Abstract
Summary
Pituitary gigantism is a rare pediatric disorder caused by excess growth hormone (GH) secretion. In almost 50% of cases, a genetic cause can be identified, with pathogenic variants in the aryl hydrocarbon receptor-interacting protein (AIP) gene being the most common. We present a case of an 11-year-old boy who exhibited progressive vision loss, associated with accelerated linear growth, and weight gain. On physical examination, he had enlarged hands, right eye amaurosis, and was already above his target height. Increased GH and IGF-I concentrations confirmed the diagnosis of pituitary gigantism. Magnetic resonance imaging showed a giant sellar lesion with supra- and para-sellar extensions. He underwent two surgeries which did not achieve a cure or visual improvement. Histopathological analysis revealed a sparsely granulated tumor, negative for somatostatin receptor type 2 (SST2) and an immunoreactivity score of 6 for somatostatin receptor type 5 (SST5). Our published artificial intelligence prediction model predicted an 83% chance of not responding to first-generation somatostatin receptor ligands. Pasireotide was therefore prescribed, and afterward cabergoline was added on. IGF-I concentrations decreased but did not normalize. We discovered a novel germline single nucleotide variant in the splicing donor region of intron 2 of the AIP gene (NM_003977.4:c.279+1 G>A), classified as likely pathogenic according to the American College of Medical Genetics and Genomics guidelines.
Learning points
-
Young-onset AIP-related pituitary tumors are associated with invasive, large, and treatment-resistant somatotroph tumors.
-
Intronic splice variants can lead to truncated proteins, possibly explaining the clinical findings.
-
Pasireotide biochemical response depends upon the presence of SST (SST2 and SST5) and possibly post-receptor mechanisms, such as AIP.
Background
Gigantism is a rare pediatric disorder caused by excess growth hormone (GH). In almost 50% of patients, a genetic cause can be identified, mainly familial isolated pituitary adenoma caused by the aryl hydrocarbon receptor-interacting protein (AIP) variants, and X-linked acro-gigantism (X-LAG) or McCune Albright syndrome (Rostomyan et al. 2015). Over 100 germline variants have been reported in the AIP gene, with less than 20 splice site variants described (Caimari et al. 2018). We report a novel germline splice-site variant in AIP in a young boy with gigantism.
Case presentation
An 11-year-old boy presented with progressive vision loss for at least 8 months, accelerated linear growth, and weight gain. On physical examination, he was above his target height (154.5 cm (p85-97), target height of 171.7 cm/p23.6), weight 66 kg (p10-p25), with enlarged hands, Tanner stage G2P3 (Fig. 1).
Physical findings. (A) Photograph of the patient’s enlarged left hand compared to Dr Lamback’s hand (adult female) that measures 17 cm (arrow). (B) Growth chart showing the patient’s height (dot) and target height.
Citation: Endocrine Oncology 4, 1; 10.1530/EO-24-0003
Increased GH of 32.7 ng/mL and insulin-like growth factor I (IGF-I) concentrations of 901 ng/mL (reference range 69–316) confirmed the diagnosis of gigantism. A visual field revealed right eye amaurosis and loss of superior quadrants in the left eye. Sellar magnetic resonance imaging showed a large and invasive tumor (Fig. 2).
Magnetic resonance imaging. Sagittal T1 post-contrast (A), coronal T1 post-contrast (B), and coronal T2 (C) imaging prior to surgery demonstrate a 5.6 × 4.7 × 5.8 cm (transverse × anterior-posterior × craniocaudal) invasive sellar lesion (Knosp 4) with suprasellar extension, compressing the optic chiasm, isointense on T2 and with heterogeneous contrast enhancement, suggestive of cystic or necrotic degeneration. Sagittal T1 post-contrast (D), coronal T1 post-contrast (E), and coronal T2 (F) imaging after the two surgeries exhibit partial tumor resection. Sagittal T1 post-contrast (G), coronal T1 post-contrast (H), and coronal T2 (I) after 5 months of pasireotide and 2 months of cabergoline show no change in signal intensities or tumor shrinkage. The fat graft in the middle of the tumor reabsorbed partially.
Citation: Endocrine Oncology 4, 1; 10.1530/EO-24-0003
Investigation
All exons and splicing sites of AIP were sequenced using Sanger sequencing in blood leukocytes. We encountered a novel single nucleotide variant (SNV) in the splicing donor region of intron 2: NM_003977.4:c.279+1 G>A (Fig. 3). Sequencing of the patient’s tumor DNA showed loss of heterozygosity (LOH).
Germline AIP sequencing. (A) Reference sequence NG_008969. (B) The patient’s blood sample sequencing exhibiting a single nucleotide variant in the splicing donor region of intron 2 with a nucleotide substitution of G>A in position 279 + 1. (C) The mother’s sequencing demonstrating the same variant. (D) The father was wild type. (E) The patient’s tumor sample sequencing demonstrating LOH.
