Outcome of Gamma Knife Radiosurgery in 82 Patients with Acromegaly: Correlation with Initial Hypersecretion

Transcription

1 X/05/$15.00/0 The Journal of Clinical Endocrinology & Metabolism 90(8): Printed in U.S.A. Copyright 2005 by The Endocrine Society doi: /jc Outcome of Gamma Knife Radiosurgery in 82 Patients with Acromegaly: Correlation with Initial Hypersecretion Frédéric Castinetti, David Taieb, Jean-Marc Kuhn, Philippe Chanson, Manabu Tamura, Philippe Jaquet, Bernard Conte-Devolx, Jean Régis, Henry Dufour, and Thierry Brue Federation of Endocrinology, Diabetes, Metabolic Diseases and Nutrition (F.C., D.T., P.J., B.C.-D., T.B.), Department of Neurosurgery and Functional Neurosurgery (M.T., J.R., H.D.), and Department of Neurosurgery (H.D.), Hôpital de la Timone, Centre Hospitalier Universitaire de Marseille and Faculté de Médecine, Université de la Méditerranée, Marseille, France; Department of Endocrinology and Metabolic Diseases (J.-M.K.), Centre Hospitalier Universitaire de Rouen, Rouen, France; and Department of Endocrinology and Reproductive Diseases (P.C.), Assistance Publique- Hôpitaux de Paris, Hôpital de Bicêtre and Faculté de Médecine Paris XI, Le Kremlin Bicêtre, France Context: Because surgical and medical therapies of acromegaly all have specific limitations, radiotherapy has been used as an adjunctive strategy. Stereotactic radiosurgery has not yet been widely evaluated. Objective: The objective was to perform an analysis of long-term hormonal effects and tolerance of gamma knife radiosurgery. Design: Eighty-two patients were prospectively studied over a decade, with a mean follow-up of 49.5 months. Setting: All patients were treated at the Department of Functional Neurosurgery of Marseille, France. Patients: The patients included 82 with active acromegaly, of whom 63 had previous transsphenoidal surgery. Intervention: Intervention included radiosurgery using the Leksell Gamma Unit B model. GH levels were less than 2 ng/ml and IGF-I was normal for age off somatostatin agonists (at least 3 months). Results: Seventeen percent of the patients were in remission without any treatment. Twenty-three percent previously uncontrolled on somatostatin agonists fulfilled the same criteria after gamma knife while maintained on medical treatment. Initial GH and IGF-I levels off somatostatin agonists were significantly higher in uncured than in remission group (P 0.01 and 0.047, respectively). Withdrawal of somatostatin agonists at the time of radiosurgery had no incidence on the outcome. No significant difference was found in success rate whether patients had previously been treated or not. Long-term side effects included complete (n 2) or partial (n 12) hypopituitarism diagnosed 1 7 yr after gamma knife. Conclusions: Gamma knife radiosurgery may represent a therapeutic approach in patients with moderate initial or residual GH hypersecretion. (J Clin Endocrinol Metab 90: , 2005) Main Outcome Measures: Remission was diagnosed when mean First Published Online May 17, 2005 Abbreviations: GK, Gamma-knife radiosurgery; MRI, magnetic resonance imaging. JCEM is published monthly by The Endocrine Society ( endo-society.org), the foremost professional society serving the endocrine community. THE AIMS OF the treatment of acromegaly are to control tumor mass, to decrease hormonal levels of GH and IGF-I, and to alleviate clinical symptoms. Mortality in this disease can be reduced to that of a reference population if GH concentration is less than 1 2 g/liter and IGF-I level is normal (1). Surgery remains the first-line reference treatment, but remission rate ranges from 44 74% because of the high prevalence of invasive tumors (2). Currently available somatostatin agonists control GH and IGF-I excess in about 50 60% of patients (2). The more recent GH antagonist pegvisomant, although usually able to decrease IGF-I levels to the normal range, has no antitumoral effect (3, 4). Postoperative conventional fractionated radiotherapy controls the disease in 5 78% of cases (5, 6) but not before several years. Moreover, complications associated with radiotherapy have limited its usefulness; these include hypopituitarism, cranial nerve neuritis, visual-field defects, possible cognitive disturbances or increased cerebrovascular disorders, radiation-induced gliomas, and delayed brain necrosis (7, 8). The adverse consequences of conventional radiotherapy have been attributed to the inability to accurately deliver adequate doses to a small tumor volume. Such a shortcoming may be addressed by using gamma knife radiosurgery (GK). The hallmark of this procedure is to use cobalt-60 to deliver a single high dose of high-energy beams targeted to stereotactically defined intracranial sites (9, 10). This kind of radiation induces circumscribed cytotoxic effects on both neoplastic and vascular cells (11). Several authors have detailed their experience with GK, with efficacy ranging from 23 90% (12, 13). Most studies reported on short-term effects of GK. Except for a study on few patients (12), no one, to our knowledge, has addressed the long-term effects of GK in acromegaly in terms of efficacy and tolerance and using strict criteria for assessing endocrinological remission. The aim of the present study was to analyze our 10-yr experience with GK in a single radiosurgical center as either 4483

