Distant metastases are found at diagnosis or followup in 10%–15% of patients with differentiated thyroid cancer [
Many of the lesions in the appendicular skeleton can be effectively managed by external beam radiation therapy (EBRT) or radioactive iodine [
Complete resection of the BM is thought to be associated with better overall survival and can be curative [
The options for limb salvage reconstructions following partial or complete bone resections have increased in the recent years [
The goal of the present study was to determine the length of survival and prognostic factors for patients with thyroid carcinoma BMs who had surgery, as well as the type and the survival of the various means of surgical reconstruction. In particular, we wished to evaluate whether the surgical implants lasted for the life of the patient, without the need for revision. We also examined the specific impact of BM resection and histologic subtype of thyroid carcinoma on survival.
We performed a retrospective review of patients treated surgically for osseous lesions secondary to metastatic thyroid carcinoma at a single institution between 1988 and 2011. Patients were identified by interrogating the Department of Orthopaedic Oncology Surgical Database. The study was performed with the approval and in accordance with guidelines by the institutional review board. Medical records, operative reports, radiographic studies, and pathology reports were reviewed.
There were 8681 patients with thyroid cancer identified from institutional records. Of these, 550 developed bone metastases and 43 required skeletal surgery. The study population was composed of 22 females and 21 males. The median age at thyroid cancer diagnosis was 54 years (mean 53, range 12–82), and at surgery for BM the median age was 62 years (mean 59, range 12–82). Forty-three patients were identified that underwent surgery for bone metastases. Two patients with surgery for lesions of the spine or cranium were excluded as they were performed by Neurosurgeons. Forty one patients underwent operative intervention by the Musculoskeletal Oncology Service at 10 different sites for disease in the pelvis or appendicular skeleton. Patient demographics are summarized in Table
Patient demographics.
Characteristic | Number (total |
---|---|
Gender | |
Female | 22 |
Male | 19 |
Age (years) | |
Mean | 59 |
Range | 12–82 |
Followup (months) | |
Median | 60 |
Range | 10–102 |
Surgery sites | |
|
|
Scapula | 1 |
Proximal humerus | 5 |
Shaft humerus | 4 |
Distal humerus | 3 |
Radius | 2 |
Metacarpal | 1 |
|
|
Acetabulum | 8 |
Proximal femur | 12 |
Shaft femur | 4 |
Distal femur | 1 |
Skeletal presentation stage | |
Solitary bone metastasis | 12 |
Multiple bone metastases | 29 |
Twelve patients presented with solitary bone metastasis (only one bone involved). Six of 12 patients with single bone metastasis had no other metastastic distant organ involvement. Two of the patients with single bone involvement had lung metastasis with no other organ involvement. The remaining 4 patients with single bone involvement had synchronous metastases in the lymph nodes, liver, and soft tissue. The remaining 29 patients presented with metastases involving multiple bones. The most common site of synchronous metastasis in these patients was the lung (Figure
The distribution of other metastatic disease sites at the time of surgical intervention for osseous metastasis is shown in the lower bars. After surgery, the sites of subsequent metastases formation are shown in the upper bars. The most common site of metastatic disease prior to surgery was lung. Following surgery, bone was the most common site of new metastatic disease.
Patients were followed for a minimum of 10 months unless they died prior to 12 months. The median duration of the followup was 60 months (10–102 range). At the time of the last followup, 11 of 41 patients were alive. All deaths were secondary to disease progression. Date of patient death was determined from patient records and the Social Security Death Index.
The site and type of surgery are summarized in Table
Types of surgery.
Tumor |
Category | Procedure | Number |
---|---|---|---|
No |
EBL | Median | 150 cc |
Range | 0–2500 cc | ||
Nail (closed) | 3 | ||
Plate | 1 | ||
Biopsy | 3 | ||
| |||
Yes |
EBL | Median | 775 cc |
Range | 0–19900 cc | ||
Curettage (19 cases) | Nail + PMMA | 9 | |
Plate + PMMA | 1 | ||
Pins + PMMA | 4 | ||
THA | 5 | ||
Wide resection (15 cases) | Endoprosthesis | 12 | |
No reconstruction | 3 |
Thyroid cancer diagnosis subtypes, bone metastases, and bone surgery sites.
