High-Risk Patients with Differentiated Thyroid Cancer T4 Primary Tumors Achieve Remnant Ablation Equally Well Using rhTSH or Thyroid Hormone Withdrawal

Abstract

Background:

Few data exist on using thyrotropin alfa (recombinant human thyroid-stimulating hormone [rhTSH]) with radioiodine for thyroid remnant ablation of patients who have T4 primary tumors (invasion beyond the thyroid capsule).

Methods:

A retrospective chart review protocol at nine centers in Europe was set up with special waiver of need for informed consent, along with a careful procedure to avoid selection bias when enrolling patients into the database. Data on 144 eligible patients with T4 tumors were collected (T4, N0–1, M0–1; mean age 49.7 years; 65% female; 88% papillary cancer). All had received ¹³¹I remnant ablation following TSH stimulation with rhTSH or thyroid hormone withdrawal (THW) since January 2000 (rhTSH n=74, THW n=70). The primary endpoint was based on evaluation of diagnostic radioiodine scan thyroid bed uptake more than six months after the ablation procedure, while stimulated serum Tg was a secondary endpoint. Safety was evaluated within 30 days after rhTSH or ¹³¹I.

Results:

Successful ablation judged by scan was achieved in 65/70 (92.9%) of rhTSH and in 61/67 (91.0%) of THW patients; the success rates were comparable, since noninferiority criteria were met. Although some patients in the initial cohort had tumor in cervical nodes and metastases, considering all evaluable patients regardless of various serum anti-Tg antibody assessments, the stimulated Tg was <2 ng/mL in 48/70 (68.6%) and 39/67 (58.2%) in rhTSH and THW groups respectively; if patients with anti-Tg antibody levels >30 IU/mL were excluded, the stimulated Tg was <2 ng/mL in 42/62 (67.7%) and 37/64 (57.8%) respectively. No serious adverse events occurred within the 30-day window after ablation.

Conclusions:

Use of rhTSH as preparation for thyroid remnant ablation in patients with T4 primary tumors achieved a rate of ablation success that was high and noninferior to the rate seen after THW, and rhTSH was well tolerated.

Introduction

An elevated thyrotropin (TSH) level in the blood is essential for effective thyroid remnant ablation. Historically, TSH has been elevated in these patients to allow remnant ablation by using thyroid hormone withdrawal (THW).

Thyrotropin alfa (recombinant human thyroid-stimulating hormone [rhTSH], Thyrogen; Genzyme, Cambridge, MA) was developed to provide TSH elevation to stimulate ¹³¹I uptake, thyroglobulin (Tg) secretion, or both while sparing patients the consequent hypothyroid morbidity and quality-of-life impairment from several weeks of THW (1 –9). In the United States, rhTSH was approved for diagnostic use in 1998, and for use for remnant ablation in 2007; these two indications were approved in Europe in 2000 and 2005 respectively. The remnant ablation indication was formally amended by regulators in Europe in 2009 to extend the indication from patients with “low-risk thyroid cancer,” a term poorly defined at that time, to all patients with well-differentiated thyroid cancer without distant metastases (European Medicines Agency Web site: www.emea.europa.eu under Thyrogen decisions, November 2009). This same indication for use of rhTSH for remnant ablation in all types of patients with well-differentiated thyroid cancer who are not M1 has been adopted by the Food and Drug Administration in the rhTSH package insert. These regulatory steps were aided by data in published patient series showing that successful remnant ablation was achieved using rhTSH in patients with larger primary tumors, tumor-node-metastases (TNM) classed T3 and T4, with presence of cervical nodes, and even with distant metastases (10 –12). Subsequently, more data about successful remnant ablation using rhTSH along with only 30 mCi ¹³¹I in low-risk patients but also in patients having cervical node involvement (13,14), and primary tumors classed T3 (14) added useful information. But published data about ablation success using rhTSH in patients with T4 primary tumors were not extensive.

Such T4 patients are rare (approximately 5% of all thyroid cancers), and are the most locally advanced stage of well-differentiated thyroid cancer. T4 patients have tumor extension beyond the thyroid capsule with invasion into surrounding local structures. In addition, these patients may or may not have involved neck nodes and distant metastases, elements that could affect the interpretation of stimulated serum Tg levels, for example when using serum Tg to assess success of remnant ablation. The subset of T4 tumors comprises a heterogeneous population, which tends to result in individualized patient treatment based on the disease progression and medical situation.

