Endocrine Service, Division of General Medicine, Department of Medicine, and Medical Library-Nathan Cummings Center, Memorial Sloan-Kettering Cancer Center, New York, New York 10021
Address all correspondence and requests for reprints to: Richard J. Robbins, M.D., Endocrinology Service, Box 296, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, New York 10021. E-mail address: robbinsr{at}mskcc.org.
The initial use of exogenous TSH to augment radioiodine uptake in thyroid cancer patients, using extracts from bovine pituitary glands, occurred over 50 yr ago (1, 2, 3). This approach fell out of favor due to the frequent development of neutralizing antibodies (4) and allergic reactions with subsequent exposures (5). The availability of bovine TSH ceased in the mid-1980s, resulting in the routine use of thyroid hormone withdrawal to stimulate radioiodine uptake in thyroid cancer patients between 1985 and 2000.
Pioneering work from the laboratories of Weintraub and colleagues (6) and Kourides and colleagues (7) in the late 1980s provided the primary sequence of the ß-subunit of human TSH. The human pituitary -glycoprotein subunit and the human TSH ß-subunit were then separately cloned and coexpressed in human embryonal cells with the recovery of recombinant human TSH (rhTSH) from the culture medium (8). Purified rhTSH was first shown to have biological activity in rat FRTL-5 cells by Thotakura et al. (9). Bioactivity was then documented in human fetal thyroid cells (10) and in rhesus monkeys (11). Early studies demonstrated that compared with TSH extracted from human pituitaries, rhTSH produced in Chinese hamster ovary cells had higher levels of sialic acid, lower binding affinity for porcine TSH receptors, reduced potency to stimulate cAMP in mammalian thyroid membranes, and slower metabolic clearance. Most of these differences were eliminated by enzymatic desialylation (9). The process of rhTSH production was modified and scaled up in Chinese hamster ovary cells with the assistance of scientists at Genzyme Genetics (Framingham, MA) with the achievement of pharmaceutical grade recombinant human TSH. The compound was subsequently developed by the Genzyme Corp. (thyrotropin alfa; Thyrogen, Cambridge, MA) and was approved for use by the U.S. FDA in December 1998.
Early clinical trials
A multicenter Phase I/II trial with Genzymes Thyrogen was conducted with 19 volunteers with differentiated thyroid cancer (DTC) (12). Rapid increases in serum TSH after im injections of rhTSH were documented. The pharmacokinetic studies showed a direct relationship between the dose administered and the serum TSH levels. Mean serum TSH levels as high as 510 mIU/liter were seen 8 h after a single 30-U dose of rhTSH. One month later, no anti-TSH antibodies could be detected in the 19 volunteers. Multiple subsequent studies confirmed the lack of apparent allergic reactions, even in subjects who received more than 10 administrations of rhTSH (13). Patients who received the highest test doses (30 U/d) complained of mild nausea and headache. None of the volunteers who received the doses currently in use (10 U/d) had any side effects. Diagnostic whole body radioiodine scans (DxWBS) were performed in the 19 volunteers after rhTSH and again, in the same patients, 3 wk later, after discontinuing thyroid hormone. Comparison of the 2 scans showed complete correlation in 63%. Significant differences were present in 7 patients, but there was no obvious advantage of one form of preparation over the other. Although serum thyroglobulin (Tg) levels rose after rhTSH treatment, Tg levels were generally higher after thyroid hormone withdrawal. Subsequent studies by Braverman et al. (14, 15) in normal human volunteers demonstrated the time course of the thyroid effects of rhTSH and established that rhTSH is a potent stimulator of T3, T4, and Tg secretion and of thyroidal radioiodine uptake in man (Fig. 1
).
