SEARCH

SEARCH BY CITATION

Keywords:

  • distant metastases;
  • follicular thyroid carcinoma;
  • prognostic factors;
  • radioiodine

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

BACKGROUND

The objective of this study was to investigate the patterns of recurrence, various prognostic factors, and the role of radioiodine in the treatment of patients with follicular thyroid carcinoma (FTC).

METHODS

The clinical outcomes of 215 patients with FTC who were treated at a single institution were analyzed retrospectively. The mean follow-up was 10.8 years.

RESULTS

The actuarial rates of cause specific survival (CSS), locoregional (LR) control, and freedom from distant metastasis (DM) at 10 years were 81%, 83%, and 72.3%, respectively. The independent prognostic factors for survival were metastasis at presentation (relative risk [RR], 47.7), radioiodine (RAI) treatment (RR, 0.25), extrathyroidal extension (RR, 3.8), and the postoperative absence of macroscopic disease in the neck region (RR, 0.06). In patients who were treated with RAI, both the LR failure rate (RR, 0.24) and the mortality rate (RR, 0.25) were reduced to about 25%. Subgroup analysis revealed that RAI improved the survival of patients with DM at presentation (RR, 0.17) and improved the LR control rate in patients who had no DM at presentation (RR, 0.13). For patients who underwent total thyroidectomy with negative resection margins, RAI significantly reduced the rate of LR recurrence (RR, 0.05). Patients with the minimally invasive type of FTC had a good prognosis. The 10-year rates for CSS, LR control, and freedom from DM were 97.6%, 100%, and 90.6%, respectively. The prognosis of patients with frankly invasive FTC was much poorer. The 10-year rates for CSS, LR control, and freedom from DM were 66.7%, 100%, and 45%, respectively.

CONCLUSIONS

RAI is an effective treatment for patients with FTC. It was associated with improved survival rates and fewer recurrences. Cancer 2002;95:488–98. © 2002 American Cancer Society.

DOI 10.1002/cncr.10683

Follicular thyroid carcinoma (FTC) accounts for 10–32% of differentiated thyroid carcinomas in large published studies.1–5 FTC often is analyzed together with papillary thyroid carcinoma, and both are known collectively as differentiated thyroid carcinoma because of the similarities in their clinically indolent behavior, management, and outcome.6–12 However, there are distinctive differences between these two histologies: Patients with FTC tend to be older, present with more advanced disease, and have a poorer survival rate.10, 11, 13–15 More recent studies tend to stratify patient outcome according to histologic classification. We analyzed the clinical features and outcomes of patients with FTC who were managed at a single institution in Hong Kong over 4 decades.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The records of 1057 patients with differentiated thyroid carcinoma who were registered at the Queen Elizabeth Hospital, Hong Kong, from 1960 to 1997 were reviewed; 842 patients (79.7%) had papillary thyroid carcinoma, and 215 patients (20.3%) had FTC. In this study, 13 patients with a diagnosis of Hurthle cell carcinoma were not included. In our pathology review, if the patient did not undergo surgery in a government-based hospital, then the slide of the pathology specimen was reviewed by our consultant pathologists. The mean follow-up was 10.8 years. Thirty patients (14%) were lost to follow-up. The demographic data are summarized in Table 1. Surgical treatment was classified into three groups: total or near total thyroidectomy (TT), lobectomy, and no surgery (or biopsy only). Postoperative locoregional (LR) residual disease was classified as macroscopic, microscopic, nil, or undetermined. The presence of macroscopic residual disease was assessed by clinical, operative, and pathologic findings. Microscopic disease was present if the pathologic resection margin revealed tumor, but operative findings showed no macroscopic tumor left in the neck.

Table 1. Clinical Characteristics of 215 Patients with Follicular Thyroid Carcinoma
Patient characteristicsNo. of patients (%)
  1. SD. Standard deviation; AJCC: American Joint Committee on Cancer.

