| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Departments of Nuclear Medicine (M.-O.B., L.L., A.A.), Pathology (C.H.), General and Gastrointestinal Surgery (F.M.), Endocrinology (G.T.), Neuroradiology (J.C.), and Orthopedic Surgery (E.E., G.S.), Hôpital Pitié-Salpêtrière, 75013 Paris, France; Epidemiology and Information Science, INSERM, U-444, Université Paris 7 (J.-Y.M.), 75005 Paris, France; and Quantitative Medical Imaging Unit, INSERM, U-494, Hôpital Pitié-Salpêtrière (M.-O.B., L.L., G.H., A.A.), 75651 Paris, France
Address all correspondence and requests for reprints to: Dr. Marie O. Bernier, Service de Médecine Nucléaire, Hôpital Pitié-Salpêtrière, 47–83 boulevard de l’Hôpital, 75651 Paris Cedex 13, France. E-mail: gilles.hejblum{at}imed.jussieu.fr
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
3.7 gigabecquerels) in 95 cases.
Survival rates at 5 and 10 yr were 41% and 15%, respectively.
Univariate analyses indicated that a young age at BM discovery
(P < 0.005) and the discovery of BM as a revealing
symptom of DTC (P < 0.05) were features
significantly associated with improved survival as well as radioiodine
therapy (P < 10-4) and BM
complete surgery (P < 0.02). Using multivariate
analysis, the detection of BMs as a revealing symptom of thyroid
carcinoma (P < 0.0005), the absence of metastasis
appearance in other organs than bones during the follow-up
(P < 0.03), the cumulative dose of radioiodine
therapy (P < 0.0001), and complete BM surgery in
young patients (P < 0.04) appeared as independent
prognostic features associated with an improved survival. |
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Among 1977 patients with DTC treated and followed at the Department of Nuclear Medicine of the Groupe Hospitalier Pitié-Salpêtrière from 1958 to 1999, 135 presented BMs originating from DTC with a high probability. Patients were diagnosed during life, either at the time of their initial evaluation or during subsequent follow-up. The thyroid origin of the BM metastases was assessed either by positive 131I uptake in the metastasis area or by analysis of BM tissue after surgery. Therefore, 19 patients with no 131I uptake or no BM histological tissue were excluded from the study even when the thyroid origin of BMs was highly probable. Seven patients were lost to follow-up and were also excluded. Information concerning patient was obtained from medical record. Follow-up information and date of death were obtained either from the medical record or from town council registers. Finally, 109 patients were retrospectively included in our study.
The study population was composed of 77 females and 32 males. The median age at diagnosis of BMs was 63 yr (range, 20–87 yr). In all, 92 (84%) patients died. The follow-up for the 17 patients (16%) who remained alive at the end of the study was greater than 5 yr for 7 patients, among whom 4 underwent follow-up for more than 10 yr and 1 for more than 20 yr. To test the impact of diagnosis and treatment improvements along the time, the study was subdivided in 4 separate periods according to BM diagnosis: 1958–1970, 1971–1980, 1981–1990, and 1991–1999. Patient median age at diagnosis increased over the 4 decades from 59 yr in the period 1958–1970 to 69 yr in the period 1991–1999. All patients except 14 (see below, radioiodine ablation of thyroid remnants and therapy) were subjected to the same protocol of treatment: total or near-total thyroidectomy completed by an ablative 131I dose of 3.7 gigabecquerels (GBq; 100 mCi). After treatment, all patients were subjected to T4 treatment at suppressive dose.
Surgical treatment of primary tumor and corresponding histology
Total or near total thyroidectomy was performed in 108 patients (99%), and cervical lymph node partial or total dissection was performed in 32 patients (29%). Information on histological type and differentiation was obtained from pathology reports and was not pathologically reviewed. The histological diagnosis of the thyroid tumor, available in 96 cases (88%), was papillary in 19 cases (17%) and follicular in 77 cases (71%). No tumor tissue (and therefore no histological diagnosis) was available for 13 patients (12%): 12 patients had previously undergone partial thyroidectomy elsewhere 1–41 yr before our observation, with no initial diagnosis of thyroid carcinoma; 1 patient was not subjected to surgery. The tumor median diameter, available in 79 cases of 96, was 3.3 cm (mean, 4.2 cm; range, 0.1–12 cm). Among these cases, 8 concerned tumors with a diameter less than or equal to 10 mm. A thyroid capsule invasion was noticed in 27 of 83 available cases.
Radioiodine ablation of thyroid remnants and therapy
After primary thyroid surgery, 14 patients (13%) were not subjected to an ablative 131I dose of thyroid remnants for various reasons: premature death due to surgery before the radioiodine treatment (3 cases), loss of autonomy related to the pathological state (2 cases), risk of aggravation of neurological symptoms (4 cases), no thyroid surgery (1 case), too advanced age to support withdrawal of T4 treatment (1 case), overloading of iodine (1 case), and bad health state (2 cases). At the end of the study, 1 of the 14 patients was still alive, and the longest follow-up among this group was 3.17 yr. The 95 remaining patients received either a single 131I ablative dose [23 patients received a dose of 3.7 GBq (100 mCi) and 3 patients received a dose of 7.4 GBq (200 mCi)] or an iterative 131I therapy (69 patients) for BM metastases, as assessed by periodic (6–18 months) 131I whole body scanning (WBS) after withdrawal of T4 treatment. The therapeutic dose of 131I ranged from 3.7–7.4 GBq, and the cumulated dose received by patients varied from 3.7–44.4 GBq (median, 14.8 GBq).
BM diagnosis and BM surgery
BMs were detected by plain roentgenogram, WBS, computerized tomography, or magnetic resonance imaging. All patients presented positive 131I uptake in bone(s), and/or symptoms indicative of skeletal disease, and/or an elevated thyroglobulin (Tg) level (dosage of Tg is only available since 1982). In the text, complete BM surgery refers to a surgical procedure that allowed macroscopic removal of any bone tumoral tissue, with pathological confirmation. Palliative surgery refers either to partial removal of bone tumoral tissue (i.e. removal of a single BM when several BMs were present) or total removal of BMs but leaving tumoral tissue in other organs. The diagnosis of BMs originating from DTC was confirmed at histology for 84 patients (77%).
Criteria of remission
Patients were considered free of disease if one of the two following criteria was fulfilled: 1) after withdrawal of T4 treatment, a) for patients for which a WBS was available, no abnormal fixation on WBS and either Tg level below 10 ng/mL (since 1982) or normality of standard radiography (before 1982), b) for patients for which WBS was unavailable, Tg level less than 10 ng/mL and normality of standard radiography; and 2) under T4 treatment, normality of standard radiography and no detectable Tg.
Statistical analyses
Survival curves were estimated using the Kaplan-Meyer method
(17). We considered initial time as the date of BM
diagnosis and final time as either the date of death (due to thyroid
cancer or any other cause) or the date of last news from survivors
(censored cases). The following variables were studied: sex,
category of age at the diagnosis of BM (<45 or 45 yr
old), number and site of BMs, discovery of BM as a revealing symptom of
thyroid carcinoma, discovery of BM exclusively during follow-up, period
of BM diagnosis, orthopedic complications, other metastasis sites
present at the initial BM diagnosis, other metastasis sites appearing
exclusively during follow-up (handled as a time-dependent variable)
either in all other organs than bones or exclusively in lungs, BM
surgery (yes/no), complete BM surgery (yes/no), other therapies of BMs
(external radiotherapy and selective arterial embolization),
radioiodine status (radioiodine uptake of metastases), the total
radioiodine therapeutic dose received was handled as a three-category
variable: 1 (reference), dose of 7.4 GBq or less; 2, no therapeutic
dose; and 3, dose greater than 7.4 GBq. The prognostic value of each
variable for survival was studied separately in a univariate analysis
using the Cox proportional hazards model (18). A variable
with P < 0.05 was considered significantly associated
with the survival function. Variables in the univariate analysis
presenting P 
0.20 were then entered into a multiple
regression Cox analysis with backward elimination of variables to
identify a small set of variables with independent prognostic
significance. The age of the patient reflects in part his state of
health. Consequently, the decision to perform a complete BM surgery is
likely to be dependent on patient age. Therefore, we included in the
multivariate analysis a term of interaction between these two
variables. All analyses were performed using SPSS statistical software
(SPSS, Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. Each of the prognostic variables was
tested for its association with survival by separate univariate
analyses. The results of the analyses are shown in Table 1 (nonsignificant prognostic variables)
and Table 2 (significant prognostic
variables).
|
|
|
The overall survival rates (all causes) at 5, 10, and 20 yr were,
respectively, 41%, 15%, and 7%, with a median survival of 3.9 yr
(mean, 5.6 yr; Fig. 1
). There was no significant difference in survival
among the four selected periods of this study even after adjusting for
age at BM diagnosis (Table 1). Survival appeared significantly improved
in patients aged less than 45 yr at the time of BM diagnosis (Table 2).
According to the criteria defining patients free of disease (see Materials and Methods), three living (after T4 treatment withdrawal Tg level was undetectable in two cases and was equal to 5 ng/mL in one case) and one dead (after T4 treatment withdrawal Tg level was undetectable) patients (4%) were without evidence of any BM.
Distribution and treatment of initial BMs
The overall prevalence of BMs in our series was 109 of 1977
patients (5.5%). BM discovery was the revealing symptom of DTC for 68
patients (62%), and this feature was significantly associated with an
improved survival (Table 2). BMs were discovered in 15 patients (14%)
at the first WBS. BMs were discovered exclusively during the follow-up
in 26 patients (24%), with a median period of 56 months (range, 4–216
months), and this feature was significantly associated with reduced
survival (Table 2). The distribution of sites involved with BMs at the
time of initial BM diagnosis is shown in Table 1. Orthopedic
complications were observed in 55 patients (51%; Table 1):
pathological fractures were observed in 14 cases (13%), radiculalgia
in 4 cases (4%), spinal cord compression in 30 cases (28%), and both
spinal cord compression and pathological fracture in 7 cases (6%). As
shown in Table 3, positive BM
131I uptake was observed in 80 cases (73%; see
also Table 1). Only 4 patients presented both positive radioiodine
uptake at the WBS and negative BM detection at standard x-ray
examination.
|
).
As shown in Table 2, 131I therapy was associated
with a significant improved survival, and the improvement was also
dependent on the cumulative dose received by the patient.
External radiotherapy was performed in 39 cases (36%). Selective arterial embolization concerned 34 patients (31%), and was performed as a preoperative procedure and/or as a palliative treatment. Two patients received chemotherapy (2%).
Sites of other metastases
The presence of metastases in other organs than bones was noticed in 53 patients. In all, 76 occurrences of metastases in other organs than bones were encountered. The lung was the most frequent site of metastases and concerned 37 patients, 22 of them presenting metastases only in lungs. Local recurrence in the thyroid area (12 occurrences) and mediastinal (8 occurrences), cerebral (7 occurrences), hepatic (5 occurrences), cutaneous (3 occurrences), retroorbital (2 occurrences), pleural (1 occurrence), and renal (1 occurrence) localizations were also observed.
As shown in Table 2, metastases in organs other than bones, either
present at the time of BM diagnosis or appearing exclusively during
follow-up, were features significantly associated with reduced
survival. In contrast, metastases exclusively in lungs (either present
at the time of BM diagnosis or appearing exclusively during follow-up)
were not significantly associated with survival (Table 1).
Prognostic variables: multivariate analysis
The results of the Cox multivariate analysis are summarized in
Table 4. BM as a revealing symptom
(P < 0.0005) and the cumulative dose of
131I therapy (P < 0.0001), were
variables that remained significantly associated with improved
survival. The appearance of metastases in other organs than bones
exclusively during follow-up was associated with reduced survival
(P < 0.03). Complete BM surgery and category of age
did not remain significant variables by themselves, but a term of
interaction combining these two variables was associated with survival;
the analysis indicates that patients less than 45 yr old who were
subjected to complete BM surgery had significantly improved survival
than all other patients (P < 0.04).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The overall prevalence of BMs in patients with DTC observed in our department (5.5%) is in the range of those previously reported in other series (2.3–12.7%) (4, 6, 9, 12, 14, 16, 19, 20). Differences in the patient’s referral pattern and in the diagnosis procedures used to detect BMs may account for the discrepancies between the studies. The fact that 8 of 109 cases included in the present study concerned tumors with a diameter smaller than or equal to 10 mm raises some comments. Thyroid microcarcinomas are known to follow a benign course, but description of fatal course or systemic metastases of such tumors has been reported (21, 22). In our series the responsibility of microcarcinomas is highly probable in 6 cases (2 papillary cases and 4 follicular cases), as the corresponding patients had been subjected to a total thyroidectomy. In 2 cases, the responsibility of microcarcinomas for the arising BM cannot be clearly established because of a possible undetected tumor removed with the initial partial thyroidectomy realized prior to our observation. Because of the BM discovery, both of the above patients were then subjected to complete thyroidectomy, and papillary microcarcinoma was discovered.
The mean patient age at the discovery of BMs (61 yr) was similar to that reported in other series (5, 6, 9, 11, 20) and higher than that of all patients with DTC, whose peak incidence occurs between 45 and 50 yr (3). The increased aggressiveness of thyroid cancer in old patients and the delay of appearance of distant metastases in DTC may explain this age difference. In our study young age at the diagnosis of BMs appeared as a prognostic factor of improved survival, as also reported in other studies (8, 10, 11, 12).
The distribution of BM sites in our series was similar to that observed by others (4, 9). Multiple BM sites occurred frequently, and BMs were also frequently associated with metastases in other organs than bones, especially in the lung, as reported in other studies (8, 9, 12). Our multivariate analysis indicated that the appearance of metastases in other organs than bones exclusively during follow-up was associated with reduced survival.
The cumulative survival rates from all causes were 41% and 15% at 5 and 10 yr, respectively. However, not all DTC deaths occurred early after BM diagnosis; one patient survived 16 yr after BM diagnosis before succumbing to thyroid cancer. Several studies have reported 10-yr overall survival rates varying from 13–33%, considering the initial date as the date of distant metastasis diagnosis (6, 9, 11, 12, 16, 23, 24). When considering exclusively BM survival, the overall survival at 10 yr was reported to range from 13–21% (9, 12, 16, 23, 24). Differences in patient age and BM treatments in the various studies may contribute to the observed differences in mortality from DTC. In our series, patients for whom BM discovery was the revealing symptom of a DTC presented improved survival compared with all other patients (i.e. the group composed of patients with BMs either diagnosed at initial WBS after thyroidectomy or appearing exclusively during the follow-up). It may be hypothesized that a BM developing after initial radioiodine treatment corresponds to a more aggressive evolution of the disease than a BM revealing a DTC, the latter BM being related to radioiodine-untreated patients. Even if such a hypothesis is plausible, it has to be pointed out that in studies such as ours, the additional survival time of patients for whom the diagnosis of BM is made after the diagnosis of DTC is not taken into account, as the initial time of the survey is considered the date of BM discovery.
An analysis of the literature on several large series of patients (4, 5, 6, 8, 9, 11, 12, 14, 16, 23, 24, 25, 26, 27) does not provide clear information on the impact of the different BM therapies. Some studies showed no impact of radioiodine therapy on survival (6, 8, 11, 16), whereas others found a valuable effect of such therapy (12, 23, 27, 28, 29, 30, 31). Prospective randomized studies are unavailable; therefore, the understanding of DTC natural history, a rare disease, relies only on retrospective studies. However, confounding bias may obscure the statistical analyses, because prognostic variables are interrelated. Radioiodine uptake and radioiodine therapy are such variables, because almost all patients with positive radioiodine uptake, and only such patients, will be subsequently subjected to radioiodine therapy. A multivariate analysis becomes essential to understand combinations of potential prognostic factors that might predict survival in such patients. Only five series studying prognostic factors of survival in the case of distant metastases used a multivariate analysis model (8, 10, 11, 12, 16), and among these, only one considered exclusively BMs (16). Three studies found a significant association between radioiodine uptake and survival (10, 12, 16), and one did not (8). One study did not include radioiodine uptake status in the multivariate analysis (11). It is difficult to compare the results of the above studies, because the corresponding patient populations are different. Indeed, Pittas et al. (16) refer to patients with BMs originating from both undifferentiated and differentiated thyroid carcinomas, and survival is known to be worse in patients with undifferentiated carcinomas (32, 33, 34), for which a negative radioiodine uptake is likely. The other studies refer to patients with metastases originating from DTC, but pooled lung and BM metastases. In our study radioiodine uptake was not significantly associated with survival. This observation may be due to a lack of statistical power. Beside sample size, the value of radioiodine uptake may appear reduced because 9% of the patients with positive radioiodine uptake were not subjected to radioiodine therapy because of their bad clinical state. Contrary to radioiodine uptake, radioiodine therapy had a significant impact on survival in our study (P < 0.0001). Survival also significantly increased with the cumulative 131I dose received. Our results contrast with the multivariate analyses previously reported, where such a therapy was not significantly associated with better survival (8, 11, 16) even if the univariate analysis showed a significant association between survival and radioiodine therapy in these three studies (8, 11, 16). It could be hypothesized that radioiodine therapy was significantly beneficial by univariate analysis, but that it was not so strong in multivariate analysis to outweigh other variables. Other studies indicated an association between survival and radioiodine therapy (10, 12); this affirmation was based only on the association of positive radioiodine uptake with survival.
Concerning the impact of bone surgery, in our univariate analysis a BM complete surgery was associated with improved survival, whereas surgery from a more general point of view (i.e. considering both patients subjected to complete surgery and those subjected to palliative surgery), was not. Our results indicate that the survival of the group of patients subjected to palliative surgery is intermediate, neither significantly different from the group of patients subjected to complete surgery nor from the group that did not receive any surgery. This result is probably due to a lack of statistical power. Many features regarding patient health state are taken into account by the surgeon for guiding the type of BM surgery performed. Among these, age is likely to reflect at least in part patient health. As shown in our multivariate analysis, although complete BM resection and age category were both significant variables in the univariate analyses, they remained significant in the multivariate analysis only in a term of interaction. This result indicates that the benefit of complete BM surgery is different according to the age category of the patient; younger patients subjected to complete BM resection had significantly improved survival compared with all other patients. Several studies have reported the importance of BM complete resection on survival (9, 14, 15, 24, 35). To assess more precisely the benefits of different surgery strategies, other studies on this topic should be performed, even in a retrospective way, and should take into account patient health state. A grading clinical score has been proposed and validated in the case of vertebral metastases (15). Surgical strategy could then be evaluated according to such a clinical score. Such a strategy, when adapted to all BM sites originating from DTC, could result in a more standardized decision, especially concerning the aggressiveness of the surgery, and might modify patient outcome.
In our study external radiotherapy was not associated with improved survival, although the association of external radiotherapy and radioiodine therapy has been reported to have an impact on cancer recurrence, pain relief, and recalcification of osteolytic abnormalities (30). Selective arterial embolization, reported previously as effective on pain and hemorrhage risk occurring during metastases surgery (36, 37, 38), did not appear significantly associated with improved survival in our study.
