Factors affecting disease control after pituitary tumor resection in acromegaly: What is the current evidence?

Abstract

Acromegaly, characterized by persistent hypersecretion of growth hormone (GH), is most often caused by a pituitary neuroendocrine tumor (PitNET), though, less often, ectopic GH or GH-releasing hormone secretion from various neoplasms outside the pituitary gland could cause it. Nearly 70% of somatotroph PitNETs are macroadenomas at diagnosis. Transsphenoidal surgery, the most effective treatment modality for acromegaly, could achieve remission in 73%. However, the remission rates could reach 87% if surgery is followed by medical therapy. Due to variable therapeutic responses to surgical and medical therapy, pre-treatment awareness regarding the best therapeutic modality based on clinical, biochemical, radiological, histopathological and genetic parameters would help in accurate pretreatment decision-making. Earlier studies have identified poor prognosis markers like tumor size, tumor invasion, T2-weighted hyperintensity, granulation, and pretreatment GH and/or insulin-like growth factor 1 levels. In a recent study, published by Alvarez et al identified that preoperative PitNET volume is a good predictor of control of acromegaly following surgical treatment and the likelihood of requiring more aggressive additional therapies after surgery. They found that PitNET volume exceeding 3697 mm³ was associated with poorer disease control in patients with somatotroph PitNETs.

Key Words: Acromegaly; Pituitary neuroendocrine tumor; Somatostatin receptor; Somatostatin receptor ligands; Transsphenoidal surgery

Core Tip: Transsphenoidal surgery is the most effective treatment for acromegaly. Alone, it could achieve a remission rate of 73%. However, when followed by medical therapy, the remission rates could reach 87%. The biochemical remission rates following surgery and medical therapy are highly variable depending on various clinical, biochemical, radiological, histopathological and genetic parameters. Earlier studies have identified tumor size, tumor invasion, T2-weighted hyperintensity, granulation, and pretreatment growth hormone and/or insulin-like growth factor 1 levels as important prognostic markers. A study published by Alvarez et al noted that preoperative pituitary neuroendocrine tumor volume could predict postoperative remission.

INTRODUCTION

Acromegaly is an uncommon chronic multisystem disease, characterized by persistent increased secretion of growth hormone (GH) leading to a progressive illness with excessive tissue growth, development of physical deformities and multiple comorbidities with an impaired quality of life[1]. It is most often (> 99%) caused by a pituitary neuroendocrine tumor (PitNET) and less often by GH or GH-releasing hormone (GHRH), or both secreted by neoplasms of extra-pituitary origin[2]. Ectopic GHRH secretion from various tumors (hypothalamic tumors, neuroendocrine tumors originating from the lung, pancreas, thymus, and intestine, medullary thyroid cancer, small-cell lung cancer, or phaeochromocytoma) results in pituitary somatotroph hyperplasia, presenting as an enlarged pituitary gland on magnetic resonance imaging (MRI), often resembling a PitNET. On the other hand, the extremely rare ectopic GH secretion from pancreatic neuroendocrine tumors or lymphoma results in a normal or small-sized pituitary gland on MRI[2]. GH promotes the synthesis of insulin-like growth factor 1 (IGF-1), the key mediator of GH effects, primarily from the liver[1]. GH secretion is stimulated by GHRH and is inhibited by somatostatin signaling through the somatostatin receptor subtypes 2 (SSTR2) and 5 (SSTR5)[1].

THE GENETIC BACKGROUND OF ACROMEGALY

Acromegaly is most commonly diagnosed in the fourth to fifth decades of life, affecting both men and women, though it is slightly more common in women[1]. Approximately 70% of somatotroph PitNETs are macroadenomas at diagnosis[2]. Although most somatotroph PitNETs are sporadic, a recognizable genetic defect has been identified in approximately 46% to 49% of cases[3]. The familial somatotroph PitNETs can present as familial isolated pituitary adenomas either from loss-of-function mutations in the AIP gene or from a gain-of-function duplication in the GPR101 gene, the latter causing X-linked acrogigantism. The familial somatotroph PitNETs can also be a component of known syndromes such as multiple endocrine neoplasia types 1 or type 4 (MEN1 and CDKN1B), McCune-Albright syndrome (GNAS), Carney complex (PRKAR1A), or the pheochromocytoma/paraganglioma-pituitary adenoma association, also known as the 3P association (SDHx,MAX, VHL, MEN1, and RET)[3].