No. (total)215 (100)
Follow-up (yrs) 
 Mean10.8
 Median9.3
Age (yrs) 
 Mean ± SD49 ± 17
 Range9.3–87.3
Gender 
 Female159 (74)
 Male56 (26)
Size (mean, 3.9 cm) 
 ≤ 1 cm11 (5.1)
 > 1 cm152 (70.7)
 Not stated52 (24.2)
Multifocal disease 
 No132 (61.4)
 Yes39 (18.1)
 Not stated44 (20.5)
Extrathyroidal extension 
 No174 (80.9)
 Yes30 (14)
 Not stated11 (5.1)
Lymph node metastases 
 No186 (86.5)
 Yes26 (12.1)
 Not stated3 (1.4)
Distant metastasis at presentation 
 No183 (85.1)
 Yes32 (14.9)
Sites of distant metastasis at presentation 
 Lung9 (28.1)
 Bone18 (56.3)
 Lung and bone2 (6.3)
 Brain1 (3.1)
 Bone, mediastinum, and subcutaneous soft tissue1 (3.1)
 Mediastinum1 (3.1)
AJCC/TNM stage 
 I88 (40.9)
 II59 (27.4)
 III23 (10.7)
 IV28 (13)
 Not stated17 (7.9)
T stage 
 03 (1.4)
 17 (3.3)
 287 (40.5)
 342 (19.5)
 430 (14)
 X46 (21.4)
N stage 
 0186 (86.5)
 126 (12.1)
 X3 (1.4)
M stage 
 0183 (85.1)
 132 (14.9)

Management Strategy

The management of patients FTC included primary surgery followed by an evaluation for radioiodine (RAI) treatment and/or external radiotherapy. TT or completion thyroidectomy was the preferred surgical procedure. According to our protocol, RAI treatment was indicated in patients with any of the following characteristics: primary tumor size > 1 cm, lymph node metastasis, age > 40 years, the presence of extrathyroidal extension, macroscopic disease in the neck, and/or the presence of distant metastasis (DM). RAI administration was scheduled at 4 weeks after surgery or after cessation of thyroxin treatment. A neck scan (tracer dose, 10 megabecquerels [MBq]) was performed to detect any uptake in the neck before RAI treatment. Recombinant human thyrotropin was not available in our hospital. The usual dose for patients with no DM was 2.96 GBq (80 mCi). For patients who had DM, the dose was usually 5.55 gigabecquerels (GBq; 150 mCi). There was minor variation in the RAI dose and indications according to individual physician or patient preferences. The patients were treated as inpatients to avoid unnecessary radiation exposure to the public, as required by hospital policy. A total body RAI scan was performed 3–6 months after RAI treatment to document the success of ablation. Patients received 80 MBq of RAI, and the total body was scanned at 72 hours. Success of the ablation was documented by a negative scan. Repeated doses were given if the scan showed uptake.

External radiotherapy to the thyroid bed and bilateral cervical lymphatics was given as part of initial therapy if there was macroscopic LR disease in the neck, extensive extrathyroidal extension, and/or extensive lymph node metastases. The treatment was divided into two phases: Phase 1 consisted of an anterior cervical field using 4.5–6.0 MV photons up to 30–36 grays (Gy), and Phase 2 consisted of an anterior electron field using 12–15 MeV electrons. The spinal cord dose was limited to < 45 Gy. The median total dose was 60 Gy prescribed at 90% isodose lines in 30 fractions, with 5 daily fractions per week. The initial treatment for the patient population is outlined in Table 2; 81.4% of patients had RAI as part of their initial therapy, and 10 patients (4.7%) had external radiotherapy to the thyroid bed.

Table 2. Type of Primary Treatment for Follicular Thyroid Carcinoma (N = 215 patients)
Type of primary treatmentNo. of patients (%)
  1. SD: standard deviation; Gy: grays.

