| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Clinical Research Center Studies |
Section of Endocrine Neoplasia and Hormonal Disorders, University of Texas, M. D. Anderson Cancer Center (S.I.S., M.J.S.), Houston, Texas 77030; the Division of Endocrinology and Metabolism, Johns Hopkins University School of Medicine (M.D.R., P.W.L.), Baltimore, Maryland 21287; and the Section of Cardiology, Baylor College of Medicine (H.A.K., W.A.Z.), Houston, Texas 77030
Address all correspondence and requests for reprints to: Steven I. Sherman, M.D., University of Texas, M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Box 15, Houston, Texas 77030.
Abstract
A thyroid hormone analog with organ-selective effects could have therapeutic application for disorders such as hyperlipidemia and osteoporosis. We performed a randomized clinical trial to determine the specific thyromimetic effects of tiratricol. Twenty-four athyreotic patients underwent detailed metabolic and physiological evaluation after a 2-month baseline period, taking TSH-suppressive doses of L-T4. They were then randomized to blinded treatment with either tiratricol (24 µg/kg twice daily) or L-T4 (1.9 µg/kg daily). The dose of hormone was increased until the TSH level was less than 0.1 mU/L, and the metabolic and physiological testing was repeated. Comparing the change from baseline to the study drug periods, when serum TSH levels were equivalently suppressed, there were no significant differences between the two groups in resting metabolic rate, weight, urea nitrogen excretion, or symptom score. Plasma total and low density lipoprotein cholesterol levels declined 13 ± 4% and 23 ± 6% in the tiratricol group compared with 2 ± 2% and 5 ± 3% in the L-T4 group (P = 0.015 and P = 0.0066, respectively). Serum sex hormone-binding globulin levels increased 55 ± 13% with tiratricol compared with a 1.7 ± 4% decline with L-T4 (P = 0.0006), indicating an augmented hepatic response to tiratricol. Skeletal metabolic activity was enhanced, with increased levels of serum osteocalcin and urinary excretion of calcium and pyridinium cross-links. Tiratricol and L-T4 had comparable effects on cardiovascular function. Tiratricol has distinct augmented hepatic and skeletal thyromimetic actions of potential therapeutic value.
THYROID HORMONES have widespread effects, including regulation of hepatic, cardiovascular, central nervous system, and skeletal functions. Clinical investigators have long attempted to capitalize on certain actions of thyroid hormones to treat disease states such as obesity, hypercholesterolemia, heart failure, and depression. However, demonstration of the efficacy of thyroid hormone therapy for these conditions has been limited by its wider multisystem actions and by the presence of endogenous thyroid hormone. Consequently, analogs of thyroid hormone with more selective actions have been sought. One of the earliest thyroid hormone analogs studied, D-T4, was reported to decrease serum cholesterol levels in hyperlipidemic patients, but proved to increase cardiovascular mortality (1). However, both the beneficial and deleterious effects of D-T4 were later attributed to mild thyrotoxicosis caused by L-T4 contaminating the pharmacological preparation (2, 3). More recently, the synthetic analogs SKF L-94901 (4) and CGS 23425 (5) have been demonstrated to lower serum cholesterol levels in laboratory animals without broader thyromimetic effects, although human trials have yet to be reported. Additional rationales for the development of such thyromimetic compounds have been to avoid the side-effects of L-T4 treatment, such as osteoporosis, and to treat thyroid hormone resistance states.
Tiratricol (3,5,3'-triiodothyroacetic acid, Triac), a metabolite of T4, is 10–20 times less calorigenic than T3 (6, 7), but has greater affinity for nuclear thyroid hormone receptors (8, 9). Controversy exists as to whether tiratricol has the same relative effects as T4 in thyroid hormone-responsive tissues (10, 11, 12, 13, 14, 15, 16). A previous randomized, double blind, cross-over trial in athyreotic patients demonstrated that tiratricol administered with L-T4 has increased organ-specific actions in the liver and skeleton, but is not a pituitary-selective therapy (17). We now report the results of the first clinical trial to examine the organ-specific thyromimetic responses to tiratricol therapy in the absence of exogenous or endogenous L-T4. This randomized, double blind trial provides evidence that tiratricol has enhanced hepatic and skeletal actions of potential therapeutic value.
