The Center for Health Outcome Policy and Evaluation Studies (E.L.M.) and Divisions of Endocrinology, Diabetes and Metabolism and Nuclear Medicine, The Ohio State University, Columbus, Ohio 43210-1228
Address correspondence and requests for reprints to: Ernest L. Mazzaferri, M.D., MACP, 4020 SW 93rd Drive, Gainesville, Florida 32608-4653. E-mail: mazz01{at}bellsouth.net
![]() |
Introduction |
---|
![]() |
Background |
---|
The annual incidence of thyroid cancer has risen nearly 50% since 1973, afflicting some 18,000 people and causing about 1,200 deaths in the United States in 1999 (1). Ten-year relative survival rates in a large cohort of United States patients with papillary and follicular cancer were, respectively, 93% and 85% (2). In 1,528 of our patients who were, as a group, slightly younger than those in the aforementioned study, 40-yr relative survival rates for papillary and follicular cancer, respectively, were 94% and 84% (P = 0.00011; Ref. 3). Patients with follicular cancer tend to be older with more advanced tumor stage at the time of diagnosis than those with papillary cancer. About half who die from thyroid cancer suffer a pulmonary death, either respiratory failure from pulmonary metastases or suffocation from airway compromise (5).
Recurrence rates
In our cohort, 40-yr recurrence rates are about 35%, two thirds
of which occurred within the first decade after initial therapy (Fig. 1A). These rates, including those for
distant recurrence, are higher under age 20 and over 60
yr2 (Fig. 1B
). Thirty-year cancer mortality
rates were about 12% with local recurrence and 43% with distant
recurrence (P = 0.001). Local disease comprised 68% of
the recurrences in our study; among the 170 patients in whom the exact
site of local recurrence was known, the 30-yr cancer mortality rate was
twice as high with recurrence in the neck soft tissues (30%) compared
with those in cervical lymph nodes or the contralateral thyroid (16%,
P < 0.05). Distant metastases, mostly to the lungs,
comprised 32% of the recurrences; after 40 yr of follow-up half have
died of cancer.
|
It is important to briefly summarize risk factors for tumor
recurrence and mortality because it is here that most clinicians
disagree, recommending therapy and follow-up according to their view of
risk. Certain prognostic factors indicate how, on average, DTC will
proceed (Table 1; Ref. 6).
The first set are patient characteristics, age at diagnosis, gender,
and family history (3). Mortality rates are low among
patients under age 40 yr but rise incrementally thereafter (Fig. 1
);
however, recurrence rates are especially high (
40%) during the
first two decades of life and after age 60 yr (Fig. 1
)
(3). Men develop DTC with about half the frequency of
women but have about twice the risk of dying from it (Table 2
; Refs. 1 and
3).
|
|
Staging systems
Staging systems usually correctly stratify mortality rates among
subsets of patients. Still, some in the lowest risk groups die of
thyroid cancer (2, 8), particularly when risk is simply
defined as being low or high (9, 10). In a large study
(5), over 10% of the patients dying of DTC had an
American Joint Commission on Cancer staging classification
(TNM3) of 1 or 2 (9).
Systems that use age to stratify risk tend to be inaccurate in
predicting recurrence-free survival, mainly because young patients have
high recurrence rates (Fig. 1B). Most staging systems have been derived
from multivariate analyses that do not consider recurrence or the
effect of therapy, and all rely on information that is often available
only after surgery. Disease-free status and survival cannot be assured
by low stage in most systems, thus providing imperfect guidance in
selecting therapy. Staging systems are best used in epidemiologic
studies. The TNM system uses age to stratify risk but many clinicians
do not use age to influence decisions about therapy
(6).
Contemporary views concerning therapy
Several studies from the United States and Europe indicate that most patients at risk for relapse or death from thyroid cancer are treated with total or near-total thyroidectomy, usually followed by 131I and levothyroxine (L-thyroxine) therapy (6, 11, 12, 13, 14, 15, 16). External radiation, has a less prominent role in initial management and usually follows 131I therapy, but may be given postoperatively to patients over the age of 40 yr with TNM stage T4 tumors, especially those with papillary thyroid cancer (6, 17, 18, 19) and unresectable gross residual disease (6, 19).
National Comprehensive Cancer Network (NCCN) guidelines
These are the most recent and explicit guidelines for the diagnosis and management of thyroid cancer (6). In lieu of prospective randomized trials, a panel of multidisciplinary experts from the 17 NCCN member institutions convened in 19981999 to discuss diagnosis and treatment approaches and to establish practice guidelines for the management of thyroid nodules and thyroid cancer. Version 1.2000 of the guidelines, which was published in November 1999 (6), will be reviewed and updated annually.
Delayed diagnosis and treatment
Prompt diagnosis has a bearing on outcome. Although fine-needle
aspiration was not done preoperatively in about 40% of 5584
thyroid surgical cases in one study (13), North American
endocrinologists use it routinely to evaluate thyroid nodules,
performing it early and relying heavily on its results, thus hastening
the decision-making process (20). Not only are nodule
evaluations by primary care physicians more costly and more time
consuming than those performed by endocrinologists, but also the
patient is more likely to undergo unnecessary surgery
(21). A delay of 12 months or longer was found in almost
30% of the patients managed by primary care physicians in one study
(21). This delay increases mortality rates significantly,
worsening as the delay becomes longer and imparting a risk comparable
with that of advanced age (Table 2; Ref. 3).
![]() |
Initial Surgical Management |
---|
This is perhaps the most widely debated issue in the management of thyroid cancer. Many surgeons perform a procedure called near-total thyroidectomy, but its definition is open-ended, leaving much doubt as to the actual extent of surgery and the amount of thyroid tissue left behind. For this reason, the NCCN guidelines specifically avoid this term (6). In practice, many patients have substantial thyroid remnants when evaluated by whole body scan (WBS; 22) and thyroglobulin (Tg) determinations even after reportedly undergoing total thyroidectomy. Thyroid ultrasound may be useful when the extent of surgery is in question, because leaving a thyroid remnant smaller than 2 g facilitates postoperative 131I ablation (23).
