Serum markers of bone metabolism in a group of transsexual patients receiving estrogen therapy

Michael Thomas

F.R.C.Path., Department of Chemical Pathology and Human Metabolism, Royal Free Hospital, Pond Street, Hampstead, London NW3 2QG.

James S Barrett

M.B., B.S., Gender Identity Clinic, Department of Psychiatry, Charing Cross Hospital, Fulham Palace Road, London W6 8RF

Donald H Montgomery

F.R.A.N.Z.C.Psych., Gender Identity Clinic, Department of Psychiatry, Charing Cross Hospital, Fulham Palace Road, London W6 8RF

Gendys Conference, 1994



There has been much speculation on the role of estrogenic hormones in the metabolism of bone. Reports on the effect of estrogen in animal models and tissue culture are contradictory as to its action. Serum markers of bone metabolism such as osteocalcin and the carboxy terminal peptide of type 1 procollagen are released into the circulation during bone matrix formation or resorption and allow the assessment of bone matrix turnover in vivo in man.

We measured the effect of supraphysiological doses of estrogens on these markers of bone metabolism in a group of transsexual biologically male patients (DSM IIIR 302.5 or ICD10 F64.0). There was a significant reduction in the level of osteocalcin in this group (t=15.36, p<0.001) and a significant correlation of osteocalcin with total alkaline phosphatase (r0.432, p=0.001). Decreased levels of osteocalcin may reflect decreased osteoblastic activity and these findings suggest that at supraphysiological levels, estrogens may inhibit matrix formation in man.

Key words

  • bone metabolism markers, osteocalcin,
  • pro-collagen 1 peptide, estrogens,
  • transsexuals


ln recent years several new serum markers of bone metabolism have been described (1,2) and their clinical utility evaluated.(3) Much of the impetus in the clinical situation has been the result of increasing use of hormone replacement therapy in women to prevent or delay osteoporosis. The markers now available for assessment of bone metabolism in the clinical situation may be enzymes associated with the either osteoblastic or osteoclastic activity such as the phosphatases or they may be fragments of bone matrix released into the circulation during the formation of new matrix or the breakdown and resorption of mature matrix.

Osteocalcin, also known as Gla-protein, was the first specific marker identified in blood and is the major non-collagenous protein of bone.(4) It has been shown to be a sensitive and specific marker of osteoblastic activity in a variety of metabolic bone diseases including osteoporosis.(5-8)

While osteocalcin is the predominant non-collagenous protein of bone, type 1 collagen accounts for greater than 90% of the organic matrix.(9) Type 1 collagen is synthesised as a larger precursor molecule, type 1 procollagen, which contains peptide sequences at both ends which are removed by specific proteases before the collagen molecules are assembled into fibres.(10) The carboxyterminal end, known as the carboxyterminal propeptide of type 1 procollagen (PICP) can be measured in blood where it too has been shown to reflect changes in metabolic bone disease.(11-13) At the cellular level, bone remodelling occurs at discrete points within the skeleton known as bone-remodelling units.(14) In osteoporosis, abnormalities in bone remodelling are known to be present. It is widely accepted that estrogens act directly on bone cells through high affinity estrogen receptors and can prevent early phase bone loss if begun shortly after the menopause. A recent report(15) suggests that estrogens may assist in the prevention of bone resorption in osteoporosis by blocking interleukin-6 production in osteoblasts.

Most of the work reported to date on the effects of estrogens on bone metabolism has been derived from in vitro work using tissue culture, ovariectomized animal models or females in the peri- or post- menopausal state. Transsexual males are biologically normal males suffering from a psychiatric disorder (DSM IIIR 302.5 or ICO10 F64.0)characterised by the unshakeable belief that they are meant to be idividuals of the opposite sex. Such patients present to medical services requesting gender reassignment surgery. Before surgery, other confounding psychiatric disorders must be excluded and they are required to live for some years in the female gender role showing a stable psychological, sexual, social and employment state. During this time, they are often given estrogenic hormones. These patients therefore provide a unique group in which the effect of long-term administration of high doses of estrogenic hormones on bone metabolism may be investigated. The purpose of the present study was to determine if the estimation of serum markers can reveal any possible alterations in bone metabolism as a result of this therapy.

