pathogenesis of skin tags.
Keywords: Skin tags, Estrogen receptors, Estrogen receptors
|How to cite this article:|
Hassan AM, Mohammad LM, Nassar AM, Abd-Elazeem MA, Abaza MM. Immunohistochemical study of estrogen and androgen receptors in skin tags. J Microsc Ultrastruct 2013;1:76-83
|How to cite this URL:|
Hassan AM, Mohammad LM, Nassar AM, Abd-Elazeem MA, Abaza MM. Immunohistochemical study of estrogen and androgen receptors in skin tags. J Microsc Ultrastruct [serial online] 2013 [cited 2021 Apr 13];1:76-83. Available from: https://www.jmau.org/text.asp?2013/1/3/76/223826
| 1. Introduction|| |
Skin tags (ST) are the most common fibro-epithelial skin tumors. They are small flesh colored to dark brown sessile or pedunculated papillomas, composed of loose fibrous tissue. They grow in the natural folds of the skin such as neck, axillae, inguinal, thigh, perineal and inframammary regions. They tend to grow progressively in size and do not involute spontaneously . Etiology of skin tags includes, insulin resistance and insulin like growth factor , hormonal factor ,, high mast cell count , and human papilloma virus .
Skin tags are tumors of the middle-aged and elderly people, and they are very common, particularly in women at the menopause. Skin tags may be aggravated during pregnancy and may resolve post partum . However, in women, the age of 50 is suggested to be the turning point of life at which the development of skin tags comes to a stop. A major event that occurs around this period of life is menopause/andropause, after menopause, estrogen receptors amounts decrease significantly .
Estrogen receptors (ER) are class I members of the steroid receptor gene superfamily, which includes the androgen receptors (ARs), progesterone receptors, glucocorticoid receptors, thyroid hormone receptors and retinoic acid receptors, they are defined as ligand-inducible transcription factors . All receptors of the nuclear receptor superfamily have a similar architecture with independent but interacting functional domains . There are two subtypes of ER; estrogen receptor-alpha (ERα) and estrogen receptor-beta (ERβ). Both ERα and ERβ bind 17β-estradiol with high affinity . Estradiole binds to keratinocytes with high affinity, and in its normal physiological range increases the regulation of the level of ER receptors and induces keratinocyte proliferation . In addition, estrogen targets human skin fibroblasts. 17β-Estradiol has been shown to increase fibroblast proliferation in human skin .
ERβ is more widely distributed within the skin and skin structures than ERα . The variation in distribution of receptors within the skin suggests that each has a different, cell-specific role . ERα and ERβ co-express in human skin fibroblasts and likely play a role in the regulation of estrogen action in the skin . The presence of estrogen receptor in some skin tumors has led to several studies assessing the association between sex hormone status and cutaneous malignancies. Estrogen receptor positivity has been demonstrated in most skin tumors by both histochemical and biochemical assays .
1.1. Aim of the work
This study aimed to assess the role of estrogen and androgen receptors in the development of skin tags by assessment of their immunohistochemical expression.
1.2. Patients and methods
Twenty five patients were included in this study. They were selected from the out-patient clinic of Dermatology and Venereology Department, Tanta university hospital during the period from April 2011 to October 2012. Informed consent was taken from all the patients and also approval from the research ethics committee was done. Patients perceiving sex steroid hormonal therapy have been excluded. For all subjects the followings were done; thorough complete history taking, general and dermatological examination, estimation of body mass index, snip excision of skin tags was done for each patient, punch biopsy of 2 mm size from adjacent sites from skin tags have been taken as perilesional skin biopsy only for twelve patients. Specimens were fixed in 10% formalin. First, hematoxylin and eosin stained histopathological sections were performed to confirm the diagnosis. Normal skin, as paraffin blocks which were of matching age, sex, and site to the patients group, has been collected to be used as controls from twelve blocks. Immunohistochemical staining was done to detect ER and AR for skin tags, perilesional skin, and control normal skin. Immunostaining was carried out by a standard streptavidin–biotin peroxidase technique with diaminobenzidine chromogen. The antibody used is mouse monoclonal antihuman antibody for androgen receptor and rabbit monoclonal antihuman antibody for estrogen receptor.