2 4484 J Clin Endocrinol Metab, August 2005, 90(8): Castinetti et al. Radiosurgery in Acromegaly first nonmedical treatment or an adjunctive treatment after transsphenoidal surgery for acromegaly. We thus evaluated the clinical and hormonal outcome in 82 patients, with a mean follow-up of 49.5 months. Patients Patients and Methods A total of 99 patients with active acromegaly were consecutively treated with GK between February 1993 and May 2003 at the Department of Functional Neurosurgery of Hôpital de la Timone (Marseille, France). The diagnosis of acromegaly was based on the association of clinical features of the disease, elevated baseline plasma GH not suppressible less than 1 g/liter after a glucose load, and increased IGF-I levels. Active disease was defined on the basis of GH levels in excess of 2 ng/ml and elevated age-adjusted IGF-I. These patients had been referred from different centers in France (n 89, 23 centers) and abroad (n 10, 6 centers). Each center agreed to monitor the patients outcome according to the standardized investigation described below. Immediate preradiosurgical and postradiosurgical evaluations were performed in the same Endocrinology Department in Marseille. In the present study, we did not analyze data from the 10 patients referred from foreign centers and from seven French patients who were all lost to follow-up. For data analysis, 82 patients were thus entered into this prospective study. Sixty-three of them were treated after unsuccessful transsphenoidal surgery. Two also previously received conventional radiotherapy. Nineteen patients had received only medical treatments (somatostatin or dopamine agonists) before GK in case of unresectable adenomas (such as small intrasellar adenomas with laterosellar extension), of contraindication (due to the patient s general condition), or of refusal of surgery. Only two patients had two consecutive GK. At the time of GK, 42 patients (51.3%) were treated with somatostatin agonists. In the others, the treatment with somatostatin analogs was discontinued at least 3 months or 2 wk before surgery for long-acting and sc forms, respectively. In the only patient treated with cabergoline, treatment was stopped 6 wk before the procedure. Radiosurgical procedure For the radiosurgical procedure, the Gamma Unit B model of the Leksell stereotactic system (Elekta Instruments, Stockholm, Sweden) was used. Irradiation was performed by cross-firing the target region with 201 beams of cobalt-60 distributed within a spherical sector. To delineate the lesion, a Leksell stereotactic head frame was used to perform both magnetic resonance imaging (MRI) and computed tomography scans to minimize metric distortion. For 43 patients treated between February 1993 and July 1997, dose planning was done with the KULA dose-planning software, and, for the 39 remaining patients treated after May 2000, it was done with the more recent Leksell Gamma-Plan system (Elekta Instruments), 18 of them with Automated Positioning System. In each case, the decision to treat by GK was made collectively by the endocrinological, neurosurgical, and radiosurgical teams. Before treatment, the target was delineated by two neurosurgeons. The radiosurgical planning was conducted jointly as part of the GK procedure. The treatment parameters and dose selection varied with the tumor size and the proximity of the tumor to the optic apparatus. Tumor margin was covered within the 50% isodose line (margin dose). Multiple isocenters were used for all of the patients. The prescribed margin dose ranged from Gy according to several factors: tumor volume, distance to the visual pathways, age, and previous treatment with fractionated radiotherapy. In all cases, the dose delivered to the optic chiasm and nerves was less than 9 Gy. In each case, marginal dose and dose delivered to the optic chiasm were recorded. Clinical and hormonal evaluation Each patient had a complete clinical and hormonal evaluation before GK, including ophthalmological evaluation using Goldman s campimetry in the same department (Endocrinology Department, Hôpital de la