Methods and Materials.

The patient population was drawn from that attending the Gender Identity Clinic at Charing Cross Hospital and under the care of one of us (DHM) All the patients (53 in total) included in the study were receiving estrogen drugs. In 36 patients conjugated estrogens in the form of Premarin was given alone at a dose of between 1.5 mg per day up to 10 mg per day (median 7.5 mg per day). Two of the 36 patients had also been castrated. In 5 patients Premarin was given in combination with cyproterone acetate (50 mg per day). Eight patients received Premarin in combination with medroxyprogesterone (5 mg per day). Two patients were taking ethinyl estradiol (50µg tds), one in combination with medroxyprogesterone (5 mg per day). One patient was taking Premarin (5 mg per day), cyproterone acetate (50 mg per day) and medroxyprogesterone (5 mg per day).

The patients ages were between 21 and 65 years (mean age 41.65 years) and each had been receiving estrogenic drug therapy for a minimum of two years (average 6 years, maximum 26 years).

Blood samples were taken at the time of attendance at clinic and allowed to clot. The serum was then immediately separated and frozen for storage at -20°C prior to analysis.

Osteocalcin was measured using reagents obtained from INCStar Ltd (Winnersh, Berkshire).This is a radioimmunoassay method using a rabbit anti-bovine osteocalcin antibody and 125|bovine osteocalcin. Phase separation is achieved using a complex of goat anti-rabbit serum, carrier rabbit serum and polyethylene glycol.(16)

The carboxyterminal peptide of type 1 procollagen was measured using a commercially supplied kit from Farmos Diagnositca through Pharmacia Biosystems Ltd (Milton Keynes). The radioimmunoassay uses a rabbit antiserum to the carboxyterminal peptide and second antibody phase separation using a goat anti-rabbit antibody covalently bound to solid particles.(17)

Total alkaline phosphatase and acid phosphatase were measured using established methods on a Hitachi 717 analyser and reagents supplied by Boehringer Mannheim UK Ltd (Lewes).(18,19)


Serum alkaline phosphatase, acid phosphatase, osteocalcin and PICP values for the transsexual group are shown in Figure 1 and compared with reference population distributions.

Osteocalcin in the transsexual group had a mean value of 1.3 µg/l (sd=0.80, sem=0.11, n=53).

The reference population gave a mean osteocalcin value of 4.2 µg/l (sd=1.13, sem=0.16, n=53). As assessed by Students t Test, serum osteocalcin for the transsexual group was significantly less than that of the reference population (t=15.36, p<0.001).

The average value for the PICP was 98 µg/l (sd=46.7, sem=6.4, n=53) for the transsexual group and was not significantly difference to a reference population of either males (mean=120 µg/l, sd=48.3, sem=8.3 n=34) or females (mean=110 µg/l, sd=0.6, sem=5.0, n=38).

The mean serum alkaline phosphatase was 46 IU/l (sd=6.4, sem=0.9, n=53) and the mean serum acid phosphatase 3.7 IU/l (sd=1.2, sem=0.16, n=53) for the transsexual group. The values for both phosphatases fell within the reference populations used routinely by the laboratory (alkaline phosphatase 35-130 IU/l; acid phosphatase 2.6-6.3 IU/l).

There was a significant correlation between serum alkaline phosphatase and osteocalcin in this group (r=0.4318, p=0.001) but no correlation with PICP (r=.2143, p=0.123). Acid phosphatase in the transsexual group did not correlate with either osteocalcin (r=0.0174, p=0.902) or PICP (r=0.0017, p=0.991).

Figure 1


Osteocalcin is commonly accepted as a useful marker of bone formation and a reflection of osteoblastic activity. Decreased levels of serum osteocalcin have been reported in untreated hypothyroidism, hypoparathyroidism(7) and in patients on corticosteroids(20) where it has been said to reflect bone loss due to decreased matrix formation. Bone loss has been reported even in patients on low dose steroids.(21) We found a significant decrease in osteocalcin in the transsexual group relative to our reference population. It is not possible to say if this is a direct effect of estrogens on bone or the result of secondary effects but it suggests that in transsexuals receiving high dose estrogen therapy there is a decrease in bone matrix formation.