Focusing the debate
The main disagreement centers on the extent of surgery that
is optimal for tumors from 14 cm in diameter (T2) without metastases.
The dispute stems from staging systems that use patient age without
regard for its differing effects on cancer recurrence and mortality
(Fig. 1B). Young patients, who have low mortality rates but high
recurrence rates are, thus, judged to be low risk by TNM or Age,
Metastases, Extent, Size (AMES) classifications, which some use
to justify unilateral lobectomy for the majority of patients
(24). Advocates of this approach emphasize the high
complication rates with total thyroidectomy (25, 26).
Nonetheless, analysis of surgical procedures performed in over 1500
United States hospitals reveals that among 5584 patients with thyroid
cancer the majority (77.4%) underwent total thyroidectomy regardless
of tumor histology and stage (13). There are compelling
reasons to do this.
Recurrence rates with lobectomy
Performing lobectomy alone may result in a 510%
recurrence rate in the opposite thyroid lobe (27, 28), a
high tumor recurrence rate (Fig. 2), and
a high (11%) incidence of subsequent pulmonary metastases
(29). High recurrence rates in patients with cervical
lymph node metastases and multicentric tumors (3) also
justify bilateral thyroidectomy and 131I
ablation. If lobectomy is performed, microscopic metastases in the
contralateral lobe or found by 131I imaging may
be missed when they are most treatable. Surgery more than lobectomy was
an independent variable affecting cancer recurrence in our study (Table 2
). DeGroot et al. (8) reported a similar
beneficial effect of more extensive surgery on the rate of tumor
recurrence.
|
It is more difficult to demonstrate that the extent of surgery
influences survival with DTC, although there is considerable proof that
it influences disease-free survival. For example, Hay et al.
(28) reported that patients treated for low-risk papillary
cancers [Age, Grade, Extent, Size (AGES) score 3.99] had no
improvement in survival rates after undergoing more than lobectomy.
Later, they reported the results of a study designed to compare
outcomes after unilateral or bilateral lobectomy for papillary cancer
considered to be low risk by AMES criteria (30).
Although there were no significant differences in cancer-specific
mortality or distant metastasis rates between the two groups, the 20-yr
rates for local recurrence and nodal metastasis after unilateral
lobectomy were 14% and 19%, respectively, significantly higher
(P = 0.0001) than the 2% and 6% rates, respectively,
seen after bilateral thyroid resection. As a result, Hay
et al. (30) concluded that bilateral thyroid
resection is the preferable initial surgical approach to patients with
low-risk papillary cancer. DeGroot et al. (8)
demonstrated a reduction in cancer mortality by more extensive surgery
in a group of patients without distant metastases; however, when T4
tumors were excluded, a benefit on survival could not be demonstrated.
Postoperative treatment with 131I also
obscures the therapeutic influence of surgery (8).
Patients in our study treated with total or near-total thyroidectomy
plus 131I ablation and
L-thyroxine had significantly fewer recurrences
and distant recurrences than those treated with any other combination,
including total thyroidectomy and L-thyroxine
(Fig. 2
). Surgery and 131I therapy had
independent effects on recurrence and cancer mortality (Table 2
). After
a median follow-up of 16.6 yr, surgery more extensive than lobectomy
was an independent variable that reduced the likelihood of cancer death
by 50% (Table 2
).
The NCCN guidelines on thyroid and lymph node surgery
The NCCN guidelines recommend total thyroidectomy and, if lymph
nodes are involved, bilateral central compartment dissection or lateral
modified radical neck dissection as the primary treatment for high-risk
DTC (Table 1; Ref. 6). Cervical lymph node metastases, the
most common site of papillary cancer metastases, are found in 5080%
of cases, most often in the central compartment (paratracheal),
followed in descending order by mid-jugular, supraclavicular, and
subdigastric nodes (31). We found that lymph node
metastases, especially bilateral cervical and mediastinal, were an
independent variable that affected recurrence and survival (Table 2
). Although not all find this to be true, others report that
systematic compartment-oriented dissection of lymph node metastases
significantly improves recurrence (P < 0.0001) and
survival (P < 0.005) rates in patients with T1T3
tumors (32). There was disagreement among NCCN panelists
about the initial surgery and the indications for completion
thyroidectomy for patients at moderate or low risk of cancer mortality
(T2, N0; Ref. 6).
Other consensus recommendations
Most United States and European clinicians opt for total or
near-total thyroidectomy when the diagnosis is known (6, 8, 11, 13, 14, 15, 16). This also applies to children and young adults because
6080% have regional lymph node involvement and 1020% have distant
metastases (30, 33, 34, 35, 36, 37). Of 50 children aged 15 or younger
in our study, 28% developed distant metastases, usually to the lung;
eight percent were found on initial presentation, and 20% were
recurrences (Fig. 1B). Although the overall long-term survival rate in
children is greater than 90%, progression-free survival is enhanced by
complete tumor resection (32, 35).
Surgery for highly invasive tumors
There is disagreement about the optimal surgery for DTC with laryngotracheal invasion (38). Shaving tumor off laryngotracheal structures or the recurrent laryngeal nerve is generally recommended to reduce morbidity (32). However, more aggressive surgery is unlikely to cure highly invasive DTC but may help maintain the integrity of the airway (38). Macroscopic residual disease has an unfavorable outcome, even in young patients (35).
Surgery for papillary microcarcinoma and minimally invasive follicular cancer
Lobectomy alone is adequate surgery for papillary microcarcinoma (<1 cm) discovered after surgery for benign thyroid conditions, provided the patient has not been exposed to radiation and the tumor is unifocal and confined to the thyroid without vascular invasion or other histological features suggesting a poor prognosis (3, 39, 40). The same is true for small (<4 cm) minimally invasive follicular cancers without vascular invasion that cannot be identified as malignant by fine-needle aspiration or frozen section (41). Because a large thyroid remnant hampers long-term follow-up with Tg and WBS, the NCCN guidelines recommend discussing the option of completion thyroidectomy with the patient but do not recommend it in this situation (6).
