Growth Hormone Receptor Variants and Response to Pegvisomant in ...

2 downloads 42 Views 143KB Size Report
Dec 7, 2011 - Pegvisomant in Monotherapy or in Combination with. Somatostatin Analogs in Acromegalic Patients: A Multicenter Study. M. Filopanti, L. Olgiati ...
J C E M

O N L I N E

Hot Topics in Translational Endocrinology—Endocrine Care

Growth Hormone Receptor Variants and Response to Pegvisomant in Monotherapy or in Combination with Somatostatin Analogs in Acromegalic Patients: A Multicenter Study M. Filopanti, L. Olgiati, G. Mantovani, S. Corbetta, M. Arosio, V. Gasco, L. De Marinis, C. Martini, F. Bogazzi, S. Cannavo`, A. Colao, D. Ferone, G. Arnaldi, F. Pigliaru, A. Peri, G. Angeletti, M. L. Jaffrain-Rea, A. G. Lania, and A. Spada* Context: The influence of full-length GH receptor (GHR) and exon 3-deleted GHR (d3GHR) on responsiveness to pegvisomant (PEG-V) in acromegalic patients is uncertain. Objective: The aim of the study was to assess the distribution of GHR genotypes in a large series of patients on PEG-V therapy and their influence on treatment efficacy and adverse effects. Design and Setting: A cross-sectional multicenter pharmacogenetic study was conducted in 16 Italian endocrinology centers of major universities and tertiary care hospitals. Patients: The study included 127 acromegalic patients enrolled from 2009 to 2010 not cured by previous surgery, radiotherapy, and long-acting somatostatin (SST) analogs, treated with PEG-V. Intervention and Main Outcome Measure: Sixty-three of 127 patients received combined PEG-V ⫹ SST analog therapy. Clinical and hormonal data at diagnosis and before and during PEG-V therapy were inserted in a database. GHR exon 3 deletion and other polymorphisms were genotyped by the coordinator center. Differences in PEG-V dosage required for IGF-I normalization and occurrence of adverse effects between carriers and noncarriers of GHR variants were evaluated. Results: d3GHR variants were not in Hardy-Weinberg equilibrium (P ⫽ 0.008). No association of these variants with PEG-V dose required for IGF-I normalization, adverse effects occurrence, and tumor regrowth was found in patients on PEG-V and on PEG-V ⫹ SST analog treatment. Similar data were obtained considering the GHR variant rs6180. Conclusions: This study did not confirm a better response of d3GHR to PEG-V treatment in acromegaly. Other studies are needed to determine whether deviation from Hardy-Weinberg equilibrium may indicate an association of d3GHR genotype with poor response to usual treatments. (J Clin Endocrinol Metab 97: E165–E172, 2012)

H receptor (GHR) is a transmembrane receptor belonging to the cytokine superfamily. A GHR variant that differs for the genomic retention [full-length GHR (flGHR)] or exclusion (d3GHR) of exon 3, most likely due to an ancestral recombination between two retroelements flanking this exon, has been described in humans (1). Exon 3 deletion results in the lack of amino acid residues from 7 to 28 with the loss of a potential glycosylation site and

G

an amino acidic change in position 6 (alanine to aspartic acid), with changes in size charge and hydrofobicity. In vitro transfection experiments showed that these modifications do not affect GH binding affinity and internalization (2). These data are consistent with the three-dimensional model of GH/GHR complex present in a public database (http://www.ncbi.nlm.nih.gov/structure), showing the peptide sequence coded by exon 3 located upstream

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2012 by The Endocrine Society doi: 10.1210/jc.2011-1769 Received June 15, 2011. Accepted November 3, 2011. First Published Online December 7, 2011

* Author affiliations are shown at the bottom of the next page. Abbreviations: d3GHR, Exon 3-deleted GHR; flGHR, full-length GHR; GHR, GH receptor; HWE, Hardy-Weinberg equilibrium; PEG-V, pegvisomant; rhGH, recombinant human GH; SDS, SD score; SST, somatostatin; ULN, upper limit of normal.

J Clin Endocrinol Metab, February 2012, 97(2):E165–E172

jcem.endojournals.org

E165

E166

Filopanti et al.

Growth Hormone Receptor Variants and Pegvisomant

to the residues involved in GH binding. Although in vitro and in silico data indicate that d3GHR binds and internalizes ligand as efficiently as flGHR, transfection experiments showed that activation of d3GHR resulted in a 30% enhanced signal transduction through the Stat5-dependent pathway in comparison with flGHR (3). Moreover, d3GHR effect was dominant over flGHR (3). This genomic polymorphism has a relatively high frequency in the general population, with 8 –10% of homozygous individuals (4, 5). In the last 5 yr, several pharmacogenetic studies investigated the impact of d3GHR on responsiveness to recombinant human GH (rhGH) treatment in patients with short stature, leading to controversial results (6 –16). A recent systematic review and meta-analysis of the available data supported the association of d3GHR variant with increased response to rhGH therapy in GHdeficient children (17). This pharmacogenetic approach was also applied to acromegaly. In fact, when the first-line therapies fail, i.e. surgical removal of the pituitary adenoma, somatostatin (SST) analogs administration, or radiotherapy, pegvisomant (PEG-V) may be effectively used to control the disease (18). PEG-V is an analog of human GH with antagonistic properties due to residue substitution and pegylation (19). These modifications, which do not involve the peptide sequence coded by exon 3, increase binding affinity to the first GHR binding site and disrupt the binding to the second binding site, blocking the transduction pathway of GHR (19). The mechanisms determining the different PEG-V dosage required to normalize IGF-I levels as well as possible escape to treatment and adverse effects occurring in some patients are largely unknown. In particular, two studies carried out on 19 and 44 acromegalic patients, respectively, suggested that d3GHR variant is associated with a better response to PEG-V therapy (20, 21). The aim of this study was to assess the distribution of different GHR genotypes in a large multicentric series of patients on PEG-V therapy as well as their influence on treatment efficacy and side effects. The secondary objective was to investigate the influence of GHR genotypes on these parameters in patients treated with PEG-V in association with longacting SST analogs.

