| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Medicine (P.U.F., C.M.R.), Columbia University, College of Physicians and Surgeons, New York, New York 10032; Departments of Neurosurgery and Medicine (L.K.), Stanford University Medical Center, Stanford, California 94305; Department of Internal Medicine (A.J.v.d.L.), Erasmus Medical Center, The Netherlands; and Department of Statistics (S.Z., D.R.), Columbia University, New York, New York 10027
Address all correspondence and requests for reprints to: Pamela U. Freda, M.D., Department of Medicine, Columbia College of Physicians and Surgeons, 630 West 168th Street, New York, New York 10032. E-mail: puf1{at}columbia.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: We conducted a meta-analysis to derive definitive estimates of their efficacy for biochemical control and tumor shrinkage.
Data Sources: A search of literature was conducted through 2003, primarily via PubMed.
Study Selection: Inclusion criteria, met in 44 trials, included at least 3 months of secondary octreotide long-acting release (LAR) or lanreotide slow release (SR) therapy or of primary octreotide LAR, lanreotide SR, or sc octreotide therapy and clearly reported data on biochemical efficacy and/or tumor shrinkage. Fifty other trials screened did not meet analysis inclusion criteria.
Data Extraction: Data were extracted by three independent observers.
Data Synthesis: Among subjects not selected for somatostatin analog responsiveness before study entry, both GH efficacy criteria and IGF-I normalization were met in a greater proportion of those treated with octreotide LAR vs. lanreotide SR (GH: B = 0.2310, P = 0.016; IGF-I: B = 0.2325, P = 0.007). Prestudy selection for somatostatin analog responsiveness was not a significant predictor of meeting GH efficacy criteria (B = 0.0992; P = 0.12). Preselection was a positive predictor of IGF-I normalization rate (B = 0.1213; P = 0.04), which was greater among preselected than unselected subjects (B = 0.1472; P = 0.0475). IGF-I normalization occurred in a greater proportion of secondary octreotide LAR- vs. primary octreotide-treated subjects (B = 0.2056; P = 0.009). The odds of tumor shrinkage more than 10% were lower in the unselected vs. preselected subjects. However, the effect of drug type was an important predictor of shrinkage; such that regardless of preselection or not, the odds of shrinkage with lanreotide SR were lower than with octreotide LAR (P = 0.003). Shrinkage greater than 10% occurred in a higher percentage of primary octreotide LAR-treated vs. primary octreotide sc-treated subjects (odds ratio = 9.4; P < 0.0001). The overall rate of tumor increase was 1.4%.
Conclusions: In this meta-analysis, we have shown that the efficacy of octreotide LAR is greater than lanreotide SR among subjects unselected for prior somatostatin analog responsiveness. Preselection is a significant positive predictor of IGF-I normalization and is associated with increased odds of tumor shrinkage, which is also greatest with octreotide LAR. Biochemical efficacy is similar, but tumor shrinkage is greater when these drugs are given as primary vs. secondary therapy.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The English language literature, through 2003, was searched in PubMed for studies reporting treatment of patients with acromegaly with the somatostatin analogs octreotide and lanreotide SR. References of these papers were then reviewed for additional papers. Studies included in the analysis were those that reported a prospective treatment of active acromegaly with the long-acting somatostatin analogs, either octreotide long-acting release (LAR) or lanreotide slow release (SR) as secondary therapy, and those that reported use of either long-acting analog, octreotide LAR or lanreotide SR, or sc octreotide as primary therapy for acromegaly. The analysis focuses on the data with depot formulations of these analogs because these are the forms primarily used in clinical practice, but it was decided to include studies of primary therapy with sc octreotide because scant data are available on the use of octreotide LAR or lanreotide SR as primary therapy. Only studies that reported somatostatin analog therapy of at least 3 months duration, that included at least five study subjects, and that clearly reported specific GH and/or IGF-I data and/or criteria for efficacy along with the numbers of subjects meeting these criteria were included in the analysis. The tumor shrinkage analysis included only those studies that reported the exact numbers of subjects of the total assessed who had tumor shrinkage while on somatostatin analog therapy and also reported the amount of tumor shrinkage.
Fifty additional studies were screened for inclusion but were rejected for the following reasons: treatment duration was less than 3 months (n = 7), retrospective nature of the study (n = 1), study size was less than five subjects (n = 4), sc octreotide was administered as secondary therapy only (n = 21), somatostatin analog and dopamine agonist therapy were combined (n = 3), primary and secondary octreotide sc data were combined (n = 2), octreotide LAR and lanreotide SR data were combined (n = 1), tumor shrinkage was recorded as "some" (n = 3), studies were case reports (n = 4), and octreotide administration was intranasal (n = 4).
Data collection
Each paper was reviewed independently by three reviewers (L.K., A.J.v.d.L., and P.U.F.), and the following data were recorded on uniform data collection forms: 1) study subjects: number of subjects, gender distribution, ages, biochemical criteria used to establish the diagnosis of acromegaly, prior therapy for acromegaly, and study subject selection criteria, in particular whether or not screening for somatostatin analog responsiveness was required for study entry; 2) study design: open label, prospective, retrospective, randomized, or blinded; 3) study drug: type of drug (octreotide LAR, lanreotide SR, or sc administered octreotide), dosages administered, proportion of subjects at each dose, and dosing intervals used; 4) biochemical efficacy: criteria used to determine biochemical efficacy in the study, number of subjects overall and within subgroups studied (if available) that met the study’s efficacy criteria, and means and SD values of baseline and poststudy GH and IGF-I levels for subjects overall and within each subgroup; efficacy was recorded for final time point examined in the study as well as each major time point of the study at which data were reported; 5) pituitary imaging: imaging protocol, duration of follow-up, number of subjects experiencing tumor shrinkage with therapy, and percent tumor shrinkage; and 6) dropout rates and reasons for dropout.
Agreement among reviewers on data was greater than 90%. Agreement of two of the three reviewers was met in 98% of the data. In cases of missing or discrepant data, the original paper was rereviewed by one of the authors (P.U.F.) for verification of the data. In all cases, this verification was consistent with the majority view of the reviewers.
