Effect of Methandrostenolone on Postmenopausal Bone Wasting as Assessed by Changes in Total Bone Mineral Mass Charles
H. Chesnut III, Wil B. Neip, David J. Baylink,
To assess the efficacy of methandrostenolone in the treatment of osteoporosis a 26-mo double-blind study was performed with 13 treated and 13 control (placebo) postmenopausal osteoporotic females. Drug effect was assessed primarily by determinations of total body calcium (TBC) by neutron activation analysis, essentially a measurement of total bone mineral mass. Results in the 16 patients completing the study ( 10 treated and 6 placebo), as well as in all 26 patients participating in the study, showed significant ( p < 0.01) differences in the change in TBC between treated and control groups. In patients dropping out, TBC changes through the
and John D. Denney
time of dropout were similar to those in patients completing the study. In those patients completing the study, TBC increased 2% in the treated group and decreased 3% in the placebo group. An approximate sixfold difference in extraskeletal calcium balance would be required to explain the magnitude of the observed intergroup TBC difference. The drug effect appeared to persist throughout the 26-mo observation period. Thus these data strongly suggest that long-term use of methandrostenolone in postmenopausal osteoporosis prevented bone loss; the possibility that it increased bone mass above initial values is less certain.
P
OSTMENOPAUSAL OSTEOPOROSIS frequently results in significant morbidity. It is thought to be due to an inadequate bone mass prior to entering menopause, an excessive loss of bone mass following menopause, or both. A number of agents have been utilized in its therapy, both alone and in various combinations, including calcium,‘.* vitamin D,’ fluoride,2.3 calcitonin4 and androgens and synthetic anabolic diphosphonates,’ estrogens,‘*6-‘2 steroids.‘,6,9,‘0,‘3,‘4 Treatment, however, is less than satisfactory, and owing to the absence of knowledge of the specific etiology of the disease it is largely empirical. Additionally, some of the uncertainty regarding treatment probably stems from the lack of sensitive techniques for assessing prolonged therapeutic effect. Methods used to assess treatment in osteoporosis have included calcium metabolic balance studies,6 qualitative and quantitative radiographic techkinetics,‘,4,‘0,‘4 bone biopsy analysis,2~3~5~9*‘0 and regional niques,8v’2 radiocalcium bone mass measurement by photon absorption. 7,” A recently developed tech-
From the Division of Nuclear Medicine, Deparrments of Medicine and Radiology, University of Washington School of Medicine. and the Mineral Melabolism Laboratories, Departments of Medicine, Veterans Administration Hospital, Seattle, Wash. Received for publication October 29, 1975. Supported in part by Atomic Energy Commission Grant A T(45-I)-2225 and USPHS Grams I FO3 AM53150-01 and HD 04872. Dr. Chesnur is the recipient of a Picker Scholar Award: Dr. Baylink is the recipient of Research Career Development A ward DE 19108. Presented in part at the American Federation for Clinical Research. Atlantic City, New Jersey, Ma,v 5, 1974. Reprinr requests should be addressed IO Dr. Chesnut at rhe Division of Nuclear Medicine, University of Washingron School of Medicine, Seattle, Washingron 98195. D I977 by Grune & Stratton. Inc.
Metabolism, Vol. 26, No. 3 (March), 1977
267
268
CHESNUT
ET At.
nique for evaluating therapeutic effect is the measurement of total body calcium (TBC) by total body in vivo neutron activation analysisI I8 (NAA). Serial TBC enables measurement of long-term calcium balance with good precision. In the present study we used serial TBC to evaluate the effect of the synthetic anabolic steroid methandrostenolone (Dianabol-CIBA*) on long-term (26mo) calcium balance in patients with postmenopausal osteoporosis. This agent was chosen for study because of data from previous short-term but inconclusive trials suggesting a possible beneficial effect of anabolic steroids on bone mass loss 1.10.14 MATERIALS
AND METHODS
P&ems s&died. Twenty-six postmenopausal osteoporotic females (24 white and 2 black), age 55-75 yr, were selected for study from patients attending the Internal Medical Outpatient Clinic of the University Hospital. Informed consent was obtained from each patient according to procedures established by the Human Subjects Review Committee of the University of Washington, A primary criterion for selection was the presence of atraumatic vertebral compression fractures and/or vertebral osteopenia. as evidenced by cortical thinning, loss of horizontal trabeculation, and anterior wedging. Secondary causes of bone wasting. such as hyperparathyroidism. osteomalacia, uremia, renal tubular acidosis, hypercorticism, hyperthyroidism, multiple myeloma. longterm immobilization, or malabsorption, were excluded by history and physical examination, and by appropriate laboratory evaluations. The latter included normal serum calcium. phosphorus, alkaline phosphatase, and carotene, and a normal urinalysis and stool fat examination. In addition the following laboratory evaluations were normal in all patients: red and white blood cell counts, blood urea nitrogen and serum creatinine, serum electrolytes, glucose, total protein, and albumin/ globulin ratio. Serum SCOT (normal < 25 IU) was elevated in one patient (36 IU) but was normal in all others. Patients selected for the study had received no therapy for osteoporosis within 6 mo of study onset.? Initial features ofthe treated and placebo groups. Each patient selected for study was randomly assigned in a double-blind fashion to treated (methandrostenolone) or control (placebo) groups of 13 patients each. As noted in Table I. age, years since menopause, height. and serum calcium and phosphorus were similar in the two groups. Similarly, disease severity as assessed by the presence and number of compression fractures was comparable, with three fractures in one treated patient. two fractures each in one treated patient and two placebo patients. and one fracture each in the remaining four treated and placebo patients (Table 1). The difference in initial total body calcium in the two groups was not statistically significant (p > 0.10). In addition. diet and physical activity were considered comparable. Patients were instructed not to vary their diet or exercise habits during the study. Treated patients received 5 mg methandrostenolone daily for 3 of every 4 wk; control patients received identically packaged placebo medication on the same dosage schedule. Measurement of TBC by NAA was performed initially and Total bod.v calcium measuremenls. at approximate 6-mo intervals throughout the study. The techniques of TBC-NAA have been fully The physical principle of the method involves the conversion by neutron capture described.‘6,‘7 4’Ca, a y particle emitter of 8.8 min half-life (4xCa [n.-r] of stable 48Ca in the patient to radioactive 4yCa). in this procedure. the patient and liquid calcium comparator standards are exposed to a uniform 4- I2-MeV neutron flux (200 mrad total body dosage) over I .5 min. The neutrons are generated by bombardment of a beryllium target with cyclotron-produced 22-MeV deuterons. Immediately following irradiation the patient and the comparator standards are measured in a whole body counter for 4yCa y particle emission. The ratio 49Capatient/49CastandardServes as an index of TBC.The
precision
for the method
is ;t2”/;,.16 To obtain
the absolute
amount
of TBC in grams,
Division. *Kindly supplied by CIBA-Geigy Research Dept.. Pharmaceuticals Summit. N .J. tTwo patients (one treated. one placebo) had been previously treated with estrogens mo. respectively.