ACROMEGALY: CLINICAL FEATURES AND COMORBIDITIES

Although somatotroph PitNETs are almost always benign, they can be locally invasive and continue to grow progressively despite treatment[2]. Persistent GH and IGF-1 excess results in various acromegaly-related comorbidities including obstructive sleep apnoea (60%-80%), osteoarthritis (30%-70%), median nerve entrapment neuropathy (40%-60%), spinal disorders (40%-50%), prediabetes (16%-46%), diabetes mellitus (20%-56%), systemic hypertension (20%-50%), colonic polyps (up to 45%), and malignant tumors (10%-23%)[1]. The treatment of acromegaly aims to normalize GH and IGF-1 levels, reduce tumor size, alleviate symptoms, manage complications, lower excess morbidity, and enhance the quality of life.

Patients presenting with classical acromegalic features of the face and extremities (present in 80% of patients[2]), with several acromegaly-related comorbidities, with a pituitary mass even in the absence of typical acromegaly clinical features, and with an exponential linear growth or unusually tall stature, especially when height is at or above two standard deviations in comparison to the mid-parental height, needs evaluation for acromegaly[1]. Those with non-classical features, such as hyperhidrosis (excessive sweating), pachydermia (thickening of skin), and cutis verticis gyrata (deep folds and furrows in the skin of the scalp) may also be offered an evaluation.

BIOCHEMICAL DIAGNOSIS OF ACROMEGALY: WHAT IS NEW ABOUT IT?

The biochemical diagnosis of acromegaly is established by detecting elevated serum IGF-1 levels and confirming the failure to suppress GH levels following an oral glucose tolerance test[2]. The glucose ingestion typically leads to a decrease in GH levels through an increase in somatostatin and a decrease in GHRH levels.

The secretion of GH is pulsatile, has significant circadian variation, and has a half-life of only 14 minutes. Hence, random GH measurements are of limited clinical value[2]. On the other hand, IGF-1 has no significant circadian variation and has a half-life of 15 hours[2]. Hence, IGF-1 can be used for biochemical screening for acromegaly as a marker of GH secretion over the last 24 hours[1], provided age-specific, and if available, sex-specific reference ranges are used. It is important to remember that the IGF-1 levels can be falsely low in those with uncontrolled diabetes mellitus, advanced hepatic or renal disease, anorexia, malnutrition, severe hypothyroidism, those on oral estrogen or selective estrogen receptor modulators, and the IGF-1 levels can be falsely high in pregnancy, late adolescence and hyperthyroidism[2].

As per a latest consensus, in patients with classical acromegalic features, a level of IGF-1 exceeding 1.3-fold the upper limit of normal for age can diagnose acromegaly without further confirmatory tests[4]. The consensus also suggests that for patients with equivocal results, the options are either to repeat the IGF-1 using the same validated assay or to arrange an oral glucose tolerance test using a 75-gram glucose load, with failure to suppress the GH nadir level to below 0.4 μg/L for body mass index less than 25 kg/m² or below 0.2 μg/L for body mass index at or above 25 kg/m² confirming the diagnosis of acromegaly[4]. Nearly 25% to 30% of acromegaly patients may show a paradoxical rise in GH with the glucose load, potentially resulting from the glucose-dependent insulinotropic polypeptide secreted by the duodenal K cells acting on its receptors on the somatotrophs of PitNETs[5].