Data compilation and analysis
Study data were separated into that for patients for whom therapy was secondary (after surgery, radiotherapy, or other medical therapy) and patients for whom somatostatin analog therapy was primary therapy (no other prior therapy). However, in studies in which data on the primary therapy subgroup was not separated from that of the secondary group, the meta-analysis performed included these patients’ data as part of the secondary therapy analysis.
Characteristics of the subjects and designs of the studies analyzed
Secondary therapy with octreotide LAR was analyzed in data from 12 studies (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12), which included 612 subjects, including 242 men and 264 women (gender breakdown was not reported in 106 subjects from two studies), with a mean age of 49.6 yr (range, 18–85 yr). Biochemical criteria for entry into the studies analyzed were: elevated IGF-I along with a post-OGTT GH more than 2 µg/liter in 49.8% of subjects, and either post-OGTT GH more than 2 µg/liter or more than 1 µg/liter alone, mean GH more than 2 µg/liter along with post-OGTT GH more than 0.8 µg/liter, or IGF-I elevation alone in 12%; these criteria were not specified in 38%. Prestudy demonstration of octreotide responsiveness was a criterion for study inclusion in 424 (69.3%) of the 612 subjects analyzed. Prestudy therapy included transsphenoidal surgery and/or radiotherapy in most patients, sc octreotide in 16%, lanreotide SR in 24.5%, and no prior therapy in 5.8% (these patients were included in this secondary analysis because their data were not provided separately from that of secondary therapy subjects). Forty-two patients (6.8%) dropped out before the per-protocol study period.
All but one octreotide LAR trial were open-label prospective studies; one study was randomized, controlled, and blinded (7). Criteria for LAR efficacy included normalization of IGF-I in 10 of 12 studies (IGF-I normative data were clearly stated to be age-adjusted in four of 10). Efficacy was also defined by a mean GH less than 2.5 µg/liter in all studies and by GH less than 1 µg/liter post OGTT in one study. At the completion of dose escalation (in 498 patients for whom this information was available), the octreotide LAR dose was 10 mg in 18%, 20 mg in 48%, 30 mg in 29%, and 40 mg in 4.8% of patients. The dosing interval for octreotide LAR was every 4 wk in 98.5% of subjects. Mean study duration was 15.5 months (range, 6–36 months). Considering the number of subjects in each study, patients received an average of 15 person-months of therapy.
Secondary or adjunctive therapy with lanreotide SR was assessed in 914 subjects from 19 studies (2, 9, 10, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28), including 447 males and 467 females, with a mean age of 50.3 yr (range, 18–89 yr). The biochemical criteria for study entry were: elevated IGF-I along with a post-OGTT GH more than 2 µg/liter in 36% of subjects, elevated IGF-I along with a post-OGTT GH more than 5 µg/liter in 2.4%, IGF-I elevated along with post-OGTT GH more than 1 µg/liter in 15.5%, IGF-I elevation with a mean GH more than 2.5 µg/liter in 6.6%, and post-OGTT GH more than 1.0 µg/liter or 2.0 µg/liter in 21.5%; criteria were not specified in 16.2%. Prestudy documentation of octreotide responsiveness was a study inclusion criterion in 283 (31%) of the 914 subjects studied. Prestudy therapy included transsphenoidal surgery and/or radiotherapy in most patients, sc octreotide in 42%, lanreotide SR in 18%, and no prior therapy in 11% (these patients were included in this secondary analysis because their data were not provided separately from the secondary therapy subjects). One hundred fourteen patients (12.5%) were reported to drop out of the studies before the per-protocol study period.
All lanreotide SR trials were open-label prospective studies. Criteria for efficacy of lanreotide SR included normalization of IGF-I in 18 of 19 studies (IGF-I normative data were clearly stated to be age-adjusted in 13 of 18). Efficacy was also defined by a mean GH less than 2.5 µg/liter in 12 of 19 studies, a mean GH less than 5 µg/liter in two studies, a mean GH less than 3.75 in two studies, and 50% GH suppression in one study. At the completion of dose escalation, the dose was 30 mg in 75.3% of patients (dosed every 10 d in 36.8%, every 14 d in 56%, every 7 d in 6.9%, or every 21 d in less than 1%); 60 mg in 24.6% (dosed > 28 d in 9.4%, every 28 d in 29.2%, every 21 d in 12.5%, every 14 d in 35.4%, every 10 d in 9.4%, and every 7 d in 4%). Mean study duration was 15.5 months (range, 3–36 months). Subjects received an average of 14.9 person months of lanreotide SR therapy.
Biochemical efficacy of the octreotide LAR, lanreotide SR, and sc administered octreotide, when given as primary therapy, was examined in 17 open-label prospective studies (1, 3, 14, 15, 16, 19, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39), including 326 subjects (115 males and 147 females; gender breakdown was not specified in 64 subjects from one study), with a mean age of 48.9 yr (range, 18–81 yr). Study inclusion into the primary therapy analysis required that biochemical and other data for the primary therapy group be given separately. Of the studies, 13 included sc octreotide (266 subjects), one with octreotide LAR (15 subjects), two with lanreotide SR (25 subjects), and one with both octreotide LAR and lanreotide SR (20 subjects). The biochemical criteria for study entry were: elevated IGF-I along with a post-OGTT GH more than 1 µg/liter in 15.5% of subjects, elevated IGF-I and post-OGTT GH more than 2 µg/liter in 21%, elevated IGF-I and post-OGTT GH more than 5 µg/liter in 4%, mean GH more than 2 µg/liter and post-OGTT GH more than 0.8 µg/liter in 2.8%, OGTT GH more than 2 µg/liter in 7.4%, elevated GH in 5.7%, and elevated IGF-I and basal GH more than 4 µg/liter in 5.1%; criteria for study entry were not specified in 24%. Prestudy demonstration of octreotide responsiveness was a study inclusion criterion in 14 (4%) of the 326 subjects studied. Thirty-two subjects (9.1%) dropped out of the studies before the per-protocol study period.