Subjects and Methods
Patients
Twenty-four patients (19 women and 5 men), aged 21–72 yr (mean \ SD, 48 \ 12 yr), who had undergone thyroidectomy and radioiodine ablation for differentiated thyroid carcinoma, were recruited from the participating institutions. All patients were free of known thyroid cancer, and none had evidence of hepatic, renal, skeletal, or pituitary disease. The study was approved by the investigational review boards of all participating institutions, and informed consent was obtained from each patient.
Protocol
Responses to thyroid hormone therapy were measured at two points, after a baseline period and after a study drug intervention period. During the 2-month baseline period, patients were treated with L-T4 at their usual TSH-suppressive dose (sufficient to reduce the TSH concentration to <0.1 mU/L). Patients were counselled to follow a eucaloric, weight maintenance diet. Upon entry, 91% of patients were observing a diet consistent with American Heart Association step I recommendations, as assessed by the MEDFICTS dietary recall questionnaire (18). After the baseline period, subjects were admitted to the General Clinical Research Center for physiological and biochemical assessment of thyroid hormone actions. On 2 successive mornings, indirect calorimetry was performed, and blood parameters reflecting hepatic and skeletal metabolism were obtained at 0700 h after a 12-h fast and before the patient had arisen from bed. A resting echocardiogram was performed before phlebotomy on the first morning, and a continuous electrocardiogram was recorded during the second night of hospitalization. Urine was collected for 24 h.
Patients were then randomized to one of two treatment groups in a double blind fashion. One group received an initial tiratricol dose of 24 µg/kg twice daily, rounded to the nearest multiple of 300 µg. The daily dose of tiratricol was increased monthly by 300- to 600-µg increments until the serum TSH concentration was less than 0.1 mU/L. The other group received an initial L-T4 dose of 1.9 µg/kg each morning, rounded to the nearest multiple of 25 µg, and a placebo "L-T4" capsule each evening. For these patients, the daily dose of L-T4 was increased monthly by 25- to 50-µg increments, with a parallel modification in the evening placebo dose. After 2 months of taking the final study drug dose, patients were readmitted to repeat the physiological and biochemical assessments.
Methods
Tiratricol and L-T4 sodium were obtained from Laphal Laboratories (Paris, France) and Boots Pharmaceuticals (Lincolnshire, IL), respectively. Based on high performance liquid chromatography (sensitivity, 0.1%), neither T3 nor L-T4 was detected in the tiratricol preparation, and contamination by diiodo- and tetraiodothyroacetic acid was less than 2%. Preparation of encapsulated drugs and placebo was performed at Johns Hopkins (Baltimore, MD).
Measurements of O2 consumption and CO2 production were performed by indirect calorimetry (DeltaTrac Metabolic Monitor, Sensormedics, Anaheim, CA). Urine was collected for 24 h to determine nitrogen excretion, with subsequent calculation of the resting metabolic rate (RMR) and the rate of fat oxidation (19, 20, 21). Measured cardiovascular responses included frequency of ectopic supraventricular and ventricular premature contractions, and mean nocturnal pulse (determined by continuous ambulatory electrocardiographic monitoring between 2400–0700 h). M-mode, two-dimensional, and Doppler echocardiographic evaluations were performed using standardized procedures by one technician at each site (SONOS 1000 and 1500, Hewlett-Packard, Andover, MA). All studies were recorded and analyzed in a random order by one observer, who was blinded to clinical status. Echocardiographic and Doppler parameters of systolic function and cardiac anatomy included heart rate, ejection fraction from biplane ventricular volumes (22), time interval from initiation of the electrocardiographic Q wave to aortic valve opening (preejection period), interval between aortic valve opening and closure (left ventricular ejection time) (23), end-diastolic and end-systolic diameters, and left ventricular posterior wall and septal thicknesses (24). Doppler measurements of diastolic function included maximal early mitral flow velocity, maximal late mitral flow velocity, time interval from aortic valve closure to mitral valve opening (isovolumic relaxation time), and the fraction of ventricular filling during atrial contraction (25). Derived parameters included the preejection period/left ventricular ejection time and maximal early mitral flow velocity/maximal late mitral flow velocity ratios, left ventricular mass index, and rate-corrected velocity of circumferential shortening (26, 27). A previously described analog thyrotoxicosis symptom questionnaire was employed (17).