Thyroid surgery 
 Total or near total thyroidectomy173 (80.5)
 Lobectomy30 (14)
 Biopsy only or no surgery12 (5.6)
Lymph node surgery 
 Excision/sampling of suspicious lymph nodes195 (90.7)
 Neck dissection: Functional or radical18 (8.4)
 Not stated2 (0.9)
Postoperative local-regional residual disease 
 Macroscopic26 (12.1)
 Microscopic6 (2.8)
 Nil149 (69.3)
 Undetermined34 (15.8)
Radioiodine treatment: Mean dose ± SD in GBq (mCi) 
 Overall: 4.14 (111.9)175 (81.4)
 Dose 1: 2.7 ± 0.95 (72.8 ± 25.8)175 (81.4)
 Dose 2: 4.3 ± 1.7 (115.8 ± 46.5)31 (14.4)
 Dose 3: 5.3 ± 1.1 (144.4 ± 30)9 (4.1)
 Dose 4: 6.2 ± 0.96 (166.7 ± 25.9)6 (2.8)
 Dose 5: 6.2 ± 1.0 (166.7 ± 28.9)3 (1.4)
 Dose 6: 5.55 (150)2 (0.9)
 Dose 7: 5.55 (150)1 (0.5)
External radiotherapy (± SD)10 (4.7)
 Mean dose (Gy)49.2 ± 17.1
 Mean no. of fractions22.4 ± 8.9
 Mean duration (days)33.8 ± 14.6

After the primary treatment, all the patients had replacement thyroxin and follow-up at intervals of 6–12 months. In the past, many patients received suppressive doses of thyroxin to maintain an undetectable thyroid-stimulating hormone (TSH) level. The trend in recent years was just to keep the TSH level within the normal range (0.35–5.5 mIU/L) for the low-risk group to avoid potential cardiac16–18 and bone19 complications of excessive thyroid hormone treatment. Although more recent studies found that prolonged suppressive thyroxin would not decrease bone mineral density20, 21 or accelerate bone turnover22, 23 and osteoporosis,24 the cardiac complications alone limit the clinical usage in low-risk patients. For patients with advanced LR or DM disease, TSH would be maintained at just below the low normal range. There is evidence that thyrotropin suppression inhibits the in vitro invasion and growth of differentiated thyroid carcinoma and clinically reduces recurrences and inhibits growth of the carcinoma. Serum thyroglobulin measurements have been available in our hospital since January, 1989. If the thyroglobulin level was > 10 μg/L after received RAI ablation, then scans were repeated to detect recurrent tumors. Chest roentgenograms, whole body iodine 131 scans, computed tomography scans, and/or magnetic resonance imaging scans were performed if disease recurrence was suspected. Routine whole body scans were not obtained as part of the follow-up procedure in patients who were clinically disease free.

Statistical Analysis

In this study, we analyzed the clinical features and treatment factors in the management of patients with FTC in relation to outcome parameters: cause specific survival (CSS), LR control, and freedom from DM. LR disease was defined as clinically detectable disease in the thyroid bed or cervical lymph nodes. DM was defined as clinically or radiologically detectable disease outside the thyroid bed and cervical lymph nodes. Clinicopathologic variables that were analyzed included age, gender, tumor size, extrathyroidal extension, lymph node metastases, DM at presentation, and postsurgical LR residual disease. Treatment variables included the type of thyroid surgery and the use of RAI and/or external radiotherapy. The CSS, freedom from DM, and LR control curves were generated by using the Kaplan–Meier method. Log-rank tests were used to identify significant factors. All variables were entered into a multivariate analysis using the Cox regression model.25 SPSS 10.0 software was used in the data analyses (SPSS, Inc., Chicago, IL). Significance levels are presented as P values. It was assumed that the observed differences were statistically significant at the P ≤ 0.05 level. Subgroup analyses were performed in relevant circumstances to specifically define the effectiveness of treatment in different subgroups.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Outcome and Prognostic Factors

The actuarial rates of CSS, LR control, and freedom from DM at 10 years were 81%, 83%, and 72.3%, respectively. The CSS curves according to International Union against Cancer26/American Joint Committee on Cancer27 stages are shown in Figure 1. The 10-year CSS rates were as follows: patients with Stage I disease, 94.2%; patients with Stage II disease, 93.5%; patients with Stage III disease, 62.1%; and patients with Stage IV, 21%.

thumbnail image

Figure 1. Disease specific survival of 215 patients with follicular thyroid carcinoma stratified according to the International Union Against Cancer/American Joint Committee on Cancer TNM staging system (n = 198 patients; 17 patients with unknown staging are not shown).