J Clin Endocrinol Metab, February 2012, 97(2):E165–E172

Patients and Methods Patients The study was a multicenter pharmacogenetic study involving 127 patients with acromegaly on PEG-V treatment. Fourteen patients of the present study had been previously genotyped with the same methods and included in a series of 19 patients already reported (20). The cross-sectional study was conducted in 16 endocrinology centers of major universities and tertiary care hospitals in Italy. Ethic committees from each hospital approved the protocol, and a written consent was obtained from all patients. Patients were enrolled for participation in the study from January 2009 to September 2010. All patients were diagnosed with acromegaly according to the present guidelines and underwent surgical adenomectomy (if not contraindicated) and/or radiotherapy, without disease remission. Mean follow-up was 38 ⫾ 21 months, and the follow-up duration was homogenous between the centers. Patients did not normalize IGF-I levels after 1 yr of long-acting SST analogs at the maximum tolerable dosage. They received PEG-V by daily sc injection, and the dose was titrated to maintain IGF-I levels below the upper limit of normal (ULN) range of each laboratory. The follow-up visits were conducted every 2 months until normal IGF-I levels were reached and every 6 months thereafter. Imaging was repeated 6 months after PEG-V beginning. Online case report forms were used to collect patient demographic, clinical, and biochemical data that were inserted in the database. Each patient received an identification number, and personal information was available only to principal investigators or coinvestigators. Blood and DNA samples, labeled with respective identification numbers, were sent to the coordinating center for genotyping. At the end of collecting data phase, data of 149 patients were placed in the database. Records with incomplete reporting forms or those not followed by biological sample shipment were excluded from the study. Full data and genotypization were obtained in 127 patients.

Methods Hormonal assays Random serum GH and IGF-I levels were measured in hospital laboratories and evaluated according to the local, decadebased reference ranges. To achieve a standardized comparability value among the centers, IGF-I levels were expressed as SD score (SDS). For each patient, values of GH and IGF-I levels both at diagnosis and before PEG-V administration were collected. There were no changes in assays or reference ranges in any hospital during the study.

Endocrinology and Diabetology Unit (M.F., A.S.), Fondazione Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS) Ca’ Granda Ospedale Maggiore Policlinico, 20122 Milan, Italy; Department of Medical Sciences (L.O., G.M., S.Co., M.A., A.G.L., A.S.), University of Milan, 20122 Milan, Italy; Endocrinology and Diabetology Unit (S.Co.), IRCCS Policlinico San Donato, 20097 San Donato Milanese, Italy; Unit of Endocrine Diseases and Diabetology (M.A.), Ospedale San Giuseppe Multimedica, 20123 Milan, Italy; Division of Endocrinology, Diabetology, and Metabolism (V.G.), Department of Internal Medicine, University of Turin, 10124 Turin, Italy; Pituitary Unit (L.D.M.), Division of Endocrinology, Department of Internal Medicine, Policlinico Universitario A Gemelli, Catholic University of the Sacred Heart, 00168 Rome, Italy; Internal Medicine (C.M.), Department of Medical and Surgical Sciences, University of Padua, 35122 Padua Italy; Department of Endocrinology (F.B.), University of Pisa, 56126 Pisa Italy; Department of Medicine and Pharmacology (S.Ca.), Section of Endocrinology, University of Messina, 98122 Messina, Italy; Departments of Molecular and Clinical Endocrinology and Oncology (A.C.), University ‘Federico II’, Naples, Italy; Department of Endocrinology and Medical Sciences (D.F.), and Center of Excellence for Biomedical Research, University of Genoa, 16126 Genoa, Italy; Division of Endocrinology (G.Ar.), Department of Internal Medicine, Polytechnic University of Marche Region, 60100 Ancona, Italy; Endocrinology and Diabetology (F.P.), Azienda Ospedaliero Universitaria of Cagliari, 09124 Cagliari, Italy; Endocrine Unit (A.P.), Department of Clinical Physiopathology, University of Florence, 50121 Florence, Italy; Internal Medicine Department (G.An.), Section of Internal Medicine and Endocrinological & Metabolic Sciences, University of Perugia, 06123 Perugia, Italy; Department of Experimental Medicine (M.L.J.-R.), University of L’Aquila, 67100 L’Aquila, Italy; and Endocrine Unit (A.G.L.), Istituto Clinico Humanitas IRCCS, 20089 Rozzano, Italy