Timone). Initial pre-gk levels of GH, IGF-I, ACTH, urinary free cortisol, prolactin, LH, FSH, testosterone or estradiol, TSH, and free T 3 and T 4 were determined. The same evaluation was then performed in each center 3 and 6 months after GK, every 6 months during 2 yr, and then yearly. During follow-up, GH and IGF-I levels were determined without any medical therapy or after at least 3 months of discontinuation of the treatment with somatostatin agonists. Results obtained earlier than 3 months or later than 3 months after the last injection of somatostatin agonists were considered as on treatment or off treatment, respectively. All pre-gk evaluations were performed off treatment, and at least one off treatment evaluation was performed during the first year after GK. Subsequently, intervals from 1 3 yr between off-treatment evaluations were decided on by each investigator according to clinical and hormonal evaluations on treatment. Patients were considered in remission if they had a mean GH level of less than 2 g/liter and a normal age-adjusted IGF-I in off-treatment period. Those who were still on somatostatin agonists were considered uncured. Mean GH levels were determined by averaging eight hourly determinations between 0800 and 1500 h. GH and IGF-I concentrations were measured in the same laboratory immediately before and after GK in all patients and at each subsequent time point in the subset of patients (n 42) followed in the Endocrine Department in Marseille. For these assays, the characteristics of the kits used were as follows. Serum GH levels were measured by an immunoradiometric assay (Immunotech, Marseille, France). The sensitivity is 0.05 ng/ml. The intraassay and interassay coefficients of variation are and %, respectively. Serum IGF-I was measured by immunoradiometric assay (Immunotech). The sensitivity is 3 ng/ml. The intraassay and interassay coefficients of variation are and %, respectively. All other determinations in the 22 other centers were done using commercial kits, and normal IGF-I values for each laboratory were taken into account. Gonadotroph deficiency was defined by low plasma testosterone with nonelevated gonadotrophin levels in men, amenorrhea with low plasma estradiol and low or normal gonadotrophins in nonmenopausal women, and a lack of increased gonadotrophins in menopausal women. Corticotroph deficiency was established by subnormal response of cortisol during an insulin tolerance test (peak, 550 nmol/liter). The diagnosis of thyrotroph deficiency was based on low free plasma T 4 level with normal or diminished plasma TSH. Initial pituitary neuroimaging Pituitary imaging was performed by MRI to define dose planning and to evaluate the limits of the lesion. The MRI sequences were coronal postgadolinium T1-weighted spin-echo (SE), T2-weighted SE, and, for most patients, a postgadolinium three-dimensional magnetization prepared acquisition gradient echo sequence. Statistical analysis Data were analyzed by parametric or nonparametric tests. Continuous data with normal distribution were analyzed by t test, and those with uneven distribution were analyzed by Mann-Whitney U test. Categorical data were analyzed by Fisher s exact test. Longitudinal evaluations were performed by Kaplan-Meier method. All statistical tests were two tailed, and P 0.05 was considered significant. Results The patients were followed up for a mean period of 49.5 months (range, months). Long-term effects of GK At the end of the study, 17 patients (20.7%) had GH levels less than 2 ng/ml after an average of 35.3 months, and 21 patients (25.6%) had normal age-adjusted IGF-I after an average of 35.5 months. Based on both criteria, 14 of 82 patients (17%) were considered to be in remission after an average of 36 months (Fig. 1 and Table 1). Thanks to GK, 19 other patients (23%), considered uncured, were controlled by somatostatin agonists, whereas they were not controlled with

3 Castinetti et al. Radiosurgery in Acromegaly J Clin Endocrinol Metab, August 2005, 90(8): FIG. 1. Kaplan-Meier analysis: number of uncured patients based on GH and IGF-I criteria. Remission was diagnosed when mean GH levels were less than 2 ng/ml and IGF-I was normal for age off somatostatin agonists (at least 3 months). Numbers on the line represent the number of patients who have a follow-up at least equal to the time point. Total mean follow-up in the whole population was 49.5 months. the same treatment before GK. The two patients who had two consecutive GK treatments were not in the remission group. Initial GH and IGF-I levels off somatostatin agonists were significantly higher in uncured patients than in patients in remission. Nevertheless, the mean percentage of GH decrease at mean