The primary objective of this chart review protocol was to demonstrate noninferior thyroid remnant first ablation success (based on historical diagnostic whole body scan [WBS] records) using rhTSH plus ¹³¹I compared to THW plus ¹³¹I in patients with T4 tumors. Secondary objectives were to assess remnant ablation efficacy of the rhTSH versus THW methods in such patients based on historical stimulated Tg levels, and also based on other historical records beyond the scan and laboratory reports (such as comments in clinic progress notes), and to explore the safety profile of rhTSH in this setting.

Methods

The study was designed and funded by Genzyme. The investigators collected data from patient charts, the data were summarized, and statistical analyses were performed by Covance. Covance and Genzyme wrote the final study report and this manuscript with input from the investigators. This study was a retrospective, noninterventional study comparing the thyroid remnant ablation success of rhTSH (thyrotropin alfa; Genzyme) plus ¹³¹I versus THW plus ¹³¹I in patients with T4 tumors. The study consisted of reviewing and recording necessary efficacy and safety information from historical patient records. As this was a retrospective study, no investigational product was administered, and Good Clinical Practice guidelines did not apply. Applicable local and national regulations were followed.

Eligible patients were thyroid cancer patients (living or deceased) who had undergone a near-total or total thyroidectomy of a well-differentiated T4 tumor on or after January 1, 2000. The chart reviews were conducted between May and August 2012. A T4 tumor was defined as a primary tumor of any size that extended beyond the thyroid capsule (referred to as TNM classification T4, N0–1, or M0–1). Eligible patients must have received their first remnant ablation attempt using THW or rhTSH (0.9 mg intramuscularly [IM] on two consecutive days) followed by high ablative activity of ¹³¹I (≥28 mCi or ≥1.036 GBq).

Each site recorded study data into an electronic case report form (eCRF). There was no monitoring of data collected on the eCRF, although the site could be contacted (to seek clarification about missing, incomplete, or unclear data) by telephone or electronic communication to maintain patients' confidentiality.

Waiver of informed consent

This study was approved by the Institutional Ethics Committees of all involved sites. There was also approval of a waiver for the requirement of obtaining patient informed consent, a decision supported by the following rationale. First, there was no physical risk to the patients. This study consisted of retrospectively reviewing and recording necessary efficacy and safety information from historical patient records. Second, collection of the informed consent would make the study unfeasible due to unavailability of current addresses. Some eligible patients may have been seriously ill, very old, or have died. Not including these patients in the study would lead to a possible bias of only including patients who were not as ill. Third, not obtaining an informed consent would not adversely affect patients' rights and welfare. The information taken from patient charts did not include identifiable information.

Procedures to avoid enrollment bias

Historical records were reviewed by investigators in a systematic fashion to identify eligible patients. Each patient record reviewed but found not to be eligible was entered into the eCRF as a screen failure, and included selected limited demographic characteristics and listing of the inclusion/exclusion criteria not met. Each patient record reviewed and found to be eligible was entered into the eCRF, and included limited demographic characteristics, information about the patient's first thyroid remnant ablation treatment, and data related to ablation success or failure.

To avoid bias when looking at charts to select patients for inclusion into the study, investigators followed a systematic process, as follows:

1. Make a list of all T4-type patients fulfilling the study's inclusion and exclusion criteria.

2. Split the list into two lists, by treatment group (rhTSH vs. THW).

3. Order each list by date of ¹³¹I ablation (most recent date on top).

Patient selection for enrollment and entry into the eCRF then occurred as follows:

1. Ensure patient source data are available for the first several patients at the top of each list to make it easier to be able to select patients consecutively to complete the eCRFs.

2. Starting with one of the two lists, enter the first patient in the list into the eCRF.

3. Moving to the other list, enter the first patient on that list into the eCRF. At this point, the number of patients in the two treatment groups (rhTSH and THW) is balanced.

4. Descending down the first list, enter the next patient from the list into the eCRF.

5. Returning to the other list, enter the next patient into the eCRF.

6. Repeat the above steps until any one of the following conditions are met: (i) the rhTSH list is exhausted of eligible patients; (ii) the THW list is exhausted of eligible patients; (iii) 20 eligible patients have been entered onto the eCRF.

After written approval from Genzyme, recruitment could be continued beyond the initial number of patients to a new target number for enrollment. Because there was an enrollment cap due to budget limitations, this “stop at 20” rule gave each institution time to review charts and enter patients into the database, allowing for a broader multinational patient sample.