J Clin Endocrinol Metab, February 2012, 97(2):E165–E172

GHR genotyping d3GHR and five missense single nucleotide polymorphisms withknownheterozygosity[rs348388342(NM_000163.2:c.10T⬎ C, W4R), rs6413484 (NM_000163.2:c.484G⬎A, V162I), rs6181 (NM_000163.2:c.536G⬎A, R179H), rs35395580 (NM_000163. 2:c.569_570insG), and rs6180 (NM_000163.2:c.1630A⬎C, I544L)] found in a public database (http://www.ncbi.nlm.nih.gov/ projects/SNP), were analyzed. Genotypization of GHR gene was performed in the coordinating center laboratory. DNA was extracted from blood leukocytes by standard procedures and genotyping of GHR polymorphisms was carried out as follows. Exon 3 deletion polymorphism genotyping. A multiplex PCR was performed with the following primers: G1, 5⬘-TGTGCTGGTCTGTTGGTCTG-3⬘; G2, 5⬘-AGTCGTTCCTGGGACAGAGA-3⬘; and G3, 5⬘-CCTGGATTAACACTTTGCAGACTC3⬘, as previously described (22, 23). Briefly, PCR was carried out in a 50-␮l reaction mix using Platinum Taq (Invitrogen, Carlsbad, CA) with denaturation at 94 C for 2 min, followed by 39 cycles of 94 C for 30 sec, 57 C for 30 sec, and 68 C for 30 sec, and a final extension phase at 68 C for 10 min. Then reaction products were run on ethidium bromide-stained 1% agarose gel electrophoresis. Full-length GHR allele was shown by the presence of two bands of approximately 935 bp. Removing the annealing of G3 primer, the genomic deletion of exon 3 formed a 532-bp band. The confirmation of d3-GHR allele was carried out by another PCR, with the following pair of primers: 5⬘-TACA CAGGGTCATATCAGATTG-3⬘ and 5⬘-CTATTCCAGTTACT ACCATCCC-3⬘. Reactions were subjected to denaturation at 94 C for 2 min, followed by 39 cycles of 94 C for 30 sec, 57 C for 30 sec, and 68 C for 30 sec, with a final extension period at 68 C for 10 min. PCR products were run on ethidium bromide-stained 1% agarose gel electrophoresis and the presence of at least one flGHR allele hesitated in a 250-bp band. rs348388342 genotyping. A PCR was performed with the following primer pair: 5⬘-CAGCTCATTCATGTCTTACC-3⬘ and 5⬘-CTTGGATGTAGCGAATACAG-3⬘. The reaction was carried out in a 50-␮l reaction mix using Platinum Taq (Invitrogen) with denaturation at 94 C for 2 min, followed by 39 cycles of 94 C for 30 sec, 56 C for 30 sec, and 68 C for 30 sec, and a final extension phase at 68 C for 10 min. Then reaction products underwent to direct sequencing, both in forward and reverse direction, by ABI Prism 310 genetic analyzer (Applied Biosystems, Life Technologies Corp., Carlsbad, CA), and chromatograms were subsequently manually inspected. rs6413484, rs6181, and rs35395580 genotyping. A PCR was performed with the following primer pair: 5⬘-GGTCTTCTGA GAAGAATGCC-3⬘ and 5⬘-GCATACAGATCAACATGAG-3⬘. The reaction was carried out in a 50-␮l reaction mix using Platinum Taq (Invitrogen) with denaturation at 94 C for 2 min, followed by 39 cycles of 94 C for 30 sec, 58 C for 30 sec, and 68 C for 30 sec, and a final extension phase at 68 C for 10 min. Then reaction products underwent direct sequencing in both forward and reverse directions by ABI Prism 310 Applied Biosystems genetic analyzer, and chromatograms were subsequently manually inspected. rs6180 genotyping. A PCR was performed with the following primer pair: 5⬘-GATGTCCCAATGTGACATGCA-3⬘ and 5⬘-

jcem.endojournals.org

E167

AGAATCCATACCCCATCCTG-3⬘. The reaction was carried out in a 50-␮l reaction mix using Platinum Taq (Invitrogen) with denaturation at 94 C for 2 min, followed by 39 cycles of 94 C for 30 sec, 56 C for 30 sec, and 68 C for 30 sec, and a final extension phase at 68 C for 10 min. Then reaction products underwent to direct sequencing in both forward and reverse directions by ABI Prism 310 Applied Biosystems genetic analyzer, and chromatograms were subsequently manually inspected. Except for exon 3 deletion, GHR polymorphisms were initially analyzed in a subset of 25 patients randomly selected to estimate their allele frequencies in the study population. The analysis has been extended in all the patients when allele frequencies allowed statistical tests.

Statistical analysis Data are presented as mean ⫾ SD (normal distributed) or median and interquartile range (nonnormal). Gaussian variables were compared using Student’s t test, preceded by Levene’s test to check variances equality (two groups) or ANOVA test, followed by Bonferroni post test (three groups). Nominal data were analyzed by Fisher’s exact test in 2 ⫻ 2 contingency tables or ␹2 test otherwise, whereas the other non-Gaussian variables were evaluated by Mann-Whitney U test. Linear or logistic multivariate regression analyses were performed with stepwise method. P ⬍ 0.05 and P ⬎ 0.10 were used as entry and removal criteria, respectively. Hardy-Weinberg equilibrium (HWE) has been evaluated by ␹2 test between observed and expected genotypes frequencies. Two-tailed P values were considered statistically significant when less than 0.05. In multiple comparisons, significant P value threshold was reduced after Bonferroni correction. Calculations were performed by SPSS 18.0 software (SPSS, Paris, France).