Criteria for efficacy of primary somatostatin analog therapy included normalized IGF-I in 17 studies (IGF-I normative data were clearly stated to be age-adjusted in eight of 17). Efficacy was also defined by a mean GH less than 2.5–5 µg/liter in all studies and by post-OGTT GH less than 1 µg/liter in four studies. Mean study duration was 14.3 months (range, 3–24 months). Subjects received an average of 16.2 person months of therapy.
Effect of somatostatin analog therapy on tumor size
Inclusion in the tumor size analysis required that the exact numbers of subjects of the total assessed who had tumor shrinkage and the amount of tumor shrinkage be clearly reported. The effect of octreotide LAR as secondary therapy on pituitary tumor size was assessed in 132 subjects from five studies (1, 3, 5, 11, 12). Tumor size was assessed by computed tomography scan in 35 subjects, by magnetic resonance imaging in 94, and by unspecified imaging modality in the remainder. The radiologists were blinded to subjects’ outcome in four of the five studies. Three studies provided data on shrinkage in relation to pretherapy tumor size classification (microadenoma, remnant, or macroadenoma).
The effect of lanreotide SR used as secondary therapy on tumor size was assessed in 248 patients from eight studies (14, 15, 16, 24, 26, 27, 28, 35). Of these, 18 subjects were studied by computed tomography scan, 139 by magnetic resonance imaging, and the remainder by unspecified imaging modality. The radiologist was blinded to outcome in three of the 10 studies. In four of the eight studies, tumor size was specified (microadenoma in 42 and macroadenoma in 37 subjects), but only two studies provided data on shrinkage in relation to pretherapy tumor size classification.
The effect of primary somatostatin analog therapy on tumor size was assessed in data from 26 studies (1, 3, 5, 9, 11, 12, 14, 15, 19, 20, 28, 29, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44) including 591 subjects, of which 71 were treated with lanreotide SR (six studies), 52 with octreotide LAR (five studies), and 468 with octreotide sc. Because many fewer subjects received primary therapy with octreotide LAR or lanreotide SR than with octreotide, the three drugs could not be compared directly with much assurance. Thirteen studies provided data on shrinkage in relation to pretherapy tumor size classification. Subjects were followed for an average of 34.5 months, or 2875 person years, of somatostatin analog therapy.
Tumor shrinkage was assessed, in some studies, based on percent decrease in maximal tumor diameter and, in others, by percent decrease in tumor volume. Of the secondary therapy studies, 66% of lanreotide studies and 87% of octreotide LAR studies used volumetric assessments. Of the primary therapy studies, all primary lanreotide SR and octreotide LAR studies and 27% of octreotide sc studies used volumetric assessment.
Statistical analysis
Analysis of mean biomarker measurements. A mixed-effects linear regression analysis was used to examine differences in biomarker measurements among treatment groups, and to explore the relative importance of study-specific, subject-specific, and residual components of variability. The influences of preselection and time-on-treatment were also examined. Pre- and posttreatment measurements were examined separately. In these analyses, biomarker measurements were modeled as fixed effects corresponding to treatment, a random study-specific effect, a random subject-specific effect, and an independent residual component of variability. Although subject-specific data were not available for most of the studies, measurements at multiple time points and differences in study sizes allowed modeling of subject-specific variance components; the power to discriminate subject-specific components, however, was weak. Data on the effect of primary octreotide LAR or lanreotide SR on mean biomarker levels were only available on 58 patients from three studies, which did not enable statistical comparison of these levels with those achieved with primary octreotide therapy or comparison with their use as secondary therapy, but descriptive data on their efficacy as primary therapy were collected. The minimum number of subjects providing data in any study at any time point was two, and the maximum was 149; the average number was approximately 38. Approximately 25% of the measurements used in the analysis were taken at 3 months on treatment, approximately 30% were taken from after 3 months through 6 months on treatment, approximately 20% were taken after 1 yr of treatment, and the remaining (approximately 25%) were taken after 1 yr through up to 3 yr. Only measurements where there was a corresponding pretreatment mean were used in the analysis.
Two studies (3, 19) providing subject-level data were also examined: the distribution of the measurements was examined informally using plots and summary statistics to gain insight into the results of the analysis of the mean measurements.
Analysis of proportions meeting efficacy criteria. Logistic regression models were used to examine the influence of drug, time-on-treatment, and preselection on the likelihood of meeting the efficacy criteria. In these models, an overdispersion parameter (deviance divided by degrees of freedom) was estimated as a way of examining whether the variation between proportions was consonant with random variation or whether it indicated heterogeneity between studies. Time-on-treatment was included in the model as a linear function of time.
Statistical analysis of tumor size. Because most studies did not provide tumor shrinkage data for each individual patient, and for most studies only the overall range of tumor shrinkage was reported, the tumor shrinkage data required simplification for analysis. The reported shrinkage in each study was taken with the best fit to the following categories: none (or <10%), at least 10% but less than 25%, 25–50%, and more than 50%. The response variables were the percentages of patients who had clinically relevant shrinkages within these shrinkage categories. Pooled percentage rates of shrinkage and the corresponding exact 95% confidence intervals (CIs) were calculated. The odds ratios (ORs) and the associated 95% CIs for the OR, for comparisons of the effect of the different somatostatin analogs and primary vs. secondary therapy on tumor shrinkage, were calculated. To explore possible significant factors influencing the percentage of shrinkage, logistic regression analyses were performed. For these exploratory analyses, the independent variables were: the drugs (octreotide LAR, octreotide sc, lanreotide SR), prior somatostatin analog therapy (yes, no), months of treatment, and preselection (yes, no). For another response variable, tumor size increase, pooled percentages of patients who had an increase in tumor size were calculated, and a logistic regression analysis was attempted on the tumor size increase data. Analyses were performed with SPSS 11.0 for the Macintosh (SPSS, Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Analysis of mean biomarker (GH and IGF-I) levels.