Number of patients planned and analyzed

Concerning enrollment at all centers, chart reviews to identify eligible patients were planned to continue until one of the following scenarios was achieved: (i) 63 or more eligible patients were enrolled who had undergone ablation by THW and 63 or more eligible patients were enrolled who had undergone ablation with appropriate use of rhTSH (0.9 mg IM on two consecutive days), or (ii) a maximum of 225 total eligible patients were enrolled altogether in the study.

Inclusion and exclusion criteria

Historical records from patients who met all of the following criteria were eligible for inclusion in this retrospective evaluation: male or female patients living or deceased, aged 18 years or older at the time of first ablation for thyroid cancer; diagnosed with a well-differentiated T4 tumor (a primary tumor of any size that extends beyond the thyroid capsule; TNM classification T4, N0-1, M0-1), excluding unusual histological types such as oncocytic (Hürthle cell carcinoma), tall cell, sclerosing, or cribriform thyroid cancers; undergone a near-total or total thyroidectomy on or after January 1, 2000; undergone first ablation of thyroid remnants with high activity ¹³¹I (≥28 mCi or ≥1.036 GBq); undergone first ¹³¹I remnant ablation using either rhTSH (0.9 mg IM on two consecutive days) or THW stimulation, excluding nonstandard rhTSH regimens; historical records available confirming ablation results by (i) diagnostic WBS (DxWBS) using a small (≥2 mCi or 74 MBq) activity of iodine ¹³¹I (or ¹²³I) performed at least six months after administration of the first ablation activity of ¹³¹I, and/or (ii) Tg measured at least six months after administration of the first ablation activity of ¹³¹I.

Historical records from patients who met any of the following criteria were excluded from this retrospective evaluation: received propylthiouracil, methimazole, vitamins, or supplements containing kelp or iodine (taking a multivitamin that does not contain iodine or kelp is acceptable); received medications that significantly affect iodine handling such as high-dose corticosteroids, high-dose diuretics, or lithium in the 45 days before administration of first ablative activity of ¹³¹I; received any iodine-containing contrast agents within three months prior to first ablative activity of ¹³¹I administered; or used amiodarone within the two years prior to first ablative activity of ¹³¹I administered.

Efficacy assessments

Primary endpoint

This was the first ablation success rate by DxWBS using a small activity (≥2 mCi or 74 MBq) of iodine ¹³¹I (or ¹²³I) performed at least six months after administration of first ablation activity of ¹³¹I. Success was defined as historical records with no visible uptake in the thyroid bed, an uptake in the thyroid bed of <0.1% of the applied activity, or a trace amount. The word “trace” need not be present in the chart as long as the word used has the same meaning.

Secondary endpoints

These were (i) first ablation success based on stimulated Tg level of <2 ng/mL performed at least six months after administration of first ablation activity of ¹³¹I; and (ii) first ablation success based on other historical records (such as comments in clinic progress notes) performed at least six months after administration of first ablation activity of ¹³¹I.

Safety evaluation

All serious adverse events (SAEs) within the safety review period were recorded on the eCRF. Events within the safety review period were defined as events that started within 30 days after rhTSH administration for rhTSH patients, or within 30 days after ablative ¹³¹I activity for THW patients.

Statistical methods

Efficacy

The primary efficacy variable—ablation success rate—was analyzed by means of a generalized linear model (GLM) to compare the rhTSH and THW groups. The model was adjusted for the following covariates using a covariate selection process: center, N1 status (N1 vs. not N1), age, sex, and year of treatment. Covariate adjustment was performed by fitting each covariate separately to the model and retaining those covariates with p≤0.2. The retained covariates were included in the model together, and those with p≤0.1 were kept in the final model. The result used for inference was the confidence interval (CI) of the risk difference (rhTSH–THW), which was compared with the prespecified noninferiority margin of −0.2. If the lower bound of the confidence interval was >−0.2, noninferiority was concluded. The primary analysis was conducted on the full analysis set (FAS), which consists of all patients who were treated and who received a DxWBS to assess ablation success. Two supportive sensitivity analyses were conducted. The first utilized a simple GLM that did not adjust for covariates. The second utilized a GLM that adjusted for propensity score conducted on the propensity set (PS). The PS is defined as those patients in the FAS retained after the following propensity trimming criteria were applied. Propensity score was first estimated for each patient in the FAS using logistic regression. The logistic regression modeled the distribution of treatment (rhTSH, THW as the outcome variable) given the following covariates: center, N1 status (N1 vs. not N1), age, sex, and year of treatment. The propensity score was then ordered, ignoring treatment group. Subjects in the THW group with propensity scores higher than the maximum propensity score in the rhTSH group were removed from analysis. Similarly, subjects in the THW group with propensity scores lower than the minimum propensity score in the rhTSH group were removed. The patients remaining after this trimming process constitute the PS. The propensity score was used to balance the covariates in the two groups, and therefore reduce the bias in the comparison of two treatment groups in a nonrandomized study.