this
CIBA-Geigy. for 6 and
13
METHANDROSTENOLONE
AND
BONE
WASTING
269
ratio is compared to a previously established cadaver standard of similar body form; the accuracy is f 5.2%.” Since in the majority of normal and diseased subjects 98%-997; of TBC is within the skeleton,‘9320 TBC is in effect a measure of total bone mass. Radiographic spine examination was performed prior to and X-ray and laboratory evaluations. at the conclusion of the 26-mo study to document possible occurrence of compression fractures; additional radiographs were obtained during the study as clinically indicated. Serum calcium and phosphorus were obtained at 6-mo intervals throughout the study. Serum SCOT, alkaline phosphatase, bilirubin, and total protein with albumin/globulin ratio were also obtained at 6-mo intervals to assess possible hepatotoxic effects of methandrostenolone. In addition, a complete blood count, urinalysis, serum electrolytes, glucose, creatinine, and BUN were obtained yearly. Clinical evaluation. At monthly visits height was measured, and patients were evaluated for features (edema, weight gain, acne, hirsutism, etc.) suggestive of drug side effects. Statistical procedures. Student’s t test, paired t test (two observations on the same group of patients), and regression analysis were performed by standard statistical methods.19
RESULTS
Serial Measurement ofTBC The participation of all patients during the entire study is shown in Fig. 1, with serial TBC measurements in individual patients noted in Table 2. Ten of the treated and six of the placebo subjects completed the 26-mo protocol. Although the number of patients withdrawing from the study was different in the two groups (three treated, seven placebo), no factors (such as subjective awareness of drug effect) could be identified to account for the greater withdrawal in the placebo group. In the ten treated patients completing the study an average 2”/ gain in TBC occurred over an average 2.4 yr, while the six placebo patients completing the full protocol displayed an average 3.1% TBC loss over an average 2.3 yr. The final change in TBC (last point in Fig. 1) between the two groups of patients was significantly different (p < 0.01). In absolute terms the treated patients gained an average of 15 g of calcium, or 18 mg/day/patient. In contrast, the placebo patients lost an average of 22 g of calcium, or 27 mg/day/patient. In the treated group, final TBC was significantly greater than initial TBC (p < 0.05); in the placebo group, final TBC was less than the initial, but the difference was not significant (p < 0. lo), perhaps owing to the small sample size. Individual TBC changes at study completion in each of the above 16 patients is shown in Fig. 2 (solid circles). Although the mean change in TBC was significantly different between the treated and placebo groups, there was considerable interpatient variation within each group. However, it is noteworthy that only one treated patient lost TBC (0.7x), and increases as much as 8% (i.e., 73 mg calcium/day) were observed. Moreover, only two of the six placebo patients gained TBC (0.5% and l.l%), and losses as great as 9% (i.e., 75 mg calcium/ day) occurred. Analysis of the percentage change in TBC at each NAA procedure (- 6-mo intervals) in the treated and placebo groups completing the study is shown in Table 3. Although the difference between the two groups was not statistically significant until the fifth NAA procedure, the difference increased in a linear fashion with time (Fig. 3). To further examine the effect of methandrostenolone, the changes in TBC were also analyzed in all 26 patients (13/group) entering the study, since all had
Group
66.0
13
4.6
zk 6.26
f
Age
were statistically
65.3
13
N
None of the differences
Placebo
stenolone
Methandro-
significant.
6.2 5.2
14 f 16 f
157.0
156.2 f
f
(cm)
0.06
0.07
13 13
4
Radiograph
by
Patients With Osteopenia
4
Radiograph
Fracture by
MenOpaU%
Compression Height
Since
Patients With
1. Initial Features of the Treated and Control Groups
Years
Table
668 f
707 f
(9)
84.3
103.2
Calcium
Total Body
Serum
9.8 ziz 0.5
9.7 * 0.4
@g/dU
Calcium
Serum
3.4 f
3.5 f
0.3
0.3
@g/d!)
Phosphorus
METHANDROSTENOLONE
AND BONE WASTlNG
271
1
1
-3.
Fig. 1. Percentage change in TBC in all patients porticipoting in the study, mean f SE. the number of patients remaining in the study is shown in parentheses.
-4 +
I 4
0
9
I2
I6
20
8 24
28
Months
been observed for a period of at least 6 mo and had participated in at least two NAA procedures (Fig. 1). This analysis included the data from the ten patients at the time they withdrew from the study as well as the data from the 16 patients completing the study. There was a 1.8% gain in TBC in the 13 treated patients (p < 0.10) and a 3% loss of TBC in the 13 placebo patients (p < 0.05). The TBC change between the two groups was highly significant (p < 0.01). However, Table 2. Eeriol Measurements
of TBC in Individual
Patients Change
NAA (g Calcium) Patients
1
2
3
4
5
782.4
Abrolute (g Calcium)
%
+9.4
+1.2
+24.4
+3.6
Treated 1
773.0
712.8
782.7
848.8
2
683.2
706.7
696.0
707.6
3
760.0
783.9
773.8
756.9
4
704.5
684.6
682.8
658.4
5
662.6
687.6
679.1
716.9
662.9
+0.3
+0.1
6
750.1
710.3
710.0
720.8
744.6
-5.5
- 0.7 +4.5
796.4
-136.4
+4.a
-46.1
- 6.5
7
703.3
707.5
714.9
681.6
735.0
+31.7
8
677.7
668.1
694.5
672.3
677.9
+0.2
NC
9
767.2
791.2
839.7
818.5
828.2
+61.0
+a.0
10
772.3
758.0
781.9
741.2
773.1
+0.8
+0.1
11
467.9
453.9
482.3
477.8
467.7
-0.2
NC
12
585.6
582.0
627.8
627.2
+41.6
+7.1
13
895.5
935.9
941.7
893.7
911.3
+15.a
+1.a
701.5
676.2
650.9
-61.6
-8.7
Placebo 1
712.5
687.0
2
614.0
581.0
- 33.0
-5.4
3
572.6
566.0
-6.6
-1.2
4
684.1
650.3
687.5
+3.4
+0.5
5
596.8
610.0
592.9
583.5
i-2.7
+0.5
6
660.6
639.0
608.9
595.0
7
749.5
704.0
8
626.5
587.6
581.7
9
697.8
662.3
674.3
10
550.4
541.0
549.6
545.6
550.4
NC
11
730.6
758.9
765.0
715.1
738.3
-t-7.7
12
632.0
664.2
13
861.7
866.2
868.6
855.2
828.5
586.6
599.5
581.3
-65.6
- 9.9
-45.5
-6.1
-45.2
- 7.2
-23.5
- 3.4 NC +1.1
+32.2
+5.1
- 33.2
-3.9
272
CHESNUT
ET AL.