follow-up (49.5 months) did not differ between the two groups (data not shown). To assess whether previous hormonal evaluations might predict later outcome of the treatment, we compared the percentage of decrease of GH hypersecretion 6 12 months after GK, off somatostatin agonists: no significant difference was observed at these or at any time points. Indeed, as shown on Fig. 2, the profile of GH decrease was similar in both outcome groups, who differed mainly by their initial hormone levels. No significant difference was observed in sex ratio, median age, initial volume of the adenoma (microadenoma or macroadenoma), type of secretion (pure GH or mixed GH and prolactin secretion), or its localization (intracavernous, suprasellar, or laterosellar). No significant difference was found in marginal dose. Interestingly, we also found no significant difference in the outcome of patients who were on or off somatostatin agonists at the time of the GK (Table 2). GK in patients without previous surgical treatment Nineteen patients had GK as a primary nonmedical treatment. In Table 3 are presented the data on patients treated primarily with GK or treated with GK as adjunct of transsphenoidal surgery (n 63). No difference was found between the two groups in terms of sex ratio, marginal dose, prevalence of macroadenomas, mixed or isolated secretion, and initial GH or IGF-I levels. The only difference was median age, reflecting that older patients often had contraindications to transsphenoidal surgery. GK given as a primary TABLE 1. Characteristics of patients in remission (n 14) Sex/age (yr) Volume/secretion/ localization Initial GH/IGF-I (ng/ml) Primary/ adjunctive SSA status Margin dose (Gy) Time to remission (months) Follow-up (months) Complication deficiency 1 F/32 M/GH-PRL/IC 4/330 A On F/33 M/GH-PRL/IS 3.5/709 A On F/53 M/GH-PRL/IC 8.5/641 P Off F/53 M/GH/IC 12/640 A On M/56 M/GH/IC-SS 8/510 A Off F/61 m/gh/ic 2.3/480 P Off TSH 7 F/76 M/GH/IC 7.5/661 P On F/50 M/GH/IC 7.5/433 A Off F/41 M/GH/IC 4.5/367 A Off F/53 M/GH/IC 9/600 A On M/55 M/GH/IC 7/396 P On F/51 M/GH-PRL/IC 4/119 A Off M/54 M/GH-PRL/IC 7.2/350 A On LH/FSH 14 F/40 M/GH/IC 14.8/700 A On Remission was diagnosed when mean GH levels were less than 2 ng/ml and IGF-I was normal for age off somatostatin agonists (SSA) for at least 3 months. Patient characteristics include sex (F, female; M, male), age, initial volume of adenoma (M, macroadenoma superior to 10 mm; m, microadenoma inferior to 10 mm), secretion (GH, pure GH hypersecretion; GH-PRL, mixed GH-prolactin hypersecretion), localization of the adenoma (IC, intracavernous; IS, intrasellar; SS, suprasellar), initial GH and IGF-I, primary nonmedical treatment (P), and adjunctive treatment (A). SSA status, Patients treated by somatostatin agonists at the time of GK treatment or who had received long-acting SSA less than 3 months before treatment were considered on SSA treatment (On); untreated patients or patients who had received long-acting SSA more than 3 months before GK were considered off SSA treatment (Off). Complication deficiency, Type of new hormonal deficiency appearing after GK.

4 4486 J Clin Endocrinol Metab, August 2005, 90(8): Castinetti et al. Radiosurgery in Acromegaly FIG. 2. Mean GH in remission (n 14) and uncured (n 68) groups. Allocation into each group was based on the following criteria: remission was diagnosed when mean GH levels were less than 2 ng/ml and IGF-I was normal for age off somatostatin agonists for at least 3 months. Mean GH levels of uncured patients are represented by a broken line with filled squares; mean GH of remission patients is represented by a dotted line with filled circles. SD is shown for each group. Comparisons between groups were performed by ANOVA. treatment or in adjunct of transsphenoidal surgery appears to be similarly effective, resulting in remission in 4 of 19 (21%) and 10 of 63 (16%) patients, respectively. Adverse effects At least one new pituitary deficiency was observed after GK in 14 patients: TSH deficiency in three (1, 3, and 7 yr after GK), ACTH deficiency in three (4, 5, and 6 yr after GK), LH and FSH deficiency in seven (between 5 and 7 yr after GK), and complete hypopituitarism in two (3 and 2 yr after GK). One of these two cases had conventional radiotherapy 10 yr before. The mean margin dose for these 14 patients was 27.5 vs. 26 Gy for patients TABLE 2. Comparison between in remission and uncured groups In remission Uncured P No. of patients Mean age (yr) 50.5 [32 76] 53.5 [21 74] ns Sex ratio (F/M) 11/3 37/31 ns Marginal dose (Gy) 24.2 [15 35] 26 [12 40] ns Macroadenoma 13 (93%) 61 (90%) ns Adenoma with pure GH 9 (64%) 54 (79%) ns hypersecretion Tumor extension Intracavernous ns Suprasellar 