The secondary efficacy parameters were ablation success rate based on Tg level of <2 ng/mL and ablation success rate based on other historical records, and were analyzed in the same way as for the primary efficacy parameter.

Sample size

Although this was a retrospective study, a sample size calculation was performed to provide an understanding of what power could be expected if a randomized study was conducted. This included the following assumptions:

• The success rate for THW under the null hypothesis. Success levels between 60% and 90% were modeled. Historically across a number of studies, the rate of ablation success has been around 80%. More recent studies comparing rhTSH to THW have seen ablation success rates around 90% or higher. These studies have generally not included T4 patients, and based on clinical experience, a value of 80% success was more realistic and assumed to be the rate for THW.

• Noninferiority margin. This was a heterogeneous population. A value of 20% was proposed.

• α=0.05.

• 80% power. It was determined that a sample size of 63 patients per treatment arm would provide 80% power, while 84 patients per treatment arm would provide 90% power.

Results

A total of 153 patients were entered into the study database. Nine of these patients did not meet all eligibility criteria and were therefore excluded from the study populations.

Of the 144 eligible patients in the treated/safety set, 74 received rhTSH plus ¹³¹I administration (rhTSH group) and 70 received THW plus ¹³¹I administration (THW group). Demographics are presented in Table 1. The mean age was 49.7 years, 65% female, with 88% papillary cancer and 12% follicular cancer. Demographics of patients in the two treatment groups were similar, although more patients in the THW group had involved cervical nodes, and the mean amount of radioiodine used for ablation was slightly lower in the rhTSH group (107.8 mCi) than in the THW group (128.6 mCi), p=0.0003. Seven of these 144 patients did not receive a DxWBS for detecting ablation success, resulting in 137 (95.1%) eligible patients in the FAS. Of these patients, 70 received rhTSH and 67 received THW. One patient who was in the THW group had a calculated propensity score lower than the minimum propensity score in the rhTSH group and was therefore removed from the PS. This resulted in 136 (94.4%) of originally eligible patients in the PS, with 70 patients in the rhTSH group and 66 patients in the THW group.

Table 1.

Demography and Baseline Variables (All Patients Treated/Safety Set)

	rhTSH (n=74)	THW (n=70)	p-Value
Age (years)^a
n	74	70
Mean±SD	50.6±17.3	48.7±13.7	0.4698
Median (min/max)	50.5 (20/86)	46.5 (20/78)
Sex, n (%)
n	74	70
Female	50 (67.6)	44 (62.9)	0.6016
Male	24 (32.4)	26 (37.1)
Histological tumor type, n (%)
n	74	70
Papillary thyroid cancer	64 (86.5)	63 (90.0)	0.6093
Follicular thyroid cancer	10 (13.5)	7 (10.0)
Other thyroid cancer	0 (0.0)	0 (0.0)
Tumor node classification, n (%)
n	74	70
N0	27 (36.5)	14 (20.0)	0.0444
N1	33 (44.6)	45 (64.3)
Nx	14 (18.9)	11 (15.7)
Tumor metastasis classification, n (%)
n	74	70
M0	30 (40.5)	28 (40.0)	0.9800
M1	5 (6.8)	5 (7.1)
Mx	39 (52.7)	37 (52.9)
Length of time between confirmed well-differentiated T4 tumor diagnosis and thyroidectomy (days)^b
n	74	70
Mean±SD	5.8±8.5	5.8±8.1	0.9504
Median (min/max)	2.0 (0/37)	2.0 (0/40)
Length of time between thyroidectomy and rhTSH use for ablation preparation (days)^c
n	74
Mean±SD	90.1±70.7
Median (min/max)	78.5 (6/508)
Length of time between thyroidectomy and first remnant ablation with high activity ¹³¹I (days)^d
n	74	70
Mean±SD	92.1±70.7	87.7±59.6	0.6887
Median (min/max)	80.5 (8/510)	76.0 (14/245)
Activity of ¹³¹I used for first remnant ablation for each unit type (mCi)^e
n	74	70
Mean±SD	107.8±22.5	128.6±40.3	0.0003
Median (min/max)	100.0 (56/205)	113.7 (50/300)

The denominator for percentages is the number of patients in the treated/safety set for the relevant treatment group.

Calculated as (date of confirmed well-differentiated T4 tumor diagnosis−date of birth+1)/365.25.