AR positivity was defined as nuclear staining while cytoplasmic staining was non-specific. Tumors with 10% or more nuclear stained cells were considered positive for AR according to . Estrogen receptors staining was also nuclear. In the nuclear staining, the percentage of the stained cells was determined and was scored using  scoring system. The staining reaction was subdivided according to the percentage of stained nuclei as follows: Negative (0) = no nuclear immuonostaining, weak staining (+1) = less than 10% positive nuclear staining, moderate staining (+2) = 10–50% positive nuclear staining and strong staining (+3) = more than 50% positive nuclear staining.
Statistical presentation and analysis of the present study was conducted, using the mean, standard deviation, unpaired Student t-test, analysis of variance [ANOVA] test, chi-square test and Monte-Carlo (MCp) test by computer statistical program SPSS V, 16.
| 2. Results|| |
In the current study, the age of skin tags cases ranged from 17 to 65 years with mean age of 40.32 ± 14.01 years. Correlation between skin tags number and age groups was done, it was statistically significant (P= 0.030). There was female predominance; 20 cases were females and 5 were males. Neck (96%) and axilla (44%) respectively were the most common sites. Thirteen patients with skin tags (52%) had positive family history. All patients complained of disfiguring skin lesions (100%), four patients (16%) complained of itching, while only one patient (4%) complained of pain. Body mass index (BMI) has been calculated by this equation: BMI = weight (kg)/height (m2 ). In this study, 2 patients (8%) were normal (BMI: 18.5–24.9), 5 patients (20%) were overweight (BMI: 25–29.9), 9 patients (36%) were obese grade I (BMI: 30–34.9), 4 patients (16%) were obese grade II (BMI: 35–39.9), and 5 patients (20%) were obese grade III (BMI: 40 and above) with a mean 2.20 ± SD 1.22 [Table 1].
|Table 1: Distribution of the studied cases according to clinical data, site and BMI.|
Click here to view
Regarding medical condition in our study, the main occurrence was obesity in 18 patients (72%) then diabetes mellitus in 7 patients (28%) and hypertension in 7 patients (28%). Also there were other dermatologic diseases that have been observed in this study, the most common was acanthosis nigricans in 8 patients (32%) [Table 2].
|Table 2: Distribution of the studied cases according to other medical conditions and associated dermatological diseases.|
Click here to view
2.1. Histopathological results
Heamatoxylin and eosin staining of skin specimens sections showed that skin tags are histologically composed of loose collagen fibers and dilated capillaries. The multiple small furrowed papules usually show papillomatosis, regular acanthosis and hyperkeratosis. The common filiform smooth growths show slight to moderate acanthosis and occasionally mild papillomatosis. The connective tissue stalk is composed of loose collagen fibers and often contains numerous dilated capillaries. Nevus cells are found in many of the filliform growths, indicating that some of them represent involuting melanocytic nevi [Figure 1] and [Figure 2].
|Figure 1: A case of skin tag showing loose connective tissue core, hyperke-ratosis and papillomatosis (H&E 100×).|
Click here to view
|Figure 2: A case of skin tag showing loose connective tissue containingdilated blood vessels and fat cells (H&E 200×).|
Click here to view
2.2. Immunohistochemical results
2.2.1. Estrogen receptors
After confirming our clinical diagnosis by Hematoxylin and Eosin staining of all sections, immunohistochemical staining for estrogen receptor (ER) by rabbit monoclonal antihuman antibody was done to all sections and staining pattern was nuclear. In the skin tags 9 cases (36%) were (0) negative and 16 cases (64%) were positive. Moreover, in the perilesional skin 9 cases (75%) were (0) negative and 3 cases (25%) were (+1) weakly stained. However the normal skin 12 cases (100%) were all (0) negatively stained. There was a statistically significant difference in estrogen receptors staining pattern between skin tags, perilesional skin and normal skin (P= 0.003) [Table 3] [Figure 2],[Figure 3],[Figure 4],[Figure 5],[Figure 6],[Figure 7],[Figure 8].