1 2 ns Intrasellar 1 6 ns Initial GH (ng/ml) 7.1 [ ] 25.3 [ ] 0.01 Initial IGF-I (ng/ml) 495 [ ] 673 [ ] Patients off SSA at the 6 (43%) 34 (50%) ns time of GK treatment Follow-up (months) 62 [6 108] 50 [8 99] ns Allocation into each group was based on the following criteria: remission was diagnosed when mean GH levels were less than 2 ng/ml and IGF-I was normal for age off somatostatin agonists (SSA) for at least 3 months. Extreme values are between brackets. Sex ratio is presented as number of female (F)/number of male (M) patients. Tumor extension was determined by pretreatment MRI as detailed in Patients and Methods. One patient in the remission group had both suprasellar and cavernous extensions. Percentages indicate the proportion of patients in each subgroup. ns, Nonsignificant. P 0.05 was considered significant. without adverse effects (P 0.05). One patient, previously treated by conventional radiotherapy, presented 1 month after GK (marginal dose, 40 Gy; dose delivered to the optic chiasm, 8 Gy) a trigeminal neuralgia that disappeared under carbamazepine treatment, and, 2 months later, the same patient displayed an impaired central visual field, which was alleviated after 3 months of steroid treatment. Overall, patients in remission (4.7 Gy) or uncured (4 Gy) did not differ in terms of dose delivered to the optic chiasm. Only a few patients had other transient adverse effects, including headaches or vomiting. Discussion When GH hypersecretion persists in patients with acromegaly, the risk of mortality was reported to be up to three to four times higher than that of patients in whom GH secretion is controlled (14 16). Thus, one of the primary aims of the treat- TABLE 3. Comparison between GK as primary nonmedical treatment (without previous surgical treatment) and as adjunctive treatment (after surgical treatment) Primary treatment Adjunctive treatment No. of patients 19 (23%) 63 (77%) Mean age (yr) 62 [35 76] 48 [21 68] 0.01 Sex ratio (F/M) 12/7 36/27 ns Marginal dose (Gy) 30 [20 35] 25 [12 40] ns Macroadenoma 16 (84%) 58 (92%) ns Adenoma with GH 15 (79%) 48 (76%) ns hypersecretion only Initial GH (ng/ml) 15 [2.3 49] 24 [4 240] ns Initial IGF-I (ng/ml) 695 [ ] 640 [ ] ns Patients off SSA at the 10 (53%) 30 (48%) ns time of GK treatment Follow-up (months) 61 [19 99] 50 [6 108] ns In remission 4 (21%) 10 (16%) ns Uncured 15 (79%) 53 (84%) ns Extreme values are between brackets. Sex ratio is presented as number of female (F)/number of male (M) patients. Percentages indicate the proportion of patients in each subgroup. SSA, Somatostatin agonists; ns, nonsignificant. P 0.05 was considered significant. P

5 Castinetti et al. Radiosurgery in Acromegaly J Clin Endocrinol Metab, August 2005, 90(8): ment of acromegaly is to normalize GH and IGF-I levels. For patients to be considered in remission, a recent consensus conference recommended achieving a GH plasma level lower than 2 ng/ml (or a suppressibility 0.3 ng/ml after oral glucose tolerance test) and a normal age-adjusted IGF-I level (17). With reference to these stringent basal GH and IGF-I criteria, 17% of the patients were in remission after an average of 36 months. In addition, based on the same criteria for GH and IGF-I levels, 23% of our patients (n 19) achieved normal values after GK while treated with somatostatin agonists, whereas they were not controlled with the same treatment before. Although some of them might have been found in remission after an appropriate withdrawal of their medical treatment, we considered all of them as uncured. A longer follow-up would obviously be helpful to better evaluate the real effectiveness of GK. In this regard, 36 of our patients had a final off-treatment evaluation less than 36 months after GK, i.e. less than the average time for normalization. In a recent review of currently published series (18), 37.7% of 361 patients reported in 19 studies on 4 68 patients were found to have complete response to GK after a corrected median follow-up of 29 months. These series, however, differed widely in the types of tumors treated, the time between last administration of medical treatment and hormonal evaluation, and the criteria for biological remission, among other factors (9, 12, 13, 19). As underlined by another recent review (20), comparison of results is indeed difficult, mainly because of the variable criteria of cure: in the series with at least 10 patients and a median follow-up of 2 yr, success rates ranged from 20 96%. For example, in the single largest series on 68 patients in 2000 by Zhang et al. (13), patients were considered in remission when GH levels were less than 12 ng/ml, with a reported cure rate of 96%, whereas the second largest series reported