RADIOLOGICAL DIAGNOSIS OF ACROMEGALY

After confirmation of biochemical confirmation of acromegaly, a MRI pituitary should be undertaken using a 1.5 Tesla or 3 Tesla scanner, with 2 mm slices. MRI of high quality is essential for accurate diagnosis, prognostication, and therapy planning. The standard sellar MRI protocol includes doing a coronal and sagittal T1-weighted spin-echo sequencing with and without administering a gadolinium-based contrast and doing a coronal T2-weighted fast-spin echo sequencing[6]. A standard MRI report should accurately describe the location of the PitNET (intrasellar, infrasellar, suprasellar, parasellar, or extrasellar) in addition to the size, and it should describe the presence or absence of cysts or haemorrhage within the PitNET, the T2-weighted signal intensity of the PitNET, proximity to the optic apparatus, its relationship with the sphenoid sinus and the cavernous sinus invasion, the latter using the modified Knosp classification. Though this standard protocol is enough for most patients, some advanced techniques, including diffusion-weighted imaging, susceptibility imaging, high-resolution 3D T2-weighted imaging, magnetic resonance elastography, and perfusion-weighted imaging, may be needed for some[6].

CORRELATION BETWEEN RADIOLOGICAL FEATURES AND RESPONSE TO TREATMENT

All somatotroph PitNETs with a diameter below 20 mm are associated with lower acromegaly-related morbidities and better treatment response, whereas all somatotroph PitNETs with a diameter ≥ 20 mm, regardless of the actual size, exhibited almost similar, but worse clinical features, treatment response, and prognosis[7]. Based on the signal intensity on T2-weighted MRI sequences, the somatotroph PitNETs can be hypointense, isointense, or hyperintense. The hypointense somatotroph PitNETs are smaller and less invasive, with the presence of densely granulated somatotroph adenomas in histopathology, with a higher expression of SSTR2[8,9]. On the other hand, hyperintense somatotroph PitNETs are larger and more invasive, with sparsely granulated somatotroph adenoma in histopathology with a higher expression of SSTR5[8,9].

A retrospective analysis of MRI T2-weighted signal intensity of 70 patients with somatotroph PitNETs for whom histopathology results are available confirmed that 86.1% of the densely granulated somatotroph adenomas appeared hypointense, whereas 38.2% of the sparsely granulated somatotroph adenomas appeared hyperintense on T2-weighted images[10]. Moreover, the sparsely granulated somatotroph adenomas showed a 4.649-fold contrast enhancement on MRI compared to the densely granulated somatotroph adenomas[10].

The hypointense somatotroph PitNETs have a greater reduction in GH and IGF-1 following treatment with the first-generation somatostatin receptor ligands (SRLs), including octreotide and lanreotide, thereby achieving a superior therapeutic response to first-generation SRLs with an 80% accuracy to predict > 80% reduction in GH levels[11-13]. On the other hand, the hyperintense somatotroph PitNETs achieve only an insufficient therapy response to first-generation SRLs. A multicentre international study that compared the T2-weighted signal intensity to GH reduction, IGF-1 reduction and tumor volume reduction observed that the hypointense somatotroph PitNETs achieved a median random GH and IGF-1 reduction and tumor shrinkage of 88%, 59%, and 38% respectively, whereas the corresponding percentages for the hyperintense somatotroph PitNETs were 36%, 33%, and 3%, and that for the isointense somatotroph PitNETs were 24%, 20%, and 8%, respectively[13]. In contrast to first-generation SRLs, both hypointense and hyperintense somatotroph PitNETs achieved equal biochemical response and tumor shrinkage to the second-generation SRL pasireotide[14-16].

A recently published systematic review and meta-analysis of twelve real-world trials comprising 409 patients on pasireotide long-acting release observed a 57.9% IGF-1 reduction and 33.3% tumor shrinkage[17]. Surprisingly, only a 30% IGF-1 reduction was observed in the randomized controlled trials. The meta-analysis reminded us that the selection of appropriate patients with higher potential to respond to pasireotide compared to the first-generation SRLs (younger patients, with hyperintense somatotroph PitNET on T2-weighted MRI, with sparsely granulated somatotroph adenoma in histology, and with raised SSTR5/SSTR2 ratio) is vital[17].