Thyroid cancer subtype | Total number of patients | Patients with bone |
Patients who had surgery on bone metastases (percentage of pts with bone mets) | Surgery sites | Other treatments |
---|---|---|---|---|---|
|
264 | 22 (8.3) | 2 (9.1) | Femur (2) | XRT, chemo |
|
637 | 96 (15.1) | 21 (21.9) | Hand metacarpal (1), radius (1), humerus (6), pelvis (5), femur (8) | XRT |
|
342 | 44 (12.9) | 10 (22.7) | Humerus (5), scapula (1), pelvis (2), femur (2) | XRT, chemo |
|
866 | 134 (15.5) | 2 (1.5) | Humerus (1), femur (1) | Chemo |
|
6572 | 254 (3.9) | 6 (2.4) | Radius (1), humerus (1), femur (4) | None |
| |||||
|
8681 | 550 (6.3) | 41 (7.5) |
Thirty-seven of the 41 surgical patients were treated with radioactive iodine (131I). At our institution, the standard treatment for patients that present without metastasis (Stage I disease) is thyroidectomy and lymph node dissection, followed by radioactive iodine for the initial treatment. Patients presenting with bone metastasis (or other distant organ involvement) also have thyroidectomy and then receive radioactive iodine if there is no prior history of treatment, and/or iodine tracer uptake at metastatic sites.
Seven of the 41 surgical patients had preoperative radiation to the bone metastatic site. Four patients received chemotherapy instead of 131I. Thirty-eight of the patients had complete thyroidectomies (one patient had a partial thyroidectomy). Surgical treatment of bone metastases included excision in 34 cases (16 with en bloc resection and 18 with curettage). In 7 cases, the tumor was not removed, and the bone was simply stabilized.
Patient overall survival and local progression (recurrence) free survival was determined by Kaplan-Meier analysis, and the log rank test was used to compare the survival curves for different groups. Cox proportional hazards model were also fitted. Local progression was assessed by imaging studies. Local recurrence was defined as the reappearance on imaging studies of osseous of soft tissue tumor after prior excision. The Student’s
Overall patient survival probability by Kaplan-Meier analysis after surgery for bone metastasis was 72% at 1 year, 29% at 5 years, and 20% at 8 years. Median survival from time of skeletal surgery was 1.9 years (range 1.2–4.2, Figure
Univariate Cox model.
Prognostic factors | Overall survival | Time to bone |
Survival after bone |
Recurrence free survival | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Hazard |
95% CI |
|
Hazard |
95% CI |
|
Hazard |
95% CI |
|
Hazard |
95% CI |
|
|
Age | 1.06 | 1.02–1.09 |
|
1.06 | 1.005–1.116 |
|
1.002 | 0.975–1.03 | 0.864 | 1.012 | 0.98–1.04 | 0.41 |
Gender |
1.19 | 0.55–2.38 | 0.66 | 0.63 | 0.21–1.83 | 0.39 | 1.63 | 0.76–3.50 | 0.21 | 1.795 | 0.87–3.7 | 0.113 |
Follicular subtype | 1.1 | 0.51–2.37 | 0.81 | 1.26 | 0.43–3.71 | 0.68 | 0.83 | 0.40–1.73 | 0.63 | 1.33 | 0.63–2.81 | 0.45 |
Radioactive iodine Y v N | 0.18 | 0.05–0.61 |
|
0.07 | 0.005–1.186 | 0.066 | 0.54 | 0.18–1.62 | 0.27 | 0.504 | 0.16–1.53 | 0.23 |
Other treatment Y v N | 1.51 | 0.63–3.62 | 0.36 | 1.721 | 0.55–5.34 | 0.35 | 0.95 | 0.40–2.25 | 0.91 | 0.72 | 0.29–1.78 | 0.48 |
Skeletal presentation |
2.44 | 0.93–6.44 | 0.07 | 3.03 | 1.03–8.92 |
|
||||||
Preoperative radiation to bone Y v N | 1.985 | 0.73–5.38 | 0.18 | 1.74 | 0.65–4.68 | 0.27 | ||||||
Tumor excised from bone |
0.21 | 0.075–0.58 |
|
0.28 | 0.11–0.75 |
|
Multivariate Cox model.