IO8-
l 0
6$ s
42-
8 g (s w
o_2_ -4D
-6-
a
0
0
-8-
0
-lO-
0
Methondrostenolone
Placebo
Fig. 2. Percentage change in TBC in each patient Participating in the study. 0, change (%) at study completion, 5 NAA procedures; 6, change (%) at withdrawal after 4 NAA procedures; 4, change (%) at withdmwal after 3 NAA procedures; h, change (%) at withdrawal after 2 NAA procedures; -, mean change (%) TBC for oil 26 patients, 13 treated and 13 placebo.
time of participation differed between the two groups, 2.1 yr (treated) and 1.5 yr (placebo), and, as in patients completing the study (Fig. 3), time was a significant covariable in patient response. Based on the relationship between treatment time and difference in per cent of TBC change for these 26 patients, the difference in treatment time did not entirely account for the observed TBC differences. Thus for all patients entering the study the mean difference in TBC change from 1.5 to 2.1 yr was about 1.47;, while the mean difference in TBC change was 4.8%. In addition, as can be seen in Fig. 2, when individual changes in all 26 patients are compared with those in the 16 patients completing the study a similar pattern is evident. Table 4 shows that in patients completing and patients dropping out (to time of dropout) of the study the direction of the differences between treated and placebo groups are similar. Consequently, the treated and placebo dropouts were, respectively, gaining or losing TBC through the time of their withdrawal in a fashion similar to treated or placebo patients completing the study. Thus whether one examines all participating patients (26), completing patients (16). or withdrawing patients (lo), the treated groups tended to gain and the placebo groups tended to lose TBC.
Table 3. Serial Changes in TBC in Patients Completing the Study MeonChange From Pretreatment’ Treated (N = 10) NAA
1 (Pretreatment)
Placebo (N = 6)
Mean Difference in Percentage TBC Change Between Treated and Placebo Patients -
0
0
NAA 2
- 0.35%
(+ 1.34)?
- 0.75%
(’
1.54)
0.40%
(NS)$
NAA
3
+2.27%
(11.18)
- 0.67%
( f 1.58)
2.94%
(NS)
NAA
4
+1.38%
(f
- 2.92%
(+0.94)
4.30%
(p < 0.10)
NAA
5
+ 1.99%
(lt0.87)
-3.05%
(lt1.71)
5.04%
(p < 0.01)
*Represents
percentage
change
in T8C from NAA
tSE. f Significance
of difference
1.62)
(t test).
1.
METHANDROSTENOLONE
AND BONE
273
WASTING
Fig. 3. Mean intergroup difference in percentage TBC change with time for the treated (ten patients) and placebo (six patients) groups completing the study. The data points represent results at NAA 1, 2, 3,4, and 5 from Table 3.
0
4
a
12
16
20
24
26
Month,
X-Ray, Laboratory, and Clinical Evaluations One methandrostenolone-treated patient experienced a compression fracture of the fifth thoracic vertebra 7 mo after onset of therapy. One placebo patient experienced a femoral neck fracture 1 mo after study onset. TBC changes in patients with compression fractures did not differ significantly from changes in patients without compression fractures, either in the treated or placebo groups. In addition, changes in TBC were not significantly correlated with patient age in either the treated or placebo groups. Serum calcium, phosphorus, and alkaline phosphatase did not change significantly in the two groups. SGOT increased in 5 of the 13 treated patients and in 2 of the 13 placebo patients during the 26-mo study. In four treated and two placebo patients this increase was a single event. In the one treated patient with an elevated SGOT (34 IU) prior to treatment subsequent 6-mo interval values were minimally elevated (26 IU). At study conclusion SGOT values were normal in all patients. Analysis of the SGOT changes in the treated and placebo groups revealed no significant difference (p > 0.10) between the two groups during the period of study participation. Other laboratory tests, including other liver function evaluations, displayed no significant changes in either group throughout the study. No significant change in height was found in either the placebo or treated groups. No fluid retention, hirsutism or other evidence of masculinizing effect, or postmenopausal menstrual bleeding were noted in the treated patients. DISCUSSION
In this study patients receiving methandrostenolone on an intermittent drug schedule (3 out of every 4 wk) experienced very few adverse side effects. Subjectively none were apparent. Elevations of the serum SGOT did occur sporadically in five treated subjects, but these were mild and transient. Although hepatomaz3 and peliosis hepatisz4 have been noted in patients treated with longterm androgenic-anabolic steroids, this is a rare occurrence and in general has followed continuous therapy with dosages at least three times those utilized in the current study. Patients have received such therapy for conditions other than osteoporosis.
*Four
P(N)
(N
(7)
placebo
patients
T-P
Comparison
withdrew
the study
after
NAA
of patients;
2.94%
from
0.40%
the study
4.
N, number
- 2.94%
10) withdrawing
P, placebo;
=
(6)
- 0.75%
(N = 16) completing
T, treated;
(3)
0%
Patients
(10)
- 0.35%
Patients
T(N)
Table
Through NAA 2
2, two
T -
(3)
to study
placebo
patients
after
between patient
after
and
(1)
- 9.97%
(6)
-2.91%
treated
and
NAA
4; three
T(N)
the
patients
(10)
+ 1.99%
From,
treated
patients.
11.37%
4.29%
T-P
Withdrawing
placebo
NAA 4
P(N)
Through
Completing,
one placebo
change
(3)
+ 1.4%
(10)
+ 1.38%
T(N)
in Patients
3, and
TBC
5.60%
2.94%
NAA
in percentage
(3)
- 3.60%
conclusion*
(6)
- 0.67%
P (NJ
T-P
TBC Changes
NAA 3
Percentage Through
P, difference
+24/o
prior
(10)
+ 2.27%
T(N)
of Mean
NAA 5
withdrew
after
(6)
- 3.05%
P(N)
Through
Study
NAA
4
5.04%
T-P
METHANDROSTENOLONE
AND BONE
WASTING
275
This is the first study utilizing the TBC-NAA technique in the assessment of anabolic steroid therapy in postmenopausal osteoporosis. A significant difference was noted in percentage change in TBC in the methandrostenolone-treated group as compared to the placebo group. In the placebo group TBC decreased, whereas in the treated group TBC did not decrease but tended to increase. The TBC loss rate in the placebo group, 1.5% of TBC/yr, is similar to the rate of age-related bone loss noted by other methods.25m2g In previous studies on the effect of synthetic anabolic steroids in postmenopausal or idiopathic osteoporosis relatively few patients were evaluated and osteoporotic control subjects were not included. L~‘“~‘3~‘4 After short-term treatment, metabolic balance studies revealed modest calcium retention,] and radiocalcium kinetic and microradiographic resultslO provided evidence of decreased bone resorption. With longer treatment (up to 12 mo), however, the positive effect was not sustained, since bone formation became reduced, as did the inhibitory effect on bone resorption. Consequently, the potential benefit to be derived from anabolic steroids has been considered only a short-term slowing or arrest of the progressive bone mass loss with little or no overall effect (short or long term) in restoring bone mass previously lost. In the present work, however, the drug effect on TBC appeared to be maintained throughout the 26-mo period. While the reason for this discrepancy is not entirely clear, the serial TBC measurements made in this study provide reliable data on net long-term changes in TBC. Moreover, a real difference in TBC is strongly supported by the magnitude of the final difference and by the consistency of the direction of this difference observed with serial TBC measurements. Whether or not the observed positive effect would be maintained for longer periods is unknown (however, see Ref. 22). There are, nevertheless, two points of consideration regarding the interpretation of the TBC data in this study. First, although we assume that the change in TBC represents a change in total bone mineral rather than a change in extraskeletal calcium (an assumption also inherent in metabolic balance studies), this cannot be established. In support of this assumption is the following: (1) 98%99% of TBC is normally skeletal calcium, 2’,22 (2) there is no evidence that osteoporotics have an increased propensity for extraskeletal calcium deposition, (3) no radiographic evidence of calcification of intervertebral discs or soft tissue was noted in the study patients, and (4) the magnitude of the intergroup TBC difference would require only a So/:,difference in bone calcium balance but about a sixfold difference in extraskeletal calcium balance.22 Second, the large number of dropouts could bias the final results. This seems unlikely, since among both dropouts (up to the time of dropout) and completers treated patients tended to gain and placebo patients tended to lose TBC. One might argue that the observed differences in percentage change in TBC, while statistically significant, were too small to improve bone strength. In this regard the number of patients developing fractures in this study was too small to establish or refute this assertion. Furthermore, even if the observed increases in percentage TBC in the treated group were real and sustained indefinitely, full repletion of skeletal mass would not be achieved within the lifespan of these patients. In this regard, it should be emphasized that osteoporosis can be a
CHESNIJT ET AL.