Estimated pre- and posttreatment mean GH and IGF-I levels in subjects who were not preselected for somatostatin analog responsiveness, those who were preselected, and all subjects combined are shown in Table 1
.
|
Preselection was not a significant predictor of posttreatment mean GH or IGF-I levels; posttreatment mean biomarker levels were somewhat, but not significantly, lower among preselected vs. unselected subjects. The expected effect of preselection vs. no preselection on GH levels was estimated to be –2.03 ± 1.59 µg/liter (P = 0.21). The corresponding estimate for IGF-I was –70.9 ± 38 ng/ml (P = 0.08). In other words, study subjects preselected for octreotide responsiveness could be expected to have an IGF-I of 70.9 ng/ml less than nonselected subjects. Inclusion of pretherapy GH or IGF-I levels as fixed effects in the model did not change the significance of these results.
We did consider whether potential posttreatment differences in mean biomarker levels could have been obscured by the substantial subject- and study-specific variability that our analyses detected. Analysis of individual patient data from two studies revealed the presence of outlying large values; posttreatment median and maximum levels of GH and IGF-I in the first study were 1.5 and 12.0 µg/liter and 210 and 750 ng/ml; and in the second, these were 3.0 and 33.0 µg/liter and 280 and 720 ng/ml. Sample correlations between pre- and posttreatment means in the first and second studies, respectively, were 0.15 and 0.49 for GH and 0.39 and 0.19 for IGF-I, indicating substantial subject-specific variability. It is not unexpected, therefore, that the variance components estimation was poor. The estimates of the SD values of the subject- and study-specific components of variance for GH were unreliable (most likely because outliers within each study led to instability in the analysis), and these estimates for IGF-I had SD values of 382 ng/ml and zero. The residual variability was estimated to have SD values of 12 ng/ml for GH and 262 ng/ml for IGF-I. The estimated coefficient for the addition of pretreatment means as a fixed effect is nominally statistically significant and positive. Its inclusion in the analysis decreases the proportion of variability explained by subject- and study-specific components of variance but does not have any important effect on inferences about the relative efficacy of the different drug treatments. The apparent positive association between pre- and posttreatment measurements is further evidence for nontrivial subject- or study-specific effects.
The mixed-effects linear regression analysis did not find significant changes in IGF-I (P = 0.66) or GH (P = 0.66) means beyond that reached at the 3-month minimum time-on-trial represented in the data set. Again, the variability described above may have limited the detection of further, yet smaller, reductions in mean levels with continued treatment beyond 3 months.
Analysis of proportions meeting GH or IGF-I efficacy criteria.
The proportions of subjects who met GH or IGF-I criteria for efficacy with secondary octreotide LAR or lanreotide SR therapy are shown in Table 1
.
Among subjects unselected for somatostatin analog responsiveness, criteria for GH suppression were met in a greater proportion of subjects treated with octreotide LAR (54 ± 0.002%) vs. lanreotide SR (48 ± 0.002%) (B = 0.2310; P = 0.016). This was also true among all subjects combined (B = –0.1568; P = 0.01) but not among the preselected subjects (B = –0.1474; P = 0.1209). Overall, somatostatin analog responsiveness as a criterion for study entry was not a significant predictor of meeting GH criteria for efficacy (B = 0.0992; P = 0.12), and the rates of meeting GH efficacy criteria did not differ in unselected vs. preselected subjects (B = 0.05133; P = 0.515). However, there was evidence of substantial interstudy heterogeneity (overdispersion estimate = 1.76) that could have affected these results. The proportion of subjects meeting efficacy criteria for GH increased significantly over time (B = 0.0046; P = 0.04); the estimated OR was 1.46/yr.
IGF-I normalization was achieved in a greater proportion of unselected subjects treated with octreotide LAR (63 ± 0.002%) vs. lanreotide SR (42 ± 0.002%) (B = 0.2325; P = 0.007). This was also true among all subjects combined (B = 0.2056; P = 0.0009) but not among the preselected group (B = –0.08667; P = 0.3310). Overall, preselection was a positive predictor of IGF-I normalization (B = 0.1213; P = 0.04), and the rate of IGF-I normalization was greater among preselected than unselected subjects (B = 0.1472; P = 0.0475). However, although a significantly larger proportion of patients in the octreotide LAR trials than in the lanreotide SR trials had been preselected for study entry based on somatostatin analog responsiveness (70 vs. 31%; P < 0.0001), preselection was not found to be a modifier for drug effect; and when the effect of preselection was controlled for, IGF-I normalization rate remained significantly greater among octreotide LAR-treated subjects (B = 0.1213; P = 0.04). The proportion of subjects achieving IGF-I normalization also increased significantly with time (P < 0.001); the estimated OR was 1.34/yr. The time rate of change in IGF-I normalization did not differ among drugs (P = 0.17). A logistic regression analysis of subject-specific data from 139 patients (3, 15, 16, 19, 26) found IGF-I normalization rate to be unrelated to pretherapy pituitary tumor size (B = –0.251; P = 0.53). Study size had no significant effect on the proportion of subjects meeting efficacy criteria for GH or IGF-I.
In a subanalysis, the relationship between efficacy and final study visit drug dose was examined in subject-specific data in 117 patients (3, 15, 16, 19, 26). This analysis demonstrated that average drug dose was higher in subjects whose IGF-I did not normalize (n = 32) vs. those in whom it did (n = 82) (P < 0.001). In data summarized from three studies (2, 9, 10), the rate of IGF-I normalization tended to be lower as octreotide LAR dose was raised: 10 mg (90%), 20 mg (61%), and 30 mg (53%) (P = 0.15). In data from two studies (27, 28), increased frequency of lanreotide SR administration was associated with reduced IGF-I normalization: every 21 d (100%), every 14 d (83%), every 10 d (50%), and every 7 d 20% (P = 0.06). These findings are not unexpected, given that, in general, these studies conducted dose escalation and shortening of dosing interval only in subjects whose initial response to somatostatin analogs was inadequate.