Download figure to PowerPoint

After primary treatment (including surgery, RAI, and external radiotherapy), 17 patients (7.9%) had uncontrolled LR neck disease. During the subsequent course of the disease, 23 patients (10.7%) had LR recurrences. Sixty-two patients (28.8%) had DM during the course of the disease: 32 patients had DM at the time of presentation, and 30 patients had DM at the time they developed recurrent disease. Overall sites of DM were lung in 30 patients (14%), bone in 41 patients (19.1%), liver in 3 patients (1.4%), and brain in 2 patients (1%). At the last follow-up, 148 patients (68.8%) were alive and disease free, 1 patient (0.5%) was alive with LR disease, 16 patients (7.4%) were alive with DM, 4 patients (1.9%) were alive with both LR disease and DM, 37 patients (17.2%) had died of FTC, and 9 patients (4.2%) had died of other, unrelated causes.

With regard to disease specific mortality in the 37 patients who died of FTC, the cause of death was LR disease in 7 patients (18.9%), DM in 21 patients (56.8%), and both LR and DM in 9 patients (24.3%). Bone was a more common site of DM than lung (41 patients vs. 30 patients) and also was a more important cause of fatal events (bone metastasis was a chief contributing factor in the deaths of 19 patients).

The independent prognostic factors for survival were metastasis at the time of presentation (relative risk [RR], 47.7), RAI treatment (RR, 0.25), extrathyroidal extension (RR, 3.8), and the absence of postoperative gross disease in the neck region (RR, 0.06) (Table 3). Table 4 summarizes the outcome of multivariate analyses on CSS, failure from DM, and LR control. RAI treatment reduced both the LR failure rate (RR, 0.24; 95% confidence interval [95% CI], 0.12–0.51) and the mortality rate (RR, 0.25; 95% CI, 0.11–0.57) of patients with FTC to about 25%. External radiotherapy was given to 10 patients. Due to the small number of patients who received this treatment, no conclusion could be drawn about its therapeutic effect on FTC.

Table 3. Cause Specific Survival: Univariate and Multivariate Analyses (N = 215 patients)
Prognostic factorsNo. of patients who died of disease (%)Cause specific-survival
Univariate analysisMultivariate analysis
P valueRelative risk (95% CI)P value
  1. 95% CI: 95% Confidence interval; NS: not significant.

Age (yrs)    
 ≤ 456/88 (6.8)0.0002NS
 > 4531/127 (24.4)
Gender    
 Female27/159 (17.0)0.69NS
 Male10/56 (17.9)
Cervical lymph node metastases    
 No25/186 (13.4)0.0005NS
 Yes10/26 (38.5)
 Not stated2/3 (67.7)
Size of primary thyroid tumor (cm)    
 ≤ 13/27 (11.1)0.0002NS
 > 114/136 (10.3)
 Not stated20/52 (38.5)
Extrathyroidal extension    
 No23/174 (13.2)< 0.00011.00.006
 Yes12/30 (40.0)3.8 (1.5–10.0)
 Not stated2/11 (18.2)
Local-regional residual disease    
 Gr Macroscopic15/26 (57.7)< 0.00011.0< 0.001
 Nil8/149 (5.4)0.06 (0.02–0.2)
 Microscopic1/6 (16.7)
 Unknown13/34 (38.2)
Metastases at presentation    
 No15/183 (8.2)< 0.00011.0< 0.001
 Yes22/32 (68.7)47.7 (17.8–128)
Type of thyroid surgery    
 Total/near-total thyroidectomy21/173 (12.1)<0.0001 NS
 Lobectomy7/30 (23.3)
 Biopsy/no surgery9/12 (75.0)
Radioiodine treatment    
 No17/40 (42.5)< 0.00011.00.001
 Yes20/175 (11.4)0.25 (0.11–0.57)
External radiotherapy    
 No33/205 (16.1)0.009NS
 Yes4/10 (40.0)
Table 4. Summarized Results of Multivariate Analysis of 215 Patients with Follicular Thyroid Carcinoma
Prognostic variableCause specific survivalLocal-regional controlFailure from distant metastasis
Relative risk (95% CI)P valueRelative risk (95% CI)P valueRelative risk (95% CI)P value
  1. 95% CI: 95% Confidence interval.