a 25% remission rate based on normalization of age- and sex-matched IGF-I (21). In contrast with a recent series with fewer cases (12), we found that initial plasma GH and IGF-I levels were significant predictive factors of outcome. Studies evaluating conventional radiotherapy (22, 23), surgery (24, 25), and somatostatin agonist therapy (26, 27) had also shown a better outcome in acromegalic patients whose pretreatment plasma GH levels were lower. The importance of initial hormonal level was confirmed in our study by the lack of difference in mean percentage of GH decrease between the two groups. In our patients, the final outcome depended mainly on GH and IGF-I initial levels off somatostatin agonists. No difference in the outcome was observed between patients who received somatostatin agonists at the time of GK (20% patients in remission) and those who did not (15% patients in remission). This result does not confirm that of a nonrandomized study by Landolt et al. (28) concluding that the use of octreotide might decrease the efficacy of radiation therapy if used at the time of the procedure. Our results are in accordance with those of a recent study (12). Nevertheless, we share other investigators viewpoint that the treatment should be stopped shortly before GK (9, 13, 29): somatostatin agonist withdrawal indeed allows further comparison of off-treatment GH secretion. In most patients, GK was indicated because transsphenoidal surgery failed to achieve remission. Our study found no significant difference in success rate whether GK was used as primary treatment (16%) or after surgery (21%). When performed in other centers as a primary treatment, GK was mainly used for microadenomas (9, 30). In our study, GK was used as primary treatment for microadenomas or small (10 15 mm) enclosed macroadenoma with laterosellar extension. The overall success rate remains low, in contrast with the study of Zhang et al. (13) in which more than 90% of patients treated with GK as primary nonmedical treatment were in remission. In that study, however, the criteria for remission were not precisely defined. To our knowledge, no other study has tried to compare the efficacy of GK as primary or adjunctive treatment on a large number of patients. As observed in other series, our patients showed no increase in tumor volume (12, 31, 32). Tumor size, however, was not systematically monitored during long-term follow-up because the treatment was not primarily aimed at reducing tumor volume of usually small adenoma remnants. According to a recent review, tumor growth control was obtained in % of patients in all but one series (20). The mean margin dose we used was 25.8 Gy. Higher success rates were observed with higher margin dose but with the drawback of more severe complications (13, 33). The doses used in our center were chosen by comparing the benefits of decreasing GH secretion and the risks of secondary effects, i.e. onset of new pituitary deficiency. Dose prescription was also adapted to the tumor volume, the distance separating the tumor and the visual pathway, the aggressiveness of the disease, and previous treatment. We found a mean time of GH and IGF-I normalization of 36 months. This latency is longer than that in most series evaluating GK in acromegaly (13, 29, 33, 34). This difference can probably be explained by the higher marginal doses used by other researchers. Because GK patients often have lower pretreatment GH concentrations than conventionally treated patients and different types of tumors, it is difficult to compare the rate of decline between GK and conventional radiotherapy. As shown in many studies (12, 16, 35), GK induces far fewer complications than conventional radiotherapy: 82% patients in our series had no side effects after GK vs. less than 50% with conventional radiotherapy in most studies (7, 36). A longer follow-up would probably be useful to better evaluate longterm effects of GK. In the present series, only one severe visual regressive trouble with cavernous syndrome was observed (dose to the optic chiasm, 8 Gy). As recommended previously (37), the dose delivered to the optic chiasm was inferior or equal to 8 Gy in all but one patient. A 9-Gy dose had to be delivered in one case because of the conformation and size of the lesion. In conclusion, after a mean follow-up of more than 4 yr, GK was able to decrease GH and IGF-I concentrations to so-called safe (1) levels in 40% (17% in remission and 23% under somatostatin agonists) of our patients. This treatment was thus helpful in this population who had failed to achieve hormonal control with either surgery alone or a combined treatment of surgery and somatostatin agonists at the time of their irradiation. GK represents a safe and efficacious treatment in welldefined indications: small postsurgical remnants, sufficiently distant from the optic chiasm, and with moderate initial GH hypersecretion. In view of the high rate of IGF-I control obtained with currently available medical therapies, the place of GK in the treatment of acromegaly may, however, need to be redefined in the future. In particular, the cost effectiveness of a