Other workers have shown the contrary effect of estrogen, an increase in bone formation, but these reports are usually where small or physiological amounts of estrogen are concerned. Yamamoto and Rodan(22) administered small amounts (2.4-24 µmol/day) of 17β estradiol directly into the tibiae of ovariectomized rats and demonstrated an inhibition of bone resorption, a stimulation of bone formation and the restoration of bone loss associated with ovariectomy. Estrogens administered in physiological amounts appear to maintain bone volume in rats not only through inhibition of bone resorption but also through stimulation of bone formation.(23) The same group of workers have also previously demonstrated in rats an effect of estrogens at higher concentrations stimulating trabecular bone formation(24) but caution that rats do not undergo sequential remodelling as humans do. Acid phosphatase, our only marker of osteoclastic activity in this study, showed no significant alteration relative to established reference populations but is probably too insensitive. This study can not therefore exclude the possibility of either an inhibition or stimulation of bone resorption.

Available evidence of a direct effect of 17β estradiol on bone formation in man remains inconclusive and its further modulation by cytokines at the tissue level as described by Girasole et al(19) complicates matters further. In our own study, estrogens were administered at supraphysiological doses and appear to suggest a decrease in bone formation. An inhibitory effect of estrogens on radial bone growth in rats has been described(25) and was associated with an inhibition of the expression of bone matrix protein genes, which included osteocalcin, in periosteal cells. Egrise et al(26) using cultured osteoblast-like cells to investigate the interaction of estrogens, triiodothyronine and caltitriol (1,25 (OH)2D3) also observed a decrease in osteocalcin production when supraphysiological concentrations of 17β estradiol was used in the culture medium. Reductions in the production of bone proteins, including osteocalcin, PICP and total alkaline phosphatase have also been described in a small group of osteoporotic females receiving a combined estrogen-progestagen treatment over a 150 week period.(27) In rats, the intermittent administration of 17β estradiol has been shown to result in a significant increase in cancellous bone volume over that seen when administration is continuous.(28) It is suggested that this may be due to receptor down regulation where administration of 17β estradiol is continuous.

An association between alkaline phosphatase and osteocalcin has been reported previously in patients on corticosteroids.(20) This association has been suggested as representing a correlation between two serum indices of osteoblastic activity. Our results confirm a relationship between total alkaline phosphatase and osteocalcin but do not reveal any significant difference from the total alkaline phosphatase of a reference population.

Total alkaline phosphatase has long been used as a marker of osteoblastic activity.(29)

Measurement of the bone specific isoenzyme was not done as part of this study. This may however prove to be a more sensitive indicator of osteoblastic activity as has been described by Leung et al,(30) using the method of Rosalki and Foo.(31,32)

Hasting et al(27) reports a reduction in PICP in a group of 12 osteoporotic women on a 150 week course of a combined estrogen-gestagen regimen and concludes that the therapy has no major stimulatory effect on osteoblastic type 1 collagen synthesis. We were unable to demonstrate any significant changes in the level of PICP in our transsexual group when compared with our reference populations. PICP production is expected to more closely reflect the amount of matrix laid down during bone formation than either alkaline phosphatase or osteocalcin. PICP is however rapidly cleared from the circulation via the mannose receptor in the liver endothelial cells.(33)

Whether or not estrogens at the level administered to the transsexual group can affect this disposal route remains unknown.

This study shows that transsexual males receiving estrogen therapy at supraphysiological levels have altered bone metabolism. The decreased levels of osteocalcin suggest that matrix formation is decreased although the effects of osteocalcin metabolism and its dependence on both Vitamin K and 1,25 (OH)2D3 for synthesis are factors which remain to be clarified. It is not possible to say whether or not there is a corresponding decrease in resorption of bone. It would seem prudent to review bone metabolism in greater detail in this group of patients particularly as they will be given hormone therapy for many years and, in all probability, for the rest of their life.