Completion thyroidectomy
The contralateral lobe should be resected in all patients with tumors that have the potential for recurrence (23). Ablating a large remnant with 131I is not recommended because it may cause radiation thyroiditis with serious pain and swelling, sometimes with thyrotoxicosis. A large remnant may suppress high TSH levels necessary for tumor 131I uptake (42) and often cannot be completely ablated (23, 43).
The rationale for completion thyroidectomy
This procedure has a low complication rate when performed by an experienced surgeon, may help unmask hidden metastases, and may enhance survival. Cancer is found in the contralateral lobe in about half of the cases (44, 45). Unrecognized lung or lymph node metastases that could only be identified after completion thyroidectomy were reported in more than 60% of a group of irradiated children from Chernobyl (36). In another study, multivariate analysis found that patients who underwent completion thyroidectomy within 6 months of their primary operation developed significantly fewer lymph node and hematogenous recurrences and survived significantly longer than those in whom the second operation was delayed for longer than 6 months (46). Completion thyroidectomy should be performed for any patient who has undergone lobectomy for tumors larger than stage T1 (>1 cm) or has metastases, recurrent cancer, or tumor in the resection margins (6).
Surgical complications
Hypoparathyroidism and recurrent laryngeal nerve damage occur most commonly after total thyroidectomy and cervical lymph node dissection (13). The rates of hypoparathyroidism immediately after total thyroidectomy are as high as 10% in adults (13) and may be twice this high in children (47, 36, 37). The rates of persistent hoarseness and hypoparathyroidism are much lower. A review of seven surgical series of adults found the average rates of permanent laryngeal nerve injury and hypoparathyroidism, respectively, were 3% and 2.6% after total thyroidectomy and 1.9% and 0.2% after subtotal thyroidectomy (48). One study found hypocalcemia in 5.4% of patients immediately after total thyroidectomy that persisted a year later in only 0.5% (49). In children, persistent hypoparathyroidism is more common and may be over 10% (50). A study of 5860 patients treated in the state of Maryland found that surgeons who performed more than 100 thyroidectomies a year had the lowest overall complication rates (4.3%); complications were 4-fold higher for those who performed less than 10 cases annually (51). Others also find that complication rates are affected by the surgeons experience (52).
![]() |
Radioiodine Ablation of Residual Normal Thyroid Tissue |
---|
Rationale for thyroid remnant ablation
Although debate continues about 131I remnant
ablation after near-total thyroidectomy (53, 54), there
are several compelling reasons to do it. First, a large remnant can
obscure 131I uptake in cervical or lung
metastases (36, 55). Second, high TSH levels necessary to
enhance tumor 131I uptake cannot be achieved with
a large thyroid remnant (42); in fact, large remnants
should undergo surgery. Third, Tg measurement made under TSH
stimulation, which is the most sensitive test for cancer when there is
no normal thyroid tissue present, usually requires ablation of thyroid
bed uptake (56). Fourth, lung metastases may be seen only
on the posttreatment WBS after remnant ablation (57, 58).
Lastly, remnant ablation may destroy residual normal follicular cells
destined to become malignant (59) and occult cancer that
might recur years later (Figs. 2 and 3
).
|
Although their prognostic importance is questioned, multifocal papillary thyroid cancers are common (3). Long considered intraglandular metastases, not individual tumors arising de novo, recent evidence suggests the opposite. A study found that only 2 of 17 patients had multifocal tumors with identical RET/PTC rearrangements, whereas the other 15 patients had diverse RET/PTC rearrangements within their multifocal tumors (59). This suggests that individual tumors arise independently in a background of genetic or environmental susceptibility, perhaps explaining contralateral lobe recurrences years after initial therapy.
Indications for thyroid 131I remnant ablation
This decision is tightly linked to that for performing total or near-total thyroidectomy. Remnant ablation should be done when the patient has uptake in the thyroid bed and no known foci of cancer after resection of a tumor that has the potential for recurrence (54). Six to 12 months after 131I ablation, if uptake in the thyroid bed is less than 0.5% at 48 h, a second ablation is unlikely to be of further benefit because such a small amount of uptake is unlikely to represent residual cancer if the Tg is low and is unlikely to be the sole source of a high serum Tg (>10 ng/mL) (60). As a practical matter, most patients who have undergone total or near-total thyroidectomy have thyroid bed uptake that requires ablation. Although some advise a more selective approach for 131I ablation based on tumor stage (61), the same authors report that 38% of the patients who had undergone ablation in their clinic had low-risk tumors (T1 or T2 and N0; Ref. 11). Once L-thyroxine has been withdrawn and the patient has followed a low-iodine diet for imaging, remnant ablation can be done on the same day as WBS, often as an outpatient (3, 8, 18, 62).
Therapeutic impact of 131I remnant ablation
Although many (54) report lower recurrence rates after 131I ablation, sometimes with reduced cancer mortality rates, not all find this, perhaps because more extensive thyroidectomy had been done (53). A 25-yr prospective study found that none of 44 patients in whom total 131I tumor ablation was achieved and maintained died of cancer; however, 70% died of cancer when this was not possible (63). Another study found that remnant ablation decreased recurrence of tumors larger than 1 cm, including those predicted to have a good prognosis (patients with class I or II disease); however, it reduced the risk of death only in patients with more advanced disease (class III or IV; Ref. 64). In another study, the rates of pulmonary metastases among 58 patients with DTC were 11% after partial thyroidectomy, 5% after subtotal thyroidectomy and 131I, 3% after total thyroidectomy, and only 1.3% after total thyroidectomy and 131I (29). Among 321 patients treated in 13 Canadian hospitals with 131I, mainly to ablate residual normal thyroid tissue in those with microscopic residual papillary or follicular cancer, local disease was controlled more often with either postoperative external radiotherapy or 131I therapy, or both together, than with L-thyroxine alone (P < 0.001; Ref. 19). Survival at 20 yr of patients with microscopic residual disease treated by surgery alone was less favorable (about 40%) than after treatment with either 131I or external radiation (about 90%, P < 0.01), whereas 131I treatment without obvious residual disease did not increase survival (19). In a later study from Canada of 382 patients with DTC, thyroid ablation with total thyroidectomy and 131I was associated with a significantly lower rate of local relapse that was independent of tumor stage (18).