Citation: Endocrine Oncology 4, 1; 10.1530/EO-24-0003
The parents’ sequencing showed that the 48-year-old asymptomatic mother also had the same variant. The patient’s mother has normal IGF-I and monomeric prolactin serum concentrations. The family does not have any known history of pituitary disease, tall stature, or infertility. The patient has no siblings from the mother’s part of the family.
Using FATHMM-XF (Rogers et al. 2018) and Eigen bioinformatics tools, the variant was considered pathogenic (non-coding scores of 0.991 and 0.913, respectively). Considering the American College of Medical Genetics and Genomics (ACMG) guidelines, it is considered likely pathogenic (one very strong and one moderate pathogenic criterion (PVS1 and PM2)).
We performed in silico prediction using SpliceAI (Jaganathan et al. 2019). The SNV results in the loss of the canonical splice-donor site of exon 2 (SpliceAI ΔScore = −0.99) and can possibly activate one of two nearby cryptic splice sites (CSS) (Fig. 4). The first CSS occurs at position NM_003977.4:c.279+11 (ΔScore = 0.27) and leads to an insertion of 11 bases and a frameshift. The second CSS occurs at position NM_003977.4:c.279+24 (ΔScore = 0.28) and results in an insertion of 24 bases from intron 2 (8 amino acids inserted in the final protein). A ΔScore of 0.20–0.35 has a validation rate of 20–40%, and a ΔScore 0.80–1.00 has a validation rate >80%, indicating that changes with higher scores are more likely to occur and cause impact.
Effects of variant NM_003977.4:c.279+1 G>A on splice-donor region. The presence of the variant results in the loss of the canonical splice site (red column, SpliceAI ΔScore = −0.99) and increases the chance of activating two possible cryptic splice sites (green columns, ΔScore = 0.27 and ΔScore = 0.28).
Citation: Endocrine Oncology 4, 1; 10.1530/EO-24-0003
Polymerase chain reaction from complementary DNA of the AIP RNA from the tumor sample was performed and showed several different sequences: i) we were still able to identify an AIP mRNA equivalent to the NM_003977.4 reference sequence; ii) we confirmed the predicted NM_003977.4:c.279+24 CSS; iii) we identified a 2-base insertion from intron 2 that results in a frameshift and would produce a truncated protein; iv) a smaller sequence that is more expressed than all others and lacks exon 2 and 3; and v) a sequence that lacks exon 2 (Supplementary Materials, see the section on supplementary materials given at the end of this article).
Treatment
He underwent transcranial and transsphenoidal surgeries which did not achieve a cure or visual improvement. Gross total resection was not expected as the patient had a very invasive tumor (Knosp 4). Unfortunately, not even vision was improved. The histopathological report is shown in Fig. 5. The diagnosis of a sparsely granulated tumor was made, with negative SST2, moderate expression in around 30% for SST3 (immunoreactive score; IRS 4), and moderate expression in around 60% for SST5 (IRS 6). Prolactin was positive in only sparse cells (around 3%) and was considered negative. This percentage of positive cells was not sufficient to consider the adenoma as a co-secreting tumor (Dottermusch et al. 2024). Our artificial intelligence model predicted an 83% chance of not responding to first-generation somatostatin receptor ligand (fg-SRL) (Wildemberg et al. 2021). Intramuscular pasireotide was started at a dose of 60 mg every 28 days, with IGF-I concentrations falling >20%, but not normalizing (before: GH 5.0 ng/mL, IGF-I 736ng/mL (69-316); 4 months after pasireotide: GH 12.9 ng/mL, IGF-I 575ng/mL (143-506)). Oral cabergoline was added at a dose of 0.5 mg 3x/week, without IGF-I normalization (1 month after cabergoline: GH 12.3 ng/mL; IGF-I 595ng/mL (143-506)).
Main histological findings: (A) Hematoxylin and eosin-stained section showing monomorphic tumor cells, displaying round to oval nuclei with conspicuous nucleoli and eosinophilic cytoplasm. A mitotic figure is shown in the insert in the upper right of the panel (up to three mitoses were found in ten 0.237 square mm high-power fields). (B) Immunohistochemistry for growth hormone depicts cytoplasmic positivity in part of tumor cells. (C) CAM5.2 immunostaining highlights cytoplasmic fibrous bodies in tumor cells, characterizing a sparsely granulated tumor. (D) The proliferation index assessed semi-quantitatively (Ki-67 immunostaining; MIB-1 clone) was around 4% overall. (E) Somatostatin receptor type 2 (SST2; UMB1 clone) immunostaining shows no expression in tumor cells. (F) Somatostatin receptor type 5 (SST5; UMB4 clone) immunostaining shows moderate expression in around 60% of tumor cells, corresponding to an immunoreactivity score (IRS) of 6 (reference range: 0–12) (Gatto et al. 2013). Scale bars equal 30 micrometers in A and B, and 60 micrometers in D to F.