276
severely disabling and progressive disease for which there is no currently available cure. Therefore any agent that will arrest bone loss is an advance. Moreover, it is possible that this drug, in combination with other drugs, could produce additive effects. In addition, any drug that arrests bone loss could be more effective if instituted before severe skeletal depletion. Finally, the effectiveness of a given drug may vary in different osteoporotic patients, particularly if osteoporosis is a heterogeneous disease. A wide variation in drug response in this study is consistent with this last possibility. This observation emphasizes the importance of further studies to determine the optimum therapy for a given subgroup of osteoporotic patients. ACKNOWLEDGMENT The authors wish to acknowledge the valuable assistance of Dr. William G. Weitkamp. Techmcal Director of the University of Washington Nuclear Physics Laboratory. and his staff, the valuable suggestions of Dr. Clayton Rich regarding study design. and the technical assistance of Robert Murano and Gervas M. Hinn.
REFERENCES 1. Lafferty FW, Spencer GE, Pearson OH: EiTects of androgens, estrogens and high calcium intakes on bone formation and resorption in osteoporosis. Am J Med 36:514-528, 1964
synthetic
2. Jowsey J, Riggs BL, Kelly PJ, et al: Effect of combined therapy with sodium fluoride, vitamin D and calcium in osteoporosis. Am J Med 53:43-49. 1912 3. Bernstein DS. Cohen P: Use of sodium fluoride in treatment of osteoporosis. J Clin Endocrinol Metab 27:197-210. 1967 4. Jowsey J, Riggs BL. Goldsmith RS. et al: Effects of prolonged administration of porcine calcitonin in postmenopausal osteoporosis. J Clin Endocrinol Metab 33:752--758, 1971 5. Jowsey J, Riggs BL, Kelly PJ, et al: The treatment of osteoporosis with disodium ethane- 1-hydroxy- 1, I-diphosphonate. J Lab Clin Med 78:574-584, 6. Reifenstein
1971
EC, Albright
effects of steroid hormones Clin Invest 26:24-56, 1947
F: The metabolic in osteoporosis.
J
7. Davis ME, Strandjord NM, Lanzl LH: Estrogens and the aging process: The detection, prevention, and retardation of osteoporosis. JAMA 196219-224, 1966 8. Meema HE, Meema S: Prevention of postmenopausai osteoporosis by hormone treatment of the menopause. Can Med Assoc J 99: 248-25 1, 1968 9. Riggs BL, Jowsey J, Kelly PJ, et al: Effects of sex hormones on bone in primary osteoporosis. J Clin Invest 48:106-1072, 1969 10. Riggs BL, Jowsey J, Goldsmith RS, et al: Short- and long-term effects of estrogen and
anabolic
hormone
in postmenopausal
osteoporosis. J Clin Invest 51:1659-1663, 1972 11. Aitken JM, Hart DM, Lindsay R: Oestrogen replacement therapy for prevention of osteoporosis after oophorectomy. Br Med J 3:515-518, 1973 12. Gordon GS: Postmenopausal osteoporosis. Am Fam Physician 8:74-83, 1973 13. Melick RA, Baird CW: The effect of on patients with osteoporosis. “Parenabol” Med J Aust 21:960-962, 1970 14. Harrison JE, Hitchman AJW, Finlay JM, et al: Effect of treatment on calcium kinetics in metabolic bone disease. Metabolism 20:1107-1118, 1971 15. Anderson J, Osborn SB, Tomlinson RWS, et al: Neutron activation analysis in man in viva. Lancet 2:1201-1205, 1964 16. Nelp WB, Palmer HE, Murano R, et al: Measurement of total body calcium (bone mass) in vivo with the use of total body neutron activation analysis. J Lab Clin Med 76: 15 I - 162, I970 17. Nelp WB, Denney JD, Murano R, et al: Absolute measurement of total body calcium (bone mass) in vivo. J Lab Clin Med 79:430438, 1972 18. Cohn SH, Dombroski CS, Fairchild RG: In-vivo activation analysis of calcium in man. Int J Appl Radiat Isotopes 21:127-137, 1970 19. Snedecor GH, Cochran WC: Statistical Methods (ed 6). Ames, Iowa State Univ Pr, 1961 20. Frost HM: Bone remodeling and the relationship to metabolic bone disease. Springfield, Thomas, 1973. pp 85-l 14
METHANDROSTENOLONE
AND BONE
277
WASTING
21. Mitchell HH, Hamilton TS, Steggerda FR: The chemical composition of the adult human body and its bearing on the biochemistry of growth. J Biol Chem 158:625-637, 1945 22. Heaney RP: Evaluation and interpretation of calcium-kinetic data in man. Clin Orthop Relat Res 31:153-183, 1963 23. Farrell CC, Wren RF, Perkins KW, et al: Androgen-induced hepatoma. Lancet 7904: 430-432, 1975 24. Bagheri SA, Boyer JL: Peliosis hepatis associated with androgenic-anabolic steroid therapy. Ann Intern Med 81:610-613, 1974 25. Newton-John HF. Morgan DB: The loss of bone with age, osteoporosis, and fractures. Clin Orthop Relat Res 71:229-252, 1970
26. Doyle F: Involutional Endocrinol Metab l:l43-167,
Osteoporosis. 1972
Clin
27. Johnson CC, Smith DM, Nance WE, et al: Evaluation of radial bone mass by the photon absorption technique in Frame B, Parlitt AM, Duncan H (eds): Clinical Aspects of Metabolic Bone Disease. Amsterdam, Excerpta
Medica,
pp 28-36
28. Garn SM, Rohmann CC, Wagner B: Bone loss as a general phenomenon in man. Fed Proc 2611729-1736, 1967 29. Adams P, Davies CT, Sweetnam P: Osteoporosis and the effects of aging on bone mass in elderly men and women. Q J Med 39:601-615, 1970.