Assays
The serum TSH concentration was quantified by chemiluminometric immunoassay (sensitivity, 0.005 mU/L; interassay coefficient of variation, 11% at 0.1 mU/L; Nichols Institute, San Juan Capistrano, CA). The serum free T4 level was determined by indirect immunoassay (interassay coefficient of variation, 3.8% at 1.39 ng/dL; Abbott Diagnostics, Abbott Park, IL). Plasma total cholesterol and triglycerides were measured enzymatically (Boehringer Mannheim Diagnostics, Indianapolis, IN), with subsequent calculation of low density lipoprotein (LDL) cholesterol (28, 29, 30). Plasma high density lipoprotein (HDL) cholesterol was determined by measuring cholesterol in the supernatant liquid after MgCl2-dextran precipitation (31). Plasma apoproteins AI and B and lipoprotein(a) were assayed by competitive enzyme-linked immunoassays using specific monoclonal antibodies (32). Radiometric assay of serum sex hormone-binding globulin (SHBG) and osteocalcin, and high performance liquid chromatography measurement of urinary pyridinium collagen cross-links, were performed at Nichols Institute Reference Laboratories. The serum nonesterified fatty acid concentration was determined by titrimetric assay at Mayo Medical Reference Laboratories (Rochester, MN). Serum 5'-nucleotidase activity was spectrophotometrically measured by SmithKline Beecham Clinical Laboratories (Houston, TX). Serum intact PTH and ferritin were assayed with commercially available radiometric assay kits [Nichols Institute, and Bio-Rad Laboratories (Richmond, CA), respectively].
Statistical analysis
Statistical analyses were based on parameter changes from the baseline period to the study drug intervention period, comparing the tiratricol-treated group with the L-T4-treated group. For parameters measured twice during each hospitalization, the two results were averaged before analysis. Serum TSH levels were logarithmically transformed before analysis. Significance testing was performed using multivariate ANOVA.
Sample size estimates were based upon expected differences in LDL
cholesterol, total cholesterol, and SHBG, for which the expected
SDs were estimated to be 10%, 10%, and 15%,
respectively. With 10 evaluable subjects/group, the study was designed
to have a power of at least 90% (ß = 0.1) to detect a difference
between treatment groups of 15% for LDL and total cholesterol, and
23% for SHBG, with a 2-sided significance level (
) of 0.05 or less.
Allowing for a potential 20% dropout rate, the study was designed to
recruit up to 25 subjects.
Results
Twenty-four subjects completed the baseline evaluation and were
randomized to 1 of the treatment groups. The results of the
prerandomization baseline evaluation are summarized in Table 1
. One patient was subsequently dropped from the study
after randomization because of episodic hypertension and palpitations;
after study completion and unblinding of drug assignments, she was
found to have been treated with L-T4.
The remaining 23 patients were treated with study drug for 92 \ 7
days (mean \ SEM).
|
)Patients in the tiratricol group (n = 12) required 48 \ 3 µg/kg tiratricol daily to maintain their serum TSH concentrations similar to those during the baseline period, when they had been treated with L-T4 (2.8 \ 0.2 µg/kg daily). Therefore, the ratio of tiratricol to L-T4 doses needed to maintain a subnormal TSH level was 17:1. Serum free T4 levels fell below the detectable limit of the assay during tiratricol therapy. In contrast, L-T4-treated patients (n = 11) maintained constant serum free T4 levels in the two phases of the study despite lower doses of L-T4 (2.6 \ 0.2 µg/kg daily at baseline; 2.2 \ 0.1 µg/kg daily during the study drug period; P = 0.006).
|
Hepatic responses
Lipid and apolipoprotein levels responded quite differently in the
two treatment groups (Fig. 1). In the tiratricol group,
mean plasma total cholesterol levels declined 13%, from 211 to 184
mg/dL (P = 0.015 compared with the
L-T4 group). Of the eight patients in
this group with fasting plasma cholesterol levels greater than 200
mg/dL during the baseline period, the average decline was 17%. The
primary contributor to this decrease was a decline in the mean plasma
LDL cholesterol concentration, which fell 23%, from 132 to 102 mg/dL
(P = 0.0066 compared with the
L-T4 group). The eight
hypercholesterolemic patients had an average reduction in the plasma
LDL cholesterol concentration of 26%. Plasma triglyceride levels also
declined, but plasma HDL cholesterol concentrations did not change.