Calculated as date of thyroidectomy−date of confirmed well-differentiated T4 tumor diagnosis.

Calculated as date of first injection of rhTSH−date of thyroidectomy.

Calculated as date of first remnant ablation with high activity ¹³¹I−date of thyroidectomy.

Units of Megabecquerels (MBq) were standardized into mCi by multiplying by 0.027; units of Gigabecquerels (GBq) were standardized into mCi by multiplying by 27.

rhTSH, recombinant human thyroid-stimulating hormone (rhTSH) and ¹³¹I administration; THW, thyroid hormone withdrawal and ¹³¹I administration.

Efficacy results

A summary of efficacy results is presented in Table 2. For the prespecified primary efficacy objective, 65/70 (92.9%) of rhTSH and 61/67 (91.0%) of THW patients had successful remnant ablation, defined as no uptake or only trace uptake in the thyroid bed on the DxWBS done more than six months later. The risk difference (rhTSH−THW) was 0.034 [CI−0.043, 0.112]. The covariate sex (p=0.054) was retained in the final model following the covariate selection process. As the lower bound value of the CI for the risk difference was >−0.2, noninferiority was successfully met (p=0.386), showing that thyroid remnant first ablation success of rhTSH plus ¹³¹I was noninferior to THW plus ¹³¹I in patients with T4 tumor based on historical DxWBS records.

Table 2.

Summary of Efficacy Endpoints(Full Analysis Set)

	rhTSH (n=70)	THW (n=67)	Total (n=137)
DxWBS showing visible uptake in the thyroid bed,^a n (%)
n	70	67	137
No uptake seen	59 (84.3)	53 (79.1)	112 (81.8)
Significant uptake	5 (7.1)	6 (9.0)	11 (8.0)
Trace uptake	6 (8.6)	8 (11.9)	14 (10.2)
Clinicians' judgment of ablation success based on historical records,^b n (%)
n	70	67	137
Successful	57 (81.4)	57 (85.1)	114 (83.2)
Unsuccessful	13 (18.6)	10 (14.9)	23 (16.8)
Stimulated thyroglobulin levels,^c n (%)
n	70	65	135
<2 ng/mL	48 (68.6)	39 (58.2)	87 (63.5)
≥2 ng/mL	22 (31.4)	26 (38.8)	48 (35.0)

The denominator for percentages is the number of patients in the full analysis set for the relevant treatment group.

Ablation success for the primary endpoint is defined as historical records of no visible uptake in the thyroid bed; an uptake in the thyroid bed (if measured quantitatively) of <0.1% of the applied activity; or a small amount of visible uptake seen in the thyroid bed but deemed to be a trace amount.

Secondary efficacy variable: first ablation success based on historical records performed at least six months after administration of first ablation activity of ¹³¹I.

Secondary efficacy variable: first ablation success based on stimulated Tg level of <2 ng/mL performed at least six months after administration of first ablation activity of ¹³¹I.

rhTSH, rhTSH and ¹³¹I administration; THW, thyroid hormone withdrawal and ¹³¹I administration.

The results from the first sensitivity analysis without adjusting for covariates were similar to the primary efficacy analysis. The risk difference (rhTSH−THW) was 0.018 [CI−0.073, 0.109] and p=0.697. The results from the second sensitivity analysis adjusting for the covariate propensity score were also similar to the primary efficacy analysis. The risk difference was 0.017 with CI [−0.071, 0.105] and p=0.706. The p-value for the covariate propensity score was 0.322. The results from the two sensitivity analyses demonstrated that any potential covariate imbalances between treatment groups were minimal and that the choice of covariates had minimal effect, supporting the primary efficacy endpoint result that rhTSH plus ¹³¹I was noninferior to THW plus ¹³¹I in patients with T4 tumor based on historical DxWBS records.

Because different types of anti-Tg antibody tests (or other such tests) were used at the different centers over the years, as a “first look” at ablation success based on the stimulated Tg result, no patient was excluded from the analysis because of anti-Tg antibody status. We suspected that tumor in neck nodes or in distant metastases also could confound this analysis of Tg values. Thyroid remnant first ablation success based on stimulated Tg level of <2 ng/mL (done more than six months after the ablation procedure) was achieved in 48/70 (68.6%) and 39/67 (58.2%) patients in the rhTSH and THW groups respectively. The risk difference was 0.036, with CI [-0.117, 0.188] and p=0.645.