H. Chesnut III, Wil B. Neip, David J. Baylink,
To assess the efficacy of methandrostenolone in the treatment of osteoporosis a 26-mo double-blind study was performed with 13 treated and 13 control (placebo) postmenopausal osteoporotic females. Drug effect was assessed primarily by determinations of total body calcium (TBC) by neutron activation analysis, essentially a measurement of total bone mineral mass. Results in the 16 patients completing the study ( 10 treated and 6 placebo), as well as in all 26 patients participating in the study, showed significant ( p < 0.01) differences in the change in TBC between treated and control groups. In patients dropping out, TBC changes through the
and John D. Denney
time of dropout were similar to those in patients completing the study. In those patients completing the study, TBC increased 2% in the treated group and decreased 3% in the placebo group. An approximate sixfold difference in extraskeletal calcium balance would be required to explain the magnitude of the observed intergroup TBC difference. The drug effect appeared to persist throughout the 26-mo observation period. Thus these data strongly suggest that long-term use of methandrostenolone in postmenopausal osteoporosis prevented bone loss; the possibility that it increased bone mass above initial values is less certain.
P
OSTMENOPAUSAL OSTEOPOROSIS frequently results in significant morbidity. It is thought to be due to an inadequate bone mass prior to entering menopause, an excessive loss of bone mass following menopause, or both. A number of agents have been utilized in its therapy, both alone and in various combinations, including calcium,‘.* vitamin D,’ fluoride,2.3 calcitonin4 and androgens and synthetic anabolic diphosphonates,’ estrogens,‘*6-‘2 steroids.‘,6,9,‘0,‘3,‘4 Treatment, however, is less than satisfactory, and owing to the absence of knowledge of the specific etiology of the disease it is largely empirical. Additionally, some of the uncertainty regarding treatment probably stems from the lack of sensitive techniques for assessing prolonged therapeutic effect. Methods used to assess treatment in osteoporosis have included calcium metabolic balance studies,6 qualitative and quantitative radiographic techkinetics,‘,4,‘0,‘4 bone biopsy analysis,2~3~5~9*‘0 and regional niques,8v’2 radiocalcium bone mass measurement by photon absorption. 7,” A recently developed tech-
From the Division of Nuclear Medicine, Deparrments of Medicine and Radiology, University of Washington School of Medicine. and the Mineral Melabolism Laboratories, Departments of Medicine, Veterans Administration Hospital, Seattle, Wash. Received for publication October 29, 1975. Supported in part by Atomic Energy Commission Grant A T(45-I)-2225 and USPHS Grams I FO3 AM53150-01 and HD 04872. Dr. Chesnur is the recipient of a Picker Scholar Award: Dr. Baylink is the recipient of Research Career Development A ward DE 19108. Presented in part at the American Federation for Clinical Research. Atlantic City, New Jersey, Ma,v 5, 1974. Reprinr requests should be addressed IO Dr. Chesnut at rhe Division of Nuclear Medicine, University of Washingron School of Medicine, Seattle, Washingron 98195. D I977 by Grune & Stratton. Inc.
Metabolism, Vol. 26, No. 3 (March), 1977
267
268
CHESNUT
ET At.
nique for evaluating therapeutic effect is the measurement of total body calcium (TBC) by total body in vivo neutron activation analysisI I8 (NAA). Serial TBC enables measurement of long-term calcium balance with good precision. In the present study we used serial TBC to evaluate the effect of the synthetic anabolic steroid methandrostenolone (Dianabol-CIBA*) on long-term (26mo) calcium balance in patients with postmenopausal osteoporosis. This agent was chosen for study because of data from previous short-term but inconclusive trials suggesting a possible beneficial effect of anabolic steroids on bone mass loss 1.10.14 MATERIALS
AND METHODS
P&ems s&died. Twenty-six postmenopausal osteoporotic females (24 white and 2 black), age 55-75 yr, were selected for study from patients attending the Internal Medical Outpatient Clinic of the University Hospital. Informed consent was obtained from each patient according to procedures established by the Human Subjects Review Committee of the University of Washington, A primary criterion for selection was the presence of atraumatic vertebral compression fractures and/or vertebral osteopenia. as evidenced by cortical thinning, loss of horizontal trabeculation, and anterior wedging. Secondary causes of bone wasting. such as hyperparathyroidism. osteomalacia, uremia, renal tubular acidosis, hypercorticism, hyperthyroidism, multiple myeloma. longterm immobilization, or malabsorption, were excluded by history and physical examination, and by appropriate laboratory evaluations. The latter included normal serum calcium. phosphorus, alkaline phosphatase, and carotene, and a normal urinalysis and stool fat examination. In addition the following laboratory evaluations were normal in all patients: red and white blood cell counts, blood urea nitrogen and serum creatinine, serum electrolytes, glucose, total protein, and albumin/ globulin ratio. Serum SCOT (normal < 25 IU) was elevated in one patient (36 IU) but was normal in all others. Patients selected for the study had received no therapy for osteoporosis within 6 mo of study onset.? Initial features ofthe treated and placebo groups. Each patient selected for study was randomly assigned in a double-blind fashion to treated (methandrostenolone) or control (placebo) groups of 13 patients each. As noted in Table I. age, years since menopause, height. and serum calcium and phosphorus were similar in the two groups. Similarly, disease severity as assessed by the presence and number of compression fractures was comparable, with three fractures in one treated patient. two fractures each in one treated patient and two placebo patients. and one fracture each in the remaining four treated and placebo patients (Table 1). The difference in initial total body calcium in the two groups was not statistically significant (p > 0.10). In addition. diet and physical activity were considered comparable. Patients were instructed not to vary their diet or exercise habits during the study. Treated patients received 5 mg methandrostenolone daily for 3 of every 4 wk; control patients received identically packaged placebo medication on the same dosage schedule. Measurement of TBC by NAA was performed initially and Total bod.v calcium measuremenls. at approximate 6-mo intervals throughout the study. The techniques of TBC-NAA have been fully The physical principle of the method involves the conversion by neutron capture described.‘6,‘7 4’Ca, a y particle emitter of 8.8 min half-life (4xCa [n.-r] of stable 48Ca in the patient to radioactive 4yCa). in this procedure. the patient and liquid calcium comparator standards are exposed to a uniform 4- I2-MeV neutron flux (200 mrad total body dosage) over I .5 min. The neutrons are generated by bombardment of a beryllium target with cyclotron-produced 22-MeV deuterons. Immediately following irradiation the patient and the comparator standards are measured in a whole body counter for 4yCa y particle emission. The ratio 49Capatient/49CastandardServes as an index of TBC.The
precision
for the method
is ;t2”/;,.16 To obtain
the absolute
amount
of TBC in grams,
Division. *Kindly supplied by CIBA-Geigy Research Dept.. Pharmaceuticals Summit. N .J. tTwo patients (one treated. one placebo) had been previously treated with estrogens mo. respectively.
