| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
REVIEW |
Cancer Research UK & UCL Cancer Trials Centre (A.H.), University College London, London NW1 2ND, United Kingdom; Formerly Thyroid Cancer Unit (C.H.), Royal Marsden Hospital, London SW3 6JJ, United Kingdom; Newcastle General Hospital (U.M.), Newcastle NE4 6BE, United Kingdom; Chelsea and Westminster Hospital (M.H.), London SW10 9NH, United Kingdom; and Division of Medical Sciences (J.A.F.), University of Birmingham, Queen Elizabeth Hospital, Birmingham B15 2TH, United Kingdom
Address all correspondence and requests for reprints to: Allan Hackshaw, Deputy Director, Senior Lecturer in Epidemiology & Medical Statistics, Cancer Research UK & UCL Cancer Trials Centre, University College London, Stephenson House, 158-160 North Gower Street, London NW1 2ND, United Kingdom. E-mail: ah{at}ctc.ucl.ac.uk.
| Abstract |
|---|
|
|
|---|
Objective: A systematic review of the published literature was used to compare the success rates of remnant ablation using approximately 30 mCi with approximately 100 mCi (1.1 vs. 3.7 GBq).
Data Sources: Data were obtained from MEDLINE and EMBASE for the years 1966 to March 2006.
Study Selection: All studies that reported rates of successful ablation associated with approximately 30 or approximately 100 mCi of radioiodine were reviewed.
Data Extraction: Studies were based on reviews of patient case notes (n = 41), prospective cohorts (n = 12), and randomized trials (n = 6). We obtained the success of thyroid remnant ablation according to different administered activities of radioiodine. Where a study reported on two or more activities, the risk ratio of having a successful ablation (
30 vs.
100 mCi) was calculated and combined in a meta-analysis.
Data Synthesis: Observational studies confirmed the high ablation success rate (
80%) using approximately 100 mCi, although 22% of studies reported a rate of 90% or greater. The pooled ablation success rate in these studies was 10% lower using 30 mCi compared with 100 mCi (95% confidence interval, 317%; P = 0.01). The meta-analysis of the randomized trials produced equivocal results. For example, the rate of successful ablation in patients given 30 mCi was 8% lower compared with 100 mCi (95% confidence interval, 29% lower or up to 20% greater, P = 0.58), consistent with there being no difference or that 30 mCi is much less effective.
Conclusions: From the published data, it is not possible to reliably determine whether ablation success rates using 30 mCi are similar to using 100 mCi. Large randomized trials are needed to resolve the issue and guide clinical practice.
| Introduction |
|---|
|
|
|---|
Most centers use a single administered activity, and there is uncertainty over the lowest effective activity of radioiodine that can achieve successful ablation in the vast majority of subjects, particularly when compared with the typically administered activity of 100 mCi (or 3.7 GBq). There are several advantages to both the patient and healthcare provider for using a lower activity of radioiodine, including less time in isolation, a shorter hospital stay (when local or national regulations deem this necessary), reduced exposure of radioiodine to the environment, and lower financial cost. Furthermore, radioiodine ablation is associated with an increased risk of second primary malignancies; the lower the activity administered the lower the risk (5). An estimated 53 solid tumors are expected among 10,000 patients after 10 yr if they were treated with 100 mCi compared with only 16 if they had been treated with 30 mCi (5).
Several guidelines for the management of patients with differentiated thyroid cancer have been published recently. The UK guidelines from the British Thyroid Association (6) recommend 131I ablation after total thyroidectomy for patients with tumors larger than 1 cm in size. Although a specific amount of activity is not recommended, the guidelines state that "the usual ablation dose is 3.7 GBq. Some centers may use low-dose ablation (1.1 GBq) or dosimetric assessment of radioiodine dosage, preferably within a trial or study setting."
The European consensus report on ablation (7) recommends administration of a high-ablation activity (
100 mCi) when patients have distant metastases, incomplete tumor resection, or high risk for recurrence (for example T3 or T4 tumors). For patients with tumors larger than 1 cm who had less than total thyroidectomy or unfavorable histology and no lymph node dissection, the report recommends "high or low activity (3.7 or 1.1 GBq)." Although the report acknowledges that the evidence based on high- vs. low-activity radioiodine is inconclusive, the authors state that 1.1 GBq may be sufficient, particularly if patients have had a total thyroidectomy. A more recent consensus on overall management endorsed by the European Thyroid Association (8) admits that the administered activity throughout 25 European countries ranges between 30 and 100 mCi or even more.