Among our patients with tumors larger than 1.5 cm, cancer recurrence,
distant recurrence, and cancer death rates (the latter for patients
>40 yr) were significantly lower after remnant ablation than with
L-thyroxine alone or no medical therapy (54).
In the latest analysis, 230 patients had undergone remnant ablation,
789 had been treated with only L-thyroxine, and 163 had
received no medial therapy; their median follow-up, respectively, was
14.7, 20.8, and 21.2 yr. During L-thyroxine therapy alone,
the recurrence rate was 4-fold (Fig. 3A, P < 0.0001)
and the rate of distant recurrence was 5-fold (Fig. 3B
, P < 0.02) that of thyroid ablation. Among patients
over age 40 yr with tumors 1.5 cm or larger, there were fewer cancer
deaths 40 yr after thyroid remnant ablation than after the other
treatment strategies (Table 3
,
P < 0.0001). Based on regression modeling of 1510
patients without distant metastases at the time of initial therapy,
remnant ablation was an independent variable that reduced cancer
recurrence, distant recurrences, and cancer death (Table 2
). The
131I ablation treatments were stratified into two
groups: 62 (46%) given 2950 mCi4
(42.7 ± 1.2 SE) and 72 (54%) given 51 to
200 mCi (114.6 ± 4.0); both groups had similar 30-yr recurrence
rates (4% and 6%, respectively, P = 0.1).
|
Many use 30 mCi to ablate a remnant if the amount of thyroid tissue remaining after surgery is small. This has been a popular way to avoid hospitalization, which is no longer necessary in most states because of the 1997 change in federal regulations that permits the use of much larger 131I doses in ambulatory patients (65). A study found that radiation exposures to household members of patients given more than 30 mCi were well below the limit (5.0 mSv) mandated by United States Nuclear Regulatory Commission regulations (66). Small 131I doses have appeal because of the lower cost and the lower whole-body radiation dose, which has been estimated to be 6.1 roentgen-equivalent-man (rem) for 30 mCi, 8.5 rem for a 50 mCi, and 12.2 rem for a 60 mCi (67). Moreover, 100 mCi 131I may cause salivary injury and testicular damage.
The arguments for high-dose 131I remnant ablation
Some prefer larger 131I doses to ablate thyroid tissue and to treat residual microscopic cancer (68). A meta-analysis found that a single administration of about 30 mCi failed to fully ablate the remnant (46%) more often than did 77100 mCi (27%, P < 0.001; Ref. 69). There was, however, a wide range of failures among the low-dose cases due to the definition of ablation and variation in the extent to surgery. Both high and low activities were most likely to completely ablate the remnant after near-total thyroidectomy (69).
The arguments for low-dose 131I remnant ablation
Thyroid bed uptake can be ablated in up to 80% of patients given 3050 mCi, providing the surgeon has left a small remnant and ablation is defined by a diagnostic WBS using 23 mCi 131I (70, 71). In a randomized study of this question, the first dose ablated thyroid bed uptake in 81% of patients given 30 mCi and in 84% treated with 100 mCi (72). Another randomized study (70) that administered fixed amounts of 131I ranging from 25200 mCi found that increasing the empirical dose to more than 50 mCi resulted in a plateau of the dose-response curve; complete ablation occurred in 63% of the 30-mCi group, 78% of the 50-mCi group, 74% of the 90-mCi group, and 77% of the 155-mCi group. Another study found a dose averaging 87 mCi and ranging from 26246 mCi that delivered at least 30,000 rad (300 Gy) was successful in ablating uptake after a single initial 131I administration in 84% of inpatients and in 79% of outpatients; administered activities low enough to permit outpatient therapy (i.e. <30 mCi) were used in 47% of the patients (23). Amounts of 131I that deliver more than 30,000 rad (300 Gy) do not result in a higher ablation rate (70). Lower success rates are found when large pretreatment scanning doses are used, regardless of the therapeutic dose of 131I and are attributed to thyroid stunning (73).
![]() |
Diagnostic 131I WBSs |
---|
Thyroid stunning
Administering more than 2 mCi 131I may have a sufficiently harmful effect on the tissue in which it concentrates to interfere with subsequent uptake of 131I for several weeks (74, 75, 76). Using 2 or 3 mCi 131I or 500 µCi 123I may avoid stunning but is less sensitive than larger 131I doses in identifying remnants or metastases (75, 77). Using more 123I may improve WBS images, but the cost is great. Delaying 131I therapy may be responsible for the stunning effect (75, 76), which did not occur in 172 patients treated with 131I within 72 h of having received 5 mCi for a diagnostic scan (22). Although a WBS is usually done postoperatively to help determine the optimal 131I dose to ablate residual thyroid tissue or cancer, another approach is to perform a posttreatment scan after administering 131I given on the basis of high postoperative serum Tg levels.
Diagnostic dose of 131I for whole body scanning
Ideally, scans are done with quantitative radiation dose estimates at 24, 48, and 72 h after the oral administration of 2 mCi 131I, but this is time consuming and of little benefit (23). A metastatic lesion concentrating 0.02% of this dose per gram will contain 0.4 µCi 131I per gram of tissue, which can be detected using current imaging techniques (23). Metastatic lesions amenable to 131I therapy are likely to be seen in the athyreotic patient using a 2-mCi dose of 131I (23). Nonetheless, some pulmonary metastases are visualized only after large therapeutic doses of 131I (57). Diagnostic scanning after preparation with recombinant human TSH (rhTSH) requires 4 mCi 131I, a dose that probably has about the same effect as a 2-mCi dose after L-thyroxine withdrawal due to greater excretion of 131I with rhTSH than during hypothyroidism, which impairs 131I renal excretion (78).