Results Patient characteristics and GHR genotype Clinical and biochemical data of the127 patients included in the study are reported in Table 1. All patients had been treated by long-acting SST analogs at the maximal tolerable dosage, which was administered after unsuccessful surgical adenomectomy in 95 patients (75%), followed by adjuvant radiotherapy in 13 (10%) (Table 1). These treatments caused a reduction of GH levels to about half of those found at diagnosis, without normalization of IGF-I levels and disease remission (Table 1). The mean weight corrected PEG-V dose required to reach and maintain IGF-I levels concentration below the ULN range of each laboratory was 1.3 ⫾ 0.5 mg/kg 䡠 wk (total weekly dose, 105.0 ⫾ 54.4 mg/wk). Sixty-three patients (49.6%) did not stop SST analogs before PEG-V administration and were therefore under combined therapy with the two drugs (PEG-V ⫹ SST analogs). The long-acting analogs used were octreotide LAR (in 72%) or lanreotide. No significant differences in clinical and biochemical parameters were observed between PEG-V and PEG-V ⫹ SST analog groups, with regard to

E168

Filopanti et al.

Growth Hormone Receptor Variants and Pegvisomant

J Clin Endocrinol Metab, February 2012, 97(2):E165–E172

TABLE 1. Basal clinical and biochemical differences between 127 acromegalic patients carrying flGHR or d3GHR allele d3 carriers d3GHR (fl/d3 ⴙ d3/d3) fl/d3GHR d3/d3GHR All patients fl/flGHR n 127 68 59 41 18 Age at diagnosis (yr) 42.2 ⫾ 13.2 41.3 ⫾ 12.4 43.1 ⫾ 14.1 44.2 ⫾ 13.8 40.7 ⫾ 14.7 Male ratio (%) 57 62 46 44 72 Weight (kg) 83 ⫾ 20 79 ⫾ 17 80 ⫾ 22 78 ⫾ 20 88 ⫾ 21 Height (cm) 170 ⫾ 11 171 ⫾ 11 169 ⫾ 11 169 ⫾ 11 170 ⫾ 11 Dimension (% macroadenomas) 76 65 87 85 89 Extrasellar extent (%) 72 71 73 70 71 GH at diagnosis (␮g/liter) 25.9 (13.7–58.3) 28.0 (12.1–50.0) 24.2 (13.9 – 67.9) 24.2 (12.2–50.0) 19.5 (7.1–73.0) IGF-I SDS at diagnosis 9.0 ⫾ 4.8 9.0 ⫾ 4.4 8.5 ⫾ 4.2 8.7 ⫾ 4.5 7.7 ⫾ 3.0 Pre-PEG-V GH (␮g/liter) 13.3 (5.8 –30.4) 11.5 (7.0 –30.3) 15.5 (5.0 –31.7) 12.2 (6.0 –25.7) 12.0 (6.9 –30.1) Pre-PEG-V IGF-I SDS 7.7 ⫾ 4.9 7.1 ⫾ 4.5 8.4 ⫾ 5.4 7.9 ⫾ 5.0 5.6 ⫾ 3.9 Treatment Previous surgery (%) 75 80 69 71 67 Radiotherapy (%) 10 13 7 7 5 PEG-V monotherapy (%) 50 48 54 49 67

Pa 0.445 0.548 0.037 0.706 0.009 0.832 0.605 0.920 0.273 0.192 0.228 0.403 0.723

Fourteen patients have been previously reported (20). a

fl/flGHR vs. d3GHR (fl/d3 ⫹ d3/d3). Bonferroni correction, P ⬍ 0.004.

GH and IGF-I levels at diagnosis and pre-PEG-V treatment (data not shown). However, macroadenomas were prevalent in patients treated with PEG-V ⫹ SST analogs (87.1 vs. 65.5%) with a P value (P ⫽ 0.009) slightly above the limit of Bonferroni correction for multiple comparisons (0.003). Genotypization of GHR gene for exon 3 deletion showed that 68 patients (48.6%) were full-length GHR homozygotes (fl/fl), 41 patients (34.3%) were heterozygote (fl/d3), and 18 patients (13.9%) were homozygotes (d3/d3) for the exon 3-deleted form of GHR. Genotype frequencies were not distributed under HWE, with a major difference found in the proportion of heterozygotes (expected, 42.2%; P ⫽ 0.008). Following the hypothesis of the dominant model, patients having the d3 allele (d3/d3 homozygotes and fl/d3heterozygotes) were grouped together (d3GHR) and compared with fl/fl homozygotes (flGHR) (Table 1). Considering the whole set of patients, d3GHR and flGHR patients did not differ for sex, age, adenoma size, sellar invasion, previous surgery and radiotherapy, GH and IGF-I levels at diagnosis, and pre-PEG-V treatments (Table 1). Moreover, there were no differences in allele frequencies between patients receiving PEG-V and patients under PEG-V ⫹ SST analogs (d3GHR proportion, 48.4 vs. 44.4%; P ⫽ 0.723). Response to treatment and d3GHR genotype Of the 127 patients included in the study, normalization of IGF-I levels was not achieved in 16 patients because of drug discontinuation in six (for tumor size increase in two, lipohypertrophy in three, and low compliance in one)