Somatostatin analogs as primary vs. secondary therapy
As shown in Table 1, posttreatment mean GH levels did not differ in primary octreotide vs. secondary octreotide LAR-treated groups (B = 0.162; P = 0.1764). Posttreatment mean IGF-I levels in the primary octreotide vs. secondary octreotide LAR treatment groups also did not differ significantly (B = 7.9; P = 0.83). The proportions of subjects meeting criteria for GH efficacy were not significantly different in the unselected primary octreotide vs. secondary octreotide LAR treatment groups (B = 0.1751; P = 0.0767) or in the overall primary octreotide and secondary octreotide LAR treatment groups (B = 0.099; P = 0.15). The time rate of change in meeting GH criteria did not differ with primary octreotide vs. secondary octreotide LAR treatment (P = 0.18).
There was marginal evidence that IGF-I normalization occurred in a greater proportion of secondary octreotide LAR- vs. primary octreotide-treated subjects overall (B = 0.1298; P = 0.055) but not among unselected subjects only (B = 0.1314; P = 0.1217). The effect of preselection on efficacy with primary octreotide could not be examined because preselection was reported only in one octreotide trial.
Too few studies were available with which to directly compare the different drug preparations when used as primary therapy.
Pretherapy GH and IGF-I levels in relation to efficacy
Overall, pretherapy GH level was a negative predictor of meeting biochemical criteria for control with somatostatin analog therapy (B = –0.00472; P = 0.05). However, among the secondary therapy group, this relationship was significant among lanreotide SR (B = –0.0105; P = 0.0145) but not octreotide LAR (B = –0.00121; P = 0.83)-treated subjects. Among primary octreotide-treated subjects, there was a trend only for pretherapy GH to be a negative predictor of efficacy (B = –0.00341; P = 0.18). Mean pretherapy GH level was higher in the primary octreotide than the secondary octreotide LAR-treated group (B = –25.35; P < 0.0001). Despite these data, the proportions of subjects meeting GH criteria for efficacy were not significantly different in the primary octreotide and secondary octreotide LAR-treated groups. Baseline IGF-I was not a significant predictor of efficacy (B = –0.00008; P = 0.63) overall or within any of the treatment groups.
Analysis of the effect of study dropouts on efficacy
A sensitivity analysis was attempted to determine the potential effect of dropouts on reported study efficacy, but the efficacy analysis was not robust to this sensitivity analysis. Therefore, no conclusion can be drawn about the effect of dropouts on efficacy.
Analysis of tumor size changes with somatostatin analogs (Table 2)
Estimates of tumor shrinkage are shown in Table 2. Among unselected subjects, tumor shrinkage of more than 10% occurred in a greater percentage of LAR- vs. lanreotide-treated patients [OR = 5.78 (95% CI, 2.39 and 13.99); P = 0.0001], but the odds of tumor shrinkage greater than 25% did not differ among the two drugs (P = 0.11). Reliability of these estimates, however, needs to consider the very small number of subjects in some of the groups (Table 2). Among all subjects who received secondary somatostatin analog therapy, tumor shrinkage of greater than 10%, as well as greater than 25%, occurred in a greater proportion of octreotide LAR- vs. lanreotide SR-treated patients [more than 10%: OR = 3.07 (95% CI, 2.0 and 4.76), P <. 0001; more than 25%: OR = 1.97 (95% CI, 1.12 and 3.2), P < 0.0062].
|
Tumor shrinkage was assessed after a mean of 17.9 ± 1.52 months of therapy. The numbers of studies reporting tumor shrinkage statistics at each time point are as follows: 3 months, n = 1; 6–7 months, n = 10; 10–12 months, n = 8; 18 months, n = 4; 24 months, n = 13; 30–36 months, n = 6. Increasing number of months of therapy beyond 6 months was not associated with an increased likelihood of tumor shrinkage.
Among the secondary therapy subjects, the impact of prior somatostatin analog therapy and the duration of washout from this therapy on the odds of tumor shrinkage were also assessed. Among unselected subjects, prior somatostatin analog therapy was associated with a decreased odds of tumor shrinkage [OR = 0.259 (95% CI, 0.102 and 0.662); P = 0.0005]; whereas among subjects selected for somatostatin analog responsiveness, prior somatostatin analog use increased the odds of tumor shrinkage [OR = 3.238 (95% CI, 1.668 and 6.29); P = 0.0047]. Among unselected subjects, increasing length of washout from prior somatostatin analog therapy did not significantly affect the odds of tumor shrinkage [OR = 0.669 (95% CI, 0.40 and 1.09); P = 0.10]; whereas among selected subjects, increasing length of washout from prior somatostatin analog therapy was associated with significantly increased odds of tumor shrinkage [OR = 2.21 (95% CI, 1.425 and 3.45); P = 0.004].
Among subjects who received somatostatin analog as primary therapy, tumor shrinkage of at least 10% occurred in a greater percentage of octreotide LAR- than lanreotide SR-treated subjects [OR = 7.04 (95% CI, 2.67 and 18.5); P = 0.01]. Tumor shrinkage of at least 10% was also reported more often in subjects treated with primary octreotide LAR than primary octreotide sc [OR = 9.4 (95% CI, 3.95 and 22.7); P < 0.0001]. There was a trend for shrinkage greater than 50% to occur more often with primary octreotide LAR than primary lanreotide SR therapy [OR = 16.6 (95% CI, 142 and 2.05); P = 0.06]. Primary octreotide and primary octreotide LAR therapy did not differ significantly in the proportion of patients who had tumor shrinkage more than 25% or more than 50%. However, many fewer subjects received either octreotide LAR or lanreotide SR as primary therapy than those who received octreotide sc as primary therapy. It is unknown whether this difference could have contributed to these findings. Preselection for somatostatin analog responsiveness was not a significant predictor of tumor shrinkage with primary therapy (P = 0.67). However, this analysis is also limited by the small number of preselected subjects who received primary therapy with octreotide LAR or lanreotide SR and the fact that none of the octreotide sc subjects were preselected.
Secondary therapy with octreotide LAR was associated with lower odds of tumor shrinkage than primary octreotide sc [OR = 0.432 (95% CI, 0.329 and 0.567); P < 0.0001]. This finding remained significant after accounting for the positive effect of preselection on tumor shrinkage in the secondary therapy group.