Age (yrs)      
 ≤ 451.00.001
 > 453.1 (1.6–7.8)
Extrathyroidal extension      
 No1.00.006
 Yes3.8 (1.5–10.0)
Cervical lymph node metastasis      
 No1.00.002
 Yes2.6 (1.4–4.8)
Postoperative residual disease      
 Macroscopic1.0< 0.0011.0< 0.001
 Nil0.06 (0.02–0.2)0.037 (0.13–0.11)
Distant metastasis at presentation      
 No1.0< 0.0011.00.002
 Yes47.7 (17.8–128)4.4 (1.76–11.2)
Type of thyroid surgery      
 Total or near-total thyroidectomy1.00.001
 Lobectomy
 Biopsy or no surgery3.6 (1.65–7.8)
Radioiodine      
 No1.00.0011.0< 0.001
 Yes0.25 (0.11–0.57)0.24 (0.12–0.51)

Patients with DM at Presentation

The 32 patients who had DM at the time of presentation were significantly different compared with the whole cohort of patients with FTC, with an older mean age at diagnosis (62.5 years vs. 49.0 years), a larger mean tumor size (4.9 cm vs. 3.9 cm), and a greater incidence of multifocal disease (28.1% vs. 18.1%), extrathyroidal extension (28.1% vs. 14%), and lymph node metastasis (31.3% vs. 12.1%). The sites of DM and the numbers of patients were as follows: lung, 9 patients; bone, 18 patients; both lung and bone, 2 patients; brain, 1 patient); diffuse metastases to bone, 1 patient, and subcutaneous soft tissue and mediastinum, 1 patient. Twenty-two patients died of disease. The CSS rate was 27% at 10 years. The independent prognostic factors that were identified for survival were RAI treatment (RR, 0.17; 95% CI, 0.06–0.51) and extrathyroidal extension (RR, 3.3; 95% CI, 1.2–9.2). The risk of dying of FTC was reduced to about one-sixth by the use of RAI. Figure 2 shows the actuarial CSS curves stratified by RAI treatment. The 5-year survival rates of patients with and without RAI were 49% and 15%, respectively. Concerning the site of DM, patients who had bone metastasis at presentation had a higher mortality rate compared with patients who had lung metastasis: 16 of 18 patients (88.9%) compared with 4 of 9 patients (44.4%), respectively.

thumbnail image

Figure 2. Disease specific survival of 215 patients with follicular thyroid carcinoma stratified according to those who did or did not receive radioiodine (RAI) treatment (P = 0.001).