and submaximal PEG-V dosage (15–20 mg/d) in 10, due to the cross-sectional design of the study. Considering this subset of patients, six patients were flGHR homozygotes, eight were heterozygotes, and two were d3GHR homozygotes, without any difference of genotype frequencies (P ⫽ 0.940) as well as gender, tumor dimension, previous radiotherapy, and GH and IGF-I SDS at diagnosis in comparison with the rest of patients. The overall IGF-I normalization was 92%, with the exclusion of the six patients who discontinued the drug for adverse effects. The influence of d3GHR on the response to PEG-V therapy was evaluated in patients who received PEG-V at the dose titrated to maintain IGF-I concentration below the ULN range of each laboratory (n ⫽ 111). In these patients, no significant difference in IGF-I levels during PEG-V treatment between d3GHR and flGHR groups was observed. Moreover, PEG-V dosage per kilogram of body weight as well as total weekly dosage was similar between d3GHR and flGHR groups (1.4 ⫾ 0.6 vs. 1.3 ⫾ 0.6 mg/kg 䡠 wk, P ⫽ 0.620; and 106.9 ⫾ 58.1 vs. 103.1 ⫾ 51.0 mg/wk, P ⫽ 0.708). Finally, no significant difference on PEG-V dosage was observed in the different genotypes, the PEG-V dosage being 1.3 ⫾ 0.6 in fl/fl, 1.4 ⫾ 0.7 in fl/d3, and 1.3 ⫾ 0.5 mg/kg 䡠 wk in d3/d3 (ANOVA test, P ⫽ 0.688). As far as patients in PEG-V monotherapy were concerned, no statistically significant difference in responsiveness to treatment between flGHR and d3GHR patients, as well as between genotypes, was observed (Table 2). Concerning adverse effects, local reaction was observed in eight patients (pain in two and lipohypertrophy in six),

J Clin Endocrinol Metab, February 2012, 97(2):E165–E172

jcem.endojournals.org

E169

TABLE 2. Clinical and biochemical differences between acromegalic patients carrying flGHR or d3GHR allele treated with PEG-V in monotherapy or combined therapy with long-acting SST analogs PEG-V ⴙ SST analogsa

PEG-V monotherapy d3 carriers

n IGF-I SDS PEG-V dosage (mg/wk 䡠 kg) PEG-V dosage (mg/wk) Tumor/residue increase (n) Adverse effects (n)

d3 carriers

fl/flGHR

d3GHR fl/d3 ⴙ d3/d3

fl/d3GHR

d3/d3GHR

30 0.2 ⫾ 1.0 1.3 ⫾ 0.4 103.5 ⫾ 32.5 1 5

26 0.1 ⫾ 0.8 1.5 ⫾ 0.6 115.0 ⫾ 54.2 3 6

18 0.1 ⫾ 0.6 1.5 ⫾ 0.7 116.1 ⫾ 60.5 2 5

8 0.4 ⫾ 1.0 1.4 ⫾ 0.4 112.8 ⫾ 42.1 1 1

a

Octreotide LAR and lanreotide.

b

fl/flGHR vs. d3GHR (fl/d3 ⫹ d3/d3). Bonferroni correction P ⬍ 0.004.

whereas mild increases in liver enzyme and creatine phosphokinase were reported in two and one patient, respectively. No association between adverse effects and GHR genotype was found. A significant regrowth (⬎3 mm) of the adenoma or postsurgery residue occurred in four patients, three of them being d3GHR. Similarly, clinical and biochemical data of patients treated with PEG-V ⫹ SST analog combined therapy did not show significant differences between d3GHR and flGHR groups as well as between genotypes (Table 2). In this set of patients, the only adverse effect was lipohypertrophy in injection site, which was reported in five patients, two of them being d3GHR. When comparing patients in PEG-V monotherapy and combined therapy, the only difference was a nonsignificant tendency of lower PEG-V dosage in PEG-V ⫹ SST analog patients (1.2 ⫾ 0.5 mg/wk 䡠 kg) in comparison to patients in PEG-V monotherapy (1.4 ⫾ 0.6 mg/wk 䡠 kg), together with the absence of tumor regrowth (Table 2). To evaluate various potential predictors of PEG-V dosage, a linear regression analysis was carried out. d3GHR allele, sex, age at diagnosis, tumor size, GH and IGF-I SDS levels at diagnosis, previous radiotherapy, and pre-PEG-V and SST analog cotreatment were taken into account in the regression model, using a backward stepwise method. Only age at diagnosis was retained in the model as a significant predictor both in weight-corrected weekly PEG-V dosage (B ⫽ ⫺0.012 ⫾ 0.006; P ⫽ 0.044) and in total weekly PEG-V dosage (B ⫽ ⫺1.477 ⫾ 0.517; P ⫽ 0.006). Other GHR polymorphisms, basal characteristics, and response to treatment rs348388342, rs6413484, rs6181, and rs35395580 GHR polymorphisms did not show significant variability in a randomly selected subgroup of 25 acromegalic patients, all patients being homozygous for T, G, G, and delG allele, respectively. Only rs6180 (Ile544Leu) had a significant heterozygosity to allow statistical tests. In

Pb 0.099 0.393 0.377

fl/flGHR

d3GHR fl/d3 ⴙ d3/d3

31 0.3 ⫾ 1.7 1.2 ⫾ 0.4 104.3 ⫾ 50.5 0 3

24 0.2 ⫾ 0.9 1.3 ⫾ 0.7 105.9 ⫾ 53.2 0 2

fl/d3GHR

d3/d3GHR

Pb

17 0.1 ⫾ 0.8 1.3 ⫾ 0.7 97.7 ⫾ 53.9

7 0.7 ⫾ 0.8 1.3 ⫾ 0.7 123.6 ⫾ 50.9

0.276 0.482 0.912

2

0

particular, 29 (22.8%) patients were Ile homozygotes, 67 (52.8%) were heterozygotes, and 31 (24.4%) were Leu homozygotes. Genotype distribution was consistent with HWE (P ⫽ 0.533). There were no associations between rs6180 genotypes and clinical or hormonal parameters, both at diagnosis and during PEG-V treatment, with or without SST analog coadministration (data not shown). Linear regression analysis was repeated with the parameters reported above and including rs6180 alleles between the predictor factors. No role of this polymorphism in predicting PEG-V dosage resulted from the analysis.