|Table 3: Comparison between the skin tag, perilesional skin and normal skin according to ER and AR immuonohistochemical (IHC) staining.|
Click here to view
|Figure 3: Normal skin showing negative (0) staining for estrogen receptors(IHC 200×).|
Click here to view
|Figure 4: A perilesional skin showing negative (0) staining for estrogenreceptors (IHC 400×).|
Click here to view
|Figure 5: A perilesional skin showing mild positive (+1) nuclear staining ofestrogen receptors (IHC 400×).|
Click here to view
|Figure 6: A case of skin tag showing mild (+1) nuclear staining of estrogenreceptors (IHC 400×).|
Click here to view
|Figure 7: A case of skin tag showing moderate (+2) nuclear staining of estro-gen receptors (IHC 400×).|
Click here to view
|Figure 8: A case of skin tag showing strong positive (+3) nuclear staining forestrogen receptors (IHC 400×).|
Click here to view
2.2.2. Androgen receptors
Immunohistochemical staining for androgen receptor (AR) was done to all sections and staining pattern was also nuclear. In the skin tags, 15 of cases (60%) were positively stained, and 10 of cases (40%) were negatively stained. However, in the perilesional skin, 4 of cases (33, 3%) positively stained, and 8 of cases (66, 6%) were negatively stained. In the control normal skin, androgen receptors were weakly stained as the target cells (follicular keratinocytes, sebocytes) in all specimens stained but lower than 10%. So according to the scoring system, these specimens were 100% negatively stained for androgen receptors. There was a statistically significant difference in staining pattern between skin tags, perilesional skin and normal skin (P= 0.001) [Table 3] [Figure 9],[Figure 10],[Figure 11].
|Figure 9: Normal skin showing negative staining for androgen receptors (IHC200×).|
Click here to view
|Figure 10: A case of skin tag showing positive nuclear staining of androgenreceptors (IHC 400×).|
Click here to view
|Figure 11: A case of skin tag showing negative nuclear staining of androgenreceptors (IHC 400×).|
Click here to view
It was noticed that no correlation between age and sex and estrogen receptor staining or androgen receptor staining in skin tags. Regarding body mass index (BMI), estrogen receptor staining revealed no relation with obesity grades, however androgen receptor staining showed a significant relation with BMI (P= 0.017). In a correlation between DM and both ER and AR immunostaining, there was no significant relation between estrogen or androgen receptors and DM in patients with skin tags. Regarding acanthosis nigricance (AN), which were associated with skin tags in 8 patients, there was no relation between estrogen or androgen receptors and AN association with ST.
Correlation between site and estrogen receptor staining revealed no significant relation between them. Moreover, there was no significant relation between site and androgen receptor staining [Table 4]. Considering the number of skin tags and estrogen receptor staining, there was no relation in the current study between them. A similar result has been obtained from correlation between number of skin tags and androgen receptor staining [Table 5].
|Table 4: Relation between site and estrogen and androgen receptor staining in skin tags.|
Click here to view
|Table 5: Relation between number of skin tags and estrogen and androgen receptor staining in skin tags.|
Click here to view
| 3. Discussion|| |
As skin is considered as the largest nonreproductive target on which estrogens and androgens act, El Safoury et al.  assumed a possible relationship between the pathogenesis of skin tags and sex steroid balance. Another phenomenon is the association of skin tags in obese patients, which may also be explained by the interplay of sex steroids and their receptors in skin tags.
In the current study, the age of skin tags cases ranged from 17 to 65 years with mean age of 40.32 ± 14.01 years. Correlation between skin tags number and age groups was done, it was statistically significant. These results were similar to the findings of other studies which reported that the risk of getting skin tags and the mean number of STs increased with age ,. However, these studies found that ST increase in frequency, reached a peak value up to the 5th decade, then they noticed stagnation or decline in this frequency. This decline after the fifth decade was not detected in the current study; this may be because of that the patients older than 50 years were all obese, same finding was reported by El Safoury and Ibrahim .