False positive 131I scans
A false positive scan that might lead to unnecessary
131I treatment may be caused by
131I in body secretions, pathologic transudates,
and areas of inflammation (79). It also can be caused by
physiologic secretion of 131I from the
nasopharynx, salivary and sweat glands, stomach, and genitourinary
tract, and from skin contaminated by urine, sputum, or tears
(80). Diffuse hepatic 131I uptake
(Fig. 4) as a result of
131I-labeled Tg is seen in 40% of 30-mCi
posttherapy scans and in 70% of 150200-mCi scans (81).
However, it may rarely represent liver metastases when there is no
uptake in the thyroid bed or metastases (81).
|
![]() |
Radioiodine (131I) Therapy for Residual Disease |
---|
Therapeutic efficacy of 131I is related to the capacity of a tumor to concentrate and retain iodine. Half to two thirds of metastases concentrate 131I, but even after meticulous preparation and large amounts of 131I the others may not concentrate or retain enough 131I to achieve a therapeutic benefit (82, 83, 84). This is more common after age 40 and with Hürthle cell cancers (84). Sodium-iodide symporter (hNIS) expression is low in some thyroid cancers, especially of high tumor stage (85, 86), and posttranscriptional events may cause hNIS dysfunction in others (87).
Low-iodine diet
A daily iodine diet of about 50 µg can raise thyroidal 131I uptake and can double the Gy per 100 mCi 131I administered (88); however, it may increase total body radiation as a result of delayed 131I clearance. The diet can be very tedious, but a daily iodine intake of 50 µg can be achieved by restricting the use of iodized salt, dairy products, eggs, and seafood (89). The diet should be started 2 weeks before 131I therapy and continued for several days thereafter. Diuretics can be used but are usually unnecessary (89). A free low-iodine cookbook is available from the Thyroid Cancer Survivors Association (90).
Efficacy of postoperative 131I treatment of residual cancer
Radioiodine therapy refers to the treatment of thyroid cancer
within the thyroid bed and in metastatic sites (91).
Surgery is the preferred treatment, but if it cannot be done then
131I is treatment of choice if the tumor
concentrates the isotope (6). In one large study of 1599
patients with DTC, 131I therapy was the single
most powerful factor accounting for disease-free survival
(92). Patients with low-risk tumors had significantly
fewer recurrences and deaths after 131I therapy
than those treated with L-thyroxine alone
(P < 0.001); however, 131I
conferred only a slight advantage to patients with high-risk tumors
(92). Based on regression modeling on 1510 patients
without distant metastases, we found that 131I
therapy of residual disease was an independent variable that favorably
reduced the likelihood of recurrence, distant recurrence, and death
from thyroid cancer (Table 2).
Approaches to radioiodine (131I) therapy
There are three approaches to 131I therapy: empiric fixed doses, upper bound limits that are set by blood and whole-body dosimetry, and quantitative tumor dosimetry (65).
Fixed 131I doses
This is the most widely used and simplest method. Its main advantages are simplicity and safety; its main disadvantage is that insufficient 131I may be administered to adequately treat tumor (23). Lymph node metastases too small to excise are treated with about 100175 mCi. Cancer growing through the thyroid capsule is treated with 150200 mCi, and patients with distant metastases are usually treated with 200 mCi, which will not induce severe radiation sickness or produce serious damage to critical structures (44, 65). Diffuse pulmonary metastases that intensely concentrate 131I are treated with an amount estimated to result in a whole body activity of less than 80 mCi after 48 h to avoid lung injury, which as a practical matter is about 200 mCi (91). Experience with 131I therapy for children is limited, but it seems to be effective in most with nodal or pulmonary metastases, with only a small increased risk of developing other cancers (91). Reynolds (93) has shown that the amount of 131I resulting in the same absorbed dose to a child as an adult given 1 mCi is linearly related to body weight and to body surface area. The fraction of an adult dose is 0.2, 0.4, 0.6, 0.8, and 1.0 for children with a body surface area5, respectively, of 0.4, 0.8, 1.2, 1.4, or 1.7 m2 or a body weight, respectively, of 10, 25, 40, 55, or 70 kg (91). Although it is generally held that to eradicate a tumor at least 0.1% of the 131I scanning dose must be concentrated by the tumor at 24 h, pulmonary metastases with negative diagnostic scans and high serum Tg concentrations can be treated effectively when uptake is much lower (57).
Quantitative tumor 131I dosimetry
A second approach is to use quantitative dosimetry to estimate tumor 131I uptake and retention, which some favor because radiation exposure from arbitrarily fixed doses of 131I can vary considerably from ineffective to excessive. The therapy dose is calculated to deliver an acceptable radiation dose to the lesion (usually nodal or discrete soft tissue metastases) without exceeding safety limits to the blood (bone marrow) and whole body. The data are collected one or more times daily over 7296 h as an outpatient (91). The cancer is unlikely to respond to 131I therapy if the tumor dose is less than 3,500 rad (35 Gy), in which case it should be considered for surgery, external radiation, or medical therapy (23, 65). In one study (23), lymph node metastases were treated successfully in 74% of patients with a single administration of 131I calculated to deliver at least 8,500 cGy (rad). For nodal metastases, success was achieved in 86% of patients at tumor doses of at least 14,000 cGy. Below 8,000 cGy, success starts to become questionable (23). Successful ablation of thyroid remnants occurred after a single initial administration of 131I in 84% of inpatients and in 79% of outpatients when treatment was standardized to a radiation dose of at least 30,000 cGy (23). It is necessary to estimate tumor or remnant size to make these calculations, which may be difficult, for example with diffuse microscopic lung metastases. In this case, the dose is based on the calculated safety limit to the bone marrow and whole body (91).