  1. Deftos.L.J., Bone Protein and Peptide Assays in the Diagnosis and Management of Skeletal Disease. Clin Chem 1991; 37; 1143-1148
  2. Price.P.A, New Bone Markers, Triangle 1988; 27 (1/2): 21-26
  3. Delmas. P.D, Biochemical Markers of Bone Turnover Methodology and Clinical Use in Osteoporosis. Am J Med 1991; Suppl 56: SSS-63s
  4. Deftos.L.J, GIowacki,J. Mechanisms of bone metabolism. in Kem DC, Frohlich E, editors. Pathophysiology. Philadelphia: JB Lippincott Co, 1984:3rd ed:445-468
  5. Slovik.D.M, Gundbeg.C.M, Neer.R.M, Lian.J.B. Clinical Evaluation of Bone Turnover by Serum osteocalcin Measurements in a Hospital Setting. J Clin Endocrinol Metab 1984; 59: 228-230
  6. Brown, J.P, Delmas, P.D, Malavat, L, Edourad, C, Chapuy, M.C, Meunier, P.J, Serum Bone Gla-protein: A Specific Marker for Bone Formation in Postmenopausal Osteoporosis. The Lancet 1984; i: 1091-1093
  7. Delmas, P.D, Wahner, H.W, Mann, K.G, Riggs, B.L, Assessment of bone turnover In postrnenopausal osteoporosis by measurement of serum bone Gla -protein, J Lab Clin Med 1983; 102: 470-476
  8. Price, P.A, Parthemore, J.G, Deftos, L.J, New Biochemical Marker for Bone Metabolism, J. Clin Invest 1980; 66: 878-883
  9. Burgeson, R.E, New collagens, new concepts. Annu Rev Cell Biol 1988; 4: 551-577
  10. Prockop, D.J, Kivirikko, K.I, Tuderman, L, Guzman, N.A, The biosynthesis of collagen and its disorders. N Engl J Med 1979; 301: 13-23, 77-85
  11. Taubmann, M.B, Kammerman, S, GoIdberg, B, Radioimmunoassay of procollagen in serum of patients with Paget's disease of bone. Proc Soc Exp Biol Med 1976; 1:52: 284-287
  12. Simon, L.S, Krane, S.M, Wortrnan, P.D, Krane, I.M, Kovitz K.L, Serum levels of Type I and Type III procollagen fragments in Paget's disease of bone. J Clin Endocrinol Metab 1984; 58: 110-120
  13. Parfitt, A.M, Simon, L.S, Villanueva, A.R, Krane, S.M, Procollagen type 1 carboxy-terminal extension peptide in serum as a marker of collagen biosynthesis in bone. Correlation with iliac bone formation rates and comparison with total alkaline phosphatase. J Bone Miner Res. 1987; 2: 427-436
  14. Riggs, B.L, Melton,L.J, The Prevention and Treatment of Osteoporosis. N Engl J Med 1992; 327: 620-627
  15. Girasole, G, Jilka, R.L, Passed, G, Boswell, S, Boder, G, Williams, D.C, Manolagas, S.C, 17β eEstradiol Inhibits Interlukin-6 production by Bone Marrow-derived Stromal Cells and Osteoblasts in Vitro: A Potential Mechanism for the Antiosteoporotic Effect of Estrogens. J Clin Invest 1992; 89: 883-891
  16. Price, P.A, Nishimoto, S.K, Radioimmunoassay for the Vitamin K dependent protein of bone and its discovery in plasma. Proc Natl Acad Sci USA 1980; 77; 2234-2238
  17. Melkki, J, Niemi, S, Risteli, L, Risteli, J, Radioimmunoassay of the Carboxyterminal Propeptide of Human Type 1 Procollagen, Clin Chem 1990; 36: 1328-1332
  18. Tietz, N.W, Rinker, A.D and Shaw, L.M, IFCC methods for the measurement of catalytic concentration of enzyme. Part 5. IFCC method for alkaline phosphatase (orthophosphoric-monester phosphorylase, alkaline optimum, EC J Clin Chem Clin Biochem 1983; 21: 731-748
  19. Hillmann, G, Continuous photometric measurement of prostatic acid phosphatase activity. Z Klin Chem Klin Biochem 1971; 9: 273-274