6 4488 J Clin Endocrinol Metab, August 2005, 90(8): Castinetti et al. Radiosurgery in Acromegaly potentially lifelong medical approach will have to be compared in the context of each particular country with an approach that offers long-term remission and possibly cure in a subset of patients with acromegaly. Acknowledgments We thank the following French physicians who referred patients for GK and provided updated information on their outcome: Dr. Jean- Michel Andrieu (Montpellier), Dr. Bernard Aubert (Metz), Prof. Ivan Bachelot (Grenoble), Dr. Catherine Basin (Paris), Prof. Benoit Bataille (Poitiers), Dr. Jean-Gerard Baudet (Le Creusot), Dr. Antoine Bennet (Toulouse), Prof. Xavier Bertagna (Paris), Dr. Basile Bigorie (Meaux), Prof. Philippe Bouchard (Paris), Prof. Didier Dewailly (Lille), Dr. Marie- Laure Dumuis (Paris), Dr. Philippe Emy (Orleans), Dr. Alexandre Fredenrich (Nice), Dr. Philippe Giraud (Angers), Dr. Didier Gouet (La Rochelle), Dr. Edgar Kaloustian (Compiègne), Prof. Véronique Kerlan (Brest), Dr. Albert Martinand (Dijon), Dr. Catherine Mattei (Marseille), Prof. Charles Oliver (Marseille), Dr. Françoise Peillon (Suresnes), Dr. Anne Priou (Avignon), Prof. Eric Renard (Montpellier), Dr. Elisabeth Requeda (Corbeil), Dr. Michel Rodier (Nîmes), Prof. Jean-Louis Sadoul (Nice), Dr. Franck Schilla (Besançon), Dr. Dominique Simon (Paris), Dr. Gilbert Simonin (Marseille), Prof. Antoine Tabarin (Bordeaux), Prof. Bernard Vialettes (Marseille), and all of the coinvestigators of the French Pituitary Society (Club Français de l Hypophyse, Société Française d Endocrinologie). We thank Dr. Vincent Pradel for statistical analysis and Dr. Christine Nicolino for fruitful discussions. Received February 11, Accepted May 5, Address all correspondence and requests for reprints to: Prof. T. Brue, Department of Endocrinology, Hôpital de la Timone, 264 rue St. Pierre, cedex 5, Marseille, France. thierry.brue@mail.ap-hm.fr. References 1. Holdaway IM, Rajasoorya RC, Gamble GD 2004 Factors influencing mortality in acromegaly. J Clin Endocrinol Metab 89: Freda PU 2003 How effective are current therapies for acromegaly? Growth Horm IGF Res 13(Suppl A):S144 S Van der Lely AJ, Hutson RK, Trainer PJ, Besser GM, Barkan AL, Katznelson L, Klibanski A, Herman-Bonert V, Melmed S, Vance ML, Freda PU, Stewart PM, Friend KE, Clemmons DR, Johannsson G, Stavrou S, Cook DM, Phillips LS, Strasburger CJ, Hackett S, Zib KA, Davis RJ, Scarlett JA, Thorner MO 2001 Long-term treatment of acromegaly with pegvisomant, a growth hormone receptor antagonist. Lancet 358: Trainer PJ, Drake WM, Katznelson L, Freda PU, Herman-Bonert V, van der Lely AJ, Dimaraki EV, Stewart PM, Friend KE, Vance ML, Besser GM, Scarlett JA, Thorner MO, Parkinson C, Klibanski A, Powell JS, Barkan AL, Sheppard MC, Malsonado M, Rose DR, Clemmons DR, Johannsson G, Bengtsson BA, Stavrou S, Kleinberg DL, Cook DM, Phillips LS, Bidlingmaier M, Strasburger CJ, Hackett S, Zib K, Bennett WF, Davis RJ 2000 Treatment of acromegaly with the growth hormone-receptor antagonist pegvisomant. N Engl J Med 342: Biermasz NR, Dulken HV, Roelfsema F 2000 Postoperative radiotherapy in acromegaly is effective in reducing GH concentration to safe levels. Clin Endocrinol (Oxf) 53: Barkan AL, Halasz I, Dornfeld KJ, Jaffe CA, Friberg RD, Chandler WF, Sandler HM 1997 Pituitary irradiation is ineffective in normalizing plasma insulin-like growth factor I in patients with acromegaly. J Clin Endocrinol Metab 82: Wass JA 1996 Long-term effects of radiotherapy for acromegaly. Metabolism 45: Barrande G, Pittino-Lungo M, Coste J, Ponvert D, Bertagna X, Luton JP, Bertherat J 2000 Hormonal and metabolic effects of radiotherapy in acromegaly: long-term results in 128 patients followed in a single center. J Clin Endocrinol Metab 85: Jackson IM, Noren G 1999 Role of gamma knife radiosurgery in acromegaly. Pituitary 2: Yamamoto M 1999 Gamma knife radiosurgery: technology, applications, and future directions. Neurosurg Clin N Am 10: Landolt AM, Haller D, Lomax N, Scheib S, Schubiger O, Siegfried J, Wellis G 1998 Stereotactic radiosurgery for recurrent surgically treated acromegaly: comparison with fractionated radiotherapy. J Neurosurg 88: Attanasio R, Epaminonda P, Motti E, Giugni E, Ventrella L, Cozzi R, Farabola M, Loli P, Beck-Peccoz