Download figure to PowerPoint

Patients Without DM at Presentation

One hundred eighty-three patients had no DM at presentation. The 10-year rates of CSS, LR control, and freedom from DM were 90%, 85%, and 85%, respectively. LR recurrence and DM occurred in 13 patients and 30 patients, respectively. Fifteen patients died of disease. Results of the multivariate analyses that included the three outcome variables are depicted in Table 5. The independent variables that influenced survival were extrathyroidal extension (RR, 10) and postoperative LR disease (RR, 0.12). The absence of postoperative LR residual disease was the single most important independent prognostic factor for predicting better survival (RR, 0.12), LR control (RR, 0.28), and freedom from DM (RR, 0.03). Patients age > 45 years had a 2.6-fold increased risk of DM (95% CI, 1.1–6.0). Patients who received RAI had less LR recurrence (RR, 0.13; P < 0.001). Figure 3 shows the actuarial LR control curves by RAI treatment. The rate of LR control at 10 years was 95% in RAI-treated patients compared with 60% in untreated patients. Figure 4 shows the results for the 32 patients with DM stratified according to those who did or did not receive RAI.

Table 5. Summary of independent Prognostic Factors in 183 Patients with Follicular Thyroid Carcinoma with No Distant Metastasis at Presentation
Prognostic variableCause specific survivalLocal-regional controlFailure from distant metastasis
Relative risk (95% CI)P valueRelative risk (95% CI)P valueRelative risk (95% CI)P value
  1. 95% CI: 95% confidence interval.

Age (yrs)      
 ≤ 451.00.022
 > 452.6 (1.1–6.0)
Extrathyroidal extension      
 No1.00.03
 Yes10 (1.2–83.0)
Postoperative residual disease      
 Macroscopic1.00.0491.0< 0.0011.00.049
 Nil0.12 (0.02–0.99)0.03 (0.008–0.108)0.28 (0.08–1.0)
Radioiodine      
 No1.0< 0.001
 Yes0.13 (0.05–0.35)
thumbnail image

Figure 3. Locoregional control curves for 183 patients who had follicular thyroid carcinoma with no distant metastases at the time of initial presentation. Results are stratified according to patients who did or did not receive radioiodine (RAI) treatment (P = 0.0001).

Download figure to PowerPoint

thumbnail image

Figure 4. Disease specific survival of 32 patients with distant metastases from follicular thyroid carcinoma. Results are stratified according to patients who did or did not receive radioiodine (RAI) treatment (P = 0.001).

Download figure to PowerPoint

Role of RAI in the Pure Ablation Setting

One hundred thirty-five patients with no DM at presentation completed thyroid surgery to remove all LR disease with clear resection margins. One hundred thirty patients underwent TT or near TT, whereas 5 patients underwent lobectomy. Subsequently, 122 patients had RAI ablation, 1 patient had both RAI and external radiation therapy, and 12 patients had no adjuvant radiation therapy. After treatment, 15 patients developed recurrent tumors: 2 patients had LR recurrences alone, 10 patients had DM, and 3 patients had both. The 10-year rates of CSS, LR control, and freedom from DM were 98.1%, 96.4%, and 91%, respectively. Multivariate analyses did not reveal any independent prognostic factor for CSS. For freedom from DM, age > 45 years (RR, 13.5) and cervical lymph node metastasis (RR, 18.8) were significant poor prognostic factors. It was found that RAI ablation was the only important factor in predicting LR control in multivariate analysis; it decreased the risk of LR failure by 20-fold (RR, 0.05; 95% CI, 0.005–0.51; P = 0.01).

Minimally Invasive FTC

Minimally invasive (encapsulated) FTC was diagnosed by the criteria total encapsulation with no macroscopic invasion; invasion identified on histologic examination.28 This entity was diagnosed in 52 patients. The clinical features of these patients are summarized in Table 6. Compared with the whole FTC group, the mean age was younger (43.7 years), the mean size of the primary tumor was slightly smaller (3.6 cm), and the incidence of extrathyroidal extension was lower (3.8%). No lymph node metastases or DM were found at the time of diagnosis. The 10-year rates of CSS, LR control, and freedom from DM were 97.6%, 100%, and 90.6%, respectively. Of these 52 patients, 50 patients underwent TT or near TT, and 46 patients received postoperative RAI treatment. After RAI treatment, 2 of 46 patients had DM; no patient developed LR recurrence. Among the six patients who had no RAI treatment, one patient developed bone metastasis and died of disease. Overall, 3 of 52 patients (5.8%) developed DM recurrence (all bone metastases), and no patients developed LR recurrence. One patient died of diffuse bone metastases (mortality rate, 2%), and the other two patients were alive with disease and essentially symptom free at the time of their last follow-up.