  20. Lukert, B.P, Higgins, J.C, Stoskopf, M.M, Serum Osteocalcin Is Increased in Patients with Hyperthyroidism and Decreased In Patients Receiving Glucocorticoids. J Clin Endocrinol Metab 1986; 62: 1056-1058
  21. Hajiroussou, V.J, Webley, M, Prolonged low dose corticosteroid therapy and osteoporosis in rheumatoid arthritis. Ann Rheum Dis 1984; 43: 24-27
  22. Tekano-Yamamoto, T, Rodan, G.A, Direct effects of 17β estradiol on trabecular bone in ovariectomized rats. Proc Natl Acad Sd USA 1990; 87: 2172-2176
  23. Chow, J Tobias, J.H, CoIston, K.W, Chambers, T.J, Estrogen maintains Trabecular Volume In rats not only by Suppression of Bone Resorption but Also by Stimulatlon of Bone Formation. J Clin Invest 1992; 89: 74-78
  24. Tobias, J.H, Chow, J, Colston, K.W, Chambers, T.J, High Concentrations of 17β estradiol Sffmutate Trabecular Bone Formation in Adult Female Rats. Endocrinology 1991; 128: 408-412
  25. Turner, R.T, Colvard, D.S, Spelsberg, T.C, Estrogen Inhibition of Periosteal Bone Formation in Rat Long Bones: Down-Regulation of Gene Expression for Bone Matrix Proteins. Endocrinology 1990; 127: 1346-1351
  26. Egrise, D, Martin, D, Neve, P, Verhas, M, Schoutens, A, Effects and Interactions of 17β estradiol, T3 and 1.25(OH)2D3 on cultured osteoblasts from mature rats. Bone Miner 1990; 11: 273-283
  27. Hasting, C, Edksen, E.F, Melkko, J, Risteli, L, Charles, P, Mosekilde, L, Risteli, J, Effects of a Combined Estrogen-Gestagen Regimen on Serum Levels of the Carboxy-terminal Propeptide of Human Type 1 Procollagen in Osteoporosis. J Bone Miner Res 1991; 6: 1295-1299
  28. Tobias, J.H, Gallagher,A, Chambers,T.J, Prolonged intermittent but not continuous administration of estradiol-17β increases bone volumen in the rat. J Endocrinology 1993; 139: 267-273
  29. Moss, D.W, Diagnostic aspects of alkaline phosphatase and its isoenzymes. Clin Biochem 1987; 20: 225-230
  30. Leung, K.S, Fling, K.P, Sher, A.H, Li, C.K, Lee, K.M, Plasma bone-specific alkaline phosphatase as an indicator of osteoblastic activity. J Bone Joint Surg (Br) 1993; 75-B: 288-292
  31. Rosalki, S B, Foo,A Y, Two new methods for separating and quantifying bone and liver alkaline phosphatase isezymes in plasma. Clin Chem 1984; 30: 1182-1186
  32. Rosalki, S.B, Foo, A.Y, Effect of Triton X-100 on precipitation of biliary phosphatase by wheat germ lectin, [letter] Clin Chem 1989; 35: 513
  33. Smedsrod, B, Melkko, J, Risteli, L, RIsteli, J, Circulating C-terminal propeptide of type 1 procollagen is cleared mainly via the mannose receptor in liver endothelial cells. Biochem J 1990; 271: 345-350

Correspondence and Reprint Requests to: Dr M Thomas, Department of Chemical Pathology & Human Metabolism, Royal Free Hospital, Pond Street, Hampstead, London NW3 2QG. Telephone: 0171 794 0500 extension 3464 FAX :0171 794 9537

Citation: Thomas, M., Barrett, J.S., Montgomery, D.H., (1994), Serum markers of bone metabolism in a group of transsexual patients receiving estrogen therapy, GENDYS '94, The Third International Gender Dysphoria Conference, Manchester England.
Web page copyright GENDYS Network. Text copyright of the author. 30.12.06 Last amended 14.07.13