Updated guidelines from the American Thyroid Association (9) recommend radioiodine ablation for patients with stage III or IV disease (American Joint Committee on Cancer classification), all patients less than 45 yr old with stage II disease, most patients aged 45 yr or older with stage II disease, and selected patients with stage I disease (especially those with multifocal disease or nodal metastases). No specific amount of activity is preferred, and the conclusion is that ablative activities between 30 and 100 mCi (1.13.7 GBq) generally show similar successful ablation rates, although higher activities may be better. The recommendation is to use "the minimum activity (30100 mCi) necessary to achieve successful remnant ablation."
The recommendations on ablation activity from these reports are not entirely consistent with each other. Thus, there remains uncertainty over whether it is acceptable to use an activity as low as 30 mCi.
There have been several studies reporting the rates of successful ablation according to different ablation activities administered, most of which are based on retrospective reviews of patient case notes. A meta-analysis published in 2000 (10), based on 10 retrospective studies and three randomized trials, concluded that a high activity was more effective than a low one; there was a reported 27% reduction in the chance of having a failed remnant ablation [95% confidence interval (CI), 1339% reduction]. However, this effect was seen only in the retrospective studies (31% reduction, 95% CI, 1643%) and there was no evidence of a benefit in the randomized trials (10% reduction, 95% CI, 40% reduction to 30% increase).
Since the review in 2000, there have been more randomized trials and observational studies. We here review the current literature on using a single administered activity in relation to radioiodine ablation.
| Methods |
|---|
|
|
|---|
The intention was to obtain the ablation success rate in patients who received approximately 30 mCi and those who received about 100 mCi. Studies, or groups of patients within studies, were not included in the analysis if: 1) patients were previously treated with 131I therapy; 2) it was not possible to separate results for high and low activities (for example, when some patients received 30 mCi and others received 100 mCi, but only the combined successful ablation rate was reported); 3) patients received an activity greater than 200 mCi; 4) all patients had either lymph node metastases or distant metastatic disease; 5) patients had fractionated doses of radioiodine; 6) less than five patients were given a specific administered activity; 7) some patients had been given external radiotherapy; or 8) it was not possible to obtain the rate of successful ablation for a specific administered activity of either approximately 30 mCi or 100 mCi. When there were multiple publications on the same patient group, only one was included.
Studies were based on retrospective reviews of patient case notes (n = 41), prospective cohorts (n = 12), or randomized trials in which patients were randomly allocated to receive one of two or more different administered activities (n = 6). All the studies included in the analyses are listed in Tables 13![]()
![]()
. From each report the following information was extracted: extent of surgery before ablation, timing of the follow-up diagnostic scan, inclusion of patients with metastatic disease, method of TSH stimulation before ablation [off thyroid hormone replacement or on recombinant human (rh) TSH], whether a low-iodine diet was recommended before ablation, the definition of successful ablation, and the proportion of patients who achieved successful ablation. There were instances in which a group of patients included those who had lymph node metastases or distant metastases. Because patients with distant metastases usually represented a small proportion of the total number of patients, they were included in the analysis. Results are provided with and without these patients. In the randomized trials, no patient had evidence of either lymph node metastases or distant metastatic disease, except one trial (63) that included some patients with spread to the lymph nodes.
|
|
|
All six randomized trials were based on comparing two or more administered activities, so only the risk ratios were combined across trials. This was also performed using a random effects model (11).
The risk ratios were combined separately for the observational studies and the randomized trials because of the obvious difference in study design.
| Results |
|---|
|
|
|---|
Tables 1
and 2
show the results from observational studies. All were based on retrospective reviews of patient case notes except 12, which were prospective cohorts (14, 15, 21, 25, 26, 28, 29, 33, 41, 42, 53, 56). The timing of the diagnostic scan ranged from approximately 312 months after ablation. The definition of successful ablation of the thyroid remnant also differed between studies, many of which were based on a visual inspection and interpretation of the follow-up scan rather than using quantitative measurements of radioiodine uptake in the neck or whole-body scan. In total, 1436 patients were given an amount of activity of approximately 30 mCi, of whom 927 were successfully ablated. An activity of approximately 100 mCi was administered to 4573, of whom 3440 were successfully ablated.
Table 1
is based on studies in which patients were reported to have undergone a total or near-total thyroidectomy before remnant ablation. In all but four studies, all patients did not take thyroid hormone replacement before ablation. In those four studies, some of the patients discontinued thyroid hormone replacement, whereas others had rhTSH before ablation (15, 29, 33), or all patients had both rhTSH and discontinued hormone replacement (21). Eleven studies reported that patients were recommended to go on a low-iodine diet before ablation (14, 15, 16, 17, 18, 25, 28, 30, 32, 35, 37).