A subgroup analysis was performed for the secondary endpoint of first ablation success based on historical stimulated Tg levels, excluding those with anti-Tg antibody level >30 U/mL, a cut-off value that had been adopted clinically at several centers, although the authors do not vouch for the validity of this approach. Removal of these patients' data resulted in 62 patients in the rhTSH group and 64 patients in the THW group for this subgroup analysis. Of these patients, 42/62 (67.7%) of rhTSH patients and 37/64 (57.8%) of THW patients had successful remnant ablation, defined as <2 ng/mL serum Tg. The risk difference was 0.028, with CI (−0.130, 0.186] and p=0.730. The covariates regional node status (p=0.002) and year of treatment (p=0.020) were retained in the model following the covariate selection process. Thus, among patients with anti-Tg antibody level ≤30 U/mL, ablation success results were similar to those seen in the overall Tg analysis. Although these data clearly are confounded, for example by tumor in cervical nodes, had the ablation success rates (using stimulated Tg as the criterion) in the two treatment groups been very different, this would have been problematic.

Thyroid remnant first ablation success based on other (nonscan and non-Tg) historical records (such as comments in clinic progress notes) was achieved in 57/70 (81.4%) and 57/67 (85.1%) patients in the rhTSH and THW groups respectively. The risk difference was 0.040 with CI [−0.089, 0.168] and p=0.544. Thus, data collected for the prespecified secondary efficacy endpoints were consistent with the primary efficacy results. Other efficacy variables were summarized descriptively by treatment group for the FAS and showed similar trends between rhTSH and THW for: (i) frequency and percentage of patients with DxWBS showing similar amounts of regional or distant metastases, (ii) stimulated and nonstimulated Tg levels, and (iii) length of time between first remnant ablation with ¹³¹I and DxWBS. However, an increased level of anti-Tg antibodies was observed in stimulated and nonstimulated samples in rhTSH patients compared to THW patients, and this remains without explanation.

Safety results

No treatment-related SAEs were identified, suggesting that both treatments are well tolerated and that rhTSH is not different from THW with regard to these safety endpoints.

Discussion

Limited data exist about use of rhTSH with ¹³¹I for remnant ablation in patients with TNM T4 primary tumors because such differentiated thyroid cancers occur in less than 5% of newly diagnosed patients. This retrospective, noninterventional study was based on chart reviews at nine European centers to assess the efficacy and safety of rhTSH use in this setting. The study had some unusual features, including obtaining waivers so that informed consent was not required and use of a well-defined method to select which patients were entered into the study such that there would be minimal selection bias while at the same time keeping the two treatment arms numerically balanced.

The majority of the patients were middle-aged females with papillary cancers, and the treatment groups had similar distributions of sex, age, cancer type, and TNM classification. All rhTSH patients had consistent treatments, that is, two identical doses of rhTSH administered one day apart. There were more patients with involved cervical nodes in patients who received THW (p=0.0444), and the activity of ¹³¹I was greater in the THW treatment group (p=0.0003). Slight differences in demographics and treatment history were not expected to impact the study results.

The primary efficacy analysis of first ablation success rate based on historical DxWBS in the FAS showed 92.9% of rhTSH patients and 91.0% of THW patients had successful remnant ablation, defined as “no uptake” or “trace uptake” in the thyroid bed on subsequent DxWBS (rhTSH was noninferior to THW). This major finding was also supported by the two sensitivity analyses, which showed that any potential covariate imbalances between treatment groups were minimal and that the choice of covariates in the primary model made no difference to the finding of noninferiority.

Thyroid remnant first ablation success based on stimulated Tg level of <2 ng/mL, or based on other historical records (e.g., comments in clinic progress notes), which were the prespecified secondary efficacy endpoints, also showed that the rhTSH treatment was noninferior to THW.

Although care was taken to avoid investigator bias regarding which patients were captured by the chart reviews, and a propensity analysis was performed, this retrospective study has limitations inherent in any nonrandomized series of patients who were not treated prospectively. The patients prescribed use of THW could have been viewed as more ill by their physicians and so were given a higher amount of ¹³¹I for their ablation procedures, although we speculate that had the rhTSH patients received more ¹³¹I, then there might have been criticism that the extra ¹³¹I was favoring the ablation results seen after rhTSH. An additional weakness is that the study was very focused on testing just one hypothesis—assessing the ablation rates by nuclear medicine scanning after two methods of preparation for ablation. As the protocol was being designed, it became clear that over years, there was great variation in the types of Tg assays (and anti-Tg antibody assays and recovery tests) used at the institutions. Therefore, Tg data were collected but relegated to a secondary endpoint. Many other clinical care parameters (use of low-iodine diets, compliance with such diets, urine iodine measurements, etc.) were intentionally not collected in order to streamline the chart reviews and because a study of fewer than 150 patients was not expected to allow rigorous analyses of various treatment subgroups. For example, although 10 patients with M1 disease were enrolled and were equally divided between the two treatment groups, we did not collect data about the location of the metastases or their response to radioiodine treatment.