Discussion The present study investigated the role of exon 3 deletion and other GHR polymorphisms in the response to PEG-V in 127 acromegalic patients receiving the drug as either monotherapy or combined therapy with longacting SST analogs. Although in vitro and in silico data indicated that d3GHR binds and internalizes GH as efficiently as flGHR, transfection studies showed increased signaling of d3GHR in comparison to the full length receptor (3). In contrast to the large body of studies carried out to assess the role of GHR genotypes in the responsiveness to rhGH replacement therapy in patients with various diseases (6 –17, 22), few pharmacogenetic studies investigated the impact of GHR variants in the responsiveness of acromegalic patients to PEG-V. The only two studies so far available on this topic were carried out on small series of patients and reported the association of d3GHR variant with better response to PEG-V monotherapy in comparison with patients carrying the full-length receptor (20, 21). Taking into account the rarity of the disorder under investigation, i.e. resistance to SST analogs in unsuccessfully surgically treated acromegalic patients, the cohort of patients here reported can be regarded as a large series. As

E170

Filopanti et al.

Growth Hormone Receptor Variants and Pegvisomant

expected, the great majority of patients had a macroadenoma, very frequently with extrasellar extension. SST analogs were administered in combination with PEG-V in about a half of patients to control residue regrowth, which is a possible outcome in patients receiving PEG-V due to the lack of GH feedback on somatotrophs. Indeed, a macroadenoma was present at diagnosis in almost all patients who were maintained in SST analogs during PEG-V treatment. Analysis of d3GHR variants in this cohort of patients showed that genotype frequencies were not distributed under HWE. It is worth noting that by reviewing and recalculating HWE in the six publications including the present study (21, 23–26), with sufficient number of acromegalic patients to perform an accurate ␹2 test, a genotype distribution of d3GHR deviating from HWE was found in five studies (21, 24 –26). Interestingly, d3GHR genotype in Hardy-Weinberg disequilibrium was found in all studies investigating cohorts of acromegalic patients with a particularly aggressive phenotype, characterized by persistence of disease after surgical and radiological treatments and resistance to long-acting SST analogs (21, 24 – 26), whereas the only study describing genotype distribution in HWE was carried out on a series of consecutive acromegalics in which only 6% were under PEG-V therapy for disease control (23). Nevertheless, it has been stated that deviation from HWE may be viewed as chance, genotyping error or spurious associations due to population stratification (27). However, these biases in the present work seem unlikely because the data here reported were obtained by the same methods and lab analysts that found d3GHR genotype distributions consistent with HWE in healthy subjects and in a different series of acromegalic patients (23), and the data were subsequently reviewed and confirmed by an independent staff. Therefore, the deviation from HWE of the d3GHR genotype distribution found in the present and previous studies might suggest an association of d3GHR variant with a phenotype characterized by resistance to traditional treatments and difficulties in disease control. Considering the whole set of patients, GHR polymorphisms did not show any significant association with clinical and hormonal characteristics at diagnosis. These data are consistent with previous reports that failed to find a clear influence of d3GHR genotype on GH and IGF-I levels, as well as adenoma size (24, 26, 28). Accordingly, d3GHR allele frequency was not different between acromegalics treated with PEG-V alone and those in whom SST analog was maintained in consideration of a possible tumor regrowth. In the present series, almost all patients (92%) reached IGF-I normalization during PEG-V therapy, very likely

J Clin Endocrinol Metab, February 2012, 97(2):E165–E172

due to adequate dose titration carried out in the different centers. Starting from similar GH and IGF-I levels after several treatments before PEG-V administration, PEG-V treatment had similar efficacy in reducing IGF-I levels in patients with d3GHR and flGHR as well as similar potency because no significant difference in PEG-V dosage required to normalize IGF-I levels was found between the two groups. Moreover, multivariate regression analysis did not retain d3GHR as predictor of PEG-V dosage both in monotherapy and combined therapy. These data are in contrast with previous findings in which d3GHR was associated with a better response to PEG-V treatment. However, previous results were obtained in very small series of patients controlled by PEG-V therapy, i.e. 19 and 33 patients (20, 21), in comparison with the present series of 111 patients. Moreover, response to PEG-V might be influenced by previous treatments, in particular radiotherapy that was performed in a high proportion of patients (66%) reported by Bernabeu et al. (21) and in only 10% of our series (10%). However, previous radiotherapy was not retained in the linear regression analysis here performed as predictor of PEG-V dosage in the small subset of patients who underwent this treatment. Taken together, our data indicate that, although in vitro studies indicated d3GHR to be a more active receptor, the blocking of either full-length receptor or the d3GHR variant by PEG-V resulted in a similar IGF-I control, further supporting a minor, if any, role of d3GHR polymorphism on GH and IGF-I secretion in acromegaly. It is well established that liver enzyme increase, lipohypertrophy, and pain in the injection site are the most common adverse effects reported by patients in PEG-V therapy (29). The small number of events recorded in the present series did not allow statistical tests to evaluate a possible association with GHR polymorphisms. Similarly, increases in pituitary tumor residue were observed in four patients, without any striking association with d3GHR polymorphism. To assess the role of other GHR variants, five missense single nucleotide polymorphisms found in a public database, i.e. rs348388342 (W4R), rs6413484 (V162I), rs6181 (R179H), rs35395580 (NM_000163.2:c.569_570insG), and rs6180 (I544L), were analyzed. Only rs6180 had a variability sufficient for statistical analysis, but no association with clinical, biochemical, or pharmacological parameters was found. In conclusion, this multicenter study did not confirm a positive role of d3GHR or other nonsynonymous polymorphisms in the responsiveness to PEG-V treatment in a large cohort of acromegalic patients. Further investigations are needed to determine whether deviation from HWE observed in these patients may indicate an associ-

J Clin Endocrinol Metab, February 2012, 97(2):E165–E172

ation of d3GHR genotype with poor response to usual treatments.

jcem.endojournals.org

10.

Acknowledgments Address all correspondence and requests for reprints to: Anna Spada, M.D., University of Milan, Department of Medical Sciences, Via F. Sforza, 35, Milan 20122, Italy. E-mail: [email protected]. This work was supported by an Associazione Italiana per la Ricerca sul Cancro, Milan Grant (to M.G.), Ministero dell’Istruzione, dell’Universitá e della Ricerca PRIN 2009YJTBAZ_004 Grant (to A.G.L.), by Ricerca Corrente Funds of Fondazione Istituto di Ricovero e Cura a Carattere Scientifico Ca’ Granda Ospedale Maggiore Policlinico, Milan, and Pfizer Italia. Disclosure Summary: The authors have nothing to declare.

11.

12.

13.

14.

References 1. Pantel J, Machinis K, Sobrier ML, Duquesnoy P, Goossens M, Amselem S 2000 Species-specific alternative splice mimicry at the growth hormone receptor locus revealed by the lineage of retroelements during primate evolution. J Biol Chem 275:18664 –18669 2. Urbanek M, Russell JE, Cooke NE, Liebhaber SA 1993 Functional characterization of the alternatively spliced, placental human growth hormone receptor. J Biol Chem 268:19025–19032 3. Dos Santos C, Essioux L, Teinturier C, Tauber M, Goffin V, Bougne`res P 2004 A common polymorphism of the growth hormone receptor is associated with increased responsiveness to growth hormone. Nat Genet 36:720 –724 4. Strawbridge RJ, Ka¨rvestedt L, Li C, Efendic S, Ostenson CG, Gu HF, Brismar K 2007 GHR exon 3 polymorphism: association with type 2 diabetes mellitus and metabolic disorder. Growth Horm IGF Res 17:392–398 5. Kenth G, Shao Z, Cole DE, Goodyer CG 2007 Relationship of the human growth hormone receptor exon 3 genotype with final adult height and bone mineral density. J Clin Endocrinol Metab 92:725– 728 6. Jorge AA, Marchisotti FG, Montenegro LR, Carvalho LR, Mendonca BB, Arnhold IJ 2006 Growth hormone (GH) pharmacogenetics: influence of GH receptor exon 3 retention or deletion on first-year growth response and final height in patients with severe GH deficiency. J Clin Endocrinol Metab 91:1076 –1080 7. Binder G, Baur F, Schweizer R, Ranke MB 2006 The d3-growth hormone (GH) receptor polymorphism is associated with increased responsiveness to GH in Turner syndrome and short small-for-gestational-age children. J Clin Endocrinol Metab 91: 659 – 664 8. Pilotta A, Mella P, Filisetti M, Felappi B, Prandi E, Parrinello G, Notarangelo LD, Buzi F 2006 Common polymorphisms of the growth hormone (GH) receptor do not correlate with the growth response to exogenous recombinant human GH in GH deficient children. J Clin Endocrinol Metab 91:1178 –1180 9. Carrascosa A, Esteban C, Espadero R, Ferna´ndez-Cancio M, Andaluz P, Clemente M, Audí L, Wollmann H, Fryklund L, Parodi L; Spanish SGA Study Group 2006 The d3/fl growth hormone (GH) receptor polymorphism does not influence the effect of GH treatment (66 microg/kg per day) or the spontaneous growth in short non-GH-deficient small-for-gestational-age children: results from a

15.

16.

17.

18.

19.

20.

21.

22.