The rates of successful ablation were clearly variable, but they were consistently lower in the low activity group. In the 22 studies (26 estimates) using a high activity of radioiodine (
100 mCi) the success rates ranged from 50 to 100%. Approximately half (14 of 26) reported an ablation success rate of 80% or more, but the rate was 90% or more in five of the 26 estimates, indicating that one in five studies achieved a very high successful ablation rate. In the nine studies using a low activity (30 mCi), the success rates ranged from 10% to 80%, and no study reported a rate greater than 80%. However, all but two of the nine estimates from the low activity studies were based on fewer than 50 patients, and the success rate in the two largest studies was approximately 70%. The pooled overall success rate associated with using a high activity (
100 mCi) is 79%, greater than the estimate of 51% using 30 mCi, although these figures should be interpreted with caution because of the methodological differences between studies. Excluding studies in which some patients had distant metastatic disease did not materially change the combined estimates of success rates.
Table 2
is similar to Table 1
but shows studies in which not all patients had a total or near total thyroidectomy (or the extent of surgery was not reported). In all studies, patients discontinued thyroid hormone replacement before ablation. Six studies reported that patients were recommended to go on a low-iodine diet (39, 42, 46, 52, 53, 60). Ignoring studies in which the extent of thyroidectomy was not reported, a high administered activity of radioiodine was associated with rates of successful ablation ranging from 28% to 96%, with 44% of estimates (8 of 18) showing a success rate of 80% or more. Similar to the results associated with a high activity from Table 1
, approximately one in four (5 of 18) estimates had a successful ablation rate of 90% or more. Among the low activity studies, 33% (5 of 15) had a rate at least 80%. The combined overall success rate for a high activity was 79% (the same as in Table 1
), again greater than the low activity group (69%). As before, excluding studies in which some patients had distant metastatic disease did not materially change the combined estimates of success rates.
Figure 1
shows the 16 observational studies from Tables 1
and 2
that allow a direct comparison of outcome from low- and high-administered activities of radioiodine in the same patient group. A test for heterogeneity was almost statistically significant (P = 0.051), indicating that there were differences between the results. From a meta-analysis of these results, the pooled risk ratio of having a successful ablation is 0.90 (95% CI, 0.830.97, P = 0.01), indicating that the rate of successful ablation is 10% lower (95% CI, 317% lower) when using an administered activity of approximately 30 mCi compared with approximately 100 mCi. This is a moderate effect. However, this result could in part be caused by inclusion of studies in which the high-activity group included some patients who received an activity that was not much greater than 30 mCi [for example Maxon et al. (42) and Lin et al. (46)]. Including such patients may reduce the difference in successful ablation rates between the high and low activities. Excluding studies in which the high activity group included those given less than 75 mCi (37, 40, 41, 42, 45, 46, 59) resulted in a pooled risk ratio of 0.85 (95% CI, 0.760.97; P = 0.012), which represents a 15% reduction in the ablation success rate (95% CI, 324% reduction), again lending support to a beneficial outcome for a high compared with low administered activity of radioiodine.
|
|
Table 3
shows selected characteristics of the six published randomized trials. TSH was stimulated by thyroid hormone withdrawal in four trials (65, 67, 68, 69), in one trial 59% of the patients had thyroid hormone withdrawal and 41% had rhTSH (66), and the method of TSH stimulation was not reported in one trial (70). Two trial reports stated that they did not recommend a low-iodine diet (66, 69); it was not known from the other papers whether patients in those trials were recommended a low-iodine diet or not. The diagnostic scan was usually performed 6 or more months after attempted ablation of the thyroid remnant. Similar to the patient case note review and prospective cohort studies, the success rates for specified ablation activities differed between trials.
Two of the six trials examined the success of ablation in relation to several administered activities, and these are summarized in Table 4
. One report by Bal et al. (67) suggested that an activity of 50 mCi (1.8 GBq) could achieve a high success rate, but a later study by the same group (69) suggested that an activity as low as 25 mCi (0.9 GBq) had a similarly high effect. Although both trials were large, particularly the later trial (69), the number of patients in each activity group was relatively small. This means that the 95% CIs for the true successful ablation rates for each activity are wide, so it is not possible to tell whether a low activity has a moderate or high success rate. For example, the 95% CI using 40 mCi (1.5 GBq) means that the true ablation rate could be anywhere between 68% and 89%.
|
|
Extent of reported surgery before 131I ablation.