this
CIBA-Geigy. for 6 and
13
METHANDROSTENOLONE
AND
BONE
WASTING
269
ratio is compared to a previously established cadaver standard of similar body form; the accuracy is f 5.2%.” Since in the majority of normal and diseased subjects 98%-997; of TBC is within the skeleton,‘9320 TBC is in effect a measure of total bone mass. Radiographic spine examination was performed prior to and X-ray and laboratory evaluations. at the conclusion of the 26-mo study to document possible occurrence of compression fractures; additional radiographs were obtained during the study as clinically indicated. Serum calcium and phosphorus were obtained at 6-mo intervals throughout the study. Serum SCOT, alkaline phosphatase, bilirubin, and total protein with albumin/globulin ratio were also obtained at 6-mo intervals to assess possible hepatotoxic effects of methandrostenolone. In addition, a complete blood count, urinalysis, serum electrolytes, glucose, creatinine, and BUN were obtained yearly. Clinical evaluation. At monthly visits height was measured, and patients were evaluated for features (edema, weight gain, acne, hirsutism, etc.) suggestive of drug side effects. Statistical procedures. Student’s t test, paired t test (two observations on the same group of patients), and regression analysis were performed by standard statistical methods.19
RESULTS
Serial Measurement ofTBC The participation of all patients during the entire study is shown in Fig. 1, with serial TBC measurements in individual patients noted in Table 2. Ten of the treated and six of the placebo subjects completed the 26-mo protocol. Although the number of patients withdrawing from the study was different in the two groups (three treated, seven placebo), no factors (such as subjective awareness of drug effect) could be identified to account for the greater withdrawal in the placebo group. In the ten treated patients completing the study an average 2”/ gain in TBC occurred over an average 2.4 yr, while the six placebo patients completing the full protocol displayed an average 3.1% TBC loss over an average 2.3 yr. The final change in TBC (last point in Fig. 1) between the two groups of patients was significantly different (p < 0.01). In absolute terms the treated patients gained an average of 15 g of calcium, or 18 mg/day/patient. In contrast, the placebo patients lost an average of 22 g of calcium, or 27 mg/day/patient. In the treated group, final TBC was significantly greater than initial TBC (p < 0.05); in the placebo group, final TBC was less than the initial, but the difference was not significant (p < 0. lo), perhaps owing to the small sample size. Individual TBC changes at study completion in each of the above 16 patients is shown in Fig. 2 (solid circles). Although the mean change in TBC was significantly different between the treated and placebo groups, there was considerable interpatient variation within each group. However, it is noteworthy that only one treated patient lost TBC (0.7x), and increases as much as 8% (i.e., 73 mg calcium/day) were observed. Moreover, only two of the six placebo patients gained TBC (0.5% and l.l%), and losses as great as 9% (i.e., 75 mg calcium/ day) occurred. Analysis of the percentage change in TBC at each NAA procedure (- 6-mo intervals) in the treated and placebo groups completing the study is shown in Table 3. Although the difference between the two groups was not statistically significant until the fifth NAA procedure, the difference increased in a linear fashion with time (Fig. 3). To further examine the effect of methandrostenolone, the changes in TBC were also analyzed in all 26 patients (13/group) entering the study, since all had
Group
66.0
13
4.6
zk 6.26
f
Age
were statistically
65.3
13
N
None of the differences
Placebo
stenolone
Methandro-
significant.
6.2 5.2
14 f 16 f
157.0
156.2 f
f
(cm)
0.06
0.07
13 13
4
Radiograph
by
Patients With Osteopenia
4
Radiograph
Fracture by
MenOpaU%
Compression Height
Since
Patients With
1. Initial Features of the Treated and Control Groups
Years
Table
668 f
707 f
(9)
84.3
103.2
Calcium
Total Body
Serum
9.8 ziz 0.5
9.7 * 0.4
@g/dU
Calcium
Serum
3.4 f
3.5 f
0.3
0.3
@g/d!)
Phosphorus
METHANDROSTENOLONE
AND BONE WASTlNG
271
1
1
-3.
Fig. 1. Percentage change in TBC in all patients porticipoting in the study, mean f SE. the number of patients remaining in the study is shown in parentheses.
-4 +
I 4
0
9
I2
I6
20
8 24
28
Months
been observed for a period of at least 6 mo and had participated in at least two NAA procedures (Fig. 1). This analysis included the data from the ten patients at the time they withdrew from the study as well as the data from the 16 patients completing the study. There was a 1.8% gain in TBC in the 13 treated patients (p < 0.10) and a 3% loss of TBC in the 13 placebo patients (p < 0.05). The TBC change between the two groups was highly significant (p < 0.01). However, Table 2. Eeriol Measurements
of TBC in Individual
Patients Change
NAA (g Calcium) Patients
1
2
3
4
5
782.4
Abrolute (g Calcium)
%
+9.4
+1.2
+24.4
+3.6
Treated 1
773.0
712.8
782.7
848.8
2
683.2
706.7
696.0
707.6
3
760.0
783.9
773.8
756.9
4
704.5
684.6
682.8
658.4
5
662.6
687.6
679.1
716.9
662.9
+0.3
+0.1
6
750.1
710.3
710.0
720.8
744.6
-5.5
- 0.7 +4.5
796.4
-136.4
+4.a
-46.1
- 6.5
7
703.3
707.5
714.9
681.6
735.0
+31.7
8
677.7
668.1
694.5
672.3
677.9
+0.2
NC
9
767.2
791.2
839.7
818.5
828.2
+61.0
+a.0
10
772.3
758.0
781.9
741.2
773.1
+0.8
+0.1
11
467.9
453.9
482.3
477.8
467.7
-0.2
NC
12
585.6
582.0
627.8
627.2
+41.6
+7.1
13
895.5
935.9
941.7
893.7
911.3
+15.a
+1.a
701.5
676.2
650.9
-61.6
-8.7
Placebo 1
712.5
687.0
2
614.0
581.0
- 33.0
-5.4
3
572.6
566.0
-6.6
-1.2
4
684.1
650.3
687.5
+3.4
+0.5
5
596.8
610.0
592.9
583.5
i-2.7
+0.5
6
660.6
639.0
608.9
595.0
7
749.5
704.0
8
626.5
587.6
581.7
9
697.8
662.3
674.3
10
550.4
541.0
549.6
545.6
550.4
NC
11
730.6
758.9
765.0
715.1
738.3
-t-7.7
12
632.0
664.2
13
861.7
866.2
868.6
855.2
828.5
586.6
599.5
581.3
-65.6
- 9.9
-45.5
-6.1
-45.2
- 7.2
-23.5
- 3.4 NC +1.1
+32.2
+5.1
- 33.2
-3.9
272
CHESNUT
ET AL.