Blood and whole-body 131I dosimetry
A third method is to administer the largest safe dose of 131I based on dosimetric calculations. Pioneering work done by Benua et al. (94) reported serious complications when radiation to the blood exceeded 200 cGy, or more than 300 mCi 131I was administered, or whole-body retention was more than 150 mCi 131I 48 h after the treatment. The 300-mCi 131I limit was based on eight severe or fatal complications that occurred among 29 (28%) administrations of more than 300 mCi compared with six among 93 (6%) administrations of less than 300 mCi; however, just one of these complications occurred when the only procedural variation was that the administered activity exceeded 300 mCi. This occurred in a patient with skeletal metastases who was given 324 mCi that delivered 170 cGy to the blood in whom 48-h whole-body retention was 81 mCi. The currently accepted upper dose limit is calculated to deliver a maximum of 200 cGy to the whole blood while keeping the whole body retention less than 120 mCi at 48 h, or less than 80 mCi when there is diffuse pulmonary uptake (95). Severe complications from these single large amounts are infrequent (96), but not absent (97).
Lithium carbonate enhancement of 131I therapy
Given at a dose of 300 mg one to three times daily (10 mg/kg) starting about a week before 131I therapy, lithium increases uptake in metastatic lesions while only slightly increasing it in normal thyroid tissue. The drug enhances 131I retention probably as a result of its inhibitory effect on iodine release from both normal and neoplastic follicular cells (98). Radiation of tumors in which the biologic half-life of iodine is less than 6 days is maximized without increasing radiation to other organs, whereas the largest increase in tumor radiation occurs in lesions with a biological half-life of less than 3 days (98). Retention of 131I during lithium therapy increases 50% in tumors and 90% in thyroid remnants (98), the net effect being more than a 2-fold average increase in the radiation dose to metastatic tumor (98). Serum lithium levels should be measured daily and maintained between 0.8 and 1.2 nmol/L. The drug may be continued for 57 days after therapy, but lithium levels cannot be measured immediately after 131I therapy and one must carefully avoid lithium toxicity during this time.
Retinoic acid
A few patients with tumor that does not concentrate 131I may benefit from retinoic acid. The drug partly redifferentiates follicular thyroid cancer in vitro, but when it was given orally (1.18 + 0.37 mg/kg) for at least 2 months to 12 patients with DTC that could not be treated with other modalities, significant 131I uptake was induced in only 2 patients and a faint response was seen in 3 patients (99). A positive response was associated with a rise in serum Tg concentration, suggesting tumor redifferentiation.
Acute complications of 131I therapy
About two thirds of patients given 200 mCi or more develop mild radiation sickness characterized by headache, nausea, and vomiting that begins about 4 h after 131I administration and resolves within 24 h (65, 96).
Radiation-induced soft tissue reactions
The most important acute complication of 131I therapy is 131I-induced tumor edema or hemorrhage, which may occur rapidly and is most serious with tumor in the brain, spinal cord, or airway (100). Pretreatment with corticosteroids and mannitol may minimize this hazard (101), but patients must be hospitalized and closely observed. Surgical debulking of spinal lesions may be prudent before 131I is given, and surgery may be the treatment of choice for operable brain metastases (102). Pain in distant metastases can occur shortly after 131I therapy as a result of radiation-induced inflammation, which can also cause vocal cord paralysis when a large amount of functioning thyroid tissue is in close proximity to the vocal cords or recurrent laryngeal nerves (65, 101). Transient peripheral facial nerve palsy was reported in two patients after high-dose 131I therapy, presumably due to radiation of the nerve as it courses through the parotid area (103).
Radiation thyroiditis
About 20% of patients develop radiation thyroiditis when large thyroid remnants are treated with doses of 131I that deliver about 50,000 rad (500 Gy; Refs. 43 and 104). It usually appears within the first week after 131I administration and is recognized by neck and ear pain, painful swallowing, thyroid swelling and tenderness, and transient mild thyrotoxicosis. A large thyroid remnant may rarely swell enough to cause airway obstruction or can cause serious thyrotoxicosis. Mild pain can be treated with salicylate, a nonsteroidal anti-inflammatory drug, or acetaminophen, but severe pain or swelling requires corticosteroid therapy.
Radiation sialadenitis and tongue symptoms
Radiation sialadenitis involving the parotid or submandibular glands, which occurs after 131I therapy in up to 33% of patients, may be either acute or chronic (105). Symptoms may occur within 24 h of treatment and are more likely when large amounts of 131I have been given to a patient with little functioning thyroid tissue (105). Chewing gum, sucking on lemon candies, and hydration during the 131I treatment might prevent sialadenitis and xerostomia. Transient tongue pain or reduced taste also may occur (105). Intermittent tender salivary gland swelling may occur for some months due to temporary obstruction by a thick salivary plug that often occurs after eating. Decompression is often associated with a salty taste and usually occurs spontaneously. Despite these symptoms, invasive therapy is not required and it usually improves spontaneously within about a year; however, some develop chronic xerostomia. Although nearly half the patients eventually have reduced salivary gland function and some report recurrent conjunctivitis, these are almost never serious problems (105).
Acute bone marrow effects of 131I therapy
A slight reduction in platelets and white-cell counts may occur after 131I therapy but is transient and typically asymptomatic (96). More severe bone marrow suppression with anemia can follow very large doses of 131I, but this typically is reversible and does not require blood transfusions (96). Grave hematological depression is unlikely when the 131I dose delivers less than 200 cGy to the blood (94).
Late complications of 131I therapy
The serious long-term complications of 131I are damage to the gonads, bone marrow and lungs, and the induction of other cancers.
Ovarian damage
Transient amenorrhea and high serum gonadotropin concentrations develop in about 25% of perimenopausal women during the first year after 131I therapy (106). No permanent sterility is predicted in women cumulatively receiving up to 300 mCi, but it may occur in up to 60% of those receiving amounts of 131I in the range of 800 mCi (107). The miscarriage rate almost doubles during the year after surgery for thyroid cancer, both before and after 131I therapy, and doubles again after 131I therapy of more than 100 mCi. Whether this relates more to gonadal irradiation or to insufficient control of hormonal thyroid status is uncertain, but the fact that the rate of miscarriage after treatment with more than 100 mCi is twice that after treatment with less than 100 mCi suggests a role for irradiation in this phenomenon (108). In a long-term study, fertility was normal in 30 patients who were aged 30 yr or less when treated, after which they had 44 live births (109).