P, Arosio M 2003 Gamma-knife radiosurgery in acromegaly: a 4-year follow-up study. J Clin Endocrinol Metab 88: Zhang N, Pan L, Wang EM, Dai JZ, Wang BJ, Cai PW 2000 Radiosurgery for growth hormone-producing pituitary adenomas. J Neurosurg 93(Suppl 3): Swearingen B, Barker 2nd FG, Katznelson L, Biller BM, Grinspoon S, Klibanski A, Moayeri N, Black PM, Zervas NT 1998 Long-term mortality after transsphenoidal surgery and adjunctive therapy for acromegaly. J Clin Endocrinol Metab 83: Abosch A, Tyrrell JB, Lamborn KR, Hannegan LT, Applebury CB, Wilson CB 1998 Transsphenoidal microsurgery for growth hormone-secreting pituitary adenomas: initial outcome and long-term results. J Clin Endocrinol Metab 83: Mahmoud-Ahmed AS, Suh JH, Mayberg MR 2001 Gamma knife radiosurgery in the management of patients with acromegaly: a review. Pituitary 4: Growth Hormone Research Society; The Pituitary Society 2004 Biochemical assessment and long-term monitoring in patients with acromegaly: statement from a Joint Consensus Conference of the Growth Hormone Research Society and The Pituitary Society. J Clin Endocrinol Metab 89: Brada M, Ajithkumar TV, Minniti G 2004 Radiosurgery for pituitary adenomas. Clin Endocrinol (Oxf) 61: Fukuoka S, Ito T, Takanashi M, Hojo A, Nakamura H 2001 Gamma knife radiosurgery for growth hormone-secreting pituitary adenomas invading the cavernous sinus. Stereotact Funct Neurosurg 76: Laws ER, Sheehan JP, Sheehan JM, Jagnathan J, Jane Jr JA, Oskouian R 2004 Stereotactic radiosurgery for pituitary adenomas: a review of the literature. J Neurooncol 69: Laws Jr ER, Vance ML 1999 Radiosurgery for pituitary tumors and craniopharyngiomas. Neurosurg Clin N Am 10: Wass JA 1997 Evidence for the effectiveness of radiotherapy in the treatment of acromegaly. J Endocrinol 155(Suppl 1):S57 S58; discussion S67 S Epaminonda P, Porretti S, Cappiello V, Beck-Peccoz P, Faglia G, Arosio M 2001 Efficacy of radiotherapy in normalizing serum IGF-I, acid-labile subunit (ALS) and IGFBP-3 levels in acromegaly. Clin Endocrinol (Oxf) 55: De P, Rees DA, Davies N, John R, Neal J, Mills RG, Vafidis J, Davies JS, Scanlon MF 2003 Transsphenoidal surgery for acromegaly in wales: results based on stringent criteria of remission. J Clin Endocrinol Metab 88: Jenkins PJ, Emery M, Howling SJ, Evanson J, Besser GM, Monson JP 2004 Predicting therapeutic response and degree of pituitary tumour shrinkage during treatment of acromegaly with octreotide LAR. Horm Res 62: McKeage K, Cheer S, Wagstaff AJ 2003 Octreotide long-acting release (LAR): a review of its use in the management of acromegaly. Drugs 63: Newman CB, Melmed S, Snyder PJ, Young WF, Boyajy LD, Levy R, Stewart WN, Klibanski A, Molitch ME, Gagel RF 1995 Safety and efficacy of long-term octreotide therapy of acromegaly: results of a multicenter trial in 103 patients a clinical research center study. J Clin Endocrinol Metab 80: Landolt AM, Haller D, Lomax N, Scheib S, Schubiger O, Siegfried J, Wellis G 2000 Octreotide may act as a radioprotective agent in acromegaly. J Clin Endocrinol Metab 85: Kim SH, Huh R, Chang JW, Park YG, Chung SS 1999 Gamma knife radiosurgery for functioning pituitary adenomas. Stereotact Funct Neurosurg 72(Suppl 1): Hayashi M, Izawa M, Hiyama H, Nakamura S, Atsuchi S, Sato H, Nakaya K, Sasaki K, Ochiai T, Kubo O, Hori T, Takakura K 1999 Gamma knife radiosurgery for pituitary adenomas. Stereotact Funct Neurosurg 72(Suppl 1): Ganz JC, Backlund EO, Thorsen FA 1993 The effects of Gamma knife surgery of pituitary adenomas on tumor growth and endocrinopathies. Stereotact Funct Neurosurg 61(Suppl 1): Mokry M, Ramschak-Schwarzer S, Simbrunner J, Ganz JC, Pendi G 1999 A six-year experience with the post-operative radiosurgery management of pituitary adenomas. Stereotact Funct Neurosurg 72: Martinez R, Bravo G, Burzaco J, Rey G 1998 Pituitary tumors and gamma knife surgery. Clinical experience with more than two years of follow-up. Stereotact Funct Neurosurg 70(Suppl 1): Lim YL, Leem W, Kim TS, Rhee BA, Kim GK 1998 Four years experiences in the treatment of pituitary adenomas with gamma knife radiosurgery. Stereotact Funct Neurosurg 70(Suppl 1): Vance ML, Gamma-knife radiotherapy for pituitary adenomas. Program of the 84th Annual Meeting of The Endocrine Society, San Francisco, CA, 2002, p 27 (Abstract S9-2) 36. Eastman RC, Gorden P, Glatstein E, Roth J 1992 Radiation therapy of acromegaly. Endocrinol Metab Clin North Am 21: Tishler RB, Loeffler JS, Lunsford LD, Duma C, Alexander 3rd E, Kooy HM, Flickinger JC 1993 Tolerance of cranial nerves of the cavernous sinus to radiosurgery. Int J Radiat Oncol Biol Phys 27: JCEM is published monthly by The Endocrine Society ( the foremost professional society serving the endocrine community.