Table 6. Clinical Features of 52 Patients with Minimally Invasive Follicular Thyroid Carcinoma
Clinical featureNo. of patients (%)
  1. SD: standard deviation.

No.52/215 (24)
Follow-up (yrs) 
 Mean ± SD8.2 ± 4.0
 Median8.3
Age (yrs) 
 Mean ± SD43.7 ± 15.5
 Range15.6–76.3
Gender 
 Female35 (67.3)
 Male17 (32.7)
Size (median ± SD: 3.5 ± 1.9 cm) 
 ≤ 1 cm1 (1.9)
 > 1 cm50 (96.2)
 Not stated1 (1.9)
Multifocal disease 
 No45 (86.5)
 Yes7 (13.5)
Extrathyroidal extension 
 No50 (96.2)
 Yes2 (3.8)
Lymph node metastases 
 No52 (100)
 Yes0 (0)
Distant metastasis at presentation 
 No52 (100)
 Yes0 (0)
TNM stage 
 I30 (57.7)
 II20 (38.5)
 III1 (1.9)
 IV0 (0)
 Not stated1 (1.9)

Widely Invasive FTC

Widely invasive FTC is diagnosed by the following criteria: macroscopically invading the adjacent thyroid, may also include invasion of more than four blood vessels.28 Ten patients had frankly invasive FTC (Table 7). This group of patients presented at an older age (mean, 62 years), had larger primary tumors (mean, 6.3 cm), and had a higher incidence of extrathyroidal extension (30%), lymph node metastasis (20%), and DM (50%). Two patients had DM at the time of initial presentation, and three patients had DM at the time they developed recurrent disease. Despite the approach of TT and RAI administration in all patients, only five patients were disease-free at last follow-up. The status of the other five patients was as follows: Three patients died of DM, one patient died of unrelated disease but had asymptomatic lung metastasis, and one patient was alive with liver and bone metastases. The 10-year rates of CSS, LR control, and freedom from DM were 66.7%, 100%, and 45%, respectively.

Table 7. Clinical Characteristics of 10 Patients with Frankly Invasive Follicular Thyroid Carcinoma
CharacteristicNo. of patients (%)
  1. SD: standard deviation.

No.10/215 (5)
Follow-up (yrs) 
 Mean ± SD6.04 ± 3.6
 Median5.2
Age (yrs) 
 Mean ± SD62 ± 11
 Range44.1–78.8
Gender 
 Female8 (80)
 Male2 (20)
Size (cm) 
 ≤ 10 (0)
 > 110 (100)
Multifocal disease 
 No6 (60)
 Yes4 (40)
Extrathyroidal extension 
 No7 (70)
 Yes3 (30)
Lymph node metastases 
 No8 (80)
 Yes2 (20)
Distant metastasis at presentation 
 No8 (80)
 Yes2 (20)
Sites of distant metastasis at presentation 
 Lung1 (10)
 Bone1 (10)
TNM staging 
 I1 (10)
 II3 (30)
 III4 (40)
 IV2 (20)