E171

two-year controlled prospective study in 170 Spanish patients. J Clin Endocrinol Metab 91:3281–3286 Blum WF, Machinis K, Shavrikova EP, Keller A, Stobbe H, Pfaeffle RW, Amselem S 2006 The growth response to growth hormone (GH) treatment in children with isolated GH deficiency is independent of the presence of the exon 3-minus isoform of the GH receptor. J Clin Endocrinol Metab 91:4171– 4174 Carrascosa A, Audí L, Ferna´ndez-Cancio M, Esteban C, Andaluz P, Vilaro´ E, Clemente M, Yeste D, Albisu MA, Gussinye´ M 2008 The exon 3-deleted/full-length growth hormone receptor polymorphism did not influence growth response to growth hormone therapy over two years in prepubertal short children born at term with adequate weight and length for gestational age. J Clin Endocrinol Metab 93: 764 –770 de Graaff LC, Meyer S, Els C, Hokken-Koelega AC 2008 GH receptor d3 polymorphism in Dutch patients with MPHD and IGD born small for gestational age. Clin Endocrinol (Oxf) 68: 930 –934 Ra¨z B, Janner M, Petkovic V, Lochmatter D, Eble´ A, Dattani MT, Hindmarsh PC, Flu¨ck CE, Mullis PE 2008 Influence of growth hormone receptor d3 and full-length isoforms on growth hormone response and final height in patients with severe growth hormone deficiency. J Clin Endocrinol Metab 93:974 –980 Schreiner F, Stutte S, Bartmann P, Gohlke B, Woelfle J 2007 Association of the growth hormone receptor d3-variant and catch-up growth of preterm infants with birth weight of less than 1500 grams. J Clin Endocrinol Metab 92:4489 – 4493 Audí L, Esteban C, Carrascosa A, Espadero R, Pe´rez-Arroyo A, Arjona R, Clemente M, Wollmann H, Fryklund L, Parodi LA; Spanish SGA Study Group 2006 Exon 3-deleted/full-length growth hormone receptor polymorphism genotype frequencies in Spanish short small-for-gestational age (SGA) children and adolescents (n ⫽ 247) and in adult control population (n ⫽ 289) show increased fl/fl in short SGA. J Clin Endocrinol Metab 91:5038 –5043 van der Klaauw AA, van der Straaten T, Baak-Pablo R, Biermasz NR, Guchelaar HJ, Pereira AM, Smit JW, Romijn JA 2008 Influence of the d3-growth hormone (GH) receptor isoform on short-term and long-term treatment response to GH replacement in GH-deficient adults. J Clin Endocrinol Metab 93:2828 –2834 Wassenaar MJ, Dekkers OM, Pereira AM, Wit JM, Smit JW, Biermasz NR, Romijn JA 2009 Impact of the exon 3-deleted growth hormone (GH) receptor polymorphism on baseline height and the growth response to recombinant human GH therapy in GH-deficient (GHD) and non-GHD children with short stature: a systematic review and meta-analysis. J Clin Endocrinol Metab 94:3721–3730 Melmed S, Colao A, Barkan A, Molitch M, Grossman AB, Kleinberg D, Clemmons D, Chanson P, Laws E, Schlechte J, Vance ML, Ho K, Giustina A; Acromegaly Consensus Group 2009 Guidelines for acromegaly management: an update. J Clin Endocrinol Metab 94: 1509 –1517 Kopchick JJ, Parkinson C, Stevens EC, Trainer PJ 2002 Growth hormone receptor antagonists: discovery, development, and use in patients with acromegaly. Endocr Rev 23:623– 646 Bianchi A, Mazziotti G, Tilaro L, Cimino V, Veltri F, Gaetani E, Pecorini G, Pontecorvi A, Giustina A, De Marinis L 2009 Growth hormone receptor polymorphism and the effects of pegvisomant in acromegaly. Pituitary 12:196 –199 Bernabeu I, Alvarez-Escola´ C, Quinteiro C, Lucas T, Puig-Domingo M, Luque-Ramírez M, de Miguel-Novoa P, Fernandez-Rodriguez E, Halperin I, Loidi L, Casanueva FF, Marazuela M 2010 The exon 3-deleted growth hormone receptor is associated with better response to pegvisomant therapy in acromegaly. J Clin Endocrinol Metab 95:222–229 Giavoli C, Ferrante E, Profka E, Olgiati L, Bergamaschi S, Ronchi CL, Verrua E, Filopanti M, Passeri E, Montefusco L, Lania AG, Corbetta S, Arosio M, Ambrosi B, Spada A, Beck-Peccoz P 2010 Influence of the d3GH receptor polymorphism on the metabolic and

E172

Filopanti et al.

Growth Hormone Receptor Variants and Pegvisomant

biochemical phenotype of GH-deficient adults at baseline and during short- and long-term recombinant human GH replacement therapy. Eur J Endocrinol 163:361–368 23. Montefusco L, Filopanti M, Ronchi CL, Olgiati L, La-Porta C, Losa M, Epaminonda P, Coletti F, Beck-Peccoz P, Spada A, Lania AG, Arosio M 2010 D3-growth hormone receptor in acromegaly: effects on metabolic phenotype. Clin Endocrinol (Oxf) 72:661– 667 24. Kamenicky P, Dos Santos C, Espinosa C, Salenave S, Galland F, Le Bouc Y, Maison P, Bougne`res P, Chanson P 2009 D3 GH receptor polymorphisms is not associated with IGF1 levels in untreated acromegaly. Eur J Endocrinol 161:231–235 25. Bianchi A, Giustina A, Cimino V, Pola R, Angelini F, Pontecorvi A, De Marinis L 2009 Influence of growth hormone receptor d3 and full-length isoforms on biochemical treatment outcomes in acromegaly. J Clin Endocrinol Metab 94:2015–2022

J Clin Endocrinol Metab, February 2012, 97(2):E165–E172

26. Mercado M, Gonza´lez B, Sandoval C, Esquenazi Y, Mier F, Vargas G, de los Monteros AL, Sosa E 2008 Clinical and biochemical impact of the d3 growth hormone receptor genotype in acromegaly. J Clin Endocrinol Metab 93:3411–3415 27. Yu KD, Di GH, Fan L, Shao ZM 2009 Test of Hardy-Weinberg equilibrium in breast cancer case-control studies: an issue may influence the conclusions. Breast Cancer Res Treat 117:675– 677 28. Schmid C, Krayenbuehl PA, Bernays RL, Zwimpfer C, Maly FE, Wiesli P 2007 Growth hormone (GH) receptor isoform in acromegaly: lower concentrations of GH but not insulin-like growth factor-1 in patients with a genomic deletion of exon 3 in the GH receptor gene. Clin Chem 53:1484 –1488 29. Hodish I, Barkan A 2008 Long-term effects of pegvisomant in patients with acromegaly. Nat Clin Pract Endocrinol Metab 4:324 – 332

Members can search for other endocrinology professionals around the world in the online Member Directory. www.endo-society.org/directory