The first ablation success rates as assessed by scan in this study were high and similar to the rates seen in other published studies with thyroid tumors of a lower TNM status (4,13,14). In a randomized, controlled study (n=63) by Pacini et al., designed to compare the efficacy and safety of rhTSH and THW in patients staged T0, T1, T2, and T4 with minor invasion of the thyroid capsule (with 90% of the patients staged T0–T2), N0–N1, and M0 or T0–T1, N1, and M0, it was found that successful ablation occurred in 100% of patients in both groups (4). In the HiLo study (14), 438 patients with differentiated thyroid cancer (tumor stage pT1–T3, Nx, N0 or N1, and M0) were randomized to one of four arms: 30 mCi or 100 mCi, each with either rhTSH or THW as preparation. Ablation success rates (as judged by both scan and stimulated Tg <2 ng/mL) were 85% in the group receiving 30 mCi versus 88.9% in the group receiving 100 mCi. The group that was prepared using rhTSH had a success rate of 87.1%, whereas patients prepared using THW had a success rate of 86.7%. The ESTIMABL study by Schlumberger et al. (13) compared the same four strategies for thyroid remnant ablation as evaluated in the HiLo study. Patients in ESTIMABL were, however, TNM stage pT1 <1 cm with N1, or pT1 >1–2 cm with either N0 or 1, or pT2 with N0, and for all these patient types, there could not be any distant metastases present. For the four groups, rhTSH+30 mCi, rhTSH+100 mCi, THW+30 mCi, and THW+100 mCi, the rates of ablation success (as judged by fulfilling both criteria of neck ultrasound and stimulated Tg<1 ng/mL by local assay) were 90%, 93%, 92%, and 94% respectively. No T4 patients had been included in the HiLo or ESTIMABL studies, but together with the present European chart review study of T4 patients, it appears that the success rates of remnant ablation using rhTSH or THW preparation are similar for patients with T0–4 primary tumor types of differentiated thyroid cancer.

No treatment-related SAEs were identified in patient records for either treatment, indicating that rhTSH plus ¹³¹I and THW plus ¹³¹I are not different from each other with regard to this safety endpoint.

In conclusion, this retrospective data collection indicates that preparation for thyroid remnant ablation using either rhTSH or THW is comparably effective and safe in such T4 type patients.

Footnotes

Acknowledgments

This study was a Genzyme-sponsored project, THYR04910, and was accomplished with the assistance of colleagues at Covance, including a skilled writer, Jeffrey Gardner. Funding for this study was provided entirely by Genzyme, a Sanofi company.

Author Disclosure Statement

J.A.V.C. received fees for speaking for Genzyme, Bayer, and Novartis Oncology, and participated in an advisory board for Covidien. J.M. is an employee of Genzyme. None of the remaining authors have anything to disclose.

References

Ladenson

, Braverman

, Mazzaferri

, Brucker-Davis

, Cooper

, Garber

, Wondisford

, Davies

, DeGroot

, Daniels

, Ross

, Weintraub

. 1997. Comparison of administration of recombinant human thyrotropin with withdrawal of thyroid hormone for radioactive iodine scanning in patients with thyroid carcinoma. N Engl J Med, 337:888–896.

Haugen

, Pacini

, Reiners

, Schlumberger

, Ladenson

, Sherman

, Cooper

, Graham

, Braverman

, Skarulis

, Davies

, DeGroot

, Mazzaferri

, Daniels

, Ross

, Luster

, Samuels

, Becker

, Maxon

, Cavalieri

, Spencer

, McEllen

, Weintraub

, Ridgway

. 1999. A comparison of recombinant human thyrotropin and thyroid hormone withdrawal for the detection of thyroid remnant or cancer. J Clin Endocrinol Metab, 84:3877–3885.

Shroeder

, Haugen

, Pacini

, Reiners

, Schlumberger

, Sherman

, Cooper

, Schuff

, Braverman

, Skarulis

, Davies

, Mazzaferri

, Daniels

, Ross

, Luster

, Samuels

, Weintraub

, Ridgway

, Ladenson

. 2006. A comparison of short-term changes in health-related quality of life in thyroid carcinoma patients undergoing diagnostic evaluation with recombinant human thyrotropin compared with thyroid hormone withdrawal. J Clin Endocrinol Metab, 91:878–884.

