Diagnosis and Management of Acromegaly: A Consensus Statement of the Pituitary Study Group of the Portuguese Society of Endocrinology, Diabetes and Metabolism

Acromegaly is a rare, chronic, slowly progressive disorder caused by hypersecretion of growth hormone (GH, somatotropin) [1]. The term “acromegaly,” historically known as “Marie’s malady,” was first coined in 1886 by the French neurologist Pierre Marie. He linked the term to a clinical disease with a characteristic phenotype: “a condition characterised by hypertrophy of the hands, feet, and the face exists which we propose to be called acromegaly which means hypertrophy of the extremities” [2].

In most cases, GH is persistently secreted by a monoclonal tumour of the anterior pituitary gland arising from somatotroph cells – a somatotroph tumour [1, 3]. The pathogenetic mechanisms underlying tumourigenesis are not fully established and may involve genetic and/or epigenetic alterations, resulting in cell cycle dysregulation, signalling defects, or loss of tumour suppressor factors [3].

GH is synthesised, stored, and secreted by somatotroph cells in response to a complex regulatory system, including hypothalamic growth hormone-releasing hormone (GHRH) and somatostatin [4]. The secretion of GH is suppressed by somatostatin signalling, primarily through the somatostatin receptors subtype-2 (SST2) and subtype-5 (SST5). GH stimulates the production of insulin-like growth factor 1 (IGF-1), mainly from the liver but also from extra-hepatic tissues. IGF-1 is the main mediator of the somatic and metabolic effects of GH. In acromegaly, the excessive production of GH and IGF-1 leads to a multisystemic disease characterised by somatic overgrowth, physical disfigurement, metabolic disturbances, and multiple comorbidities [4]. If uncontrolled, this disorder is associated with significant morbidity and increased mortality [1, 5].

Due to the rarity of acromegaly, large population studies are needed to generate reliable epidemiological data. A recent systematic review and meta-analysis provided an update on the epidemiology of the disease [6]. The pooled prevalence of acromegaly from 22 studies was 5.9 cases per 100,000 people, and the incidence rate was 0.38 cases per 100,000 person-years; however, the authors considered that the quality of the studies was medium/low with marked heterogeneity [6]. Acromegaly is typically diagnosed between the fourth and the fifth decade of life, affecting both sexes, but its incidence appears to be slightly higher in females [1, 7].

The aims of treatment are to normalise hormonal hypersecretion, control tumour growth, manage comorbidities, improve the signs and symptoms of the disease, decrease morbidity and mortality, as well as improve the patient's quality of life. To achieve these goals, the treatment requires a multidisciplinary team using different therapeutic tools, including neurosurgery, medical therapy, e.g., somatostatin receptor ligands (SRLs), dopamine agonists (DAs), GH receptor antagonist (GHRA), and radiotherapy [8, 9].

A multidisciplinary group of experts from the Pituitary Study Group of the Portuguese Society of Endocrinology, Diabetes and Metabolism convened to develop a consensus statement on the diagnosis and management of acromegaly. The experts reviewed the current literature, focusing on key topics of diagnosis, disease control and remission, therapy with SRLs, DAs, GHRA, alone or in combination therapy, and radiotherapy. Herein, the authors present an up-to-date consensus document on the diagnosis and management of acromegaly.

The topics of interest for this consensus statement were identified through a literature search in PubMed for papers in English published up until December 2023, and the authors’ personal archives were also screened for additional references. Search terms included “acromegaly” and terms associated with each topic: “biochemical outcomes,” “tumour volume,” “clinical symptoms,” “somatostatin receptor ligand,” “dopamine agonist,” “GH receptor antagonist,” “oestrogens,” “selective oestrogen receptor modulators,” “mortality,” “complications,” “surgical outcomes,” “remission,” “cure,” and “guidelines.” The panel was composed of experts in pituitary diseases with different backgrounds from the Pituitary Study Group of Portuguese Society of Endocrinology, Diabetes and Metabolism. Panellists were selected based on their demonstrated experience, knowledge, and contributions to the field, ensuring a representation from different professional backgrounds. A modified Delphi methodology was employed to achieve consensus among the panellists [10]. This interactive process involved a first round to define the topics of interest and then the panel members were assigned to those specific topics. Panellists were asked to generate the statements and provide the supporting scientific evidence related to each topic of interest, based on the revision of the current scientific literature and on their expert opinions, when insufficient clinical evidence was available. In a second round, panellists were presented with the compiled list of statements and were then asked to vote each statement based on predefined criteria, such as “agree as is,” “agree with modifications,” “disagree,” or “not my area of expertise.” Panellists were also encouraged to provide comments or suggestions for refinement. This step was followed by the analysis of the responses and comments. The revised version of the statements were subjected to a final round for approval, and then the manuscript was prepared and approved by all authors. Each round aimed to converge towards a group consensus. Consensus was defined as ≥80% agreement among the panellists, and the resulting statements are presented and discussed in the following sections.

In acromegaly, elevated levels of GH and IGF-1 lead to somatic overgrowth, physical disfigurement, multiple comorbidities, poor quality of life, and premature mortality [4]. Acromegaly affects both women and men, although men present at a younger age, and women appear to have a higher incidence and mortality risk [11]. The presentation of acromegaly can be insidious, and despite the latest advancements in management, that have reduced the overall mortality, the diagnosis is still often delayed, and this negatively impacts patients’ quality of life and their morbi-mortality in the long term [4, 11]. Current guidelines suggest the screening of acromegaly in patients with typical clinical manifestations, especially those with facial and acral acromegaloid features, but also in patients who have several acromegaly-related conditions, even if typical manifestations are absent, and in individuals with a pituitary mass [4].

Key typical clinical features of acromegaly include enlargement of hands and feet, as well as craniofacial changes, such as coarse facies, frontal bossing, large nose and lips, prognathism, diastema, and macroglossia [4, 12, 13]. Accelerated linear growth or tall stature, particularly if the height is ≥2 standard deviations above the mid-parental height, are key manifestations of pituitary gigantism, resulting from exposure to excessive levels of GH/IGF-1 before the epiphyseal closure [14]. Other common features of acromegaly include thickened soft tissue, hyperhidrosis, oily skin, skin tags, hirsutism, hoarse voice, headache, visceromegaly, fatigue, and weight gain [4, 12, 13]. Some skin manifestations are uncommon in acromegaly, including acanthosis nigricans, pachydermia, and cutis verticis gyrata (i.e., a rare benign skin lesion characterised by thickened and folded scalp resembling brain gyri and sulci, of which acromegaly is among the main causes) [15, 16]. The following conditions and comorbidities are common in acromegaly: sleep apnoea syndrome (60–80%), arthralgia and debilitating arthritis (30–70%), carpal tunnel syndrome (40–60%), backache and vertebral disease (40–50%), glucose intolerance (16–46%), diabetes (20–56%), hypertension (20–50%), menstrual irregularities, erectile dysfunction, colonic polyps (≤45%), and malignant neoplasms (10–23%) [4, 12, 17].

Many individuals with acromegaloid appearance, tall stature or overgrowth, or acromegaly-related conditions will have normal variants of physical traits or growth. Some will have elevated GH/IGF-1 levels, leading to a diagnosis of acromegaly or pituitary gigantism. However, a few cases may display similar clinical features of acromegaly but without GH/IGF-1 abnormalities, falling into the pseudoacromegaly category. The term pseudoacromegaly describes patients with clinical features of acromegaly or pituitary gigantism but with no GH/IGF-1 abnormalities, encompassing a wide range of conditions (e.g., pachydermoperiostosis, Sotos syndrome, Weaver syndrome, Cantú syndrome, etc.) that are difficult to diagnose due to their rarity and heterogeneity, as well as due to the overlapping clinical characteristics among them [18, 19].