Regarding the gender in this study, there was female predominance: 20 cases were females and 5 were males, this may be explained by the cosmetic concern of the females who ask more for medical advice. This result was similar to what Waisman  found in subjects with skin tags, it was twice more frequent in women than in men. However some other studies did not found significant sex preference ,,.
Considering the localization of skin tags in the cases in the present study, neck (96%) and axilla (44%) respectively were the most common sites. These results went with other studies reported that skin tags were most commonly seen on the neck followed by the axilla then the groin and other sites ,. However Mackie and Quinn  found that the most frequent localization of skin tags were in the axilla region followed by the neck region.
According to family history in this study, thirteen patients with skin tags (52%) had positive family history. In a study done by Erkek et al. , a positive family history was obtained from (65.5%) of patients.
Considering the general medical condition of the patients with skin tags in this study, majority of the patients had obesity (72%); Diabetes mellitus (28%) and hypertension (28%) were equally presented. Thappa  found that skin tags have been associated with many diseases including DM and obesity. He also reported that the frequency of ST had been found to co-exist with DM patients with a significant relationship, and ST may serve as a marker for DM.
However, Kahana et al. , reported that skin tags were not associated with increased incidence of obesity compared to the general population but reported that those patients with skin tags had greater impairment of carbohydrate metabolism and detection of skin tags may serve a cutaneous marker for identifying patients at increasing risk of having DM.
Crook  conducted a study on patients who had skin tags and who also displayed an abnormal lipid profile, which is also known as the atherogenic profile and is associated with insulin resistance, type 2 diabetes mellitus and an increased risk of cardiovascular disease.
In a study done on obese patients, the percentage of those with STs increased with the severity of obesity . In a correlation between skin tags number and body mass index in this study, it was noticed that there was increase in number of skin tags in obese and overweight patients but it was statistically insignificant. Rasi et al.  had found that no correlation was found between number of skin tags and BMI, while another study reported that the mean number of skin tags was significantly higher among obese participants in comparison to the non-obes .
It was noticed in the current study that acanthosis nigricans (AN) had the higher percentage among associated skin diseases that had been detected. It was detected in eight (32%) cases. Similar findings were also reported  but others showed no statistically significant difference between cases of skin tags with AN and those without .
Hormonal factor in the pathogenesis of ST had been suggested by some authors who found that skin tags may be aggravated by pregnancy and may resolve postpartum ,,.
The role of estrogen and androgen receptors in pathogenesis of skin tags was studied only by El-Safoury et al. , they investigated androgen receptors (AR), estrogen receptors a (ERα), and estrogen receptors β (ERβ) by reverse transcription-polymerase chain reaction (RT-PCR). To the best of our knowledge the present study is the first immunohistochemical study for this subject.
In the present study, estrogen receptors were significantly positive in skin tags in contrast to perilesional skin and normal skin. Estrogen receptors scoring were moderate (+2) and strong (+3), nuclear immune staining in 24% and 12% respectively in skin tags cases compared with ER scoring in perilesional and normal skin 0%.
Moreover, androgen receptors were also significantly positive, nuclear immune-staining in skin tags in 15 patients (60%), compared with perilesional skin in 4 patients (33.3%), and normal control (0%). These significant results were in agreement with El Safoury et al. . They found that the level of estrogen receptors (both α and β) was significantly higher in skin tags than in controls. They also found the same significant upsurge for androgen receptors in skin tags relative to control.
A correlation between age, sex and estrogen receptor expression has been done in the current study and there was no significant relation between neither age nor sex and estrogen receptor staining. In another study done by Auw-Haedrich and Feltgen , there was significant relation between age and ER expression but they found no difference in the proportion of cells of Meibomian gland More Detailss expressing ERs between female and male patients. Regarding age and sex in correlation with AR expression in this study, no relation between age or sex and androgen receptor staining in skin tags had been found.
In the current study there was no correlation between DM and ER expression. On the other hand, several studies have reported a potential relationship between estrogen and glucose homeostasis in physiological and pathological states as DM . Also, there was no relation between neither BMI nor AN and ER expression in skin tags. This result was in agreement also with El Safoury et al. .