There was no clear indication from the observational studies in Tables 1
and 2
whether the reported extent of thyroidectomy influences ablation success rates. The combined success rates in patients given around 100 mCi was 79% when total/near total thyroidectomies were reported (Table 1
) and when they were given a mixture of total, near total, subtotal, and partial (Table 2
). Restricting the analysis to the high activities in Table 2
in which no patients were reported to have had a total/near total thyroidectomy (10, 16, 51, 52, 53, 54, 57, 58) produced a combined success rate of 78%. There was no conclusive evidence that a low activity is associated with a high success rate when patients have had a reported total or near-total thyroidectomy. Surprisingly, the combined success rate associated with 30 mCi tended to be higher among the group of patients in which the preablation surgical method was not always a total or near-total thyroidectomy (54% in Table 1
compared with 69% in Table 2
), although this could in part be due to methodological differences between the studies. Furthermore, in studies in which the patients had a mixture of total/near total and subtotal/partial it is not possible to determine the proportion of patients in each category, and in most studies in Table 1
only a relatively small number of patients received 30 mCi, so no clear conclusions can come from this comparison. Among the observational studies that allowed a comparison of two activities, only two (10, 22) reported that all patients had a total or near-total thyroidectomy; so, no useful information could come from this. In the randomized trials, two (67, 69) reported no evidence of a difference in successful ablation rates between total and subtotal thyroidectomy. In all the studies in Tables 13![]()
![]()
, it is possible that the extent of surgery reported by the clinician may not correlate well with the actual remnant size in some cases.
Postsurgical remnant size.
Several studies reported that the success of thyroid remnant ablation decreased with increasing postsurgical remnant size (16, 17, 18, 24, 34, 39, 70). There were two studies (one a retrospective case note review and the other a randomized trial) that compared ablation success between two administered activities according to neck uptake, as determined by a preablation scan (17, 70). Table 5
shows these results and provides the difference in ablation success rates (low activity minus high activity) in groups defined by neck uptake. The differences get larger as neck uptake increases. These data suggest that a low activity may have similar successful ablation rates to a high activity in patients with small remnants, but the rate may be much lower in patients with large remnants.
|
Use of rhTSH.
Although the effects of rhTSH on quality of life have been reported (33, 71, 72), only a few small studies have directly compared rates of successful ablation of the thyroid remnant between rhTSH and thyroid hormone withdrawal. Table 6
summarizes the results. rhTSH was given by injection on 2 consecutive days before ablation, 131I therapy was then administered on the third day in three studies (15, 33, 73) or 48 h after the last injection in one study (29). Three studies showed the rate of success in terms of remnant ablation to be similar, but one study (29) reported a lower rate among patients given rhTSH. A possible explanation for this was that ablation in the rhTSH group was not given until 48 h after the last injection of rhTSH, later than the recommended 24 h.
|
| Discussion |
|---|
|
|
|---|
This systematic review of the published literature comparing outcome in terms of ablation of the thyroid remnant after thyroid surgery highlights the relative paucity of robust evidence regarding the lowest activity of radioiodine that can be administered to achieve successful ablation. It is therefore understandable that authors of guidelines find it difficult to be clearer on which radioiodine activity to use.
Most studies identified were based on retrospective reviews of patient case notes and there were few randomized trials. Perhaps the most important methodological difference between the studies is the definition of successful ablation. Many studies used a visual inspection of the follow-up scan, others used a cut-off level associated with a quantitative measurement of neck uptake, and some studies used thyroglobulin measurements in addition to the scan result. Other study differences were due to the extent of surgery, method of TSH stimulation, whether an iodine-free diet was recommended or not, the use of an 131I preablation scan, and the length of time between ablation and the follow-up scan. The lack of blinding associated with an administered activity may also have an effect on whether patients are recorded as having a successful ablation. All of these factors, along with sample size, would have contributed to the variability in the reported rates of ablation success. It is, however, difficult to precisely quantify the effect of these factors on the success rates, either on their own or when considered in combination, using the studies identified in this review. This is largely because so few studies were conducted in the same way, with regards all the important factors. By focusing only on studies that compared two administered activities (Fig. 1
), it is possible to partly overcome some of these methodological differences because the same methods would be applied to all patients within a particular study.
The data from the studies based on retrospective case note reviews or prospective cohorts confirm the belief that a radioiodine activity of approximately 100 mCi (3.7 GBq) is associated with a high rate of successful ablation: it is approximately 80%, although the rate was 90% or more in one in four studies (11 of 49 estimates from Tables 1
and 2
).