Endoscopic transsphenoidal surgery of the somatotroph PitNETs is the safest and most effective treatment for acromegaly. By itself, surgery could achieve remission in 73%, with the rates increasing up to 87% if surgery is followed by medical therapy[18]. The larger the somatotroph PitNET, the higher the pre-operative GH and IGF-1 levels. The extent of resection depends on the size of the somatotroph PitNET and the cavernous sinus invasion. The higher the Knosp score, the lower the extent of resection. Patients with small and less invasive somatotroph PitNETs that facilitate gross-total resection, those associated with lower preoperative IGF-1, and lower first postoperative day GH levels are associated with higher postoperative biochemical remission[18]. On the other hand, large somatotroph PitNETs with extension outside the sella, invasion of the cavernous sinus, raised GH levels preoperatively, and operated at low volume turnover surgical centres are associated with poorer postoperative biochemical remission[19,20].

As the somatotroph PitNETs are associated with variable therapeutic responses to surgical and medical therapy, pre-treatment awareness regarding the potential best therapeutic modality would help the clinician in accurate decision-making. Older age, female gender, PitNET volume ≤ 1.11 cm³, low baseline GH and IGF-1 levels at diagnosis, hypointensity on T2-weighted MRI, and densely granulated somatotroph adenoma in histopathology are the most important noninvasive markers indicating a high likelihood of biochemical remission to first-generation SRLs[21]. Additionally, the Ki-67 index < 3%, higher SSTR2 expression, and higher E-cadherin expression on immunohistochemistry are also associated with better biochemical responses to the first-generation SRLs[22]. On the other hand, younger age, hyperintense somatotroph PitNET on T2-weighted MRI, sparsely granulated somatotroph adenoma in histopathology, and raised SSTR5/SSTR2 ratio are associated with better biochemical responses to the second-generation SRL[17].

The non-invasive preoperative markers of postoperative biochemical remission in patients with somatotroph PitNET include the tumor size, cavernous sinus invasion, and preoperative IGF-1 (as well as GH) levels[19,23]. A recently published retrospective study observed that 95% of patients who achieved postoperative biochemical remission had preoperative 3D tumor volume below 1.51 cm³[23]. Familial isolated pituitary adenoma patients with loss-of-function mutations in the AIP gene or gain-of-function duplication in the GPR101 gene have early onset, larger, and more aggressive tumors with higher GH secretion that exhibit lower treatment response to first-generation SRLs but respond well to pasireotide[24,25]. If these mutations are identified, it can guide to personalized treatment like more aggressive surgery and early consideration of radiotherapy. Certain instruments/tools like SAGIT or ACRODAT are used to guide the decision-making process[26,27]. The following Table 1 summarizes the clinical, radiological, histopathological, and genetic markers associated with biochemical remission in patients with somatotroph PitNETs.

Table 1 The various clinical, radiological, histopathological, and genetic markers associated with biochemical remission in patients with somatotroph pituitary neuroendocrine tumors.

	Good biochemical remission	Poor biochemical remission
Age	Older age	Younger age
Gender	Female	Male
Preoperative GH levels	Low	High
Preoperative IGF-1 levels	Low	High
Short acute octreotide test	2-hour GH < 2.7 ng/mL	2-hour GH > 2.7 ng/mL
Tumor size	< 20 mm	≥ 20 mm
Cavernous sinus invasion	Less invasion	More invasion
T2-weighted signal intensity	Hypointensity	Iso or hyperintensity
Contrast enhancement	Less marked	More marked
SSTR2 vs SSTR5	Raised SSTR2:SSTR5 ratio	Raised SSTR5:SSTR2 ratio
Granulations	Densely granulated	Sparsely granulated
E-cadherin expression	High	Low
Ki-67 index	< 3%	≥ 3%
First-generation SRL response	Complete response	Incomplete response
AIP mutations	No	Yes
GPR101 mutation	No	Yes

GH: Growth hormone; IGF-1: Insulin-like growth factor 1; SSTR: Somatostatin receptor; SRL: Somatostatin receptor ligand.