Prognostic factors | Overall survival | Survival after bone metastasis surgery | Recurrence free survival | ||||||
---|---|---|---|---|---|---|---|---|---|
Hazard ratio | 95% CI |
|
Hazard ratio | 95% CI |
|
Hazard ratio | 95% CI |
|
|
Age | 1.054 | 1.02–1.09 |
|
||||||
Radioactive iodine Y v N | 0.164 | 0.041–0.654 |
|
||||||
Follicular subtype | 1.589 | 0.646–3.910 | 0.313 | 0.832 | 0.396–1.748 | 0.627 | 1.547 | 0.706–3.39 | 0.276 |
Skeletal presentation |
2.102 | 0.783–5.643 | 0.141 | 3.097 | 1.008–9.521 |
|
|||
Tumor excised from bone | 0.242 | 0.087–0.677 |
|
0.336 | 0.125–0.903 |
|
Kaplan-Meier analysis of patient survival. (a) Following surgery for osseous metastases patient survival was 72% at 1 year (95% CI 59–87%), 29% at 5 years (95% CI 17–49%), and 20% at 8 years (95% CI 10–42%). (b) There was better survival for patients when the metastastic osseous tumor was excised versus not excised. The median survival time for patients with tumor excision was 2 years (95% CI 1.4–5 years), compared with 0.6 years (95% CI 0.1–1.8 years) for patients without tumor excision (
In comparison, median survival after thyroidectomy in the same patient group was 5 years (range 0.8–8.5). Overall survival probability was 62% at 5 years and 35% at 10 years (Figure
The majority of patients (
Progression of disease in bone. (a) Kaplan-Meier analysis of time to formation of skeletal metastasis. The median time to formation of bone metastasis after thyroid removal was 3.44 years. (b) Kaplan-Meier analysis of time to recurrence. The probability of recurrence free survival was 89% at 1 year (95% CI 80–100%), 60% at 5 years (95% CI 37–96%), and 40% at 8 years (95% CI 16–100%). (c) A diagnosis of follicular thyroid carcinoma was associated with a higher risk of recurrence versus other diagnoses (papillary, medullary, anaplastic, and Hurthle cell) (
Serum thyroglobulin (TGB) levels were measured both pre- and postoperatively in 26 of 41 patients. In 22 patients, the serum TGB levels decreased following bone metastasis surgery. Preoperative serum TGBs ranged from 100 to 13000; with the percentage of decrease after bone metastasis surgery ranging from 9%–99%. In the 4 patients where there was no TGB decrease, 3 had widely metastasis disease involving multiple bones, and the skeletal surgery only addressed one site of bone involvement. The one remaining patient with no decrease in TGB had a biopsy, wherein the bone metastasis was not removed.
Eight of 41 cases were complicated by local recurrence. The local progression free survival was 89% at 1 year (80%–100%, 95% C.I.), 60% at 5 years (37%–96%, 95% C.I.), and 40% at 8 years (16%–100%, 95% C.I.) (Figure
In all cases of local progression, additional surgery was performed with hardware revision to either intercalary prosthesis, endoprosthesis, or joint replacement. One patient who was treated with an intramedullary nail for a humerus metastasis developed increasing pain, fracture, and progression of disease 4 years after surgery. The recurrent tumor was resected, and the nail was converted to a total humerus endoprosthesis (Figures
Kaplan-Meier analysis for recurrence free survival, as determined by the probability of no recurrence or death. (a) Patients had a higher probability of recurrence free survival if the osseous metastasis was excised versus no excision (
Patients treated with radioactive iodine (131I) had better survival following thyroidectomy (
Kaplan-Meier analysis of the effect of radioactive iodine treatment on survival. (a) After thyroidectomy, patients treated with radioactive iodine had more favorable survival than those who were not candidates for radioactive iodine (
Seven patients failed prior to palliative external radiation of osseous metastases and subsequently underwent surgical treatment. Preoperative external beam radiation to the affected bone had no significant effect on overall patient survival after skeletal surgery (Table
Two patients had immediate perioperative complications. One patient with a history of smoking had respiratory insufficiency requiring delayed extubation. One other patient had atrial fibrillation. Both subsequently recovered uneventfully. There were no superficial or deep infections, nonunions, or perioperative deaths.