Citation: Endocrine Oncology 4, 1; 10.1530/EO-24-0003
Outcome and follow-up
The patient developed pre-diabetes with pasireotide (glucose 121 mg/dL, glycated hemoglobin A1C 6.1%). He gained 9.0 cm in the last 13 months (present height 163.5 cm at the age of 12 years and 5 months – Fig. 1) and the tumor did not exhibit signal intensity change or shrinkage – Fig. 2. The patient is undergoing radiotherapy and will receive pegvisomant. His treatment will include pasireotide combined with pegvisomant.
Discussion
The AIP variant reported has no registry in the Clinical Genome Resource or ClinVar database. SNV in splice sites can cause different effects on mRNA, and in our case, we could predict two possible outcomes: a frameshift that would result in a stop codon in codon 159 at the end of exon 3 and a truncated protein, or an insertion of 24 bases. Splice variants leading to truncating variations in AIP have been considered disease-causing (Boguslawska & Korbonits 2021). Also, since the predicted scores of the new splice-donor sites were weak (<0.40) we have to consider that the SNV could lead to more complex alterations during splicing and whole or multiple exon skipping.
Patients with pathogenic variants in AIP present early onset of symptoms, are more often male, with aggressive tumors, requiring multimodal therapy (Marques et al. 2020, Korbonits et al. 2024), as seen in our patient. Poor response to fg-SRL has been described (Daly et al. 2010). In our case, since the response to fg-SRL was very unlikely because of negative SST2 expression and isointensity on T2, we chose not to use it.
Pasireotide was our initial choice. In three cases, pasireotide normalized IGF-I in patients resistant to fg-SRL with AIP-mutations (Rostomyan et al. 2017, Daly et al. 2019). However, in ours and in another case, normalization of IGF-I was not seen (van Santen et al. 2021), illustrating that AIP variants might be involved in treatment resistance. Pegvisomant may be an alternative; however, AIP-mutated tumors are often more aggressive, and pegvisomant has no tumoral effect (Giustina et al. 2017). However, in cases like ours, both IGF-I control and tumor size control can be achieved with combined pegvisomant and SRL (Coopmans et al. 2022). In conclusion, we describe a novel germline splice variant in AIP leading to gigantism. The patient had a sparsely granulated tumor, with a modest response to pasireotide.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/EO-24-0003.
Declaration of interest
EL has received speaker fees from Ipsen. LK has received speaker fees from Ipsen and Novo Nordisk. LEW has received speaker fees from Ipsen and Recordati, and is sub-investigator in clinical trials from Recordati and Crinetics. MRG has received speaker fees from Recordati, Ipsen and Novo Nordisk, has served as a member of the advisory board of Recordati, Ipsen, Novo Nordisk and Crinetics, and as principal investigator in clinical trials from Recordati and Crinetics. The other authors report no conflicts of interest.
Funding
This work received a grant (308488/2019-9) from Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq) to MRG.
Patient consent
Written informed consent was obtained from the patient’s mother for publication of the submitted article and accompanying images.
Author contribution and acknowledgements
EL gathered the data and wrote the initial draft, RLM did the Sanger sequencing and pathogenic variant analysis, FA and LC were responsible for the histological slides, LEW is the patient’s main physician and, with LK and MRG, critically reviewed the draft. All authors accepted the final version of the draft.
References
Bogusławska A & Korbonits M 2021 Genetics of acromegaly and gigantism. Journal of Clinical Medicine 10 1377. (https://doi.org/10.3390/jcm10071377)
Caimari F, Hernández-Ramírez LC, Dang MN, Gabrovska P, Iacovazzo D, Stals K, Ellard S, Korbonits M & International FIPA consortium 2018 Risk category system to identify pituitary adenoma patients with AIP mutations. Journal of Medical Genetics 55 254–260. (https://doi.org/10.1136/jmedgenet-2017-104957)
Coopmans EC, van der Lely AJ & & Neggers SJCMM 2022 Approach to the patient with treatment-resistant acromegaly. Journal of Clinical Endocrinology and Metabolism 107 1759–1766. (https://doi.org/10.1210/clinem/dgac037)
Daly AF, Tichomirowa MA, Petrossians P, Heliövaara E, Jaffrain-Rea ML, Barlier A, Naves LA, Ebeling T, Karhu A, Raappana A, et al.2010 Clinical characteristics and therapeutic responses in patients with germ-line AIP mutations and pituitary adenomas: an international collaborative study. Journal of Clinical Endocrinology and Metabolism 95 E373–E383. (https://doi.org/10.1210/jc.2009-2556)
Daly AF, Rostomyan L, Betea D, Bonneville JF, Villa C, Pellegata NS, Waser B, Reubi JC, Waeber Stephan C, Christ E, et al.2019 AIP-mutated acromegaly resistant to first-generation somatostatin analogs: long-term control with pasireotide LAR in two patients. Endocrine Connections 8 367–377. (https://doi.org/10.1530/EC-19-0004)
Dottermusch M, Ryba A, Ricklefs FL, Flitsch J, Schmid S, Glatzel M, Saeger W, Neumann JE & & Schuller U 2024 Pituitary neuroendocrine tumors with PIT1/SF1 co-expression show distinct clinicopathological and molecular features. Acta Neuropathologica 147 16. (https://doi.org/10.1007/s00401-024-02686-1)
Gatto F, Feelders RA, van der Pas R, Kros JM, Waaijers M, Sprij-Mooij D, Neggers SJ, van der Lelij AJ, Minuto F, Lamberts SW, et al.2013 Immunoreactivity score using an anti-sst2A receptor monoclonal antibody strongly predicts the biochemical response to adjuvant treatment with somatostatin analogs in acromegaly. Journal of Clinical Endocrinology and Metabolism 98 E66–E71.
Giustina A, Arnaldi G, Bogazzi F, Cannavo S, Colao A, De Marinis L, De Menis E, Degli Uberti E, Giorgino F, Grottoli S, et al.2017 Pegvisomant in acromegaly: an update. Journal of Endocrinological Investigation 40 577–589. (https://doi.org/10.1007/s40618-017-0614-1)
Jaganathan K, Kyriazopoulou Panagiotopoulou S, McRae JF, Darbandi SF, Knowles D, Li YI, Kosmicki JA, Arbelaez J, Cui W, Schwartz GB, et al.2019 Predicting splicing from primary sequence with deep learning. Cell 176 535–548.e24. (https://doi.org/10.1016/j.cell.2018.12.015)
Korbonits M, Blair JC, Boguslawska A, Ayuk J, Davies JH, Druce MR, Evanson J, Flanagan D, Glynn N, Higham CE, et al.2024 Consensus guideline for the diagnosis and management of pituitary adenomas in childhood and adolescence: part 2, specific diseases. Nature Reviews. Endocrinology 20 290–309. (https://doi.org/10.1038/s41574-023-00949-7)
Marques P, Caimari F, Hernández-Ramírez LC, Collier D, Iacovazzo D, Ronaldson A, Magid K, Lim CT, Stals K, Ellard S, et al.2020 Significant benefits of AIP testing and clinical screening in familial isolated and young-onset pituitary tumors. Journal of Clinical Endocrinology and Metabolism 105 e2247–e2260. (https://doi.org/10.1210/clinem/dgaa040)
Rogers MF, Shihab HA, Mort M, Cooper DN, Gaunt TR & & Campbell C 2018 FATHMM-XF: accurate prediction of pathogenic point mutations via extended features. Bioinformatics 34 511–513. (https://doi.org/10.1093/bioinformatics/btx536)
Rostomyan L, Daly AF, Petrossians P, Nachev E, Lila AR, Lecoq AL, Lecumberri B, Trivellin G, Salvatori R, Moraitis AG, et al.2015 Clinical and genetic characterization of pituitary gigantism: an international collaborative study in 208 patients. Endocrine-Related Cancer 22 745–757. (https://doi.org/10.1530/ERC-15-0320)
Rostomyan L, Daly AF, Potorac I, Betea D & & Beckers A 2017 Long-term management of resistant acromegaly with pasireotide LAR in a patient from an AIP mutation positive FIPA family. Problems of Endocrinology 63 29–30.
van Santen SS, Daly AF, Buchfelder M, Coras R, Zhao Y, Beckers A, van der Lely AJ, Hofland LJ, Balvers RK, van Koetsveld P, et al.2021 Complicated clinical course in incipient gigantism due to treatment-resistant aryl hydrocarbon receptor-interacting protein-mutated pediatric somatotropinoma. AACE Clinical Case Reports 8 119–123. (https://doi.org/10.1016/j.aace.2021.12.003)
Wildemberg LE, da Silva Camacho AH, Miranda RL, Elias PCL, de Castro Musolino NR, Nazato D, Jallad R, Huayllas MKP, Mota JIS, Almeida T, et al.2021 Machine learning-based prediction model for treatment of acromegaly with first-generation somatostatin receptor ligands. Journal of Clinical Endocrinology and Metabolism 106 2047–2056. (https://doi.org/10.1210/clinem/dgab125)