Acromegaly and pituitary gigantism are most often caused by a somatotroph tumour. GH excess due to pituitary hyperplasia is less common and mainly occurs as part of genetic disorders, such as the Carney complex, X-linked acrogigantism (X-LAG), or McCune-Albright syndrome [20]. Rarely, GHRH-secreting neuroendocrine tumours or ectopic GH-secreting tumours may also lead to acromegaly [21]. Most somatotroph tumours occur sporadically, with ∼40% of cases associated with somatic mutations in the GNAS gene. However, a subset of these cases may be familial [20]. To date, the percentage of familial forms of acromegaly remains unestablished, although it is known that nearly 50% of childhood-onset somatotroph tumours, leading to gigantism, have an identifiable genetic abnormality [14]. More broadly, about 5% of all pituitary tumours occur in the background of a predisposing germline mutation, more often in aryl hydrocarbon receptor-interacting protein (AIP) or menin 1 (MEN1) genes [22, 23]. Hereditary acromegaly may be accompanied by other conditions, as part of a syndromic disease, such as in multiple endocrine neoplasia type 1 or 4 (MEN1, MEN4), Carney complex, or the so-called “3 PAs syndrome” (i.e., paragangliomas, phaeochromocytomas, and pituitary tumours). It may also present as part of familial isolated pituitary adenoma (FIPA), a heterogeneous condition characterised by the occurrence of two or more pituitary tumour cases in the same family, in the absence of other syndromic manifestations (shown in Fig. 1) [22, 23].

FIPA is the most common cause of familial acromegaly or pituitary gigantism and can be categorised into three groups, depending on the genetic background: (i) AIP mutation-positive cases; (ii) X-LAG cases (rare); and (iii) FIPA with unknown genetic cause (most common scenario) [22, 23]. FIPA due to AIP mutations comprises about 10–20% of all FIPA families, predisposing individuals to young-onset pituitary tumours, that are typically aggressive, invasive, large, and refractory to conventional treatments, including SRLs [22‒25]. About 50% of AIP mutation-positive kindreds present as sporadic cases due to the incomplete and relatively low penetrance of the disease, with only 20–23% of AIP mutation carriers developing a pituitary tumour [22, 23]. The prevalence of AIP mutations in apparently sporadic young-onset cases ranges from 4 to 14%, while ≤20% of apparently sporadic paediatric pituitary tumours may be due to an AIP mutation [25‒30]. The risk of an AIP mutation is particularly high for patients ≤18 years old with a GH-secreting macroadenoma and a family history of pituitary tumours [31]. A lower frequency of AIP germline mutations (i.e., 1.8%) was reported in a Portuguese cohort of patients with sporadic pituitary macroadenomas diagnosed before the age of 40; however, the frequency of AIP mutations almost doubled (i.e., 3.4%) if only patients <30 years were considered and reached 5% in patients <18 years [32]. These data, combined with the observation that ∼90% of all patients with clinically presenting AIP mutated pituitary tumours have disease onset at ≤30 years, support the use of this cut-off as probably the best for identifying young-onset sporadic cases [22, 23, 25, 33, 34]. X-LAG, caused by GPR101 gene microduplications, is rare and is the main cause of gigantism in very young patients (i.e., below the age of 5), resulting from GH excess due to a somatotroph tumour or pituitary hyperplasia [33, 35].

MEN1 syndrome occurs due to germline mutations in the MEN1 gene and typically involves the parathyroids (i.e., primary hyperparathyroidism), duodeno-pancreatic neuroendocrine tumours, and the pituitary, with pituitary tumours occurring in about 30–50% of subjects carrying a germline MEN1 mutation [20]. Pituitary tumours may be the first manifestation of MEN1 syndrome in ≤15% of patients, with prolactinomas representing the majority of MEN1-related pituitary disease; however, 7–13% may be somatotroph tumours [36‒39]. Acromegaly may also occur in other rare syndromes, such as MEN4, Carney complex, “3 PAs syndrome”, neurofibromatosis type 1, and Lynch syndrome [20, 22].

Multifocal pituitary tumours are rare in sporadic settings but have been described more often in familial settings, such as Carney complex, MEN1, or AIP mutation-positive disease [40‒42]. The presence of two or more pituitary tumours in the same patient may raise suspicion for an inherited condition; thus, genetic testing may be considered. Resistance to SRLs is a typical feature of AIP mutation-positive acromegaly [24, 25]. However, it was shown that AIP (or MEN1) testing in patients with acromegaly resistant to octreotide or lanreotide does not improve the detection rates of AIP (or MEN1) germline mutations [43]. Thus, genetic testing for patients with GH-secreting pituitary tumours resistant to SRLs and/or DAs cannot be generally recommended.

The identification of germline mutations in patients with acromegaly will enable genetic and clinical screening of at-risk family members, allowing for earlier identification and treatment of the disease, which can lead to better outcomes. For syndromic forms, such as MEN1 or “3 PAs syndrome,” it facilitates the early screening, diagnosis, and treatment of other syndromic manifestations in the patient and their affected family members [20, 25, 44, 45].

Patients with clinical symptoms of acromegaly, especially those with acral and facial features, should undergo biochemical screening, starting with the measurement of serum IGF-1 levels. The Endocrine Society also suggests measuring IGF-1 in patients without typical acromegaly clinical features but with several associated conditions, e.g., sleep apnoea syndrome, type 2 diabetes mellitus, debilitating arthritis, carpel tunnel syndrome, hyperhidrosis, or hypertension [4]. It is also recommended to assess IGF-1 in patients with a pituitary mass [4, 46].

Serum IGF-1 is a reliable biochemical marker for assessing integrated GH secretion over the previous 24 h and is a crucial tool for the diagnosis of acromegaly [46]. In contrast to the pulsatile secretion of GH, serum IGF-1 levels remain stable throughout the day and do not have significant circadian variation. However, the reliability of serum IGF-1 may be affected by technical aspects, including issues with the validation and standardisation of IGF-1 laboratory methods and the establishment of normal reference ranges, as well as biological factors and medical conditions that influence IGF-1 concentrations (shown in Fig. 2) [47]. Some intrinsic characteristics of IGF-1 can introduce sources of error in its laboratory assessment. These include the strong binding of IGF-1 to high-affinity IGF-binding proteins (IGFBPs), glycosylation at the recognition site, IGFBP3 proteolysis, and the influence of other factors, such as age, sex, ethnicity, body mass index (BMI), thyroid hormone levels, among others [47, 48].

Serum IGF-1 levels are physiologically higher during adolescence and pregnancy, thereby posing difficulties in the interpretation of IGF-1 levels in these groups [49]. Thus, age-specific normal range intervals must be considered when interpretating serum IGF-1. On the other hand, serum IGF-1 levels can be blunted in patients with acromegaly who exhibit resistance to GH action related to advanced liver or kidney disease, severe hypothyroidism, malnutrition, anorexia, and poorly controlled diabetes mellitus, or in women receiving oral oestrogens [12, 49, 50]. Additionally, variability in IGF-1 assays across different laboratories may pose further challenges, as demonstrated in a multicentric UK-based study that found more than a twofold difference in the serum IGF-1 levels in the same sample among different laboratories, leading to diagnostic failure of acromegaly in 30% of the centres [51]. To overcome such variability and obtain accurate results, it is recommended that all pre-analytical variables be controlled, and serum IGF-1 should be assayed in the same laboratory for serial follow-up [50, 52]. Currently, there is no evidence that IGF-1 measurement by mass spectrometry is superior to immunoassays [53].

According to a recent consensus, in patients with typical signs and symptoms of acromegaly, IGF-1 levels >1.3 times the upper limit of normal for age confirm the diagnosis without the need for further testing [53]. In patients with equivocal results, it is advisable to repeat the IGF-1 measurement. Additionally, an oral glucose tolerance test (OGTT), assessing the GH nadir after administration of 75 g of oral glucose, may also be useful [53].