Although significant differences between the drugs, on average, were found, tumor shrinkage rates varied considerably, so that most individual study results were outside of the 95% CI for the pooled estimates. In addition, for our model, r2 = 0.102, which demonstrates that important explanatory variables for tumor shrinkage remain unaccounted for.
A subanalysis of the differential effect of somatostatin analogs in patients with baseline microadenomas vs. those with macroadenomas was also conducted. Secondary somatostatin analog therapy led to more than 20% tumor shrinkage in 26.3% (95% CI, 12.3 and 40.3) of microadenomas (n = 17) vs. 39.5% (95% CI, 28.7 and 50.8) of macroadenomas (n = 43) (P = 0.29). Of the primary somatostatin analog therapy group, more than 20% tumor shrinkage occurred in 35% (95% CI, 22 and 44.7) of microadenomas (n = 75) vs. 47.4% (95% CI, 39.5 and 55.3) of macroadenomas (n = 259) (P = 0.05).
Of 468 subjects in the primary octreotide therapy group, an increase in tumor size during therapy was reported in seven subjects (1.6%; 95% CI, 0.7 and 3.32). In the secondary therapy group, tumor size increase was reported in four of 286 subjects (1.4%; 95% CI, 0.1 and 3.5) treated with lanreotide SR and in one of 145 patients (0.7%; 95% CI, 0.0 and 3.8) treated with octreotide LAR. With all trials combined, the overall rate of tumor increase was 1.4% (95% CI, 0.1 and 2.8). Too few cases of tumor size increase were reported and too little biochemical data on these cases were available for a logistic regression model to identify predictors of tumor increase.
Although our meta-analysis included only trials of at least 3 months duration, an additional large trial examined tumor size with 1 month of lanreotide SR therapy in 73 patients (45). In a supplementary analysis that included this trial, secondary therapy with lanreotide was associated with an increase in tumor size in 19 of 145 patients (4.7%; 95% CI, –0.25 and 12); and overall, tumor increase was reported in 25 of 937 patients (2.67%; 95% CI, –0.2 and 5.5) treated with any of the three somatostatin analogs.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our meta-analysis has revealed that, among subjects unselected for somatostatin analog responsiveness, IGF-I levels normalized and GH criteria for efficacy were met in a higher proportion of octreotide LAR- than lanreotide SR-treated subjects. These results were qualitatively similar to the results of the analysis of the mean biomarker measurements, and together they suggest somewhat greater efficacy of octreotide LAR than lanreotide SR for the secondary treatment of acromegaly.
Another important finding of our meta-analysis was that preselection for octreotide responsiveness appeared to confer no efficacy advantage with respect to meeting GH efficacy criteria but did confer a modest, but significant, benefit for IGF-I normalization. We were also able to show that the greater efficacy of octreotide LAR could not be explained by a higher rate of preselection among subjects studied with this analog. It might have been expected that preselection would have conferred a greater efficacy advantage, in particular on GH normalization. Failure to demonstrate this advantage could not be explained by differences in baseline GH levels because these were not significantly higher in the unselected group. It is also possible that the criteria used to define somatostatin analog sensitivity in the trials we analyzed were not strict by today’s standards. These criteria were an acute response to short-acting octreotide of a suppression of GH by 50% in three studies and by 20% in one study and a fall of GH to less than 5 µg/liter in three studies or to less than 10 µg/liter in three others (two of which were the largest studies, totaling approximately 200 patients). It is unknown whether more stringent criteria for somatostatin analog sensitivity would better predict somatostatin analog efficacy. In one study that directly examined this question, acute octreotide responsiveness, classified as a 50% fall in GH, did not predict long-term normalization of GH and IGF-I (46).
Our analysis also examined the differential efficacy of primary octreotide vs. secondary octreotide LAR therapy. We found that efficacy with respect to GH levels did not differ and found only marginal evidence for a greater IGF-I normalization with secondary octreotide LAR therapy. We do need to consider whether higher pretherapy GH levels and the overall negative relationship between pretherapy GH levels and efficacy we found could have confounded the results of our primary vs. secondary therapy analysis. The fact that posttherapy levels were similar in the two groups, yet baselines were higher, does suggest that primary therapy was more effective at lowering GH levels. However, individual subject data with subjects matched for characteristics such as pretherapy GH levels and tumor size will be needed to examine this possibility further. Unfortunately, a drawback to our analysis is that we were unable to examine each drug as both primary and secondary therapy or compare all three drugs as primary therapy, because too few studies to date have assessed primary octreotide LAR (3, 39) and primary lanreotide therapy (14, 15, 39) to allow for a statistically valid analysis.
In contrast to baseline GH levels, baseline IGF-I did not predict response to somatostatin analogs. There are a number of potential explanations for this finding. One is that washout from prior therapy in the secondary treatment groups may have been insufficient for IGF-I, with a much longer half-life than GH, to rise to true pretreatment levels. Also, IGF-I values plateau despite increasingly higher GH levels (47), so baseline GH levels may be more predictive of disease severity and thus be more predictive of reduced somatostatin analog efficacy than baseline IGF-I levels.
In our meta-analysis, we found a significant impact of treatment duration on efficacy. Our analysis showed that maximal effect on mean GH and IGF-I levels seemed to occur by the 3- to 6-month treatment point, when these levels reached a plateau. By contrast, analysis of the proportions meeting efficacy, which is less susceptible to intersubject variations, discerned an appreciable effect of time-on-trial, the majority of which also occurred by 6 months but which also continued beyond this time. Although we were unable to quantify the effect of dropout rate on reported efficacy, substantial numbers of study dropouts may have biased the results to greater efficacy in studies of the longest duration. Dropouts may have been difficult to quantify, because studies varied considerably in how dropouts were reported. Many did not comment on dropout rate, raising the possibility that this was underreported.
Despite the value of the efficacy estimates that we have derived in this meta-analysis, there was evidence for substantial subject-specific or study-specific effects that could have obscured differences in biochemical efficacy among the treatment types. Some possible nonquantifiable subject-to-subject variability may be attributed to differences in prior therapy. Potential study-to-study differences may have included nonuniform subject entry criteria (especially for GH), variability in the efficacy criteria as demonstrated by the substantial overdispersion estimates, variability of GH and IGF-I assays (very likely variable but impossible to correct for), and some difference in drug dosing among studies.