|
Two Genzyme-sponsored Phase III studies were carried out with volunteers with DTC. They had similar experimental designs, in which each subject received a 0.9-mg rhTSH injection on d 1 and 2, a small amount of 131I on d 3, and DxWBS on d 5. Thyroid hormone was then stopped until the serum TSH level was greater than 25 mU/liter, and then another low dose of 131I was given, with DxWBS performed 48 h later. In the first study (16), the investigators confirmed that there were virtually no symptoms of hypothyroidism in the rhTSH-treated individuals, confirming previous findings (12). They found that the sensitivity of the rhTSH-assisted DxWBS was lower than that of the DxWBS obtained in the hypothyroid state. This had been suggested in the Phase I/II studies, but was previously suspected of being an artifact of the slower whole body radioiodine clearance associated with hypothyroidism (12). In the second Phase III study (17), improved quality control on DxWBS readings and full measurements of serum Tg were incorporated. This second study found no significant difference between the hypothyroid or rhTSH DxWBS scan sensitivity and documented an 89% concordance rate overall. The researchers also showed that an rhTSH-stimulated Tg level greater than 2 ng/ml identified 100% of patients with distant metastases. The only common adverse effects were headache and mild nausea, which occurred in less than 10% of patients. They concluded that rhTSH was a safe and effective means of monitoring patients with DTC without the reduced quality of life that was evident when the subjects were hypothyroid. A cohort of subjects in the second Phase III trial received three injections of rhTSH before the first scan, but it was determined that this schedule did not result in significant improvements in diagnostic sensitivity. In aggregate, the Phase III studies were somewhat limited by the fact that the DxWBS scans were negative in more than 50% of the patients and by the low prevalence (<20%) of patient with distant metastases. However, due to the unique cooperation of an international group of thyroid experts and Genzyme Corp., rhTSH went from a novel laboratory accomplishment to a fully FDA-approved medication within 10 yr.
Recent clinical studies
Subsequent independent studies have examined the validity of the initial clinical observations. A large retrospective study from our cancer center in which approximately 50% of the patients had residual thyroid cancer, confirmed that the sensitivity and specificity of DxWBS alone, stimulated Tg alone, or the combined DxWBS/stimulated Tg test were similar whether the patients were prepared by rhTSH treatment or thyroid hormone withdrawal (18). We also found that the sensitivity of the DxWBS was slightly (69% vs. 80% for withdrawal), but not significantly, lower in the rhTSH-treated patients, and that the sensitivity of the stimulated Tg was slightly higher in the rhTSH group (86% vs. 79% for withdrawal; P = NS). The sensitivity of the combined DxWBS/Tg outcome after rhTSH treatment was higher (98%) than that for either test alone. Pacini et al. (19) conducted a prospective study in 72 DTC patients. In this study each patient served as his/her own control, evaluated after rhTSH and then subsequently while hypothyroid. All were low risk patients with undetectable suppressed serum Tg. Eighty-eight percent of those with undetectable Tg after rhTSH treatment also had undetectable Tg while hypothyroid, and none had evidence of residual cancer on a hypothyroid DxWBS. In the 12% who had detectable hypothyroid Tg but undetectable rhTSH Tg, the hypothyroid DxWBS was either negative or showed only faint thyroid bed uptake. The height of the stimulated Tg, on the average, was higher after thyroid hormone withdrawal compared with rhTSH. Confirming previous studies, this group found that all low risk patients with residual cancer were identified by a rise in Tg after rhTSH. Mazzaferri and Kloos (20) reported that 10% of 107 DTC patients who had no obvious residual disease and very low Tg levels on suppression, ultimately had evidence of residual DTC. All of the patients with occult disease had an rhTSH-stimulated Tg that was more than 2 ng/ml (sensitivity and negative predictive value, 100%). Importantly, they found that the DxWBS after rhTSH seldom provided any important additional information. In a recent study of 366 DTC patients, Robbins et al. (21) reported that the stimulated Tg alone was a sufficient means of monitoring residual thyroid carcinoma in low risk patients who have had at least one prior negative DxWBS. This study also reported that residual thyroid carcinoma could be discovered in 8% of low risk patients whose stimulated Tg was less than 2 ng/ml using tests such as ultrasonography or positron emission tomography scanning. Table 1 summarizes the published reports that contain diagnostic accuracy data. Overall, preparation for testing with rhTSH has excellent diagnostic accuracy compared with preparation by thyroid hormone withdrawal, without the need for the reduced quality of life attendant to the hypothyroid state.
|
Several studies have shown that the absolute serum Tg level is higher in patients after thyroid hormone withdrawal for 4 wk or more compared with 72 h after two doses of rhTSH (12, 16, 17, 18, 19). This is probably a combination of enhanced tumoral Tg production caused by prolonged TSH stimulation as well as reduced clearance of Tg in the hypothyroid state. However, no studies have shown that the withdrawal Tg is more likely than rhTSH-stimulated Tg to reveal residual thyroid cancer. This may be due to the wide spectrum of hypothyroidism in large groups of patients. Minimal elevations of TSH may not stimulate Tg expression sufficiently to change the serum Tg level. Patients with a rhTSH-stimulated Tg greater than 5 ng/ml usually have residual thyroid cancer that can be localized on further testing (20, 22).