In the current study there was no relation between AR immunostaining and DM. Cao et al. , reported that AR were negatively correlated with the course of DM.
There was no relation between AR immunostaining and AN. In a study done by Charles et al. , the severity of the AN was directly related to the degree of hyperinsulinemia and was not correlated to the level of androgens. They suggested that the insulin resistance seen in obese acanthotic subjects cannot be attributed to hyperandrogenemia or obesity, since control subjects matched for severity of androgen excess or degree of obesity had much less insulin resistance. Moreover, El Safoury et al.  reported that there was no significant difference in AR level between participants with acanthosis nigricance and those without.
However AR staining showed a significant relation with BMI in this study, where 11 patients from 15 (73.3%) with positive AR staining were obese. This was similar to the results of Gustafson et al. , who showed a strong relationship between two common polymorphisms in the AR gene and indices of central obesity, particularly in women.
The results in the present study point out to a significant estrogen and androgen receptors nuclear expression, associated with the development of skin tags. According to the results of this study, the significant relationship between estrogen receptors as well as androgen receptors and skin tags, may represent a strong evidence of the pathogenesis of sex hormones in this disease. Further research involving larger series of patients is recommended and it may be useful to evaluate antiandrogen or anti-estrogen therapy effect on the course of progress of skin tags.
Conflict of interest
We have no conflict of interest to declare.
| References|| |
MacKie RM, Quinn AG. Non-melanoma skin cancer and other epidermal skin tumors. In: Burns T, Breathnach SM, Cox N, et al., editors. Rook’s Textbook of Dermatology. 7th ed. Oxford, UK: Blackwell; 2004. p. 36-45.
Ten S, Maclaren N. Insulin resistance syndrome in children. J Clin Endocrinol Metab 2004;89(6):2526-39.
Errickson CV, Matus NR. Skin disorders of pregnancy. Am Fam Physician 1994;49(3):605-10.
Tunzi M, Gray GR. Common skin conditions during pregnancy. Am Fam Physician 2007;75(2):211-8.
Zaher H, El Safoury OS, El Komy MM, Mahmoud SB, Abd El Hameed H. Study of mast cell count in skin tags. Indian J Dermatol 2007;52(4):184-7.
El-Tahlawy SR, Esmat SM, Bosseila MA, et al. Detection and localization of human papilloma virus (HPV) by PCR in situ hybridization. Egypt J Derm Androl 2000;20:81-8.
El Safoury O, Rashid L, Ibrahim M. A study of androgen and estrogen receptors alpha and beta in skin tags. Indian J Dermatol 2010;55(1):20-4.
Couse JE, Mahato D, Eddy EM, Korach KS. Molecular mechanism of estrogen action in the male: insights from the estrogen receptor null mice. Reprod Fertil Dev 2001;13(4):211-9.
Gong H, Xie W. Orphan nuclear receptors, PXR and LXR: new ligand and therapeutic potential. Expert Opin Ther Targets 2004;8(1):49-54.
Pelletier G, Ren L. Localization of sex steroid receptors in human skin. Histol Histopathol 2004;19:629-39.
Verdier-Sevrain S, Yaar M, Cantatore J, Traish A, Gilchrest B. Estradiol induces proliferation of keratinocytes via receptor-mediated mechanisms. FASEBJ 2004;18:1252-4.
Tomaszeweski J, Adamiak A, Skorupski P, Rzeski W, Rechberger T. Effects of 17β-estradiol and phytoestrogen daidzein on the proliferation of pubocervical fascia and skin fibroblasts derived from women suffering from stress urinary incontinence. Ginekol Pol 2003;74:1410-4.
Thornton MJ, Taylor AH, Mulligan K, Al-Azzawi F, Lyon C, O’Driscoll J, Messenger AG. The distribution of estrogen receptor beta is distinct to that of estrogen receptor alpha and the androgen receptor in human skin and the pilosebaceous unit. J Investig Dermatol Symp Proc 2003;8:100-3.