Our meta-analysis also revealed that preselection for somatostatin responsiveness increased the odds of tumor shrinkage, but that tumor shrinkage remained more likely with octreotide LAR than in those treated with lanreotide SR when the preselection effect was accounted for. Our analysis also revealed that tumor shrinkage occurred more often when octreotide LAR was given as primary vs. secondary therapy and that, among primary therapy groups, tumor shrinkage was greatest with octreotide LAR. However, it is important to recognize that the primary and secondary therapy groups were not matched for potential factors that could have influenced tumor shrinkage, such as baseline tumor size. In addition, macroadenomas, which were more prevalent in the primary therapy groups, showed greater shrinkage than microadenomas overall. It is unknown whether a true difference exists in the effect of somatostatin analogs on tumor shrinkage in de novo patients vs. those that have received prior therapies. Because the tumor size at diagnosis in the secondary therapy group was unknown, these differences could not be factored into the analysis. It should also be recognized that studies differed in their methods of tumor shrinkage assessment, and these differences could not be factored into the analysis. As was the case with the biomarker data, there was considerable variability among studies in terms of the percentage of patients who had tumor shrinkage as well as the amount of shrinkage. Individual subject data would have been necessary to ascertain sources of these differences.
This analysis also revealed that an increase in tumor size while on somatostatin analog therapy was very rare, being reported in less than 2% of patients in the trials analyzed in this meta-analysis. Including an additional trial, which alone reported an increase in tumor size in 18% of patients, the overall rate of tumor increase was 2.67% (45). However, the results of this trial need to be interpreted cautiously because baseline and follow-up imaging studies were separated by 1 month or less of lanreotide SR therapy, which is unlikely to be reliable information on tumor size changes.
In conclusion, our meta-analysis provides new data on the expected efficacy of long-acting somatostatin analogs in subjects unselected for somatostatin analog responsiveness. Our analysis suggests that prior efficacy estimates for IGF-I normalization may have been confounded by a positive effect of preselection on IGF-I normalization rate. We have also been able to show that octreotide LAR remains more efficacious than lanreotide SR when potential preselection effects are removed. We have also shown that preselection for somatostatin analog responsiveness positively impacts on the potential for tumor shrinkage and that primary somatostatin analog therapy is more likely to produce significant tumor shrinkage than if the analogs are given as secondary therapy. Additional studies are needed to examine prospectively the biochemical efficacy of the analogs when given as primary vs. secondary therapy.
|
Footnotes |
|---|
First Published Online May 10, 2005
Abbreviations: CI, Confidence interval; LAR, long-acting release; OR, odds ratio(s); SR, slow release.
Received February 7, 2005.
Accepted May 4, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Colao, R. S. Auriemma, M. Galdiero, G. Lombardi, and R. Pivonello Effects of Initial Therapy for Five Years with Somatostatin Analogs for Acromegaly on Growth Hormone and Insulin-Like Growth Factor-I Levels, Tumor Shrinkage, and Cardiovascular Disease: A Prospective Study J. Clin. Endocrinol. Metab., October 1, 2009; 94(10): 3746 - 3756. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ning, S. J Knox, G. R Harsh, M. D Culler, and L. Katznelson Lanreotide promotes apoptosis and is not radioprotective in GH3 cells Endocr. Relat. Cancer, September 1, 2009; 16(3): 1045 - 1055. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Giustina, S. Bonadonna, G. Bugari, A. Colao, R. Cozzi, S. Cannavo, L. de Marinis, E. degli Uberti, F. Bogazzi, G. Mazziotti, et al. High-dose intramuscular octreotide in patients with acromegaly inadequately controlled on conventional somatostatin analogue therapy: a randomised controlled trial Eur. J. Endocrinol., August 1, 2009; 161(2): 331 - 338. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Colao, R. S. Auriemma, S. Savastano, M. Galdiero, L. F. S. Grasso, G. Lombardi, and R. Pivonello Glucose Tolerance and Somatostatin Analog Treatment in Acromegaly: A 12-Month Study J. Clin. Endocrinol. Metab., August 1, 2009; 94(8): 2907 - 2914. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Mazziotti, I. Floriani, S. Bonadonna, V. Torri, P. Chanson, and A. Giustina Effects of Somatostatin Analogs on Glucose Homeostasis: A Metaanalysis of Acromegaly Studies J. Clin. Endocrinol. Metab., May 1, 2009; 94(5): 1500 - 1508. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Marazuela, T. Lucas, C. Alvarez-Escola, M. Puig-Domingo, N. G. de la Torre, P. de Miguel-Novoa, A. Duran-Hervada, R. Manzanares, M. Luque-Ramirez, I. Halperin, et al. Long-term treatment of acromegalic patients resistant to somatostatin analogues with the GH receptor antagonist pegvisomant: its efficacy in relation to gender and previous radiotherapy Eur. J. Endocrinol., April 1, 2009; 160(4): 535 - 542. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Sherlock, E. Fernandez-Rodriguez, A. A. Alonso, R. C. Reulen, J. Ayuk, R. N. Clayton, G. Holder, M. C. Sheppard, A. Bates, and P. M. Stewart Medical Therapy in Patients with Acromegaly: Predictors of Response and Comparison of Efficacy of Dopamine Agonists and Somatostatin Analogues J. Clin. Endocrinol. Metab., April 1, 2009; 94(4): 1255 - 1263. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Colao, R. S. Auriemma, M. Galdiero, P. Cappabianca, L. M. Cavallo, F. Esposito, L. F. S. Grasso, G. Lombardi, and R. Pivonello Impact of Somatostatin Analogs Versus Surgery on Glucose Metabolism in Acromegaly: Results of a 5-Year Observational, Open, Prospective Study J. Clin. Endocrinol. Metab., February 1, 2009; 94(2): 528 - 537. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Jimenez, P. Burman, R. Abs, D. R Clemmons, W. M Drake, K. R Hutson, M. Messig, M. O Thorner, P. J Trainer, and R. F Gagel Follow-up of pituitary tumor volume in patients with acromegaly treated with pegvisomant in clinical trials Eur. J. Endocrinol., November 1, 2008; 159(5): 517 - 523. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S Petersenn, M Buchfelder, M Reincke, C M Strasburger, H Franz, R Lohmann, H-J Quabbe, U Plockinger, and the Participants of the German Acromegaly Register Results of surgical and somatostatin analog therapies and their combination in acromegaly: a retrospective analysis of the German Acromegaly Register Eur. J. Endocrinol., November 1, 2008; 159(5): 525 - 532. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Colao, R. Pivonello, R. S. Auriemma, M. Galdiero, S. Savastano, L. F. S. Grasso, and G. Lombardi Growth Hormone-Secreting Tumor Shrinkage after 3 Months of Octreotide-Long-Acting Release Therapy Predicts the Response at 12 Months J. Clin. Endocrinol. Metab., September 1, 2008; 93(9): 3436 - 3442. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Beckers Does Preoperative Somatostatin Analog Treatment Improve Surgical Cure Rates in Acromegaly? A New Look at an Old Question J. Clin. Endocrinol. Metab., August 1, 2008; 93(8): 2975 - 2977. [Full Text] [PDF]
|
|
|
|
|
|
R. D. Murray and S. Melmed A Critical Analysis of Clinically Available Somatostatin Analog Formulations for Therapy of Acromegaly J. Clin. Endocrinol. Metab., August 1, 2008; 93(8): 2957 - 2968. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Colao, R. Pivonello, M. Galderisi, P. Cappabianca, R. S. Auriemma, M. Galdiero, L. M. Cavallo, F. Esposito, and G. Lombardi Impact of Treating Acromegaly First with Surgery or Somatostatin Analogs on Cardiomyopathy J. Clin. Endocrinol. Metab., July 1, 2008; 93(7): 2639 - 2646. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Plockinger, S. Albrecht, C. Mawrin, W. Saeger, M. Buchfelder, S. Petersenn, and S. Schulz Selective Loss of Somatostatin Receptor 2 in Octreotide-Resistant Growth Hormone-Secreting Adenomas J. Clin. Endocrinol. Metab., April 1, 2008; 93(4): 1203 - 1210. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. F Taboada, R. M Luque, L. V. Neto, E. d. O Machado, B. C Sbaffi, R. C Domingues, J. B Marcondes, L. M C Chimelli, R. Fontes, P. Niemeyer, et al. Quantitative analysis of somatostatin receptor subtypes (1-5) gene expression levels in somatotropinomas and correlation to in vivo hormonal and tumor volume responses to treatment with octreotide LAR Eur. J. Endocrinol., March 1, 2008; 158(3): 295 - 303. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. L Ronchi, E. Rizzo, A. G Lania, R. Pivonello, S. Grottoli, A. Colao, E. Ghigo, A. Spada, M. Arosio, and P. Beck-Peccoz Preliminary data on biochemical remission of acromegaly after somatostatin analogs withdrawal Eur. J. Endocrinol., January 1, 2008; 158(1): 19 - 25. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Abrams, O. Alexopoulou, R. Abs, D. Maiter, and J. Verhelst Optimalization and cost management of lanreotide-Autogel therapy in acromegaly Eur. J. Endocrinol., November 1, 2007; 157(5): 571 - 577. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Colao, R. Pivonello, R. S Auriemma, M. Galdiero, S. Savastano, and G. Lombardi Beneficial effect of dose escalation of Octreotide-LAR as first-line therapy in patients with acromegaly Eur. J. Endocrinol., November 1, 2007; 157(5): 579 - 587. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Bex, R. Abs, G. T'Sjoen, J. Mockel, B. Velkeniers, K. Muermans, and D. Maiter AcroBel the Belgian registry on acromegaly: a survey of the 'real-life' outcome in 418 acromegalic subjects Eur. J. Endocrinol., October 1, 2007; 157(4): 399 - 409. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Resmini, P. Dadati, J.-L. Ravetti, G. Zona, R. Spaziante, A. Saveanu, P. Jaquet, M. D. Culler, F. Bianchi, A. Rebora, et al. Rapid Pituitary Tumor Shrinkage with Dissociation between Antiproliferative and Antisecretory Effects of a Long-Acting Octreotide in an Acromegalic Patient J. Clin. Endocrinol. Metab., May 1, 2007; 92(5): 1592 - 1599. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Colao, M. Filippella, R. Pivonello, C. Di Somma, A. Faggiano, and G. Lombardi Combined therapy of somatostatin analogues and dopamine agonists in the treatment of pituitary tumours Eur. J. Endocrinol., April 1, 2007; 156(suppl_1): S57 - S63. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E Ferrante, C Pellegrini, S Bondioni, E Peverelli, M Locatelli, P Gelmini, P Luciani, A Peri, G Mantovani, S Bosari, et al. Octreotide promotes apoptosis in human somatotroph tumor cells by activating somatostatin receptor type 2. Endocr. Relat. Cancer, September 1, 2006; 13(3): 955 - 962. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Colao, R. Pivonello, R. S. Auriemma, F. Briganti, M. Galdiero, F. Tortora, F. Caranci, S. Cirillo, and G. Lombardi Predictors of Tumor Shrinkage after Primary Therapy with Somatostatin Analogs in Acromegaly: A Prospective Study in 99 Patients J. Clin. Endocrinol. Metab., June 1, 2006; 91(6): 2112 - 2118. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Cozzi, M. Montini, R. Attanasio, M. Albizzi, G. Lasio, S. Lodrini, P. Doneda, L. Cortesi, and G. Pagani Primary Treatment of Acromegaly with Octreotide LAR: A Long-Term (Up to Nine Years) Prospective Study of Its Efficacy in the Control of Disease Activity and Tumor Shrinkage J. Clin. Endocrinol. Metab., April 1, 2006; 91(4): 1397 - 1403. [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 |