The evidence in support of using a lower activity of 30 mCi (1.1 GBq) is less clear. A difficulty in data interpretation is that the estimates of success rate were found to be more variable: the SD was 26% based on 30 mCi, compared with 17% among estimates based on approximately 100 mCi. This is partly due to the fact that most low activity studies were small: only 33% (nine of 27) of studies using 30 mCi were based on 50 patients or more compared with 57% (28 of 49) of the high activity studies. Small studies are more likely to be affected by chance variation. In the observational studies, although the pooled estimates for a specific amount of activity suggested that 30 mCi had a lower ablation success rate than approximately 100 mCi (51 vs. 79% in Table 1
and 69 vs. 79% in Table 2
), a meta-analysis of the risk ratios indicated only a modest effect (a 10% reduction). It is therefore not possible to adequately judge from these data whether the true success rate is similar to or significantly lower than that associated with using 100 mCi.
Furthermore, differences between patients in the low- and high-activity groups make it difficult to determine whether a low-administered radioiodine activity is better or similar because of potential selection bias (outside the context of randomized trials) in that clinicians may choose the activity that is judged to be the most appropriate depending on the clinical or other characteristics of the patient. There may also be other factors that differ between patients given a low or high activity that affect the rates of successful ablation (i.e. confounders). Therefore, comparing success rates between low and high activities in these case note review and prospective cohort studies will be affected by both bias and confounding.
Randomized trials are the best way of comparing different ablation activities because, by design, the process of randomization minimizes the influence of the effect of bias and confounding, allowing a fair comparison of success rates between different radioiodine activities. However, the results from the meta-analysis based on the published randomized trials are equivocal. Only a few have been conducted and the comparisons of different activities are based on relatively small numbers of patients. None of the trials has sufficient statistical power to detect a difference of 10 percentage points or more. An equivalence trial of 468 patients would be required to compare a successful ablation rate of 70% (30 mCi) with 80% (100 mCi), and if the difference is less than 10 percentage points, to conclude that the two activities have a similar effect (with 85% power) (74). This is a group size larger than the published trials combined. The results of the meta-analysis show that, at present, it is not possible to determine the lowest effective activity for radioiodine ablation of the thyroid remnant after primary surgical treatment of thyroid cancer. There is, therefore, a need for further appropriately powered randomized trials with ablation success and recurrence rates as outcome measures (75).
There are currently two large trials being planned, one in the United Kingdom and the other in France. The UK trial (called HiLo and designed to recruit 468 patients) will be a factorial study comparing 30 with 100 mCi, with patients also randomized to receive rhTSH or thyroid hormone withdrawal before ablation. The French trial, based on 700 patients (personal communication, Professor Martin Schlumberger) will have a similar design (but 95% power). Both trials will determine whether a low activity is as effective as the established 100 mCi or whether it is significantly less effective and will determine whether the success of ablation is affected by using rhTSH to stimulate radioiodine uptake compared with thyroid hormone withdrawal. They will also examine side effects, long-term outcome, and undertake an economic evaluation.
In the absence of a large randomized trial, the current evidence base for guiding clinical practice, in the context of thyroid remnant ablation, is not conclusive. The published data cannot be used to reliably determine whether an administered activity as low as 30 mCi could be used instead of 100 mCi. Activities of 30 mCi should be used with caution.
| Footnotes |
|---|
First Published Online October 10, 2006
Abbreviations: CI, Confidence interval; rh, recombinant human.
Received June 21, 2006.
Accepted September 29, 2006.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Hackshaw and U. Mallick Low versus High I-131 Dose for Remnant Ablation in Differentiated Thyroid Cancer Clin. Med. Res., June 1, 2009; 7(1-2): 1 - 1. [Full Text] [PDF] |
||||
![]() |
S. A. R. Doi and A. Onitilo Author Reply Clin. Med. Res., June 1, 2009; 7(1-2): 2 - 3. [Full Text] [PDF] |
||||
![]() |
D. Focosi, S. Galimberti, and M. Petrini Acute myeloid leukemia and follicular lymphoma after very low dose radioiodine therapy for thyroid diseases Haematologica, September 1, 2007; 92(9): e96 - e97. [Full Text] [PDF] |
||||
![]() |
S. A.R. Doi, N. J. Woodhouse, L. Thalib, and A. Onitilo Ablation of the Thyroid Remnant and I-131 Dose in Differentiated Thyroid Cancer: A Meta-Analysis Revisited Clin. Med. Res., June 1, 2007; 5(2): 87 - 90. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |