Pituitary gigantism is a rare and severe form of acromegaly (1). It is caused by chronic excessive growth hormone (GH) and insulin-like growth factor 1 (IGF-1) secretion due to a pituitary adenoma and/or hyperplasia that begins before the growth plates have fused. Unlike in adult acromegaly, where germline genetic causes of pituitary adenomas are rarely encountered, nearly 50% of pituitary gigantism patients have an identifiable genetic cause (1–3). The most frequent of these are germline pathogenic variants (mutations) in the aryl hydrocarbon receptor interacting protein (AIP) gene, which account for around 30% of cases (2). Patients with pituitary gigantism due to AIP mutations are typically males that present during adolescence with large GH-secreting macroadenomas, which often co-secrete prolactin and may occur in familial isolated pituitary adenoma (FIPA) kindreds (4). The next most common genetic cause of pituitary gigantism is X-linked acrogigantism (X-LAG) (2). This very rare disease has a distinctive presentation beginning in the first 3 years of life (5). Most cases present with mixed GH and prolactin-positive macroadenomas, but individuals with hyperplasia alone have been described (5–7). In X-LAG, the young patients can have markedly increased height and weight at diagnosis compared with their age-peers (8). X-LAG is caused by duplications on chromosome Xq26.3 that include the gene GPR101, which encodes an orphan G protein-coupled receptor (GPCR) (5). In patients with X-LAG, there is a very marked over-expression of GPR101 in pituitary adenomatous/hyperplastic tissue (9). Usually, sporadic, X-LAG can occur due to constitutional or somatic mosaicism (in male patients) for the Xq26.3 duplication; three FIPA kindreds with X-LAG have also been reported (5, 10–13). As GPR101 is a constitutively active receptor that drives GH and prolactin secretion from somatotropes, this leads to hormonal hypersecretion and gigantism (5, 14, 15). Genomic duplications in X-LAG lead to disruption of the local chromatin structure around GPR101, which is normally insulated from nearby regulatory elements in its own topologically associating domain (TAD) (16). The duplication-induced reshuffling of genomic sequences and a TAD border brings the promoter of GPR101 into contact with ectopic enhancers that form a neo-TAD and drive the over-expression of GPR101 that typifies X-LAG (16).

While advances have been made in understanding the pathophysiology of X-LAG, the rarity of the disease means that there is incomplete information about longer-term responses to different management options. As X-LAG occurs in very young children, the management of pituitary adenoma is complicated in terms of surgical approach and choice of drug options and doses (5, 6, 8, 13, 17–20). Based on existing patients, many of whom were retrospectively diagnosed with X-LAG after many years of disease, GH and IGF-1 hypersecretion is challenging to control with standard approaches. New information on the long-term management of newly diagnosed X-LAG patients can help to clarify outstanding questions. In this study, we describe the presentation, diagnosis, and management of a young female patient with X-LAG from disease onset as a toddler until the age of nearly 10 years.

Somatotropinomas very rarely occur in early childhood but can have an important impact on the patient due to difficulties in treatment (1). The current study describes the complicated treatment journey that can occur in patients with the earliest-onset form of somatotropinomas, X-LAG. X-LAG is a very rare disease with fewer than 50 cases having been described in the decade since it was first identified as a distinct entity (2, 5, 6, 8, 10, 17–20, 25–29). The current patient illustrates well many of the characteristic features of X-LAG, namely, sporadic onset in a previously normally-growing female infant, with the first signs of overgrowth occurring around the first birthday. Over-growth usually becomes dramatic and drives patients onto the diagnostic pathway. By the time of the first MRI, X-LAG patients as young as 2 to 3 years old already have significant pituitary macroadenomas. Evidence indicates that tumorigenesis in X-LAG is already established perinatally. One familial case showed that excess GH and prolactin hypersecretion from an enlarged anterior pituitary (>10 mm in largest diameter) was clearly demonstrable by 3 weeks of age (10). Our patient had an established pituitary mass of 17 × 12 mm at the age of 4 years, which was later found to be a pituitary adenoma at surgery.

Hormonal hypersecretion in the current case was also typical of that seen in X-LAG in that it involved marked GH/IGF-1 and prolactin excess. While prolactin normalized rapidly with modest doses of cabergoline, the response of GH and IGF-1 to somatostatin analogs was poor during the pre-operative period. This was despite using adult-appropriate doses of octreotide and lanreotide in a child of only 3 to 4 years of age. Such resistance occurred despite a moderate-to-high expression of the SSTR2 receptor in the excised tumoral tissue, a factor that has been seen in previously reported X-LAG patients (8, 18). Other genetic forms of acrogigantism also show relative resistance to octreotide/lanreotide (30). Somatostatin analog resistance in patients with AIP mutations can have a number of explanations, both related to and independently of SSTR2 expression. A low AIP expression in mutated tissue could itself alter tumor responses to somatostatin analogs—for example, Bogner et al. showed that AIP mutations led to increased miR-34a, which, in turn, altered the inhibition of GH by octreotide independently of SSTR2 levels (31). X-LAG-related somatostatin resistance is not as yet fully explained. The circulating GH levels in these patients can be very high (5, 8, 18, 32). This may reflect a strong pro-secretory phenotype of the affected somatotropes potentially driven by protein kinase A and C-linked pathways in mice and in pituitary tumors from X-LAG patients (14). One potential mechanism for poor clinical responses to somatostatin analogs in patients is that treatment is not capable of overcoming this very high GH secretion driven by over-expressed, constitutively active GPR101 (14). In somatotropinoma patients who are resistant to first-generation somatostatin analogs, one of the main options is switching to pasireotide, which targets a wider number of somatostatin receptors (33–36). In the case of resistant acromegaly due to AIP pathogenic variants, it can be a valuable option, but not universally so, and appears to depend on tumoral somatostatin receptor expression (37, 38). In X-LAG, there is limited experience with pasireotide use and no clear evidence of improved hormonal control (26). Adjuvant radiotherapy has been used in the management of resistant hormonal hypersecretion in X-LAG, but it has the down-side of slow onset of effect, which limits its usefulness in controlling overgrowth promptly during childhood.

In X-LAG, there are two main therapeutic options that can bring about the effective control of GH hypersecretion and over-growth. The aggressive resection of the anterior pituitary or anterior hypophysectomy has been used to halt GH excess in some cases, particularly those with widespread hyperplasia (5). This comes at the obvious cost of panhypopituitarism, often accompanied by arginine–vasopressin deficiency, although these can be successfully replaced over the long term in X-LAG patients (5, 8). The other effective option is the use of the GH receptor antagonist, pegvisomant, which controls IGF-1 and growth in X-LAG (5, 8, 20). As illustrated by this case, X-LAG is remarkable in that tiny or even non-visualized post-operative tumoral remnants are capable of causing long-term GH-IGF-1 excess. The source of excess GH secretion in this and other such cases is likely to be GPR101 over-expressing cells contained within the residual normal-appearing pituitary tissue. In X-LAG and other forms of resistant pituitary gigantism, pegvisomant can be used to bring about rapid hormonal control and slow overgrowth, either alone or in combination with other medical therapies (2, 18, 20, 39–42). In the current case, pegvisomant could be used only briefly and despite an interruption of drug supplies, it successfully lowered the IGF-1 levels to close to normal. Lipohypertrophy at injection sites in the abdomen and thighs developed within 6 months of starting the pegvisomant treatment. This is a recognized side effect of pegvisomant for acromegaly, and as in other cases, the subcutaneous adipose tissue overgrowth regressed following drug withdrawal (43). As there was no overt tumor to target for re-operation or radiotherapy, this left very limited options. Cabergoline was reintroduced to lower prolactin, which was again effective. Preoperative treatment with relatively low doses of first-generation somatostatin analogs had been unsuccessful, so we hypothesized that tumor debulking (as in acromegaly) could permit hormonal control if we reintroduced octreotide LAR and dose-titrated upwards (44). The current study shows that it is possible to use post-operative, first-generation somatostatin analogs at adult doses to suppress GH excess from a non-visualized tumor remnant in X-LAG. Titration to 30 mg per month of octreotide LAR led to the control of circulating IGF-1, and the growth curve is now trending in the normal direction. As with other X-LAG cases, no tumor regrowth occurred following surgical resection (1, 5).

Pathological studies showed a Pit-1-positive mixed somatotrope–lactotrope adenoma or PitNET. Densely granulated, GH-positive acidophils predominated (80%) and were mixed with two sub-populations of sparsely granulated chromophobes—one was positive for prolactin and the other for GH. These findings are consistent with the pathological characteristics of X-LAG tumors (5, 8, 18). Adenomas are rarely accompanied by somatotrope hyperplasia, and a few cases have hyperplasia alone; no hyperplasia was seen in the tumor from the current patient (5, 8, 18). The tumor had a low Ki67 percentage, which is in line with most tumors from X-LAG patients, although a higher Ki67 level has been reported in one X-LAG family (5, 8, 10, 17, 18).

Initial genetic studies using aCGH confirmed that the patient had a 656.5-kb duplication on chromosome Xq26.3; neither of her parents carried this abnormality. The duplication included the last exon of VGLL1 and the entire CD40LG, ARHGEF6, RBMX, and GPR101 genes. We recently identified that duplications in this region cause X-LAG by interfering with a demarcated section of DNA called a topologically associating domain (TAD) (16). A TAD is a mega-base-scale genomic region within which interactions occur between genes and their regulators (enhancers and transcription factors) with higher likelihood than with factors outside of the TAD. This organization is facilitated by invariant boundaries that include binding sites for specific elements such as the transcriptional regulator CTCF, a key component of cohesin-mediated DNA looping and gene transcription. GPR101 is normally contained within its own TAD and insulated from nearby genes and enhancers. However, duplications like those in X-LAG remove one of the invariant borders and create a neo-TAD that places the GPR101 promoter under the control of ectopic pituitary enhancers located close to VGLL1, ARGHEF6, and RBMX. This drives the over-expression (>1,000-fold) of GPR101 in somatotropes which is responsible for chronic GH, IGF-1, and prolactin hypersecretion that characterizes X-LAG. The study of leukocytes from the current patient demonstrated that her unique duplication led to a neo-TAD formation and ectopic enhancer interactions with the GPR101 promoter (16).

In conclusion, this case illustrates the clinical presentation and complicated therapeutic journey in a sporadic female patient with infant-onset pituitary gigantism due to X-LAG. X-LAG is a very rare disorder due to GPR101 over-expression that is driven by ectopic enhancers in a neo-TAD. This leads to a tightly choreographed presentation of very early-onset GH and prolactin-secreting adenomas that are resistant to treatment with preoperative first-generation somatostatin analogs. Disease persistence after gross total resection of the adenoma was eventually overcome with adult doses of octreotide, whereas pegvisomant had to be withdrawn due to lipohypertrophy. The combination of tumor debulking and adult doses of somatostatin analogs post-operatively represents a practical management option in pediatric patients with X-LAG.

The datasets presented in this study can be found in the GEO database under the accession code GSE249129.

The studies involving humans were approved by Liege University—CHU Liège Faculty Ethics Committee. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants’ legal guardians/next of kin. Written informed consent was obtained from the minor’s legal guardian/next of kin for the publication of any potentially identifiable images or data included in this article.

MCar: Formal analysis, Investigation, Supervision, Writing – review & editing. DM: Investigation, Methodology, Writing – review & editing. SA: Data curation, Investigation, Methodology, Writing – original draft. GC: Data curation, Investigation, Writing – review & editing. GT: Data curation, Investigation, Writing – review & editing, Formal analysis. MF: Formal analysis, Investigation, Methodology, Writing – original draft. DA: Investigation, Writing – review & editing. JH: Investigation, Writing – review & editing. VR: Investigation, Writing – original draft. PP: Writing – review & editing, Supervision. AB: Investigation, Writing – review & editing. MCap: Investigation, Writing – review & editing. AD: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Supervision, Writing – original draft, Writing – review & editing.