current status and future directions of pharmacological therapy for

© 2015 EDIZIONI MINERVA MEDICA The online version of this article is located at http://www.minervamedica.it

Minerva Endocrinologica 2016 September;41(3):351-65

REVIEW M E D I C A L T R E AT M E N T O F P I T U I TA R Y T U M O R S

Current status and future directions of pharmacological therapy for acromegaly Moisés MERCADO 1, 2 *, Etual ESPINOSA 1, 2, Claudia RAMÍREZ 1, 2 1Experimental Endocrinology Unit, Hospital de Especialidades, Centro Médico Nacional Siglo XXI, Mexico City, Mexico; 2Neurological Center, Centro Médico ABC, Mexico City, Mexico *Corresponding

author: Moisés Mercado, Experimental Endocrinology Unit, Hospital de Especialidades, Centro Médico Nacional Siglo XXI, Sur-132, number 142, suite 210, Colonia las Américas, Mexico City, Mexico. E-mail: moises.mercado@endocrinología.org.mx

A B S T RAC T Acromegaly is a chronic systemic disorder caused in the vast majority of cases by a GH-secreting pituitary adenoma and resulting in significant morbidity and mortality if left untreated. The treatment of choice is the trans-sphenoidal resection of the adenoma, and although 80% of patients with microadenomas or confined macroadenomas achieve biochemical remission, the surgical success rate for patients harboring tumors with extrasellar extension is below 50%. Thus, a considerable proportion of patients will require some form of adjuvant treatment. Acromegaly can be approached pharmacologically by inhibiting GH secretion by the tumor (somatostatin analogues, dopamine agonists) or by antagonizing GH actions at its target tissues (GH receptor antagonists). The primary pharmacological treatment of acromegaly is increasingly gaining acceptance by both physicians and patients. The decision to use primary pharmacological treatment has to take into account the clinical characteristics of the patient (presence of comorbidities that significantly increase the surgical risk) and the biological nature of the adenoma (tumor size and location), as well as other aspects such as the availability of a pituitary surgeon and the cost of medications. This review provides a critical summary and update of the pharmacological treatment of acromegaly focusing both, on well-established agents and strategies as well as on novel compounds that are currently being developed. (Cite this article as: Mercado M, Espinosa E, Ramírez C. Current status and future directions of pharmacological therapy for acromegaly. Minerva Endocrinol 2016;41:351-65) Key words: Acromegaly - Growth hormone - Insulin-like growth factor I - Somatostatin - Octreotide - Dopamine agonists.

S

urgical resection of the growth hormone (GH)-secreting adenoma remains the initial treatment of choice in acromegaly.1 Although the surgical success rates can be as high as 80-90% for microadenomas and for confined, intrasellar macroadenomas, less than 50% of patients harboring macroadenomas with extrasellar extension and probably less than 10% of those with tumors invading the cavernous sinuses achieve a proper biochemical remission.2, 3 Thus, a significant proportion of patients will require some form of adjunctive therapy, be it pharmacological therapy or

Vol. 41 - No. 3

radiotherapy. The primary pharmacological treatment of acromegaly is increasingly gaining acceptance by both physicians and patients.1, 4 The decision to use primary pharmacological treatment has to take into account not only the clinical characteristics of the patient (presence of comorbidities that significantly increase the surgical risk) and the biological nature of the adenoma (tumor size and location), but also other aspects such as the availability of an experienced pituitary surgeon and the economic resources to pay for costly medications.1, 4 This review provides a critical summary and

Minerva Endocrinologica

351

MERCADO

PHARMACOLOGICAL THERAPY FOR ACROMEGALY

update of the pharmacological treatment of acromegaly focusing both, on well-established agents and strategies as well as on novel compounds that are currently being developed. Conventional somatostatin analogues: octreotide and lanreotide Pharmacodynamics and pharmacokinetics The hypothalamic control of GH synthesis and release involves the pulsatile and stimula-

tory signal mediated by GHRH (GH-releasing hormone) and the tonic and inhibitory effect of somatostatin (Figure 1).5, 6 The extra-hypothalamic control of GH secretion is largely due to ghrelin, an orexigenic hormone produced by the gastric fundus and the negative feedback mediated by insulin-like growth factor I (IGF-1) itself and perhaps GH as well (Figure 1).7 The GHRH and ghrelin signals occur via two distinct G protein-coupled receptors, the GHRH and the GH-secretagogue receptors, respectively.6, 7 Somatostatin, also known as SRIH (somatotro-

Figure 1.—Regulation of the GH/IGF-1 axis and mechanisms of action of the different pharmacological interventions.

352


September 2016

PHARMACOLOGICAL THERAPY FOR ACROMEGALY MERCADO

Table I.—Molecular characteristics and binding affinities of the 5 somatostatin receptor subtypes (SSTR) Chromosome Molecular weight (kDa) Expression in GH-adenomas Binding affinity (IC50, nmol/L) –– SS14 –– Octreotide –– Lanreotide –– Pasireotide –– Somatoprim –– Dopastatin

SSTR-1

SSTR-2

SSTR-3

SSTR-4

SSTR-5

14q13 43 44%

17q24 41 96%

22q13.1 46 44%

20p11.2 42 5%

16p13.3 39 86%

0.93 280 180 9.3 >1000 622

pin release-inhibiting hormone) is synthesized as a precursor, which after proteolytic cleavage yields two biologically active peptides of 28 and 14 amino acids (SS-28 and SS-14).8, 9 Besides its inhibitory effect on GH synthesis and secretion, somatostatin also inhibits insulin, glucagon thyroid-stimulating hormone (TSH), secretin and cholecystokinin (CCK).10 Both SS-28 and SS-14 interact with similar binding affinities with five different somatostatin receptor subtypes (SSTR-1 to 5), each encoded by a different gene (Table I).11-13 SSTRs are G-coupled receptors that are expressed in a tissue-specific manner (Table I).11-13 These receptors are known to homo- and heterodimerize, a phenomenon that potentially affects

0.15 0.38 0.54 1 3 0.03

0.56 7.1 14 1.5 >100 160

1.5 >1000 230 >100 7 >1000

0.29 6.3 17 0.16 6 4

signal transduction as well as receptor internalization and recycling.13-15 SSTR signaling involves mainly adenyl cyclase inhibition, which results in a reduction of cyclic adenosine monophosphate (cAMP) generation that eventually prevents protein kinase A (PKA) and CREB (cAMP response element-binding protein) signaling into the nucleus, and thus, a reduction in the transcription rate of genes encoding GH, TSH, insulin, glucagon, secretin and CCK (Figure 2).11-13 The inhibition of PKA also results in inactivation of the MAPK/ERK (mitogen-activated protein kinases, originally called extracellular signal-regulated kinases) pathway, which ultimately leads to cell cycle arrest and apoptosis induction (Figure 2).16-19

Figure 2.—Mechanisms of action of somatostatin: upon binding to its G-protein-coupled receptor, the alpha subunit binds GTP, dissociating from the beta/gamma complex; this results in inhibition of adenyl cyclase, with the consequent reduction of cAMP generation. The low intracellular levels of cAMP lead to a transcription of the GH, alpha-SU, insulin, glucagon and TSH genes. The inhibition of PKA also results in the inactivation of the MAPK/ERK pathway, which ultimately leads to cell cycle arrest and probably induction of apoptosis.

Vol. 41 - No. 3


353

MERCADO


Indirect anti-proliferative effects are achieved through inhibition of angiogenesis, which in turn are mediated by down-regulation of vascular-endothelial growth factor and basic fibroblast growth factor.20 Native somatostatin cannot be used pharmacologically because of its extremely short half-life, its concomitant inhibition of insulin, glucagon and TSH and the rebound GH secretion that occurs upon stopping it.10 Thus, somatostatin analogues with longer half-lives and capable of inhibiting GH secretion almost exclusively were developed during the 1980s.10 Subcutaneous octreotide is one of such analogues and the first one to be used in the treatment of acromegaly;21, 22 lanreotide was introduced a few years later.23 Both lanreotide and octreotide are in fact SSTR-2-preferential (if not selective) that also bind, albeit with lower affinity, SSTR-5 (Table I).24, 25 SSTR-2 and SSTR-5 are the predominantly expressed somatostatin receptor subtypes expressed by GH-secreting pituitary adenomas (Table I).24, 25 Subcutaneous octreotide has to be administered three times a day in order to appropriately suppress GH levels.10 A depot preparation known as Octreotide LAR was launched in Europe in 1997 and has, for the most part, displaced the subcutaneous formulation in the treatment of acromegaly.1 In octreotide LAR (sandostatin LAR) (long acting repeatable), the SSA is contained within microspheres made of a biodegradable polymer (poly-[DL-lactidecoglycolide] glucose).26 After intramuscular injection, octreotide levels rise sharply during the first 5 days, reflecting the SSA that lies on the surface of the microspheres; thereafter, octreotide is steadily liberated as the microspheres are degraded, and this goes on for approximately 30 days.26 Octreotide LAR exists as 10, 20 and 30 mg formulations. Although the original lanreotide formulation, Lanreotide SR, was developed as a slow release preparation that could be administered every 7, 14 or even 28 days for the 60 mg preparation, in 2002 a product known as Lanreotide autogel (Somatuline Autogel), which can be administered monthly was officially launched and is currently the form used to treat acromegaly.27 Somatuline Autogel

354

is a water-based preparation that after being deposited in the deep subcutaneous compartment, releases lanreotide over 30 days without an initial peak; it comes in prefilled syringes containing 60, 90 and 120 mg of lanreotide.27 Safety and efficacy Traditionally, a complete biochemical response was defined as the achievement of both, a GH 20%) is achieved in 60-70% of patients.41, 42, 47, 48 Although few controlled studies have formally compared Octreotide LAR and Lanreotide Autogel, the efficacy rate and safety profile of the two medications are similar.49-51 Upon switching well-controlled patients from monthly 20 mg Octreotide LAR to 90 mg Lanreotide Autogel, a significant proportion of patients will require the lanreotide dose to be escalated to 120 mg in order to maintain biochemical control highlighting the importance of carefully monitoring GH and IGF-1 levels when moving from one SSA to the other.49-51 Predictors of pharmacological response to SSA Although there is no absolutely reliable way of identifying those patients who will eventually respond to long-term therapy with SSA, some factors inherent both to the individual patient and to the tumor biology itself have been linked to the pharmacological response. Patients with more severe hypersomatotropinemia, as judged by a very high GH concentration at baseline are known to be relatively poor responders.42-44 The value of the subcutaneous octreotide test remains controversial; some studies claim that this test is highly predictive of the pharmacological response,52-54 while others conclude that it should only be used to establish tolerance.55, 56 Some characteristics of the adenoma on MRI have been associated with a better outcome upon SSA treatment. In a recent study by Puig-Domingo et al. 62% of patients harboring hypointense lesions on T2-weighted mages were able to maintain a complete pharmacological response to SSA after 12 months of treatment in contrast to only 24% of those who

Vol. 41 - No. 3

had hyperintense adenomas.57 Other imaging methods like 111In-octreotide scanning cannot distinguish responsive from resistant patients, due to the background uptake by the normal pituitary gland.58 The presence of GSP-alpha or GNAS mutations seems to be associated with a better response to SSA,59, 60 although not all studies have confirmed this finding.61, 62 The granulation pattern of the adenoma by electron microscopy has also been associated with the success of SSA, with a better response found among patients with densely granulated tumors than in those with sparsely granulated adenomas.62, 63 The granulation pattern by electron microscopy closely correlates with cytokeratin immunostaining using cam 5.2 antibodies;62 sparsely granulated tumors show perinuclear staining (the so-called fibrous bodies), whereas densely granulated adenomas present a diffuse cytoplasmic immunostaining.62 A low proliferative index as judged by the number of nuclei positive for Ki-67 has also been associated with a favorable response to SSA independently of SSTR-2 expression and in close correlation with the cytokeratin pattern of the adenoma.64, 65 Germline mutations of the gene encoding the aryl hydrocarbon receptor interacting protein (AIP) can be found in up to 40% of cases of familial acromegaly; subjects harboring such mutations are unusually young, tend to have larger and more invasive tumors and to respond poorly to SSA.66 The prevalence of somatic AIP mutations in young patients with sporadic acromegaly has been reported to be between 4 and 8%.67 A recent study has found that in sporadic GH-secreting adenomas, AIP expression by IHC directly correlates with the subsequent pharmacological response to octreotide LAR.68 SSTR-2 expression by the tumor at both the mRNA (by RT/PCR) 69 and protein (by immunohistochemistry [IHC])70-72 levels has been positively associated with the response to SSA. However, a significant number of patients with tumors expressing high levels of SSTR-2 are resistant to these pharmacological agents.71-73 Compared to SSTR-2, SSTR-5 is expressed by a greater proportion of these


355

MERCADO


tumors yet its significance in terms of predicting response to SSA is less well understood.73 Although in vitro SSTR-5 appears to be fundamental for SSA-induced GH inhibition, a low SSTR-2/SSTR-5 mRNA ratio was found to be associated with a poor pharmacological response to Octreotide LAR.69 At the protein level, while SSTR-2 immunostaining is found in tumors from SSA-responsive patients; in the majority of SSA-resistant subjects SSTR2 immunostaining is absent, whereas SSTR-5 immunostaining remains.71, 73 Despite these observations subsequent IHC studies have mostly focused on SSTR-2. Thus, the mere expression of these receptors does not guarantee an adequate pharmacological response;74 rather, other, not fully characterized functional aspects of this interaction, such as receptor homo- and heterodimerization, and the resulting signaling cascade, likely play a role in determining whether a patient with acromegaly will respond or not to these agents.75 SSTR-5 generates two truncated splice variants, SSTR5TMD4 and SSTR5TMD4, which although notoriously absent from normal pituitary tissue, are abundantly expressed in GHsecreting adenomas; the expression of these splice variants has been negatively associated with the in-vivo response to SSA.76, 77 Follow up of patients with acromegaly treated with SSA Biochemical assessment consists of periodical determinations of random GH and IGF-1 measurements; glucose-suppressed GH levels should not be used to evaluate the pharmacological treatment with SSA. In most centers, GH and IGF-1 are measured 3-4 months after the initiation of therapy. Patients who achieve a full biochemical response, i.e. a normal ageadjusted IGF-1 and a GH below 1 ng/mL, have their injection interval increased progressively to every 6, 8 or even 10 weeks, provided these goals are met and patient’s comorbidities remain controlled.78 On the other end of the spectrum, in subjects who do not show even a minimal reduction of GH and IGF-1 after 3 months we see no point in continuing therapy. Patients with a

356

partial biochemical responses, usually defined as a greater than 50% reduction in either GH and/or IGF-1 compared to baseline, can either have their SSA dose duplicated and/or have cabergoline (CBG) (1.5 to 3 mg per week) added to the treatment. When a patient shows discordant GH and IGF-1 results upon biochemical follow up, the decision to up-titrate the SSA relies on the presence or absence of symptoms and control of comorbidities. Using this algorithm routinely to treat patients with acromegaly in “real-life” conditions we have found that 35% of the patients reach the combined target of a GH