IO8-
l 0
6$ s
42-
8 g (s w
o_2_ -4D
-6-
a
0
0
-8-
0
-lO-
0
Methondrostenolone
Placebo
Fig. 2. Percentage change in TBC in each patient Participating in the study. 0, change (%) at study completion, 5 NAA procedures; 6, change (%) at withdrawal after 4 NAA procedures; 4, change (%) at withdmwal after 3 NAA procedures; h, change (%) at withdrawal after 2 NAA procedures; -, mean change (%) TBC for oil 26 patients, 13 treated and 13 placebo.
time of participation differed between the two groups, 2.1 yr (treated) and 1.5 yr (placebo), and, as in patients completing the study (Fig. 3), time was a significant covariable in patient response. Based on the relationship between treatment time and difference in per cent of TBC change for these 26 patients, the difference in treatment time did not entirely account for the observed TBC differences. Thus for all patients entering the study the mean difference in TBC change from 1.5 to 2.1 yr was about 1.47;, while the mean difference in TBC change was 4.8%. In addition, as can be seen in Fig. 2, when individual changes in all 26 patients are compared with those in the 16 patients completing the study a similar pattern is evident. Table 4 shows that in patients completing and patients dropping out (to time of dropout) of the study the direction of the differences between treated and placebo groups are similar. Consequently, the treated and placebo dropouts were, respectively, gaining or losing TBC through the time of their withdrawal in a fashion similar to treated or placebo patients completing the study. Thus whether one examines all participating patients (26), completing patients (16). or withdrawing patients (lo), the treated groups tended to gain and the placebo groups tended to lose TBC.
Table 3. Serial Changes in TBC in Patients Completing the Study MeonChange From Pretreatment’ Treated (N = 10) NAA
1 (Pretreatment)
Placebo (N = 6)
Mean Difference in Percentage TBC Change Between Treated and Placebo Patients -
0
0
NAA 2
- 0.35%
(+ 1.34)?
- 0.75%
(’
1.54)
0.40%
(NS)$
NAA
3
+2.27%
(11.18)
- 0.67%
( f 1.58)
2.94%
(NS)
NAA
4
+1.38%
(f
- 2.92%
(+0.94)
4.30%
(p < 0.10)
NAA
5
+ 1.99%
(lt0.87)
-3.05%
(lt1.71)
5.04%
(p < 0.01)
*Represents
percentage
change
in T8C from NAA
tSE. f Significance
of difference
1.62)
(t test).
1.
METHANDROSTENOLONE
AND BONE
273
WASTING
Fig. 3. Mean intergroup difference in percentage TBC change with time for the treated (ten patients) and placebo (six patients) groups completing the study. The data points represent results at NAA 1, 2, 3,4, and 5 from Table 3.
0
4
a
12
16
20
24
26
Month,
X-Ray, Laboratory, and Clinical Evaluations One methandrostenolone-treated patient experienced a compression fracture of the fifth thoracic vertebra 7 mo after onset of therapy. One placebo patient experienced a femoral neck fracture 1 mo after study onset. TBC changes in patients with compression fractures did not differ significantly from changes in patients without compression fractures, either in the treated or placebo groups. In addition, changes in TBC were not significantly correlated with patient age in either the treated or placebo groups. Serum calcium, phosphorus, and alkaline phosphatase did not change significantly in the two groups. SGOT increased in 5 of the 13 treated patients and in 2 of the 13 placebo patients during the 26-mo study. In four treated and two placebo patients this increase was a single event. In the one treated patient with an elevated SGOT (34 IU) prior to treatment subsequent 6-mo interval values were minimally elevated (26 IU). At study conclusion SGOT values were normal in all patients. Analysis of the SGOT changes in the treated and placebo groups revealed no significant difference (p > 0.10) between the two groups during the period of study participation. Other laboratory tests, including other liver function evaluations, displayed no significant changes in either group throughout the study. No significant change in height was found in either the placebo or treated groups. No fluid retention, hirsutism or other evidence of masculinizing effect, or postmenopausal menstrual bleeding were noted in the treated patients. DISCUSSION
In this study patients receiving methandrostenolone on an intermittent drug schedule (3 out of every 4 wk) experienced very few adverse side effects. Subjectively none were apparent. Elevations of the serum SGOT did occur sporadically in five treated subjects, but these were mild and transient. Although hepatomaz3 and peliosis hepatisz4 have been noted in patients treated with longterm androgenic-anabolic steroids, this is a rare occurrence and in general has followed continuous therapy with dosages at least three times those utilized in the current study. Patients have received such therapy for conditions other than osteoporosis.
*Four
P(N)
(N
(7)
placebo
patients
T-P
Comparison
withdrew
the study
after
NAA
of patients;
2.94%
from
0.40%
the study
4.
N, number
- 2.94%
10) withdrawing
P, placebo;
=
(6)
- 0.75%
(N = 16) completing
T, treated;
(3)
0%
Patients
(10)
- 0.35%
Patients
T(N)
Table
Through NAA 2
2, two
T -
(3)
to study
placebo
patients
after
between patient
after
and
(1)
- 9.97%
(6)
-2.91%
treated
and
NAA
4; three
T(N)
the
patients
(10)
+ 1.99%
From,
treated
patients.
11.37%
4.29%
T-P
Withdrawing
placebo
NAA 4
P(N)
Through
Completing,
one placebo
change
(3)
+ 1.4%
(10)
+ 1.38%
T(N)
in Patients
3, and
TBC
5.60%
2.94%
NAA
in percentage
(3)
- 3.60%
conclusion*
(6)
- 0.67%
P (NJ
T-P
TBC Changes
NAA 3
Percentage Through
P, difference
+24/o
prior
(10)
+ 2.27%
T(N)
of Mean
NAA 5
withdrew
after
(6)
- 3.05%
P(N)
Through
Study
NAA
4
5.04%
T-P
METHANDROSTENOLONE
AND BONE
WASTING
275
This is the first study utilizing the TBC-NAA technique in the assessment of anabolic steroid therapy in postmenopausal osteoporosis. A significant difference was noted in percentage change in TBC in the methandrostenolone-treated group as compared to the placebo group. In the placebo group TBC decreased, whereas in the treated group TBC did not decrease but tended to increase. The TBC loss rate in the placebo group, 1.5% of TBC/yr, is similar to the rate of age-related bone loss noted by other methods.25m2g In previous studies on the effect of synthetic anabolic steroids in postmenopausal or idiopathic osteoporosis relatively few patients were evaluated and osteoporotic control subjects were not included. L~‘“~‘3~‘4 After short-term treatment, metabolic balance studies revealed modest calcium retention,] and radiocalcium kinetic and microradiographic resultslO provided evidence of decreased bone resorption. With longer treatment (up to 12 mo), however, the positive effect was not sustained, since bone formation became reduced, as did the inhibitory effect on bone resorption. Consequently, the potential benefit to be derived from anabolic steroids has been considered only a short-term slowing or arrest of the progressive bone mass loss with little or no overall effect (short or long term) in restoring bone mass previously lost. In the present work, however, the drug effect on TBC appeared to be maintained throughout the 26-mo period. While the reason for this discrepancy is not entirely clear, the serial TBC measurements made in this study provide reliable data on net long-term changes in TBC. Moreover, a real difference in TBC is strongly supported by the magnitude of the final difference and by the consistency of the direction of this difference observed with serial TBC measurements. Whether or not the observed positive effect would be maintained for longer periods is unknown (however, see Ref. 22). There are, nevertheless, two points of consideration regarding the interpretation of the TBC data in this study. First, although we assume that the change in TBC represents a change in total bone mineral rather than a change in extraskeletal calcium (an assumption also inherent in metabolic balance studies), this cannot be established. In support of this assumption is the following: (1) 98%99% of TBC is normally skeletal calcium, 2’,22 (2) there is no evidence that osteoporotics have an increased propensity for extraskeletal calcium deposition, (3) no radiographic evidence of calcification of intervertebral discs or soft tissue was noted in the study patients, and (4) the magnitude of the intergroup TBC difference would require only a So/:,difference in bone calcium balance but about a sixfold difference in extraskeletal calcium balance.22 Second, the large number of dropouts could bias the final results. This seems unlikely, since among both dropouts (up to the time of dropout) and completers treated patients tended to gain and placebo patients tended to lose TBC. One might argue that the observed differences in percentage change in TBC, while statistically significant, were too small to improve bone strength. In this regard the number of patients developing fractures in this study was too small to establish or refute this assertion. Furthermore, even if the observed increases in percentage TBC in the treated group were real and sustained indefinitely, full repletion of skeletal mass would not be achieved within the lifespan of these patients. In this regard, it should be emphasized that osteoporosis can be a
CHESNIJT ET AL.