Pacini

, Ladenson

, Schlumberger

, Driedger

, Luster

, Kloos

, Sherman

, Haugen

, Corone

, molimaro

, Elisei

, Ceccarelli

, Pinchera

, Wahl

, Leboulleux

, Ricard

, Yoo

, Busaidy

, Delpassand

, Hänscheid

, Felbinger

, Lassmann

, Reiners

. 2006. Radioiodine ablation of thyroid remnants after preparation with recombinant human thyrotropin in differentiated thyroid carcinoma: results of an international, randomized, controlled study. J Clin Endocrinol Metab, 91:926–932.

Hänscheid

, Lassmann

, Luster

, Thomas

, Pacini

, Ceccarelli

, Ladenson

, Wahl

, Schlumberger

, Ricard

, Driedger

, Kloos

, Sherman

, Haugen

, Carriere

, Corone

, Reiners

. 2006. Iodine biokinetics and dosimetry in radioiodine therapy of thyroid cancer: procedures and results of a prospective international controlled study of ablation after rhTSH or hormone withdrawal. J Nucl Med, 47:648–654.

Borget

, Remy

, Chevalier

, Ricard

, Allyn

, Schlumberger

, De Pouvourville

. 2008. Length and cost of hospital stay of radioiodine ablation in thyroid cancer patients: comparison between preparation with thyroid hormone withdrawal and Thyrogen. Eur J Nucl Med Mol Imaging, 35:1457–1463.

Remy

, Borget

, Leboulleux

, Guilabert

, Lavielle

, Garsi

, Bournaud

, Gupta

, Schlumberger

, Ricard

. 2008. ¹³¹I effective half-life and dosimetry in thyroid cancer patients. J Nucl Med, 49:1445–1450.

Rosario

, Borges

MAR

, Purisch

. 2008. Preparation with recombinant human thyroid-stimulating hormone for thyroid remnant ablation with ¹³¹I is associated with lowered radiotoxicity. J Nucl Med, 49:1776–1782.

Vallejo Casas

, Mena Bares

, Gálvez

, Marlowe

, Latre Romero

, Martinez-Paredes

. 2011. Treatment room length-of-stay and patient throughput with radioiodine thyroid remnant ablation in differentiated thyroid cancer: comparison of thyroid-stimulating hormone stimulation methods. Nucl Med Commun, 32:840–846.

10.

Tuttle

, Brokhin

, Omry

, Martorella

, Larson

, Grewal

, Fleisher

, Robbins

. 2008. Recombinant human TSH-assisted radioactive iodine remnant ablation achieves short-term clinical recurrence rates similar to those of traditional thyroid hormone withdrawal. J Nucl Med, 49:764–770.

11.

Tuttle

, Lopez

, Leboef

, Minkowitz

, Grewal

, Brokhin

, Omry

, Larson

. 2010. Radioactive iodine administered for thyroid remnant ablation following recombinant human thyroid stimulating hormone preparation also has an important adjuvant therapy function. Thyroid, 20:257–263.

12.

Hugo

, Robenshtok

, Grewal

, Larson

, Tuttle

. 2012. Recombinant human thyroid stimulating hormone-assisted radioactive iodine remnant ablation in thyroid cancer patients at intermediate to high risk of recurrence. Thyroid, 22:1007–1015.

13.

Schlumberger

, Catargi

, Borget

, Deandreis

, Zerdoud

, Bridji

, Bardet

, Leenhardt

, Bastie

, Schvartz

, Vera

, Morel

, Benisvy

, Bournaud

, Bonichon

, Dejax

, Toubert

M-E

, Leboulleux

, Ricard

, Benhamou

; Tumeurs de la Thyroide Refractaires Network for the Essai Stimulation Ablation Equivalence Trial. 2012. Strategies of radioiodine ablation in patients with low-risk thyroid cancer. N Engl J Med, 366:1663–1673.

14.

Mallick

, Harmer

, Yap

, Wadsley

, Clarke

, Moss

, Nicol

, Clark

, Path

, Farnell

, McCready

, Smellie

, Franklyn

, John

, Nutting

, Newbold

, Lemon

, Gerrard

, Abdel-Hamid

, Hardman

, Macias

, Roques

, Whitaker

, Vijayan

, Alvarez

, Beare

, Forsyth

, Kadalayil

, Hackshaw

. 2012. Ablation with low-dose radioiodine and thyrotropin alfa in thyroid cancer. N Engl J Med, 366:1674–1685.