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Compared with papillary thyroid carcinoma, patients with FTC have a higher risk of developing distant failure and dying of FTC.3, 11, 14, 15, 29–32 Based on our experience, the rates of DM (28.8%) and mortality (17.2%) in patients with FTC were higher than in patients with papillary thyroid carcinoma (8.9% and 7.6%, respectively). Some investigators have observed that, when patients with DM at the time of initial presentation were excluded in the survival analysis, the mortality rates for patients with FTC and patients with papillary thyroid carcinoma were similar.33 Various poor prognostic factors identified in patients with FTC included older age,14, 15, 29, 34–37 DM at presentation,34–38 extrathyroidal extension,37 lymph node metastasis,37 larger tumor size,36, 37 marked vascular invasion,35 tumor differentiation,39 widely invasive FTC,40 more advanced stage,39 postoperative LR disease,37 and risk group classification.14 From the multivariate analysis of our patients, we found that the independent poor prognostic factors were extrathyroidal extension, LR residual disease, DM at presentation, and absence of RAI treatment. Despite a number of negative reports on the efficacy of RAI treatment,14, 30, 34 studies with large numbers of patients have shown that RAI can reduce recurrences10, 11, 41–43 and mortality13, 41, 43 in patients with differentiated thyroid carcinoma. RAI also is effective in treating patients with DM.42, 44–47 Multivariate analyses of our data revealed that RAI was important in improving the overall survival rate (RR, 0.25) and reducing LR recurrences (RR, 0.24). In patients with DM, RAI improved survival (RR, 0.17). For patients without metastasis at presentation, adjuvant RAI treatment or RAI ablation may improve LR control (RR, 0.13). In the subgroup of patients with complete surgery (negative resection margins) and no DM at presentation, RAI decreased the risk of LR failure by 20-fold (RR, 0.05). It was effective both in the treatment of patients with metastatic disease and in an adjuvant setting.

Despite the reported therapeutic benefit of RAI in patients with thyroid carcinoma, there is still concern related to potential RAI side effects, especially teratogenicity, infertility, secondary malignancies, and leukemia. RAI is well tolerated by oral administration with mild acute side effects like nausea, acute sialadenitis, transient neck pain (especially in patients who undergo lobectomy), and hematologic depression.48–51 The reported long-term side effects include decreased salivary production, radiation pneumonitis, and pulmonary fibrosis.52 The incidence of carcinoma is increased slightly in those areas where RAI concentrates, such as the salivary glands, colon, and bladder.52–55 A small risk of leukemia is reported in patients who received very high doses of RAI.52, 56, 57 In a large cohort of 1497 patients who received an average cumulative dose of 7.2 GBq of RAI, not a single incidence of leukemia was found after a mean follow-up of 10 years.52 Furthermore, the risk of colorectal carcinoma was related to the total activity of RAI.54 An analysis of decision-making suggested that the benefits of RAI exceed the risk of leukemia by 4–40-fold.58 At the dose levels currently used for ablation and treatment, there is no definite evidence of an increased risk of permanent infertility or teratogenicity.59–64 Judging from the results of our study and others, RAI treatment appears to be safe and effective in reducing the risk of recurrence and improving survival.

Minimally invasive FTC was diagnosed in 52 patients in our cohort. Our data are consistent with other reported series suggesting that patients with minimally invasive FTC present at a younger age,65, 66 their primary tumors are smaller,38, 40 and lymph node metastases are rare.29, 67 The reported incidence of LR recurrences and DM is low.38, 65 The three treatment failures (5.8%) in our report were all in patients with bone metastases, which is contrary to a report from the University of California, San Francisco,65 where the recurrences all were local disease (5 of 95 patients). This probably can be explained by our higher rate of bilateral surgery (50 of 52 patients) and RAI treatment (46 of 52 patients), which resulted in a higher LR control rate in our group of patients.

The group of patients with widely invasive FTC had a higher incidence of lymph node involvement and DM68, 69 and a higher mortality rate.40, 69 A substantial proportion of patients with widely invasive FTC reported in the literature actually had poorly differentiated (insular) carcinomas.28 Five patients who were diagnosed with insular carcinoma at our center were excluded from this study. In the 10 patients with widely invasive FTC who we studied, the prognosis was poor, and only 50% of patients were disease free at the last follow-up.

In summary, FTC is a tumor that is amenable to treatment by RAI. RAI reduced both the LR failure rate (RR, 0.24) and the mortality rate (RR, 0.25) to about 25%. RAI was effective for patients with or without DM at the time of initial presentation.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The authors thank Mr. Oscar Mang of the Hong Kong Cancer Registry.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES