| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Centre Hospitalier Universitaire (J.L.W., M.N., P.P.), 59037 Lille, France; Centre Hospitalier Universitaire Rangueil (P.C.), 31403 Toulouse, France; Centre Hospitalier Universitaire (A.B.), 400-35, Liège, Belgium; Centre Hospitalier Universitaire (V.R.), 49033 Angers, France; Centre Hospitalier Universitaire (J.O., F.B.-C.), 69495 Lyon, France; Centre Hospitalier Universitaire Saint Pierre, 1000 Bruxelles, Belgium; and Novartis Pharma SAS (S.B.-L., I.B., S.D.), 92506 Rueil Malmaison, France
Address all correspondence and requests for reprints to: Prof. J. L. Wémeau, Clinique Endocrinologique Marc Linquette, Centre Hospitalier Régional Universitaire, 59037 Lille Cedex, France. E-mail: jl-wemeau{at}chru-lille.fr.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Only a weak consensus exists on either the evaluation or the management of GO. Two systems are routinely used for evaluation. The NOSPECS index (N, no signs or symptoms; O, only signs, no symptoms; S, soft tissue involvement; P, proptosis; E, extraocular muscle involvement; C, corneal involvement; S, sight loss) gives a global estimate of the severity of the various symptoms, although it has been criticized for providing only a mean assessment, so that many experts propose evaluating the various symptoms separately (10). It was partly in response to this type of criticism that Mourits et al. (11) developed the Clinical Activity Scale (CAS), which gives a measure of disease activity.
It is generally accepted that corticosteroids and/or orbital radiotherapy should be considered as treatments for established moderately severe to severe GO, both of which are associated with some adverse reactions (12). Moreover, a successful outcome is far from certain (7, 13, 14). There exists, therefore, a genuine need for new effective and safe modalities that can be used in the early stages of GO to prevent deterioration to a more severe form.
Somatostatin is an endogenous cyclic peptide that has broadly inhibitory activity on a variety of different systems (15). Many different cell types, including fibroblasts and lymphocytes, are known to express somatostatin receptors, and the level of expression of high affinity somatostatin receptors in retrobulbar tissue is known to correlate with the clinical activity of GO (6). The rationale for using somatostatin analogs to treat GO is based on the putative capacity of these peptides to inhibit lymphocyte activation, proliferation, and cytokine production (16, 17). Moreover, somatostatin is known to inhibit the release of IGF, the production of which by fibroblasts has been implicated in the pathogenesis of GO (18), even if serum levels of IGF are normal in the active forms of GO, suggesting intraorbital overproduction (19). A number of preliminary open studies have suggested that somatostatin analogs might be effective against GO. In these studies the main effect of octreotide or lanreotide was the attenuation of soft tissue and extraocular muscle involvement in a majority of the patients treated (in most cases for a period of 3 months) (20, 21, 22), although in few open studies a beneficial effect was also observed on proptosis (23, 24).
However, the studies conducted to date have addressed highly disparate patient populations at different stages of the disease, they have lacked appropriate controls, and the numbers studied have been small, so any conclusions to be drawn from these studies are necessarily provisional (18). Because a significant proportion of patients with GO improve spontaneously, clinical trials designed to test the efficacy of new treatments in thyroid-associated ophthalmopathy should be scrupulously controlled to allow for the natural tendency toward remission (25).
Hence, there was a need for a randomized, placebo-controlled, double-blind study to investigate the efficacy of somatostatin analogs in the treatment of GO. In this study it was decided to use the long-acting release formulation of octreotide (octreotide-LAR) in patients with mild, active GO with the aim of preventing additional deterioration and precluding the need for more aggressive therapeutic modalities, such as glucocorticoids or radiotherapy.
| Subjects and Methods |
|---|
|
|
|---|
A double-blind, placebo-controlled, randomized study was conducted. Patients were recruited and treated at 13 centers (10 in France and three in Belgium). Randomization on a 1:1 basis to either octreotide-LAR or placebo was centralized and was stratified by sex and current smoking status (smoking being defined as the consumption of more than five cigarettes per day).
Between inclusion and the beginning of treatment, an 8-wk interval was programmed to confirm the stability of euthyroidism. After baseline monitoring on the first day of treatment, patients were examined at 4-wk intervals throughout the 16-wk treatment period and then again in a follow-up evaluation 6 months after the beginning of treatment. Adverse events occurring at any time in the 6-month study period were recorded and analyzed.
Patients
Eligible for the study were euthyroid outpatients with clinically active, mild to moderate GO. The diagnosis of GO was based on the presence of at least two of the following signs: retraction of the eyelid (a palpebral fissure wider than 11 mm or a distance of >7 mm between the edge of the lower eyelid and the center of the pupil), palpebral edema, proptosis, and impaired ocular mobility with increased intraocular pressure on upgaze associated with enlargement of at least one oculomotor muscle. Mild to moderate disease severity was defined using the NOSPECS scale as either minimal to moderate soft tissue involvement (class 2, grades ac) or minimal to moderate soft tissue involvement (grade 2a or b) coupled with mild proptosis (grade 3a), mild extraocular muscle involvement (grade 4a), or mild corneal involvement (grade 5a). Clinically active disease (based on eye pain, ocular or eyelid erythema, conjunctival or eyelid edema, and worsening of proptosis) was defined as a score of 3 or more on the 10-item CAS proposed by Mourits et al. (11). Stable euthyroidism was defined by blood concentrations of free T4, free T3, and TSH in the normal range throughout the study period.
Exclusion criteria included severe refractive abnormalities or any other eye problem that might interfere with the accuracy of the study measurements, gallstones, or a history of treatment within the preceding 6 months with systemic corticosteroids, immunosuppressive drugs, radiotherapy, or chemotherapy.
Study treatment
Octreotide-LAR (Sandostatin LAR, supplied by Novartis, Rueil Malmaison, France) was supplied in vials containing 30 mg lyophilisate for reconstitution to a volume of 2 ml. Treatments (octreotide-LAR or placebo) were administered at 4-wk intervals over 4 months (d 0 and wk 4, 8, and 12) by im injection.
Concomitant medications
Stable euthyroidism was a formal requirement in the protocol, so exogenous T4 at the appropriate dosage was permitted throughout the study period, as were antithyroid drugs (carbimazole, benzylthiouracil, and propylthiouracil), as long as the dosage of the latter was not significantly modified during the study. Also permitted were topical products (artificial tears, ocular lubricants, and antibiotic eye drops). Formally excluded were radiotherapy, chemotherapy, iv immunoglobulin, immunosuppressants, anticoagulants, antiplatelet drugs, oral iodine/iodide, lithium, and amiodarone.
Study assessments
The main end point was a binary success/failure criterion based on changes in NOSPECS classification and CAS between baseline (d 0) and the end of treatment (wk 16). Success was defined as a decrease in NOSPECS class or grade only when CAS was unchanged or decreased; failure was defined as either an increase in CAS or an absence of change or an increase in NOSPECS class or grade. If changes were recorded in two different NOSPECS classes, the higher change was used to define the result.
In addition to the main composite efficacy end point, in a secondary analysis the groups were compared on the basis of overall changes in CAS and in individual items of the ophthalmological examination, with special attention paid to objective parameters (notably proptosis, as measured using the Hertel exophthalmometer, and palpebral opening and retraction). Furthermore, self-assessment was evaluated with two validated quality of life questionnaires: one global, the SF-36 (26), and the other specific to GO, the GO-QOL (27).
Thyroid function (blood levels of TSH, T4, and T3) was monitored throughout the study period, and pertinent serological parameters (antibodies against thyroglobulin, thyroperoxidase, and the TSH receptor) were assayed at baseline and the end of treatment. All serum measurements were performed in each center with commercial kits.
Somatostatin receptor scintigraphy was performed at baseline using a
-camera to acquire data on the anterior head and both profiles 4 and 1824 h after the iv injection of [111In]pentreotide (220 MBq, 10 µg; octreoscan, Mallinckrodt, St. Louis, MO). The images were centrally interpreted by two independent observers blinded to patients data. A semiquantitative four-point score was used to classify the intensity of uptake, as previously reported (28, 29): negative scintigraphy, uptake indiscernible from background noise; ±, weakly positive scintigraphy, uptake hardly to slightly discernible from background noise at 4 or 24 h; +, moderately positive scintigraphy, moderate uptake at 4 and 24 h; and 2+, strongly positive scintigraphy, strong uptake at 4 h with enhancement at 24 h. In the event of conflicting interpretations, the two observers had to meet to agree to a consensus. The objective was to investigate the reported correlation of this parameter with the CAS (30, 31) with a view to evaluating the predictive power of this examination vis-à-vis the efficacy of octreotide treatment (32). In normal individuals, there is no orbital uptake of octreoscan.
Extraocular muscle involvement, an almost typical sign in GO, was specifically investigated at baseline and at month 4 by nuclear magnetic resonance (NMR) measurement of the lateral, medial, inferior, and superior rectus muscles (33, 34, 35) with a T1-weighted and a T2 gradient echo sequence. Results were read centrally by an independent expert and analyzed for correlations with other parameters, notably baseline scintigraphic data and CAS, antibody titers, and outcome.
For the safety analysis, details of all adverse events occurring at any time in the 6-month study period were collected.
Sample size and statistical analysis
Assuming a success rate of 30% in the placebo group and a difference of 35% in the treated group compared with placebo, it was calculated that 37 patients/treatment group had to be recruited to guarantee a power of 0.80. However, due to the very slow recruitment, the study had to be stopped after 51 patients had been randomized (inclusion duration of 18 months).
The main efficacy analysis was based on the intention to treat population (all patients who received at least one dose of the test treatment and for whom pretreatment and end of treatment data were available for at least one efficacy parameter) and comparison of the baseline reading with the last measurement during treatment. Parallel analyses conducted on the per protocol population gave similar results. All statistical analyses were two-sided, with a critical significance level of 5%. For the main end point, the effect of treatment was analyzed using the Mantel Haenszel test adjusted for sex and smoking status. Because stratification of randomization according to sex and smoking resulted in unbalanced distribution between the two groups at certain centers, no center-adjusted analysis was attempted. For the secondary end points discussed, changes in the two groups between baseline and the end of either treatment or follow-up were compared using the Wilcoxon two-sample test. All relevant ophthalmological parameters were evaluated for both eyes and were analyzed in two ways: the mean of both eyes, and with respect to the most severely affected eye at baseline (as defined by the exophthalmometer measurement). Frequency tables were compiled for adverse events, classified according to the standard WHO-ART Body System Dictionary and preferred terms. All laboratory results out of the normal range were listed.
Ethics
All patients gave written, informed consent to participate, after having been judged able to understand and comply with the requirements of the study, which was conducted in accordance with the Helsinki Declaration concerning medical research on human subjects (as amended) and European Union Directive 91/507/EC. The protocol was preapproved by the ethics committee of Lille University Hospital.
| Results |
|---|
|
|
|---|
Enrolment began in January 2001, with the last patient completing final evaluation in June 2003. Fifty-one patients were recruited. The patient populations are described in Table 1
.
|
The study population was representative of the GO population with respect to all key parameters, namely gender, age, body weight, and the preponderance of GD of recent onset. The orbitopathy in this population was of relatively long standing (Table 2
), but was still active (CAS,
3; mean, 4.2 ± 1.6). As shown in Table 2
, the stratified centralized randomization process generated two broadly comparable treatment groups. There were no significant differences between the two groups.
|
No significant treatment effect emerged from the analysis of the main composite efficacy end point (Table 3![]()
![]()
) regardless of whether the analysis was conducted on the intention to treat or the per protocol population. Considering the components of the primary end point independently, a small improvement was observed in CAS for patients treated with octreotide-LAR and also in the group receiving placebo, without any significant difference between the two groups. No significant change was detected in either group in any of the following ophthalmological parameters: palpebral edema, oculomotor function, tear secretion, diplopia, ophthalmoscopy findings, and intraocular pressure. There was a marginal tendency toward diminished soft tissue involvement, but this was not associated with any difference between groups (not shown).
|
|
|
|
|
The semiquantitative results of orbital [111In]pentetreotide uptake did not show any significant correlation with the baseline CAS (r = 0.21; P = 0.147), nor was any correlation observed between the strength of the initial signal and the response to treatment (Table 4
). It is worth noting that 46% of the patients in the octreotide-LAR group and 40% in the placebo group produced no signal. Baseline MRI detected a T2 hypersignal in only seven of 25 patients in both octreotide-LAR and placebo groups. The presence of T2 hypersignal correlated significantly with [111In]pentetreotide uptake (P = 0.032); there was also some, but a nonsignificant, correlation with the CAS (mean score, 5.00 ± 2.08 in the T2+ patients vs. 4.09 ± 1.04 in the T2 patients; P = 0.21).
|
The levels of TSH and free T4 remained stable in both groups throughout the study period. The level of antibodies directed against the TSH receptor decreased significantly in both groups (P < 0.005), with no significant difference between the groups (P = 0.35). Antibody levels did not correlate with any inflammatory parameter related to disease activity (CAS or MRI T2 signal).
Safety
Octreotide-LAR induced the expected profile of minor side-effects: essentially temporary, mild gastrointestinal perturbations (diarrhea, abdominal pain, nausea, and constipation). Nevertheless, a serious adverse event attributed to the treatment was one case of gallstones that was confirmed on coelioscopic cholecystectomy; the pretreatment ultrasound examination carried out on this patient 9 months before the study had revealed no abnormality.
| Discussion |
|---|
|
|
|---|
Mention should be made of possible bias in the study. 1) There were more active smokers in the octreotide-LAR than in the placebo group. 2) More patients in the placebo group had not received any form of treatment in the past for GD, suggesting that their disease might have been of more recent onset. 3) Four patients in the octreotide-LAR group had been previously treated with radioactive iodine compared with none in the placebo group. 4) As a result of different strategies for the management of GD in the various centers, the protocol requirement for stable euthyroidism throughout the duration of the study was violated for a short while in four cases, all in the octreotide-LAR group. All of these factors are known to exacerbate GO or compromise the outcome of treatment: smoking (36, 37), stage of disease (38), iodine radiotherapy (39), and unstable thyroid function (25). Furthermore, the disease was of relatively long standing in those patients with a mean history of almost 2 yr and a median of 12 months. Probably the longer the duration of GO, the weaker the inflammatory activity of the disease (40). Although a CAS of more than 3 was a prerequisite for inclusion, the fact that a large proportion of the patients in both groups may have been at a relatively late stage of the disease is also supported by both the baseline scintigraphic data, with 43% of the patients giving a completely negative signal, and the T2-weighted MRI scan results, with only 28% of patients giving a positive signal at the beginning of the study.
Although correlation has been reported between [111In] pentetreotide uptake and the presence of a MRI T2 hypersignal, we failed to observe any such correlation between uptake and either baseline CAS (21, 32, 41) or response to treatment with somatostatin analogs (42). In previous studies the populations were probably at a more active stage of the disease, as evidenced by scintigraphic data (29).
Somatostatin is a pleiotropic effector molecule with characterized effects on thyroid function (43). In this study no significant differences were observed between the treated and control groups in a variety of parameters of thyroid function, notably the concentrations of T4, T3, and TSH in the blood. Predictably, no effects of octreotide-LAR on thyroid function could be detected in this study, due to the stronger effects of the antithyroid drugs. Moreover, levels of antibodies directed against the TSH receptor decreased in a significant fashion in both groups, with no difference observed between them. In contrast to a prior observation (6), there was no correlation between antibody levels and any of the inflammatory parameters related to disease activity (CAS or MRI T2 signal).
Despite the lack of any difference between the two groups with respect to the main efficacy end point, a significant effect of treatment was observed in one key symptom of GO, namely proptosis. Proptosis, which is particularly amenable to objective measurement, was mildly decreased in the treated group, but not in untreated patients. The effect at the end of treatment was statistically significant for both the mean of the two eyes and the most severely affected eye alone, and the improvement was sustained through 2 months after the end of treatment. In a study such as this in which a great number of different individual variables are tested (with the ophthalmological examination alone generating five objective variables and five other analyzable variables), the dangers of multiple testing are an important issue (44). If 10 variables are tested with a critical significance level of 5%, it is likely that a significant difference will be observed in at least one of them even if the null hypothesis is true. However, in this study strong circumstantial evidence that the beneficial effect of treatment on proptosis is indeed a real one is provided by the concordance of a whole series of results for other, closely related parameters, namely NOSPECS class III grade, opening of the upper eyelid, and the difference in ocular pressure between primary position and upgaze. Furthermore, MRI dimensioning detected a tendency toward diminished soft tissue involvement, the process that underlies proptosis, during and after octreotide-LAR treatment. The total area corresponding to the cumulate surface of the intraorbital muscles was reduced, but not significantly. This suggests that the reduction of proptosis arose from a reduction of both muscular and adipose intraorbital contents.
None of these other individual differences is significant, but all are in the same direction, pointing to real therapeutic efficacy vis-à-vis proptosis, one of the most clinically significant symptoms of GO and certainly the most debilitating in cosmetic and, therefore, psychological terms. In agreement with these data, the double-blind, placebo-controlled trial of octreotide-LAR in thyroid-associated orbitopathy conducted in Great Britain and Germany also mentioned a modest benefit of the somatostatin analog, but it was only significant on exophthalmos (45). The slight effectiveness of octreotide-LAR might be dependent upon its restricted specificity [somatostatin receptor 2 (SstR2) and SstR5]. Because other SstR subtypes are also expressed in resident and infiltrating cells in the orbit of GO patients (17), the potentiality of new universal analogs, such as SOM 230, has to be tested in this condition (46).
No consensus exists on the most clinically relevant parameters for the evaluation of GO, a lack the multicenter European Group on Graves Orbitopathy is bound to correct (47). Pertinently, one of the findings of the first large-scale review conducted by this group is that the severity of proptosis tends to be underestimated in the current NOSPECS classification system, with class 3 signs found in only 38% of patients, reflecting the ill-chosen proptosis cut-off point of 23 mm. The upper limit of normal proptosis values in Caucasians is 19 mm (mean ± SD), and 63% of our patients had Hertel readings of 21 mm or greater, suggesting that the current NOSPECS classification underestimates the severity of this sign. Thus, in our study perhaps the primary end point was ill chosen, but the European Group on Graves Orbitopathy work may lead to the definition of more pertinent criteria. Nevertheless, the solid, sustained effect on proptosis of just 16 wk of octreotide-LAR treatment is an encouraging preliminary result in light of the serious lack of therapeutic options for this condition. Future larger-scale studies are needed to confirm these results and to generate complete data on the risk to benefit ratio of the somatostatin analogs in this indication.
| Acknowledgments |
|---|
| Footnotes |
|---|
First Published Online November 23, 2004
Abbreviations: CAS, Clinical Activity Scale; GD, Graves disease; GO, Graves ophthalmopathy; MRI, magnetic resonance imaging; NMR, nuclear magnetic resonance; SstR, somatostatin receptor.
Received July 9, 2004.
Accepted November 12, 2004.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. Bartalena, L. Baldeschi, A. Dickinson, A. Eckstein, P. Kendall-Taylor, C. Marcocci, M. Mourits, P. Perros, K. Boboridis, A. Boschi, et al. Consensus statement of the European Group on Graves' orbitopathy (EUGOGO) on management of GO Eur. J. Endocrinol., March 1, 2008; 158(3): 273 - 285. [Full Text] [PDF] |
||||
![]() |
M. N. Stan, J. A. Garrity, E. A. Bradley, J. J. Woog, M. M. Bahn, M. D. Brennan, S. C. Bryant, S. J. Achenbach, and R. S. Bahn Randomized, Double-Blind, Placebo-Controlled Trial of Long-Acting Release Octreotide for Treatment of Graves' Ophthalmopathy J. Clin. Endocrinol. Metab., December 1, 2006; 91(12): 4817 - 4824. [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 |