Testicular damage
The testis is even more sensitive to irradiation than is the ovary. A single administration of 50100 mCi may deliver sufficient radiation to the testes to cause transient testicular failure of uncertain long-term consequence (110). Young men may develop permanent testicular damage with reduced sperm counts roughly proportional to the amount of 131I administered (111). The only manifestation may be asymptomatic FSH elevation; however, after several 131I treatments, this may be associated with reduced sperm motility, although serum testosterone usually remains normal (112). Permanent sterility occurs in less than 10% of men cumulatively given 300 mCi and over 90% treated with 800 mCi (107). It, thus, seems prudent to consider banking sperm in young men treated with 131I, especially if the cumulative dose is anticipated to be over 100 mCi.
Developmental defects caused by 131I
There is no evidence that treating children or women during the childbearing years increases the risk of congenital abnormalities. In a long-term study of 33 children treated at an average age of 14.6 yr with a mean of 196 mCi 131I, the frequency of infertility (12%), miscarriage (1.4%), prematurity (8%), and major congenital anomalies (1.4%) after an average of almost 19 yr of follow-up was not significantly different from that in the general population (113). A study of 2113 pregnancies in females treated for thyroid cancer found that the incidences of stillbirth, preterm birth and low birth weight, congenital malformation, and death during the first year of life were not significantly different before and after 131I therapy (108).
Bone marrow damage and the induction of tumors and leukemia
Bone marrow damage and induction of other tumors are the most serious late sequelae of 131I therapy. Cumulative amounts of 131I over 1000 mCi cause a small but significant excess of deaths from bladder cancer and leukemia (109). Bladder cancer is more likely in those with relatively little 131I uptake in the neck or metastases (109). There is an increased risk of colorectal cancer 5 or more years after 131I therapy that is related to the cumulative activity of 131I administered (101). This is probably caused by accumulation of 131I in the colon, especially in hypothyroid patients, underscoring the importance of ensuring one to two large bowel movements for a few days after 131I administration, which may require a laxative that does not contain iodine. Magnesium citrate should be used with caution in patients taking lithium.
Patients may develop abnormalities of erythrocytes, platelets, and
granulocytes with very high cumulative 131I
activities (1000 mCi). In 13 large series comprising a total of 2753
patients with thyroid cancer, 14 cases of leukemia were detected
(101), resulting in a prevalence of about 5 leukemia cases
per 1000 patients (0.5%), which is higher than expected. Acute myeloid
leukemia, the type associated with 131I therapy,
usually occurs within 210 yr of treatment. It is less likely when
131I is given annually rather than every few
months, and when total blood dose per administration is less than 200
cGy (101).
Despite these reports, the lifetime risk of leukemia is so small (0.33%) that it does not outweigh the benefit of 131I therapy (114). The estimated absolute risk of life lost from recurrent thyroid cancer exceeds that from leukemia by 4- to 40-fold, depending on the age at which the patient is treated (114). When 131I is given at 12-month intervals and at lower total cumulative 131I activities (600800 mCi), long-term effects on the bone marrow are minimal (96) and few cases of leukemia occur. After a mean follow-up of 10 yr, no cases of leukemia were observed (2.5 expected) in 1771 patients treated with an average cumulative 131I activity of 195 mCi (115). The risk is small enough that a population study did not find an increased risk of leukemia in patients treated with 131I for thyroid cancer (116).
Pulmonary fibrosis
Lung fibrosis may occur in patients with diffuse pulmonary metastases treated with 131I (94, 117, 118, 119). It can be avoided by using smaller 131I doses (100200 mCi) when a diagnostic scan shows intense uptake in the lungs.
Management of DTC during pregnancy
Management of a pregnant woman may be associated with considerable anxiety, mainly regarding the timing and recommendations for treatment. Although there are case reports that pregnancy may accelerate the course of the disease, a large study shows the prognosis of newly diagnosed DTC in pregnancy is similar to that occurring in similarly aged nonpregnant women (120). Surgery should be considered in the second trimester, but 131I scans and treatment can be safely delayed until after delivery.
![]() |
Assessment after Initial Surgical and 131I Treatment |
---|
Posttreatment 131I scans
About 47 days after 131I therapy is given,
a WBS should be done to document 131I uptake by
the tumor, which may show lesions not detected by the diagnostic scan
(Table 4; Refs. 11 and
57). Posttreatment WBS is most likely to yield critical
information when the serum Tg level is elevated and a tumor cannot be
found on examination or by imaging studies such as diagnostic WBS, neck
ultrasonography, CT, MRI or PET scans.
|
During follow-up, periodic elevation of serum TSH levels to stimulate Tg release and 131I uptake for WBS is the optimal way to detect residual thyroid tissue or cancer. L-thyroxine may be withdrawn, which causes symptomatic hypothyroidism, or rhTSH may be given im, which stimulates thyroidal 131I uptake and Tg release while the patient continues L-thyroxine, thus avoiding hypothyroidism (122).
rhTSH was approved for diagnostic use after being tested in two large multicenter studies. The first found that WBS results after two 0.9-mg doses of rhTSH, given while L-thyroxine was continued, were of good quality and were equivalent to the scans performed after L-thyroxine withdrawal in 66% of the patients, superior in 5% and inferior in 29% (122). This proved that rhTSH stimulates 131I uptake for WBS, but with lower sensitivity than after L-thyroxine withdrawal (122). A second multicenter international study was done to test the effects of two rhTSH dosing schedules on WBS and serum Tg levels compared with those obtained after L-thyroxine withdrawal. The scanning method was more carefully standardized, taking into account the lower renal 131I clearance in hypothyroidism than after rhTSH (78). Scans were concordant in 89% of the patients and superior in 4% after rhTSH and in 8% after L-thyroxine withdrawal, differences that were not statistically significant. The main findings were that the two dose schedules gave similar results and the combination of WBS and Tg measurements detected 100% of the patients with metastatic DTC (78).