Our results suggest that recent advances in the diagnosis and management of this disease have a moderate impact on the average survival of patients with BMs originating from DTC. When splitting the course of the study into four periods, we did not detect a significant difference between the survival rates, even after adjusting for age at diagnosis. Complete BM surgery, used in our series since 1980, was expected to have a more pronounced impact on survival. In fact, two major problems need to be taken into consideration. First, simultaneous involvement of several bones was observed in 52% of the patients and usually impaired complete resection of BMs. Second, no surgery was preferred in 37% of the nonsymptomatic patients, especially if patients were old. In the study by Schlumberger et al., the survival of patients with BM diagnosis after 1976 was improved compared with survival of patients with BM diagnosis before 1960 (12). In contrast, Dineen et al. reported a patient outcome not significantly improved over 5 decades (1940–1989) (11). The introduction of selective arterial embolization and cementoplasty as alternative therapies in cases of impossible complete surgery have shown encouraging results for pain (36, 37, 38), but no investigation studying the impact of such therapies on survival has been yet reported. In our series one patient was subjected to iterative selective arterial embolizations with a good result on tumor growth, as shown by the decrease in Tg level after each embolization. However, the absence of long-term evaluation together with the low number of patients subjected to such therapies may explain the absence of significant impact on survival.
In summary, as assessed by multivariate analysis, the detection of BMs as a revealing symptom of thyroid carcinoma, the absence of metastasis appearance in organs other than bones during the follow-up, the cumulative dose of 131I therapy, and complete BM surgery in patients less than 45 yr old appeared as independent prognostic features associated with improved survival in patients with BMs originating from DTC. The surgery strategies need further studies that should especially focus on the dependent initial clinical state of the patients. New techniques, such as selective arterial embolization or cementoplasty, should be more investigated. Our data indicate that the therapeutic strategy should be as aggressive as possible, especially in young patients.
Received July 6, 2000.
Revised December 8, 2000.
Accepted December 19, 2000.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M M Muresan, P Olivier, J Leclere, F Sirveaux, L Brunaud, M Klein, R Zarnegar, and G Weryha Bone metastases from differentiated thyroid carcinoma Endocr. Relat. Cancer, March 1, 2008; 15(1): 37 - 49. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Hindie, P. Zanotti-Fregonara, I. Keller, F. Duron, J.-Y. Devaux, M. Calzada-Nocaudie, E. Sarfati, J.-L. Moretti, P. Bouchard, and M.-E. Toubert Bone metastases of differentiated thyroid cancer: impact of early 131I-based detection on outcome Endocr. Relat. Cancer, September 1, 2007; 14(3): 799 - 807. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Durante, N. Haddy, E. Baudin, S. Leboulleux, D. Hartl, J. P. Travagli, B. Caillou, M. Ricard, J. D. Lumbroso, F. De Vathaire, et al. Long-Term Outcome of 444 Patients with Distant Metastases from Papillary and Follicular Thyroid Carcinoma: Benefits and Limits of Radioiodine Therapy J. Clin. Endocrinol. Metab., August 1, 2006; 91(8): 2892 - 2899. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Pacini, M. Schlumberger, H. Dralle, R. Elisei, J. W A Smit, W. Wiersinga, and the European Thyroid Cancer Taskforce European consensus for the management of patients with differentiated thyroid carcinoma of the follicular epithelium. Eur. J. Endocrinol., June 1, 2006; 154(6): 787 - 803. [Full Text] [PDF]
|
|
|
|
|
|
A. Rouxel, G. Hejblum, M.-O. Bernier, P.-Y. Boelle, F. Menegaux, G. Mansour, C. Hoang, A. Aurengo, and L. Leenhardt Prognostic Factors Associated with the Survival of Patients Developing Loco-Regional Recurrences of Differentiated Thyroid Carcinomas J. Clin. Endocrinol. Metab., November 1, 2004; 89(11): 5362 - 5368. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