Most studies suggest that the DxWBS is more likely to show residual thyroid cancer deposits when the scan is performed in the hypothyroid state compared with scans performed during thyroid hormone treatment after rhTSH administration (16, 17, 18). It was postulated by Meier et al. (12) that this might be due to the higher 131I background in hypothyroid patients. They showed that the apparent higher sensitivity of a hypothyroid scan could be negated by compensating for the higher 131I background in hypothyroid scans. In a more recent, smaller, retrospective study, Durski et al. (23) reported that scans obtained after rhTSH treatment were indistinguishable from those obtained in the same subjects when hypothyroid.
Adverse effect of rhTSH
To date there are no reports that Thyrogen causes any allergic reactions, even after multiple exposures. Likewise, no tachyphylaxis has been reported, as the effectiveness of rhTSH seems to continue despite multiple exposures in the same patient. Durski et al. (23) found that 45% of patients had mild nausea and/or headache after rhTSH treatment. These are the two most common symptoms reported by many investigators. Vargas et al. (24) reported the sudden onset of hemiplegia in a hypothyroid patient who was given rhTSH, presumably due to rapid expansion of a brain lesion. A similar event was reported by Robbins et al. (25). Giovanni et al. (26) reported three patients with distant metastases who had severe bone pain over bone metastases after rhTSH. Lippi et al. (27) reported that rhTSH given to patients with metastatic disease can result in pain and swelling near the lesions, especially in those with bone metastases, similar to the symptoms experienced by the same patients when they were taken off thyroid hormone. Our own personal experience has shown that this subacute rhTSH-associated lesion swelling can be effectively reduced by pretreatment with glucocorticoids.
Therapy
Compassionate need.
Rudavsky and Freedman (39) were the first to report on the off-label use of rhTSH to assist with therapeutic uptake of radioiodine. Genzyme Corp., in conjunction with federal regulatory agencies in the United States and Europe, made rhTSH available on a compassionate need basis for individuals who could not produce TSH or for those in whom thyroid hormone withdrawal was medically contraindicated. Masiukiewicz et al. (28) used rhTSH in two patients with hypopituitarism to administer 131I therapy. Several other investigators (see Table 2) have confirmed the ability of rhTSH to stimulate radioiodine uptake into metastatic lesions (25, 29, 30).
|
The concept that rhTSH could assist in the radioiodine therapy of multinodular goiters was proposed by Ramirez et al. (14) and Adler et al. (31). Huysmans et al. (32, 33) carried out the initial studies and reported that rhTSH pretreatment of patients with nontoxic nodular goiters increases overall radioactive iodine uptake and can covert cold nodules to hot nodules. In theory, this may make 131I therapy more efficacious. Preliminary results from Silva et al. (34) suggest that this is the case.
Robbins et al. (35) reported that successful rhTSH-assisted thyroid remnant ablation can be achieved after total thyroidectomy for DTC. This same group reported a case-controlled study in which radioiodine remnant ablations were attempted in 42 patients after thyroid hormone withdrawal and in 45 patients after two 0.9-mg injections of rhTSH (36). After 11 months, DxWBS showed complete ablation in 84% of those treated after rhTSH treatment compared with 81% after thyroid hormone withdrawal (P = NS). Successful remnant ablations have also been reported by Berg et al. (37) and Perros (38).