The mean blood loss of 1014 mL (range 0–3900 mL) for the cases where the tumor was not removed (
The results of this study suggest that metastatic thyroid cancer in bone is rare, but behaves aggressively. Surgery for bone metastasis in thyroid carcinoma has been infrequently studied [
Thyroid cancer generally has an indolent nature, so that patients requiring skeletal surgery presented with bone metastasis both in late stages after treatment (16 of 41 patients) and with initial presentation (25 of 41 patients). Our study corroborates findings that survival with metastatic thyroid cancer tends to be better than some other cancers that metastasize to bone, such as renal cell carcinoma and lung carcinoma [
Despite the survival advantage afforded by more aggressive tumor removal, whether a patient underwent tumor excision did not affect whether there was local progression for patients with follicular carcinoma subtype. There was, however, considerable morbidity for patients who developed uncontrolled tumor growth at metastatic osseous sites. In our study, all of the patients with local recurrence required additional surgical procedures.
There was no survival advantage or disadvantage following thyroid cancer diagnosis or surgery for bone metastasis associated with subtype, although bone metastases are more common with the follicular subtype. Follicular thyroid cancer accounts for less than 15% of all differentiated thyroid cancers, but has a relatively high incidence of bone metastases, ranging from 7% to 28% [
Bone metastases are much less common in papillary thyroid cancer, with a reported incidence from 1% to7% [
Anaplastic thyroid carcinoma typically has a poor prognosis. It is known to be the most lethal among all thyroid cancers, with median life expectancies reported to be from 4 to 12 months [
Studies show that more than 80% of bone metastases from all thyroid tumors are located in the axial skeleton [
There are few studies on survival following surgery for bone metastasis [
Total thyroidectomy followed by radioactive iodine is the treatment most often recommended for patients with synchronous distant metastases to bone. It has been argued that 131I is the only opportunity to slow progression and to prolong survival [
The involvement of multiple bones precludes curative resection of bony disease. Of the 24 patients who had excision of the bone metastasis, 10 had solitary bone involvement. At the last followup 5 of the 10 were alive, 4 died, and one was lost to followup. Others studies have reported similar outcomes, with a significant improvement in survival with complete resection of skeletal metastases [
In patients with multiple organ involvement, the role of metastectomy is less well understood. In this study, two patients with solitary skeletal metastasis at presentation did not have surgical extirpation of the BM because of synchronous involvement in other organs. One of these patients had stabilization of a lesion in the radius with a plate, followed by external radiation. This patient was alive at 2-year followup, but went on to develop additional bone lesions in the clavicle and scapula. The other patient had synchronous lesions in the lungs and liver and a solitary lesion in the femur that was stabilized by intramedullary nail and treated with external radiation postoperatively. The patient died 8 months after surgery.
In one compendium of 13 studies, the rate of bone metastasis was found to be 25% among 1231 patients [
In conclusion, the development of osseous metastases from thyroid carcinoma is associated with poorer patient survival and prognosis. Patients with multiple bone metastases had poorer survival than those with solitary bone metastases. While 28% of patients died within a year of surgery for bone metastases, 20% remained alive at 8 years. Patients who responded to radioactive iodine had the best long-term survival. Multivariate analysis showed that both presentation with a single BM and complete BM resection surgery were significantly associated with improved patient and implant survival following surgery for skeletal metastases. Follicular thyroid cancer is the most common type associated with bone metastasis and recurs locally more often after surgical treatment. More work will be needed to fully understand and identify the factors that determine which patients survive longer and have greater need for local control. Thus, it remains necessary to allow for the surgeon’s judgment as to which patient should undergo tumor excision as a part of the metastatic tumor treatment.