Random serum GH levels are inadequate for diagnosing acromegaly due to the pulsatile secretion of GH and the marked changes of GH concentrations throughout the day. In contrast, dynamic testing with a 75 g oral glucose load followed by serial measurements of GH levels may be useful to the diagnosis of acromegaly [53]. Normal subjects will exhibit suppression of GH to undetectable levels following a 75 g oral glucose load. However, the GH response during an OGTT may vary significantly in acromegalic and non-acromegalic individuals and should be interpreted cautiously, according to the individual physiological, pharmacological, pathological, and clinical factors [46, 52]. Random and during an OGTT GH levels may vary depending on several factors, such as sex, BMI, physical exercise, circadian rhythm, medications (e.g., oestrogen-containing oral contraceptives), glycaemic status, liver, and kidney function [52]. GH may fail to suppress in healthy adults, such as adolescents or young females (particularly those on oestrogen-containing oral contraceptives), or in individuals with liver or kidney disease, poorly controlled diabetes, anorexia, or other conditions (shown in Fig. 2) [4, 47, 50].

The appropriate GH nadir cut-off during an OGTT has been extensively debated over the years. The Endocrine Society guidelines, published in 2014, adopted a suppressibility cut-off of GH <1.0 µg/L, arguing that a GH nadir <1.0 µg/L within 2 h after oral glucose load usually excludes the diagnosis of acromegaly [4]. It was noted that a GH nadir <0.4 µg/L may be more reliable in establishing or ruling out acromegaly. However, this lower cut-off was not adopted at that time due to the insufficient accuracy of many assays at GH levels <1.0 µg/L [4]. Reference levels for GH nadir using the IDS-iSYS GH assay during an OGTT, that consider BMI, sex, and oestradiol-containing oral contraceptives have been established [11, 54].

According to the latest acromegaly consensus, if an OGTT is performed, 75 g of oral glucose should be administered in a fasting state, with GH levels assessed after 30, 60, 90, and 120 min [53]. The lack of suppression of BMI-based GH nadir level to <0.4 µg/L for BMI <25 kg/m² and <0.2 µg/L for BMI ≥25 kg/m² can be considered for diagnosis [53]. These proposed cut-offs derive from a large multicentre study investigating nadir levels of GH during an OGTT with the IDS-iSYS GH immunoassay, which concluded that BMI, sex, and oestrogen-containing oral contraceptives are major determinants of the OGTT GH nadir levels [54]. Healthy premenopausal women on oestrogen-containing oral contraceptives typically have higher GH levels; hence, it is recommended to discontinue these drugs 4 weeks prior to the OGTT [53, 55]. It is also worth noting that around one-third of patients with acromegaly may show a paradoxical increase in GH after OGTT [56].

In patients with impaired oral glucose tolerance or type 2 diabetes mellitus, the OGTT appears to be safe and effective for the diagnosis of acromegaly using the BMI cut-offs mentioned above. However, due to the suppressive effect of hyperglycaemia on GH, the results should be interpreted with caution, particularly in patients with uncontrolled diabetes [53].

Biochemical discrepancies between GH and IGF-1 levels are not uncommon in acromegaly [47, 57, 58]. A large meta-analysis encompassing 39 studies (7,071 patients with acromegaly) reported a pooled discordance rate between GH and IGF-1 of ∼26%. The most frequent discrepancy (i.e., ∼15%) was elevated IGF-1 with normal GH levels (GH status was based on mean value of a multiple-sample day curve, nadir value during an OGTT, or random basal sample of GH) [58]. In situations of discrepant values between GH and IGF-1, the use of other markers related to the GH action, such as IGFBP3 and the acid-labile subunit, may be helpful [11]. Soluble klotho levels correlate with GH and IGF-1 levels and correlate with clinical symptoms and quality of life in treated patients with acromegaly [11, 59, 60]. However, more studies are needed to establish whether soluble klotho is a valid marker in clinical practice. Metabolomics may also serve as a future auxiliary tool in the diagnosis and prognosis of pituitary tumours, including somatotroph tumours [61].

Metabolic, endocrine, cardiovascular, and other systemic complications that are common in acromegaly should be actively screened (shown in Table 1) [4, 62]. Routine biochemical tests, such as glucose, glycated haemoglobin (HbA1c), lipid profile, calcium, and phosphorus, help in the screening of diabetes, dyslipidaemia, and primary hyperparathyroidism, which may occur in MEN1 or MEN4. In acromegaly patients with parathyroid hormone-independent hypercalcaemia, measuring the serum levels of 1,25(OH) vitamin D can be useful [62, 63]. A comprehensive assessment of pituitary function should be also carried out, particularly in cases of pituitary macroadenoma. This should include free T4 (FT4), thyroid-stimulating hormone (TSH), morning serum cortisol (and, if necessary, the cosyntropin stimulation test), follicle-stimulating hormone (FSH), luteinising hormone (LH), total testosterone (in males), or oestradiol (in females). Prolactin (PRL) co-secretion occurs in 30% of patients with acromegaly, and its elevation may be functional, secondary to compression of the pituitary stalk by the tumour, or due to a mixed GH-PRL tumour [12].

Given the increased risk for colonic polyps in acromegaly, a screening colonoscopy at the time of diagnosis is recommended (shown in Table 1) [11]. After treatment of acromegaly, a colonoscopy can be repeated according to the guidelines for the general population, provided the patient no longer has active acromegaly [11]. Other exams may be needed to evaluate other acromegaly-related complications on an individual basis and according to the clinical scenario, including thyroid ultrasound (e.g., for clinically suspected thyroid nodules), echocardiography (e.g., for cardiomyopathy and heart valve disease), polysomnography (e.g., for sleep apnoea syndrome), among others (shown in Table 1).

After the biochemical diagnosis of acromegaly, an imaging study should be performed to identify the pituitary tumour [4, 64]. Magnetic resonance imaging (MRI) should be the first choice, with computed tomography (CT) scan as an alternative, if MRI is unavailable or contraindicated [4, 46, 65]. MRI should be performed using 1.5 Tesla or 3 Tesla scanners, with two-millimetre slices, which is particularly important to identify small microadenomas, although most cases of pituitary acromegaly (∼77%) are caused by macroadenomas [46]. MRI reports should be standardised and routinely include relevant information, such as the tumour size, intensity signal, and tumour relationship with neighbouring structures (e.g., optic chiasm, cavernous sinus, and sphenoid sinus). A detailed description of the supra- or infra-sellar extension, and the invasion of the cavernous sinus, using the modified Knosp classification, should also be reported. T2-weighted MRI signal intensity can predict the response to SRLs, with hypointense tumours showing better response to therapy with SRLs [53, 66].

Several groups have studied the role of functional imaging in detecting small microadenomas or sites of residual disease that may be amenable for repeat surgery or targeted radiotherapy, such as stereotactic radiosurgery [67]. Among the reported ligands, 11C-methionine has been shown to be particularly useful when MRI results are indeterminate or when the true extent of lateral tumour extension is unclear [67]. In a study involving 17 patients, functional imaging with 11C-methionine positron emission tomography (PET) showed higher sensitivity in detecting somatotroph tumours than 18F-fluorodeoxyglucose, with similar specificity [68]. Additionally, 18F-fluoroethyltyrosine has been successfully used to detect a somatotroph tumour in a paediatric patient who had undergone three previous transsphenoidal surgeries [69].

Very rarely, patients with biochemically confirmed acromegaly may not have a detectable tumour in the pituitary gland. In such cases, serum GHRH should be measured, and other imaging methods may be performed to investigate ectopic disease, such as thoracic and abdominal CT scan, and 68Ga-DOTA-somatostatin analogue PET/CT. 18F-fluorodeoxyglucose-PET should be reserved for the staging of ectopic tumours with aggressive behaviour [21, 70, 71].

Although biochemical remission is the primary treatment goal, it is not the only therapeutic objective in acromegaly [53]. Disease control is achieved when different goals are met: biochemical normalisation, control of tumour mass, preservation or replacement of pituitary function, prevention and control of symptoms and signs, management of comorbidities, improvement quality of life, and reduction of morbidity and mortality [5, 8, 50, 72].

For biochemical monitoring during or after treatment, both IGF-1 and GH levels may be measured, as they yield qualitatively different types of information. IGF-1 correlates more closely with improvement in symptoms and signs, while serum GH is more indicative of tumour secretory activity [73, 74]. Postoperative random GH levels and GH nadir during an OGTT are associated with long-term remission. An immediate postoperative GH level <1.0 μg/L is highly predictive of biochemical remission, and a GH nadir level <0.4 ng/mL 1 week postoperatively showed a positive predictive value >95% for surgical remission [13]. The postoperative decline of IGF-1 is slower than that of GH likely due to the longer half-life of IGFBPs. IGF-1 levels in the early postoperative are highly variable and stabilise approximately 12 weeks after surgery [75]. Therefore, serum IGF-1 should be measured at least 12 weeks postoperatively to determine the biochemical status of the disease [53]. If preoperative SRLs were used, serum IGF-1 should be repeated at 3–6 months to confirm remission, since the SRLs carryover effect may influence postoperative IGF-1 levels [53, 76, 77]. Moreover, some patients with acromegaly undergoing successful surgery have a delayed normalisation of serum IGF-1. In a series including 46 patients with acromegaly who underwent surgery and showed no residual tumour on MRI postoperatively, along with a nadir GH level <0.4 μg/L during an OGTT, it was found that 41% experienced a delayed (i.e., >3 months) IGF-1 normalisation, and 24% of them achieved a normal serum IGF-1 levels only 12–57 months after surgery [78].

The biochemical definition of remission is difficult and not consensual. Several biochemical tests have been used across different studies to define remission, including serum IGF-1, GH nadir during an OGTT, random GH, and mean GH on a GH day curve, with different laboratory assays, reference ranges, and cut-offs, complicating the establishment of robust conclusions. Therefore, there is no definitive evidence or universal consensus on the optimal evaluation to define postoperative remission, nor on the timing for such evaluations. In general, studies favour using serum IGF-1 to monitor the disease activity in the medium/long term, with normal age-adjusted IGF-1 levels as the goal. Given the variability of results between different immunoassays and between distinct laboratories, it is advisable to measure IGF-1 using the same assay and in the same laboratory during the follow-up of a patient with acromegaly [1, 4, 51]. Previous recommendations defined a random GH level <1.0 µg/L or a GH nadir level <0.4 µg/L during an OGTT as therapeutic targets. However, shortcomings of GH assessments at baseline or after glucose load limit their use as criteria to define remission or disease control in acromegaly [4, 53, 79]. A recent consensus established the efficacy goal as a normal age-adjusted serum IGF-1, while recognising the utility of GH nadir during an OGTT in the evaluation of patients with borderline IGF-1 levels and clinical signs of active acromegaly [53]. In cases where GH and IGF-1 are discrepant or unreliable indicators of changes in comorbidities, monitoring of the disease may implicate a direct assessment of acromegaly-related comorbidities (e.g., formal sleep studies, pulmonary function testing, or echocardiography) [80].

The optimal management of patients with acromegaly requires a multidisciplinary team, including endocrinologists, pituitary neurosurgeons, neuroradiologists, neuropathologists, and radiotherapists with extensive expertise in pituitary and parasellar tumours. The therapeutic options in acromegaly are surgery, medical therapy, and radiotherapy. A proposed management algorithm is shown in Figure 3 [81, 82].

As the most effective option to achieve complete biochemical remission and provide tumour tissue for pathological characterisation, transsphenoidal tumourectomy is the cornerstone treatment for somatotroph tumours and the optimal primary treatment for eligible patients [4]. Surgical debulking should be considered when there is visual impairment and when a substantial amount of the tumour can be successfully removed to enhance the response to medical therapy and/or radiotherapy [4, 8]. Repeated pituitary surgery may also be considered in patients with acromegaly and residual intrasellar disease following the initial operation, particularly if the initial intervention was not performed at a pituitary tumour centre of excellence [4].

Currently, the choice of the surgical technique (i.e., endoscopic or microscopic approaches) depends on the neurosurgeon’s expertise, as there are no definitive data recommending one approach over the other [83]. When performed by an experienced neurosurgeon, transsphenoidal surgery (TSS) is effective in 75–100% of acromegaly patients with microadenomas, but remission rates for macroadenomas drop down to 48–75% [4, 8, 84‒86]. Perioperative complications include bleeding, cerebrospinal fluid rhinorrhoea, meningitis, arginine vasopressin deficiency, and hypopituitarism [4]. Perioperative mortality rates are <1% in expert hands [87].

Published data are contradictory regarding the potential benefits of preoperative treatment with SRLs administered 3–6 months before surgery. While patients with severe acromegaly comorbidities and high surgical risk (e.g., severe pharyngeal thickening and sleep apnoea syndrome, or high-output heart failure) may have some improvement, routine preoperative SRL treatment is not recommended for other patients with acromegaly [4, 88]. Data on postsurgical remission rates are controversial. While short-term remission rates appear to improve after preoperative medical treatment, the long-term outcomes remain unchanged [88].

Preoperative medical treatment with SRLs should also be considered when an expert neurosurgeon is not available or when surgery is expected to be significantly delayed. However, effective SRL treatment should not postpone surgery, as it remains the first-line treatment for acromegaly [88].

Although TSS is the first-choice treatment for most microadenomas and macroadenomas, around 40–60% of macroadenomas cannot be controlled by surgery alone. Some patients may refuse surgery or have contraindications due to high anaesthetic risk (e.g., severe cardiomyopathy or respiratory disease). Additionally, the lack of an experienced pituitary neurosurgeon, conducting at least 50 pituitary operations per year, has been associated with lower control rates and should prompt referral of the patient to a specialised centre [89]. The Endocrine Society guidelines, published in 2014, recommend the use of SRLs as the primary medical treatment in patients with acromegaly who cannot be cured surgically, those with tumours invading the cavernous sinus without compressing the optic chiasm, or for patients deemed poor surgical candidates [4]. The 2011 guidelines of the American Association of Clinical Endocrinology also suggest primary medical therapy in patients who have a macroadenoma without compressive effects, who have a tumour that cannot be completely removed surgically (because of extrasellar extension of the tumour, especially into the cavernous sinus), who are poor surgical candidates, or who prefer medical treatment [46, 49].

Different studies showed that primary therapy with long-acting SRLs reduces the hypersecretion of GH and IGF-1 in about one-third of patients naïve to therapy and decreases the tumour volume by 20% in two-thirds [90, 91]. When comparing head-to-head the efficacy of octreotide long-acting release (LAR) with surgery, a prospective randomised controlled trial showed that treatment outcomes were not significantly different [92].

In a series of patients with acromegaly who refused surgery or had severe comorbidities, 5 years of primary medical treatment with SRLs resulted in biochemical control and improvement of hypertension, dyslipidaemia, and respiratory complications [93]. Other studies revealed variable results but with a significant proportion of patients reducing at least 50% of tumour volume with SRLs, which was correlated with biochemical control (i.e., percent of IGF-1 and GH decrease) [94, 95].

When medical therapy fails or when it is the patient’s preference (e.g., patient does not want to receive long-term injections), radiation therapy may be an alternative treatment option (shown in Fig. 3). However, medical therapy may be needed until the effectiveness of radiotherapy is established, which can take several years [4, 96].

Biochemical control of GH and IGF-1 remains the strongest predictor of patient outcomes and is one of the main treatment goals in acromegaly [11, 97]. TSS is the first-line treatment and the most effective option to rapidly achieve normal GH and IGF-1 levels and remission of the disease.

Large tumours, with extrasellar extension or cavernous sinus invasion, high preoperative serum GH levels, and female sex are associated with poor remission rates after pituitary surgery [85, 86, 98‒101]. In cases of persistent or recurrent disease following TSS, medical therapy (i.e., SRLs, DAs, and GHRA, alone or in combination therapy) and radiotherapy are generally recommended as second-line and third-line options, respectively (shown in Fig. 3) [8, 49, 85, 86, 98‒101].

The limited data on second surgical attempts, the variable success rates, and the increased risk for surgical complications, have maintained the debate open as to whether reoperations should be performed [102‒104]. A recent meta-analysis evaluated 161 reoperations and 2,180 first-time surgeries reported a disease control rate for reoperation of ∼47% and ∼56% for first-time surgery. In microadenomas, the control rate was also similar (comparing first-time surgery to reoperation), however in macroadenomas reoperation was associated with a lower control rate in comparison to first-time surgery (i.e., ∼28% vs. ∼54%) [104]. Reoperation may be considered in patients with a potentially resectable residual tumour after an unsuccessful first surgery and in cases with significant residual tumours and poor response to postoperative medical therapy [8].

When remission is not achieved with TSS, medical therapy should be initiated. There are three classes of drugs available: SRLs, DAs, and GHRA (shown in Table 2) [8, 11, 49, 84].

First-generation SRLs (e.g., octreotide LAR and lanreotide Autogel) are considered the first-line of medical treatment in most cases [8]. These drugs are usually administered monthly, either intramuscularly (octreotide) or subcutaneously (lanreotide), and they target the SST2 expressed by pituitary tumours and have concomitantly anti-secretory and anti-proliferative effects [49, 109]. Normal GH and IGF-1 levels may be achieved in 30% and 55% of cases, respectively, while significant tumour shrinkage (i.e., tumour volume reduction >20%) can be observed in ≤66% of patients [91, 109‒115]. Among individuals previously controlled with long-acting first-generation SRLs, recent studies showed that extended-dosing intervals of 120 mg lanreotide Autogel (i.e., beyond 4 weeks) may still be effective [79, 116].

There is a new formulation of octreotide, i.e., oral octreotide capsules, approved by the US Food and Drug Administration (FDA), in 2020, and the European Medicines Agency (EMA), in 2022, respectively. This formulation has been demonstrated to be effective for patients who have achieved a biochemical response with stable doses of injectable octreotide LAR or lanreotide Autogel. So far, there is currently no data supporting the use of oral octreotide capsules as primary medical therapy in SRL-naïve patients or in patients previously treated with pasireotide LAR [11, 117].

Pasireotide LAR is a second-generation SRL with higher affinity for SST4 and SST5 receptor subtypes, and also some affinity for SST2, approved by the FDA and EMA for acromegaly, in 2014 [118, 119]. Several prospective randomised trials showed that pasireotide LAR is effective in patients with acromegaly not controlled with maximal doses of octreotide LAR or lanreotide Autogel, achieving rates of biochemical remission ≤54% and higher response in tumour shrinkage [120‒124]. However, the risk of drug-induced hyperglycaemia and/or diabetes mellitus is greater with pasireotide LAR (i.e., ≤70% and 40%, respectively), likely due to impaired insulin and incretin secretion combined with a minimal impact on glucagon production [62, 120, 125]. Considering its safety profile and efficacy, this drug should be used in patients with resistance to first-generation SRLs, especially those with normal glucose metabolism and when there is a concern with tumour mass [62, 79].

Cabergoline is a DA with a high affinity for dopamine D2 receptors and low affinity for dopamine D1, α1-and α2-adrenergic, and 5-HT1- and 5-HT2-serotonin receptors. It is more potent, long-lasting, and better tolerated than bromocriptine [126]. Although cabergoline has a modest effect on acromegaly control, low baseline serum IGF-1 levels have been considered predictors of higher likelihood of biochemical response to this drug [127‒129]. For this reason, cabergoline is recommended as first-line medical therapy in patients in whom surgery has failed to achieve biochemical control and IGF-1 levels are only mildly elevated (i.e., <1.5 times the upper limit of normal) and may be considered as an add-on in cases of partial response to SRLs or pegvisomant [79, 127, 128]. IGF-1 normalisation is achieved in 35% of patients with high-dose cabergoline as monotherapy and in 50% when combined with SRLs [127].

Pegvisomant is a pegylated GH analogue with enhanced affinity for the GH receptor, preventing the signalling cascade activated by GH receptor-binding [130, 131]. This GHRA is usually recommended as a second-line medical therapy for patients with intolerance or incomplete biochemical response to maximal doses of SRLs; in addition, it can also be considered a treatment option in certain cases of persistent disease after surgery or when a surgical procedure is not performed [132]. Since pegvisomant does not target the pituitary tumour, GH hypersecretion persists during treatment, although tumour growth is rare (i.e., <5%) [133]. For this reason, in patients with large tumours abutting the optic chiasm and other vital structures, or in those with residual tumour, other tumour-targeted treatments should be preferred [8, 133]. Compared to other medical therapy, pegvisomant is of particular interest for individuals with diabetes or glucose intolerance, as it improves glycaemic control independently of IGF-1 control [134, 135]. Pegvisomant is highly effective in blocking the GH action, normalising the IGF-1 levels in 90% of patients in clinical trials and ≤73% in real-life studies [136‒139]. The therapeutic control rate is dose-dependent and in particular situations, such as patients with higher baseline IGF-1 levels, diabetes, or higher BMI increased doses may be required to achieve IGF-1 normalisation [139].

The systematic evaluation of clinical and biochemical parameters is crucial for monitoring therapeutic response and optimising treatment decision in acromegaly. In addition, tools like SAGIT or ACRODAT can be used to assess the disease status and evolution and to guide treatment in patients with acromegaly [140, 141].

Response to either surgical or medical treatment is variable and inconsistent. Understanding which factors contribute to variability in the therapeutic response to SRLs will help clinicians to better manage patients with acromegaly. Age, sex, tumour size, genetic background, baseline GH and IGF-1, along with histological, and imaging characteristics of the tumour are possible predictors of biochemical response to SRLs [114]. Older age, female sex, lower serum IGF-1, and GH levels at baseline are associated with a higher likelihood of achieving biochemical control [49, 142]. Patients with germline mutations in the AIP gene typically have a reduced response to first-generation SRLs, but a better response to pasireotide LAR [24, 25, 143]. Patients with X-LAG due to GPR101 microduplications also have a poor response to SRLs [33].

Several histopathological and molecular markers may also help predicting the response to SRLs. A lower Ki-67 index (i.e., <3%) and a higher SST2 expression have been associated with better treatment responses [49]. Densely granulated tumours are more likely to achieve remission compared to sparsely granulated tumours [144]. Moreover, patients with sparsely granulated tumours that are not controlled with SRLs appear to respond to a switch or addition of GHRA [144].

Imaging characteristics on MRI may also predict the response to SRLs. Tumours that secrete GH can be hypo-, iso- or hyperintense on T2-weighted MRI sequences [65]. The relationship between T2-weighted signal intensity on MRI and tumour response to SRLs as primary monotherapy was evaluated in a multicentre study. The response to SRLs, evaluated through random GH levels, IGF-1 levels, and tumour volume was correlated with the calculated T2 intensity, and the lower the T2 signal intensity the greater the response to SRLs [65, 66].

At the standard starting dose of SRLs, not all patients achieve biochemical remission and control of clinical disease. One approach is to start SRLs at a lower dose and titrate upwards at 3–6 months intervals, according to the serum levels of IGF-1. For octreotide LAR, the starting dose is typically 20 mg, administered by intramuscular injection every 4 weeks. If this dose fails to achieve normal IGF-1 levels after the first 3–6 months of treatment, the dose may be increased to 30 mg every 4 weeks, and then to the maximum dose of 40 mg every 4 weeks [145]. Lanreotide Autogel is started at doses of 60, 90, or 120 mg every 4 weeks. If serum IGF-1 levels do not normalise with lower doses of 60 or 90 mg every 4 weeks after 3–6 months, the dose may be gradually increased to 120 mg every 4 weeks [146, 147]. In selected cases with inadequately controlled disease by long-term conventional SRLs, the administration of lanreotide Autogel 120 mg more frequently (i.e., every 3 weeks) may control the disease in some of those patients [148].

Biochemical monitoring of patients on SRLs includes measuring serum IGF-1 levels every 3 months. Once normal serum IGF-1 levels are achieved, the doses of octreotide LAR or lanreotide Autogel should be maintained or eventually down-titrated. For lanreotide Autogel, the dose can be reduced or the dose intervals can be extended up to 6 or 8 weeks [11, 149]. The initial dose of pasireotide LAR is 40 mg every 4 weeks, but if serum IGF-1 levels do not normalise after 3–6 months, the dose can be increased to 60 mg every 4 weeks [150].

Switching between first-generation SRLs can improve the biochemical control and reduce adverse effects. When switching from octreotide LAR to lanreotide Autogel, doses of 60, 90, or 120 mg can be administered, regardless of the previous dose of octreotide LAR. However, when switching from lanreotide Autogel to octreotide LAR, the starting dose should be 20 mg every 4 weeks. In either case, serum IGF-1 levels should be measured regularly, and if the levels are not within the normal range the dose can be increased. When normal IGF-1 levels are not achieved after at least 6 months of treatment, with maximum doses of first-generation SRLs, therapy may be switched to pasireotide LAR [150].

SRLs are safe and effective for long-term use. In patients who achieve long-term control of acromegaly, reducing the dose or decreasing the frequency of administration of SRLs should be considered [151].

SRLs are generally well tolerated and safe, but patients should be informed about possible adverse events when starting treatment (shown in Table 2). The main adverse events of SRLs stem from their inhibitory effect on the secretion of pancreatic and gastrointestinal peptides, as well as their modulation of gastrointestinal functions (e.g., intestinal motility and gallbladder contractility) [152].

Octreotide LAR and lanreotide Autogel have similar side effects, which are generally transient, modest in severity, and may onset shortly after the first administration [153]. The frequency of adverse effects decreases progressively with continued treatment [46, 153]. These adverse effects include gastrointestinal disturbances (e.g., abdominal cramps, flatulence, diarrhoea, nausea, and vomiting), malabsorption (e.g., steatorrhoea), gallbladder disease (e.g., biliary sludge and gallstones), hair loss (rarely alopecia), and sinus bradycardia [46, 153]. Occasionally, SRLs may cause local skin irritation, erythema, and pain at the injection site [4]. Discontinuation of SRLs due to side effects is rare, primarily occuring in relation to gastrointestinal complaints [46, 153]. Cholelithiasis is one of the most serious complications associated witn SRLs; it is usually asymptomatic and may occur even after SRLs withdrawal [153]. Incident gallstones, without major complications, were observed in 10–22% of cases in the studies leading to market authorisation of SRLs for treating neuroendocrine tumours [154‒157]. In studies involving patients with acromegaly treated with SRLs, the reported incidence of cholelithiasis has been higher (42%–62%), although symptomatic disease requiring cholecystectomy is not common [158, 159]. Biliary stone disease is a frequent side effect of SRLs; however, given the infrequent occurrence of symptoms and/or related complications, only patients who have signs and symptoms of cholelithiasis should be evaluated by ultrasound [4].

The effects of SRLs on glucose tolerance and inducing hyperglycaemia-related events have yielded discordant findings in the literature (reviewed in [153]). However, the overall impact of SRLs on glucose metabolism appears to be beneficial, albeit modest and transient, except for pasireotide LAR, which is associated with significant hyperglycaemia and diabetes. This can be explained by the inhibitory effects of SRLs on insulin and glucagon, as well as GH secretion [4]. The greater risk of hyperglycaemia and diabetes with pasireotide LAR (e.g., occurring in 57% of patients) likely stems from impaired insulin and incretin secretion, with a minor effect on glucagon production, as previously mentioned, and the degree of hyperglycaemia is largely dependent on glycaemic control at baseline [62, 120, 123]. Patients being considered for treatment with pasireotide LAR should be carefully monitored for glycaemic adverse effects, and this drug should preferably be used in patients with normal glucose tolerance. Glucose levels should be monitored weekly for the first 3 months of treatment, as well as in the first 4–6 weeks after any dose increase. Ongoing monitoring of glucose status (e.g., glucose, HbA1C) should continue throughout treatment, as clinically indicated [79].

Other uncommon side effects of SRLs include asthenia, headache, dizziness, loss of appetite, constipation, pruritus, increased serum bilirubin, and decreased libido [160]. Additionally, SRLs can decrease TSH secretion leading to secondary hypothyroidism; thus, thyroid function tests should be carefully interpreted.

Pegvisomant is administered subcutaneously in daily doses of 10, 15, or 20 mg (daily dosing ranging from 10 to 40 mg) and has proven to be effective in clinical and biochemical control of acromegaly [161]. Given its long half-life (i.e., up to 100 h after subcutaneous administration), alternative dosing regimens, such as less frequent administration than daily, have also been shown to be equally effective and should be considered on an individual basis [162, 163]. So, alternate-day administration, once or twice weekly, or even monthly dosing, especially when pegvisomant is combined with SRLs, may be viable alternative therapeutic regimens, depending on several factors, such as the magnitude of IGF-1 elevation, planned monotherapy or combined therapy, prior irradiation, and patient’s specific conditions. It should be emphasised that a heterogenous response in terms of efficacy (i.e., normalisation of IGF-1) has been reported in a subset of patients treated with pegvisomant [131]. This apparent lower-than-expected efficacy, observed especially in retrospective studies compared to clinical trials, may be explained by various factors, such as patient poor compliance to daily injections, inadequate dose titration, selection bias in clinical trials, technical issues with IGF-1 assay, and finally a true “biochemical resistance” to pegvisomant cannot be ruled out yet [131].

As a GHRA, pegvisomant has a shorter half-life than injectable SRLs, and IGF-1 should be monitored every 1–3 months after treatment initiation or dose adjustments to determine the optimal dosing regimen. After achieving stability, monitoring can be spaced out to every 6–12 months [53]. Most commercial GH assays (except IDS-iSYS hGH immunoassay) cannot differentiate between endogneous human GH and pegvisomant, resulting in the measurement of pegvisomant levels as well. In addition, to this mimicking effect of pegvisomant in laboratory tests, the rapid normalisation of IGF-1 levels induces an increase in endogenous GH concentrations due to the feedback loop between IGF-1 and GH secretion via the hypothalamus and pituitary, making the evaluation of GH during pegvisomant treatment uninformative, and it should not be performed. [164]. Thereby, serum IGF-1 is the sole reliable biomarker for evaluating pegvisomant efficacy.

The dose requirements of pegvisomant to achieve disease control appear to vary among specific groups of patients, and the response to pegvisomant was shown to be influenced by patient’s age, BMI, and baseline IGF-1 levels [165, 166]. The starting dose of pegvisomant should be higher, with more rapid dose titration, in younger patients, in obese subjects, and in those with a more severe endocrine profile [165, 166]. Similarly, patients with diabetes appear to require higher doses and faster up-titration of pegvisomant to reduce and normalise IGF-1 levels compared to patients without diabetes [134, 167, 168]. Despite this, patients treated with pegvisomant may show improvement in glucose tolerance; therefore, glucose levels should be monitored more frequently during the first months of treatment, and antidiabetic drugs adjusted as needed [137, 169]. Additionally, it has also been showed that IGF-1 normalisation requires lower doses in male than in female patients, and in those who have undergone irradiation compared to non-irradiated patients [170].

When combined with SRLs, the required dose of pegvisomant may be lower than when used in monotherapy [171, 172]. During combined treatment, pegvisomant levels were shown to be increased because, while acting as competitive blocker, it has fewer endogenous GH molecules to compete with and a lower number of GH receptors to block [173]. When used as a monotherapy, a weekly mean dose of pegvisomant of 130 mg was needed to control IGF-1 in >90% of patients [137]. However, when combined with SRLs, the weekly necessary mean dose of pegvisomant to normalise IGF-1 dropped to 77–80 mg [137, 174]. A low-dose SRL regimen plus weekly pegvisomant (40–160 mg, provided as once or twice weekly injections) is a cost-effective option for patients requiring combination therapy [175, 176].

The long-term safety of pegvisomant was evaluated in several studies, demonstrating a good safety profile with few serious adverse events reported [136, 137, 161, 177]. The most frequent adverse events include injection site reactions, liver dysfunction, and changes in tumour size (shown in Table 2) [161].

Injection site reactions are typically mild and transient, presenting as localised erythema and soreness, and do not require treatment or drug discontinuation [136, 137]. Although rare (i.e., ∼2%), lipohypertrophy at the injection site may occur, but it usually resolves with a rotation of the injection site and does not require discontinuation of pegvisomant [161].

A small proportion of patients (i.e., 5–8%) treated with pegvisomant develop liver dysfunction; however, it is generally mild and transient, with drug-induced hepatitis being rarely reported [131, 178]. Transaminase elevations do not appear to be dose-related and generally occur during the first year of treatment [179]. The incidence of liver dysfunction is higher in certain groups (i.e., ∼14%), such as patients with prior liver impairment, diabetes, concomitant use of hepatotoxic drugs, or combined treatment with SRLs [174]. Therefore, ongoing monitoring of liver function is advised (e.g., liver function tests should be performed every 1–2 months for the first 6 months and thereafter every 6 months), and discontinuation of pegvisomant should be considered if transaminases exceed >2-fold the upper limit of normal [4, 180].

Pegvisomant does not have anti-proliferative effects on pituitary tumours and interrupts the negative feedback loop on GH secretion, raising theoretical concerns that somatotroph tumours may enlarge in patients receiving this drug [181]. Some studies have reported an increase in tumour size in patients treated with GHRA; however, there is no clear evidence supporting this association [4, 131]. A meta-analysis of observational longitudinal studies, involving 14 studies, reported tumour growth in ∼7% of cases [182]. Both the German Pegvisomant Observational Study (∼3–6%) and the ACROSTUDY (∼7%) found similar growth rates based on local MRI scans; however, these rates decreased to ∼3% in both studies after centralised image re-evaluation [161, 183]. Therefore, careful serial evaluation of MRI imaging by expert neuroradiologists is advised to establish tumour progression. Tumour growth appears to be more frequent during the first year of treatment with pegvisomant, which may reflect the natural history of the disease and/or the effect of withdrawal of prior SRL treatment [137, 183‒185]. Patients with higher tumour expression of GH and insulin receptors appear to be associated with a greater risk of tumour progression [183, 186]. Overall, pegvisomant is a safe second-line option for patients with active acromegaly, provided that no large tumour remnants are in contact with the optic chiasm or vital structures [4].

Combination therapy for acromegaly is increasingly being used and should be considered for patients who do not respond adequately to monotherapy [151]. Both cabergoline or pegvisomant can be added to SRLs, and the association of cabergoline with pegvisomant is also possible [4, 151]. The goal of combined medical therapy is to increase the treatment efficacy, reduce the adverse effects associated with monotherapy in high doses, and decrease the frequency of administrations and the dosage of a single drug [187].

The combination of first-generation SRLs with a GHRA is the most frequently used treatment regimen. The combination of pegvisomant 40–160 mg/week with SRLs resulted in normalisation of IGF-1 levels in 95–100% of patients [174, 188‒192]. The combination of cabergoline with SRLs may be useful in patients with mild uncontrolled disease, achieving IGF-1 normalisation in 42–56% of patients with this association [190]. However, better outcomes were reported in patients with lower basal IGF-1 levels [4, 46]. For patients intolerant to SRLs, the combination of cabergoline and pegvisomant can be useful, particularly in patients with mild elevations of IGF-1 [193]. In a randomised trial including 52 patients treated with low-dose monthly octreotide LAR (10 mg) or lanreotide Autogel (60 mg) combined with weekly pegvisomant (40–160 mg/week) showed that the biochemical control rate was achieved in 96% of previously controlled and uncontrolled patients, with only 30% of cases requiring up-titration of pegvisomant [176]. This combination (i.e., low-dose monthly SRLs combined with weekly pegvisomant) had lower cost compared to high-dose monthly SRLs combined with weekly pegvisomant or low-dose monthly SRLs combined with daily pegvisomant [176]. A combination of octreotide LAR or lanreotide Autogel with pegvisomant may be considered for patients requiring control of the residual tumour with impaired glucose tolerance or diabetes [79].

The combination of pasireotide LAR with pegvisomant was also investigated in the PAPE study [194]. This study included 61 patients well controlled on octreotide/lanreotide plus pegvisomant. Patients were switched to pasireotide LAR with or without pegvisomant, depending if IGF-1 levels were or not >1.2 times the upper limit of normal, respectively. The primary outcome was treatment efficacy, and the secondary outcome was pegvisomant dose reduction compared to baseline. IGF-1 levels remained normal in ∼93% of patients who transitioned to pasireotide LAR monotherapy. In patients with IGF-1 levels >1.2 upper limit of normal, ∼67% normalised IGF-1 levels (i.e., IGF-1 ≤1.2 upper limit of normal) when treated with combination therapy of pasireotide LAR and pegvisomant. Furthermore, the pegvisomant dose was reduced by ∼66% in these patients. However, the prevalence of diabetes increased from ∼33%, at baseline to ∼69% after 24 weeks in patients treated with pasireotide LAR [194]. Only 25% of patients were receiving antidiabetic medications at baseline, though after 24 weeks of pasireotide LAR 69% of patients required at least one antidiabetic medication [194]. No significant difference was observed in HbA1c levels between patients treated with pasireotide LAR monotherapy versus combination therapy with pegvisomant, suggesting that pegvisomant did not mitigate the pasireotide-induced hyperglycaemia [195]. The combination of pasireotide LAR and pegvisomant may be an option for patients with acromegaly and tumour growth when radiotherapy is contraindicated, not available, or while awaiting irradiation-induced tumour shrinkage [8].

The use of radiotherapy has decreased over the years due to advances in surgical techniques and medical treatments. However, it may still be considered a third-line therapy, following surgery and medical therapy, particularly when surgery is unsuccessful and there is resistance or intolerance to medical therapy [8, 196]. The benefits of radiotherapy include tumour volume control and decreased hormonal secretion; however, these advantages are hindered by the long latency of therapeutic effects and the high risk of long-term adverse events, such as hypopituitarism, radiation-induced optic neuropathy, cranial nerve deficit, brain necrosis, and secondary brain tumours [197, 198].

Conventional radiotherapy is administered by a linear accelerator (4–8 MeV) with a total dose of 40–45 Gy. It should be fractionated in at least 20 sessions [198]. This approach results in long-term tumour growth control in 80–100% of cases, with 60–80% achieving normalisation of GH/IGF-1 levels [8, 196].

The development of stereotactic methods has improved the precision of tumour targeting, while minimising damage to adjacent tissues. These methods include stereotactic radiosurgery (SRS) and fractionated stereotactic radiotherapy (FSRT) [198, 199]. SRS, which includes Gamma Knife, CyberKnife, and linear accelerator, is potentially more effective than conventional radiotherapy in inducing biochemical remission [198, 200‒203]. SRS is suggested for tumours of small to moderate size (<2.5–3 cm) [4]. If the tumour is located close to the optic apparatus, SRS can still be applied but the maximum dose delivered to the optic apparatus should be maintained below 8–10 Gy [199]. In clinical practice, SRS at doses of 16–25 Gy are a convenient approach for patients with relatively small tumours that are not in proximity (i.e., at least 2–4 mm) to the optic chiasm. For tumours with >2.5–3 cm and/or involving the optic chiasm, FSRT should be preferred [204, 205]. Long-term follow-up data show that around half of the patients with acromegaly treated with SRS or FSRT achieve and maintain biochemical control [96, 206]. However, up to one-third of patients with normal pituitary function before irradiation will develop hypopituitarism, underscoring the need for monitoring the pituitary function after radiotherapy. More studies are needed to evaluate the extent of long-term hypopituitarism and brain-sparing effects of these treatments [197, 207].

Acromegaly has been associated with decreased fertility through different mechanisms, such as hypopituitarism, hyperprolactinaemia due to stalk compression or related to a mixed GH and PRL-secreting tumour, and the direct effects of excessive levels of GH and IGF-1 in the hypothalamic-pituitary-gonadal axis [208]. However, some affected women can conceive, particularly those with a preserved hypothalamic-pituitary-gonadal axis and whose disease is controlled prior to pregnancy, as well as those who underwent assisted reproductive techniques [208].

Pregnancy appears to have a transient “therapeutic effect” on the clinical and biochemical activity of acromegaly, in part due to the increased oestrogen levels acting as GH antagonists in the liver, leading to a state of relative hepatic GH resistance [209]. This results in decreased maternal levels of serum IGF-1, particularly during the first trimester, although some individual variability is seen on IGF-1 levels [210]. During pregnancy, the placenta secretes a variant of GH which cross-reacts with pituitary-derived GH in most immunoassays [211]. Therefore, measuring serum GH and IGF-1 during pregnancy does not provide useful or reliable information, and it is not recommended [4, 212].

In fact, pregnancy in women with acromegaly is frequently associated with disease control and is generally safe in terms of foetal and maternal outcomes [213]. Most pregnant patients with acromegaly do not have an increase in tumour size during gestation (i.e., the overall frequency of tumour growth is estimated at 9%); nonetheless, clinical follow-up during pregnancy is recommended, and visual field testing should be performed in patients with macroadenomas, particularly those that extend superiorly to the optic chiasm [4, 213, 214]. In addition, non-contrast MRI should be considered in patients exhibiting visual changes [211].

The metabolic complications associated with the effects of GH on the mother, such as gestational diabetes or worsening of previous diabetes and hypertensive gestational disorders, appear to be only modestly increased in pregnant women with acromegaly (i.e., overall frequency described of 9% and 6%, respectively). Therefore, these conditions do not commonly adversely affect foetal growth or maternal outcomes in women with acromegaly [213‒215]. Furthemore, the overall frequency of premature labour, spontaneous miscarriage, small for gestational age, and congenital malformations is also very low [213].

Usually, there is no need to use medications to control GH hypersecretion or tumour size in patients with acromegaly during pregnancy, and pharmacological treatment can safely be interrupted for most patients [214, 216]. Drug therapy is generally recommended only for tumour growth control and moderate to severe resistant headaches. An exception is made for women with aggressive somatotroph tumours who manage to conceive, as they require close monitoring and individualised treatment approach [211].

There are limited data documenting safety of medical therapy during pregnancy, particularly regarding potential adverse effects on foetal development. Therefore, their liberal use is not recommended, and prompt discontinuation upon confirmation of pregnancy is advised [211]. Nevertheless, cumulative data suggest that there are no major adverse consequences/teratogenicity of first-generation SRLs administered during gestation for either the mother and the foetus, even during the first trimester, when embryogenesis occurs [209]. No foetal malformations have been reported in children born to women who were treated with long-acting SRLs at the time of conception [217, 218]. Octreotide is known to bind to somatostatin receptors in the placenta and cross the placenta, potentially affecting foetal tissues, particularly the brain [219]. The experience with pegvisomant is significantly more limited, and its safety profile remains unestablished. Nonetheless, data from several women exposed to pegvisomant at some point during pregnancy appear to suggest no adverse consequences on pregnancy outcomes [212, 218, 220]. Long-acting formulations of SRLs and pegvisomant should be avoided approximately 2 months before attempting to conceive and during pregnancy [4, 216]. During the conception period, if necessary, subcutaneous injections of short-acting octreotide may be preferable until pregnancy is confirmed [4]. However, some patients may choose to continue long-acting SRLs until the confirmation of pregnancy, which may also be acceptable given the safety data thus far available [212]. If the tumour increases in size, DAs, particularly cabergoline, can be used with demonstrated safety for the developing foetus. In cases where there is no response to medical therapy, TSS may be performed during the second trimester if necessary [221].

Although the paucity of data limits evidence-based recommendations for preconception counselling and pregnancy surveillance, it is highly recommended to control tumour size and hormonal activity before pregnancy to ensure better outcomes, and surgical intervention should be attempted before pregnancy, when indicated, particularly in women with a recent diagnosis of acromegaly seeking fertility [212]. This underscores the need for prenatal counselling and a comprehensive care with frequent visits during pregnancy for all women with acromegaly who become pregnant [213].

After the delivery, a rebound in acromegaly activity is often observed; thus, restarting the medical treatment precociously may be necessary. However, this decision needs to be balanced between the activity of the disease after delivery and the desire of the women to breastfeed [214, 222]. In principle, breastfeeding is possible and not contraindicated, but medical therapy (particularly SRLs and pegvisomant) should be avoided in nursing mothers [212].

Acromegaly is a rare, chronic, slowly progressive debilitating disorder caused by the hypersecretion of GH. If left untreated, it is associated with increased morbidity and mortality. The diagnosis is established by demonstrating elevated IGF-1 levels and/or autonomous GH hypersecretion and detection of a pituitary tumour. Patients should be managed by an experienced multidisciplinary team, including endocrinologists, neurosurgeons, neuroradiologists, radiotherapists, and pathologists. Pituitary surgery is the first-line therapy for most patients, although primary medical therapy may be considered in selected cases. In patients where remission is not achieved after first surgery, reintervention or medical therapy should be considered. SRLs, DAs, and GHRA are pharmacological options that should be tailored to the individual patient and clinical scenario. Radiotherapy is, generally, the third-line of therapy and should be considered in the more challenging cases with active disease. New pharmacological agents are needed to improve disease control and patient's quality of life.

A.P.S. has received travel grants from Novartis and Ipsen. C.A. has received travel grants from Recordati and Pfizer. D.C. has received travel grants, consultancy, and speaker fees from Recordati and Pfizer. O.M. has received travel grants and consultancy fees from Recordati. All other authors declare no competing interests.

The development of this consensus statement and open-access funding were supported by the Portuguese Society of Endocrinology, Diabetes and Metabolism (SPEDM). The authors thank to the SPEDM for the funding and logistical support provided for both in-person and virtual meetings required for the development of this consensus document. No other entity provided financial support.

L.M.C., P.M., C.A., and D.C. conceptualised and coordinated the development of the consensus. D.C., C.A., D.D., I.M., J.P., L.C., L.M.C., P.M., S.A., T.M., T.P., and V.B.S. summarised recent available literature leading to the consensus statements. A.G., C.A., D.C., F.F., I.M., J.S.D., J.P., L.G., L.M.C., M.J.B., M.O., O.M., P.M., S.A., T.M., T.P., and V.B.S. participated in the Delphi process. L.M.C., P.M., C.A., and D.C. wrote the full draft manuscript from the author’s contributions. L.M.C., P.M., M.T.P., A.A., R.A., S.A., M.J.B., M.B.C., L.C., D.B.D., J.S.D., F.F., L.G., I.M., O.M., T.M., M.J.O., I.P., J.P., A.P.S., V.B.S., I.T., C.A., and D.C. critically revised the initial draft, edited, and approved the final manuscript.