Rudavsky and Freeman (39) were the first to report a favorable tumor response to rhTSH-stimulated 131I therapy in a 54-yr-old man with widespread DTC metastases. Chiu et al. (40) used rhTSH to prepare a patient with brain metastases for radioiodine therapy. Robbins et al. (25) reported rhTSH assisted 131I therapy in a follicular thyroid carcinoma patient who could not produce TSH due to radiotherapy to the hypothalamus. Follow-up studies showed good uptake of the 131I and disappearance of some metastatic lesions on subsequent scanning. Lippi et al. (27) reported that rhTSH can stimulate the uptake of 131I into metastatic thyroid cancer lesions with subsequent reductions of serum Tg levels or stabilization of disease. Luster et al. (29) found that serum Tg levels fell or disease regressed in 8 of 11 DTC patients who received 131I therapy for metastatic DTC after rhTSH preparation on a compassionate need basis. Mariani et al. (41) used rhTSH to prepare 8 patients for 131I therapy and documented good uptake in residual disease in 7 of the 8 subjects. Rotman-Pikielny et al. (30) reported a partial tumor response in a large hepatic DTC metastasis treated after rhTSH preparation. Other reports of rhTSH-assisted therapy for DTC metastases have been published (41, 42, 43).
Looking ahead
The optimum dose of rhTSH for diagnostic studies has not been determined. Future studies should be conducted and compared with the regimen of two 0.9-mg injections of rhTSH given 24 h apart. The 0.9-mg rhTSH regimen given three times over 1 wk was slightly, but not significantly, more successful in the second Phase III trial (17). It is conceivable that a single dose could provide enough sodium-iodide symporter stimulation for routine diagnostic studies; however, multiple small doses may exert a priming effect.
There will undoubtedly be growing use of rhTSH to assist radioiodine therapy. This includes both remnant ablation and therapy for metastases. In the past when therapy required a hypothyroid preparation, therapeutic administrations of 131I were usually only performed once a year to improve patient acceptability. Given that rhTSH injections do not result in hypothyroid symptoms, this raises the question of how often 131I could be given to patients with widespread metastases that concentrate radioiodine. Could rhTSH-prepared patients receive therapy every 23 months? No studies examining the optimal interval between 131I treatments have ever been performed in DTC patients with metastases. There are anecdotal suggestions that cessation of thyroid hormone supplements for 35 d before a therapeutic dose of 131I (in an attempt to reduce iodine intake) could result in better intralesional retention, although this issue has not been studied carefully.
A multicenter trial is underway to examine the rate of success of rhTSH-assisted radioiodine remnant ablations after total thyroidectomy for DTC in patients who are receiving T4 replacement. If the success rate is shown to be the same or better than that after thyroid hormone withdrawal, clinicians and patients will have an option that avoids hypothyroidism (17, 44, 45) and reduces total body radiation exposure.
Preliminary reports from Pacini and colleagues (46) in Italy suggest that the efficacy of chemotherapy for widespread aggressive thyroid carcinoma may be increased if the metabolic activity of the tumor is stimulated by endogenous or exogenous TSH. Such studies raise novel and potentially important questions for patients with widespread thyroid cancer that does not concentrate radioiodine, who have few therapeutic options. Stimulation of glucose transport in metastatic lesions by rhTSH may also make the detection of poorly differentiated lesions by [18F]-2-fluoro-2-deoxy-D-glucose-positron emission tomography scanning more effective (47, 48, 49).
Follow-up studies to confirm and extend the initial observations of Huysmans et al. (32, 33) on the usefulness of rhTSH to assist 131I treatment of multinodular goiters are needed. One of the key safety issues, as this is generally a condition of older individuals, is how to prevent hyperthyroidism due to rhTSH stimulation of T4 and T3 secretion during the preparation for radioiodine therapy. Doses of rhTSH much lower than those used for thyroid cancer patients will certainly be necessary, with careful attention to the frequency of adverse cardiovascular responses. Overall, the availability of rhTSH has opened a new era in thyroid cancer management and will enable investigators to explore questions of diagnosis and therapy that were impossible or impractical just 5 yr ago.
Acknowledgments
We thank Dr. Brian Haugen (University of Colorado, Denver, CO) for providing unpublished data from his earlier studies, and Dr. R. Michael Tuttle for editorial comments. We thank Cherryl Murray-Marone for assistance with manuscript preparation.
Footnotes
Abbreviations: DTC, Differentiated thyroid cancer; DxWBS, diagnostic whole body radioiodine scan; rhTSH, recombinant human TSH; Tg, thyroglobulin.
Received December 19, 2002.
Accepted January 23, 2003.
References