A recent study published in the World Journal of Radiology by Alvarez et al[28] identified that preoperative tumor volume is a good predictor of control of acromegaly following surgical treatment and the likelihood of using more aggressive additional therapies in a cohort of 77 patients from two centres in Colombia. In line with the previous observations from multiple studies[19-21], the authors also identified that higher pre-operative tumor volume is an important predictor of uncontrolled disease in acromegaly. They found that a tumor volume exceeding 3697 mm³ was associated with poorer disease control in patients with acromegaly from PitNETs requiring third- or fourth-line therapeutic options. They also identified that the longer duration of acromegaly was associated with a better surgical response in disease control, presumably due to less aggressive disease among these patients, resulting from slow, indolent tumours.

However, the authors couldn’t identify other risk factors mentioned in Table 1 above for poorer surgical cure, likely because of the small number of cases analyzed in their cohort and the retrospective nature of the study. This limits the generalizability of the findings. Moreover, confounding factors such as the volume turnover of PitNET surgeries, tumor characteristics such as Ki-67 index, preoperative hormone levels, invasion of the cavernous sinus, genetics studies, and histological features that could have impacted surgical response were not analyzed in a multivariate analysis, reducing the robustness of the conclusions. Despite these deficiencies, the study sheds some light on our existing knowledge, reinforcing the fact that preoperative tumor volume is an important predictor of disease control in acromegaly. Figure 1 below shows an algorithm for the management of a PitNET causing acromegaly.

Figure 1 A therapeutic algorithm for management of somatotroph pituitary neuroendocrine tumor causing acromegaly. PitNET: Pituitary neuroendocrine tumor; TSS: Transsphenoidal surgery; SRL: Somatostatin receptor ligand; IGF-1: Insulin-like growth factor 1.

Here are some recommendations to improve the robustness of the conclusions regarding various factors that could predict biochemical remission after treatment of acromegaly. Conduct prospective studies with larger cohorts to validate the predictive value of tumor volume and other factors. This would provide more robust evidence for clinical decision-making. Perform comprehensive multivariate analyses to account for multiple factors simultaneously. This would help in identifying the most significant predictors of treatment response and remission. Encourage the use of advanced imaging techniques (e.g., diffusion-weighted imaging, magnetic resonance elastography) and genetic testing in routine practice. This could improve the accuracy of preoperative assessments. Include long-term follow-up data to assess the durability of treatment effects and the impact on quality of life. This would provide a more complete picture of disease management. Develop standardized protocols for the diagnosis and management of acromegaly to reduce variability in clinical practice. This could involve the use of tools like SAGIT or ACRODAT to guide decision-making.

CONCLUSION

The safest and most effective treatment for acromegaly is surgery. However, the biochemical remission rates after surgery are variable based on various parameters like tumor size, tumor invasion, preoperative GH and/or IGF-1 levels and T2-weighted hyperintensity. Hence, there is variability in their applicability and reliability across different studies. For those with postoperative persistent disease, the surgery would be followed by medical therapy, mostly with one of the first-generation SRLs. Analogous to surgery, the biochemical remission rates after first-generation SRLs are variable based on various additional parameters, including granulation, Ki-67 index, SSTR2 expression, E-cadherin expression, mutation status, etc. A recent study published in the World Journal of Radiology by Alvarez et al[28] identified that preoperative PitNET volume is a good predictor of postoperative biochemical remission and of the likelihood of requiring more aggressive additional medical therapies after surgery.

ACKNOWLEDGEMENTS

We are thankful to Dr. Marina George Kudiyirickal, MSc, MJDF-RCS, PhD, for providing us with the audio core tip of this article.