Paralleling the lipoprotein changes, the plasma apoprotein B level
declined by 11% (P = 0.045 compared with the
L-T4 group), but the plasma
apoprotein AI level did not change. Plasma levels of lipoprotein(a) and
nonesterified fatty acids did not change in either group.
|
Skeletal responses (Fig. 2)
Serum alkaline phosphatase activity increased 12% in the tiratricol-treated group, whereas levels fell 4% in the L-T4 group (P = 0.025). The similarly altered levels of osteocalcin (27% increase vs. 8% decline; P = 0.016) and the absence of significant changes in 5'-nucleotidase suggest that these differences reflect increased osteoblast activity during tiratricol therapy. Patients treated with tiratricol had a marked increase in the excretion of pyridinoline (45%; P = 0.028 compared with the L-T4 group) and deoxypyridinoline (59%; P = 0.0052 compared with the L-T4 group), consistent with increased bone resorption. Differences in pyridinium cross-links excretion between the two drug treatment groups remained significant (P = 0.023 for pyridinoline; P = 0.010 for deoxypyridinoline) even when gender and baseline excretion levels were added as parameters in the analysis. Although changes in serum calcium levels did not differ significantly between the two groups, tiratricol produced a 28% decline in PTH (P = 0.0016 compared with the L-T4 group).
|
)There were no significant differences in the change from baseline in resting pulse, blood pressure, or the frequency of ectopic supraventricular or ventricular premature contractions between the two treatment groups. No significant differences were seen between the groups in any of the echocardiographic parameters of either systolic or diastolic function. There was a trend toward a decline from baseline in the left ventricular mass index in the L-T4 group, but the difference between the two treatment arms was not significant. In the tiratricol group, the change from baseline of the left ventricular mass index did not statistically differ from zero.
|
Tiratricol therapy in athyreotic patients produced enhanced thyromimetic actions in the liver and skeleton compared with the effects of L-T4 therapy. The substantial reductions of total and LDL cholesterol seen in tiratricol-treated patients could be associated with a 20–30% reduction in cardiovascular risk (33). Although the decrease in apoprotein B levels with tiratricol therapy was less marked than that for LDL cholesterol, a reduction in atherosclerotic risk would be anticipated from the change in this parameter as well (34). In light of the other evidence of increased hepatic thyromimetic effects of tiratricol, i.e. increased circulating SHBG and ferritin concentrations (35, 36, 37, 38), potential mechanisms for the atherogenic lipoprotein reductions include decreased apoprotein B secretion, increased LDL receptor-mediated LDL cholesterol uptake, and enhanced biliary secretion of cholesterol (39, 40, 41). Thyromimetic effects on lipoprotein metabolism also typically include reductions in serum levels of HDL cholesterol, apoprotein AI, and lipoprotein(a) (42). The absence of any exaggeration of these effects with tiratricol therapy further suggests that the hepatic responses to tiratricol differ from those to L-T4 (43, 44). The reduction in atherogenic lipoprotein levels without a reduction in HDL cholesterol suggests a potential role for tiratricol therapy in patients with hypercholesterolemia.
Tiratricol therapy was associated with biochemical evidence of increased skeletal turnover. Levels of both osteocalcin, a marker of anabolic osteoblast activity, and urinary pyridinium cross-links, a marker of catabolic osteoclast activity, were elevated in patients treated with tiratricol compared with their respective baseline values. These findings are consistent with in vitro evidence that tiratricol stimulates bone formation and resorption with at least the potency of T3 (45, 46). Although this clinical study was not designed to detect the long term effect of tiratricol on bone mineralization, the changes in serum PTH and urinary calcium excretion suggest that prolonged therapy might produce net bone resorption and osteopenia. Accelerated bone demineralization may, therefore, limit potential therapeutic applications of tiratricol.
The magnitude of physiologic effects of tiratricol on pituitary thyrotropes and integrated metabolic and cardiovascular functions were comparable to those of L-T4. With serum TSH levels held constant, tiratricol therapy produced no changes in metabolic rate, substrate oxidation, weight, symptom score, or nitrogen excretion. There were no significant differences in atrial tachyarrhythmias, systolic time intervals, velocity of ventricular fiber shortening, or diastolic time intervals and flow velocities, indicating a comparable effect of tiratricol on cardiovascular function. Although it is possible that more pronounced effects might be seen with a longer duration of therapy, even structural changes in ventricular wall thickness associated with TSH-suppressive L-T4 therapy are detectable within 4 months (47). Thus, these results demonstrate that tiratricol, administered in TSH-suppressive doses, is not pituitary specific in its actions in human subjects with normal extrapituitary tissue responsiveness to thyroid hormones, in contrast with the results of earlier nonblinded studies (10, 12, 13, 14, 16).