The recommended dose of rhTSH, 0.9 mg, is given im on 2 consecutive days, followed by at least 4 mCi 131I on the third day and a WBS and Tg measurement on the fifth day. Whole-body images are acquired after 30 min of scanning or after 140,000 counts. This is necessary because 4 mCi 131I after rhTSH has about the same effect as 2 mCi given during hypothyroidism with reduced renal clearance and raised body 131I retention (91). A serum Tg of 2.0 ng/mL or higher 72 h after the last rhTSH injection indicates that thyroid tissue or DTC is present, which almost always can be identified on the rhTSH-stimulated WBS, providing the recommended scanning procedure is followed (78). The drug is well tolerated. Transient headache (7.3%) and nausea (10.5%) are its main adverse effects (78), with almost no other symptoms and no dysphoric mood states as occur with hypothyroidism (122).
Tg measurements
Serum Tg measurement is the best means of detecting normal and malignant thyroid tissue because there are no other sources to falsely elevate it. Most patients who are free of disease have undetectable serum Tg levels (11, 123). Antithyroglobulin antibodies (TgAbs), which are found in up to 25% of patients with DTC compared with about 10% of the general population, must be measured in the same serum sample and, when present, usually invalidate the serum Tg result (56). Immunometric assay (IMA) methods are prone to underestimating the serum Tg level when TgAbs are present, increasing the risk of a false negative test (56). Conversely, when Tg remains high when measured by IMA in the presence of TgAbs, residual cancer may be present (56). Changes in posttreatment serum TgAb levels directly correlate with the presence or absence of disease (56).
The first serum Tg after surgery is a good prognosticator. An initial Tg value higher than 70 ng/mL had a 90% positive predictive value for metastases in one study (124); however, the Tg may remain detectable for up to a year after treatment before becoming undetectable (11). Thereafter, Tg should be measured when TSH has been stimulated by L-thyroxine withdrawal or rhTSH stimulation, which lowers the false negative rate well below that of WBS (11, 78, 121). A Tg messenger RNA method is more sensitive than the IMA method, particularly during L-thyroxine treatment or when TgAbs are present, but the test is not yet widely available (125).
Tg and WBS are usually considered complementary (126). Patients rarely have DTC when they have had two negative postablation scans (127) and serum Tg values less than 2 ng/mL during rhTSH stimulation (78) or less than 5 ng/mL after thyroid hormone withdrawal (123); however, results of Tg assays vary in different laboratories, even with the use of a new international standard (CRM 457; Ref. 128). Nonetheless, patients with undetectable TSH-stimulated Tg levels alone rarely have cancer (11). In lieu of a diagnostic WBS performed 1 yr after thyroid ablation, serum Tg measurement after L-thyroxine withdrawal or rhTSH stimulation may serve as a guide for the selection of patients who might have persistent cancer (11). Scanning with 25 mCi is relatively useless compared with administering 100 mCi or more when the Tg rises above some arbitrary limit suggesting the presence of metastasesusually around 10 ng/mL (11, 57).
Tg-positive, diagnostic WBS-negative patients
Pulmonary metastases sometimes may be found only after administrating large doses of 131I and obtaining a posttreatment WBS (57).
Frequency
One study found that about 6% of 283 patients with high serum Tg
levels treated with 100 mCi 131I had distant
metastases detected on posttherapy WBS that were not detected on a
2-mCi scan (83). In another study, all but 1 of 17
patients with elevated serum Tg levels and a negative 5-mCi diagnostic
scan had 131I uptake after 75140 mCi; more than
half had lung metastases (129). Among 89 consecutive pairs
of diagnostic and postablative 131I scans done on
79 patients in our clinic (Table 4), 8 of 10 patients with negative 4-
to 5-mCi diagnostic scans had a posttherapy scan revealing distant
metastases; all had a serum Tg level above 15 ng/mL when the TSH was 30
µU/mL or higher.
Results of therapy
Two to 4 yr after receiving a total of 350700 mCi 131I, 3 of our 10 patients had no uptake on the posttherapy scan and a serum Tg level less than 5 ng/mL while off L-thyroxine. The prognostic importance of finding lung metastasis at an early stage was shown by Schlumberger (4), who reported 100% 10-yr survival with lung metastases detectable only by elevated Tg levels and confirmed by posttherapy WBS, 91% when chest x-rays were normal but diagnostic WBS was positive, 63% with lung micronodules, and 11% with lung macronodules. Treatment of pulmonary metastases found only on posttherapy scans usually reduces the tumor burden, but complete eradication of metastases may nonetheless be difficult to achieve (130). The Tg level that is used for recommending treatment of scan-negative, Tg-positive patients has been coming down; it was about 30 or 40 ng/mL about a decade ago but now is about 10 ng/mL (6, 57, 131), although this is an arbitrary cutoff.
![]() |
Thyroid Hormone Suppression of TSH |
---|
![]() |
External Radiation Therapy |
---|
![]() |
Conclusion |
---|
![]() |
Acknowledgments |
---|
![]() |
Footnotes |
---|
2 Here and elsewhere, age refers to patient age at
the time of initial therapy.
3 Definition of TNM: T, primary tumor; TX, cannot
be assessed; T0, no evidence of primary tumor; T1, <1 cm; T2, 1- 4 cm;
T3, >4 cm; T4, tumor of any size beyond thyroid capsule; N, regional
lymph node; NX, cannot be assessed; N0, not present; N1, present; M,
distant metastases; M0, none; M1, present. Patients under age 45 with
any T, any N, and M0 are stage 1; and M1 are stage 2. Patients >45
with T1 are stage 1, T23 are stage 2, T4 or N1 are stage 3, and M1
are stage 4.
4 Multiply by 37 to convert to MBq.
5 Body surface area is calculated as Ht
(inches) x wt (pounds)/3131.
Received October 11, 2000.
Revised December 8, 2000.
Accepted December 13, 2000.
![]() |
References |
---|