Thornton MJ. The biological actions of estrogen on skin. Exp Dermatol 2002;11:487-502.
Haczynski J, Tarowski R, Jarzabec K, Slomczynska M, Wolczynski S, Magoffin D, Jakowicki JA, Jakimiuk AJ. Human cultured skin fibroblasts express estrogen receptor alpha and beta. Int J Mol Med 2002;10:149-53.
Cohen C, DeRose C, Patricia B, Campbell WG, Schlosnagle DC, Sgoutas D. Estrogen receptor status in malignant melanoma. Am J Dermatopathol 1990;12(6):562-4.
Park JS, Jung WH, Kim JK, Hwang HK, Cho SI, Yoon DS, Chi HS, Kim BR. Estrogen receptor alpha, estrogen receptor beta, and progesterone receptor as possible prognostic factor in radically resected gallbladder carcinoma. J Surg Res 2009;152:104-10.
Deroo BJ, Karoch KS. Estrogen receptors and human disease. J Clin Invest 2006;116:561-70.
Banik R, Lubach D. skin tags: localization and frequencies according to sex and age. Dermatology 1987;174:180-3.
El Safoury O, Ibrahim M. A clinical evaluation of skin tags in relation to obesity, type 2 diabetes mellitus, age, and sex. Indian J Dermatol 2011;56(4):393-7.
Waisman M. Cutaneous papillomas of the neck. South Med J 1957:50:725-32.
Chiritescu E, Maloney ME. Acrochordons as a presenting sign of nevoid basal cell carcinoma syndrome. J Am Acad Dermatol 2001;44:789-94.
Garcia-Hidalgo L, Orozco-Tropete R, Villa AR, Dalman JJ, Ortiz-Pedroza G. Dermatoses in 156 obese adults. Obes Res 1999;7(3):299-302.
Tamega AA, Aranha AM, Guiotoku MM. Associationbetween skin tags and insulin resistance. Ann Bras Dermatol 2010;85:1.
Erkek E, Kisa U, Bagci Y, Sezikli H. Leptin resistance and genetic predisposition as potential mechanisms in the development ofskin tags. Hong Kong J Dermatol Venerol 2011;19:108-14.
Thappa DM. Skin tags as markers of diabetes mellitus: An epidemiological study in India. J Dermatol 1995;22:729-31.
Kahana M, Grossman E, Feinstein A, Ronnen M, Cohen M, Millet MS. Skin tags: a cutaneous marker for diabetes mellitus. Acta Derm Venereol 1987;67:175-7.
Crook MA. skin tags and the atherogenic lipid profile. J Clin Pathol 2000;53(11):873-4.
Rasi A, Soltani-Arabshahi R, Shahbazi N. Skin tags as a cutaneous marker for impaired carbohydrate metabolism: a case control study. Int J Dermatol 2007;46(11):1155-9.
Kar S, Kirshnan A, Varma SP. Pregnancy and skin. J Obstet Gynecol India 2012;62(3):268-75.
Kroumpouzos G, Cohen LM. Dermatoses of pregnancy. J Am Acad Dermatol 2001;45:1-19.
Auw-Haedrich C, Feltgen N. estrogen receptor expression in meibomian glands and its correlation with age and dry-eye parameters. Graefes Arch Clin Exp Ophthalmol 2003;241:705-9.
Rodrigo PA, Machado UF, Gustafsson JA. Estrogen receptors: new players in diabetes mellitus. Trends Mol Med 2006;12(9): 425-31.
Cao J, Li J, Hao W, Li X, Wang H, Liu L, Zhu B, Guo Z. Correlation of sex hormone and androgen receptor with diabetes mellitus in elderly men. Aging Male 2011;14(3):162-7.
Charles A, Edward JS, Melvin JP. Insulin resistance with acanthosis nigricans: the roles of obesity and androgen excess. Metabolism 1986;35(3):197-205.
Gustafson DR, Wen MJ, Koppanat BM. Androgen receptor gene repeats and indices of obesity in older adults. Int J Obes 2003;27:75-81.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8], [Figure 9], [Figure 10], [Figure 11]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]