276
severely disabling and progressive disease for which there is no currently available cure. Therefore any agent that will arrest bone loss is an advance. Moreover, it is possible that this drug, in combination with other drugs, could produce additive effects. In addition, any drug that arrests bone loss could be more effective if instituted before severe skeletal depletion. Finally, the effectiveness of a given drug may vary in different osteoporotic patients, particularly if osteoporosis is a heterogeneous disease. A wide variation in drug response in this study is consistent with this last possibility. This observation emphasizes the importance of further studies to determine the optimum therapy for a given subgroup of osteoporotic patients. ACKNOWLEDGMENT The authors wish to acknowledge the valuable assistance of Dr. William G. Weitkamp. Techmcal Director of the University of Washington Nuclear Physics Laboratory. and his staff, the valuable suggestions of Dr. Clayton Rich regarding study design. and the technical assistance of Robert Murano and Gervas M. Hinn.
REFERENCES 1. Lafferty FW, Spencer GE, Pearson OH: EiTects of androgens, estrogens and high calcium intakes on bone formation and resorption in osteoporosis. Am J Med 36:514-528, 1964
synthetic
2. Jowsey J, Riggs BL, Kelly PJ, et al: Effect of combined therapy with sodium fluoride, vitamin D and calcium in osteoporosis. Am J Med 53:43-49. 1912 3. Bernstein DS. Cohen P: Use of sodium fluoride in treatment of osteoporosis. J Clin Endocrinol Metab 27:197-210. 1967 4. Jowsey J, Riggs BL. Goldsmith RS. et al: Effects of prolonged administration of porcine calcitonin in postmenopausal osteoporosis. J Clin Endocrinol Metab 33:752--758, 1971 5. Jowsey J, Riggs BL, Kelly PJ, et al: The treatment of osteoporosis with disodium ethane- 1-hydroxy- 1, I-diphosphonate. J Lab Clin Med 78:574-584, 6. Reifenstein
1971
EC, Albright
effects of steroid hormones Clin Invest 26:24-56, 1947
F: The metabolic in osteoporosis.
J
7. Davis ME, Strandjord NM, Lanzl LH: Estrogens and the aging process: The detection, prevention, and retardation of osteoporosis. JAMA 196219-224, 1966 8. Meema HE, Meema S: Prevention of postmenopausai osteoporosis by hormone treatment of the menopause. Can Med Assoc J 99: 248-25 1, 1968 9. Riggs BL, Jowsey J, Kelly PJ, et al: Effects of sex hormones on bone in primary osteoporosis. J Clin Invest 48:106-1072, 1969 10. Riggs BL, Jowsey J, Goldsmith RS, et al: Short- and long-term effects of estrogen and
anabolic
hormone
in postmenopausal
osteoporosis. J Clin Invest 51:1659-1663, 1972 11. Aitken JM, Hart DM, Lindsay R: Oestrogen replacement therapy for prevention of osteoporosis after oophorectomy. Br Med J 3:515-518, 1973 12. Gordon GS: Postmenopausal osteoporosis. Am Fam Physician 8:74-83, 1973 13. Melick RA, Baird CW: The effect of on patients with osteoporosis. “Parenabol” Med J Aust 21:960-962, 1970 14. Harrison JE, Hitchman AJW, Finlay JM, et al: Effect of treatment on calcium kinetics in metabolic bone disease. Metabolism 20:1107-1118, 1971 15. Anderson J, Osborn SB, Tomlinson RWS, et al: Neutron activation analysis in man in viva. Lancet 2:1201-1205, 1964 16. Nelp WB, Palmer HE, Murano R, et al: Measurement of total body calcium (bone mass) in vivo with the use of total body neutron activation analysis. J Lab Clin Med 76: 15 I - 162, I970 17. Nelp WB, Denney JD, Murano R, et al: Absolute measurement of total body calcium (bone mass) in vivo. J Lab Clin Med 79:430438, 1972 18. Cohn SH, Dombroski CS, Fairchild RG: In-vivo activation analysis of calcium in man. Int J Appl Radiat Isotopes 21:127-137, 1970 19. Snedecor GH, Cochran WC: Statistical Methods (ed 6). Ames, Iowa State Univ Pr, 1961 20. Frost HM: Bone remodeling and the relationship to metabolic bone disease. Springfield, Thomas, 1973. pp 85-l 14
METHANDROSTENOLONE
AND BONE
277
WASTING
21. Mitchell HH, Hamilton TS, Steggerda FR: The chemical composition of the adult human body and its bearing on the biochemistry of growth. J Biol Chem 158:625-637, 1945 22. Heaney RP: Evaluation and interpretation of calcium-kinetic data in man. Clin Orthop Relat Res 31:153-183, 1963 23. Farrell CC, Wren RF, Perkins KW, et al: Androgen-induced hepatoma. Lancet 7904: 430-432, 1975 24. Bagheri SA, Boyer JL: Peliosis hepatis associated with androgenic-anabolic steroid therapy. Ann Intern Med 81:610-613, 1974 25. Newton-John HF. Morgan DB: The loss of bone with age, osteoporosis, and fractures. Clin Orthop Relat Res 71:229-252, 1970
26. Doyle F: Involutional Endocrinol Metab l:l43-167,
Osteoporosis. 1972
Clin
27. Johnson CC, Smith DM, Nance WE, et al: Evaluation of radial bone mass by the photon absorption technique in Frame B, Parlitt AM, Duncan H (eds): Clinical Aspects of Metabolic Bone Disease. Amsterdam, Excerpta
Medica,
pp 28-36
28. Garn SM, Rohmann CC, Wagner B: Bone loss as a general phenomenon in man. Fed Proc 2611729-1736, 1967 29. Adams P, Davies CT, Sweetnam P: Osteoporosis and the effects of aging on bone mass in elderly men and women. Q J Med 39:601-615, 1970.