The potential mechanisms of organ-specific thyromimetic responses
include differences in the tissue distribution of the various thyroid
hormone receptors and their accessory proteins (17). Both the liver and
the skeleton contain a predominance of the ß-forms of the
T3 receptor (48, 49), which have a relatively higher
affinity for tiratricol than the -subtype (8, 9). In liver,
ß1- and ß2-receptors combined account for
about 80% of receptor content and hormone binding (48, 50). By causing
greater degrees of receptor activation and thyroid hormone-responsive
gene expression due to preferential binding, tiratricol may have
relatively greater activity in these ß-predominant organs. In
contrast, the -form of the receptor accounts for almost half of the
T3-binding capacity in the heart (48), which may contribute
to the lack of enhanced thyromimetic cardiac effects of tiratricol.
Tissue-specific T3 receptor-binding proteins, which
interact with the nuclear T3 receptors, may further affect
the response to hormone in a tissue-specific, ligand-specific manner
(51, 52). Alternatively, a combination of mechanisms involving both
nuclear receptors and tissue-specific cofactors may determine
individual organ responses to thyroid hormones, as has been proposed to
explain the phenotypic diversity of the syndrome of resistance to
thyroid hormone (53).
Regardless of the mechanisms responsible for its effects, this prospective randomized double blind trial demonstrates that tiratricol does have a pattern of organ-specific actions distinct from that of L-T4, with relatively greater effects in the liver and skeleton. Further study will be necessary to determine the role of tiratricol in patients requiring long term thyroid hormone therapy and potential applications for its use in the treatment of hyperlipidemia.
Acknowledgments
The authors thank Marge Ewerts, R.N., Kathy Lesh, R.N., and the staffs of the Methodist Hospital and Johns Hopkins Hospital General Clinical Research Centers for their excellent nursing support; and Elizabeth Tomalis, M.P.H., R.D. and Phyllis Tacquard, R.D. for their nutritional counselling.
Footnotes
1 This work was supported in part by NIH Grants RR-00350 (General
Clinical Research Center, Baylor College of Medicine) and RR-00035
(General Clinical Research Center, Johns Hopkins School of Medicine).
Presented in part at the 11th International Thyroid Congress, Toronto,
Ontario, Canada, September 1995. 
2 Recipient of a Clinical Associate Physician Award under the grants
mentioned above.
3 Recipient of a postdoctoral fellowship from Pfizer, Inc.
Received February 4, 1997.
Revised March 6, 1997.
Accepted March 18, 1997.
References
This article has been cited by other articles:
 |
|
 |
|
  G. Medina-Gomez, R. M. Calvo, and M.-J. Obregon Thermogenic effect of triiodothyroacetic acid at low doses in rat adipose tissue without adverse side effects in the thyroid axis Am J Physiol Endocrinol Metab, April 1, 2008; 294(4): E688 - E697. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. J. Kahaly and W. H. Dillmann Thyroid Hormone Action in the Heart Endocr. Rev., August 1, 2005; 26(5): 704 - 728. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. L. Reed, K. R. Reedy, L. A. Palinkas, N. Van Do, N. S. Finney, H. S. Case, H. J. LeMar, J. Wright, and J. Thomas Impairment in Cognitive and Exercise Performance during Prolonged Antarctic Residence: Effect of Thyroxine Supplementation in the Polar Triiodothyronine Syndrome J. Clin. Endocrinol. Metab., January 1, 2001; 86(1): 110 - 116. [Abstract] [Full Text]
|
|
|
|
 |
|
 R. J. Clifton-Bligh, F. de Zegher, R. L. Wagner, T. N. Collingwood, I. Francois, M. Van Helvoirt, R. J. Fletterick, and V. K. K. Chatterjee A Novel TR{beta} Mutation (R383H) in Resistance to Thyroid Hormone Syndrome Predominantly Impairs Corepressor Release and Negative Transcriptional Regulation Mol. Endocrinol., May 1, 1998; 12(5): 609 - 621. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
