Subcellular localization of fibroblast growth factor receptor type 2 and correlation with CTNNB1 genotype in adrenocortical carcinoma.

OBJECTIVE: Fibroblast growth factor receptor (FGFR) 2 regulates the development of the adrenal gland in mice. In addition, FGFR2-mediated signalling has been shown to prevent apoptosis and to enhance proliferation in adrenocortical precursor cells. The activation of the Wingless/Int-1 (WNT)/beta catenin pathway as a key mechanism of adrenocortical tumourigenesis has been linked to FGFR2 signalling in other cell types. Therefore we hypothesised that FGFR2 expression may also play a role in adrenocortical carcinoma (ACC). We conducted a pilot study and analysed protein expression of FGFR2 in 26 ACCs using immunohistochemistry technique. Data on the CTNNB1 mutation status and clinical data were correlated to the expression of FGFR2. RESULTS: We observed a high variability in FGFR2 expression between the different tumour samples. There was a subset of ACC with comparatively high nuclear expression of FGFR2. We did not find a clear association between the CTNNB1 mutational status or clinical features and the FGFR2 expression. We conclude that FGFR signalling plays a role in adrenocortical carcinoma. Our data encourages further investigations of FGFR signalling in ACC, especially since new inhibitors of FGFR signalling are already entering clinical trials for the treatment of other cancer types.

USP8 Mutations in Pituitary Cushing Adenomas-Targeted Analysis by Next-Generation Sequencing.

Gain-of-function somatic mutations in the ubiquitin specific protease 8 (USP8) gene have recently been reported as a cause of pituitary adenomas in Cushing disease. Molecular diagnostic testing of tumor tissue may aid in the diagnosis of specimens obtained through therapeutic transsphenoidal surgery; however, for small tumors, availability of fresh tissue is limited, and contamination with normal tissue is frequent. We performed molecular testing of DNA isolated from single formalin-fixed and paraffin-embedded (FFPE) tissue sections of 42 pituitary adenomas from patients with Cushing disease (27 female patients and 15 male patients; mean age at surgery, 42.5 years; mean tumor size, 12.2 mm). By Sanger sequencing, we identified previously reported USP8 missense mutations in six tumors. Targeted next-generation sequencing (NGS) revealed known or previously undescribed missense mutations in three additional tumors (two with two different mutations each), with mutant allele frequencies as low as 3%. Of the nine tumors with USP8 mutations (mutation frequency, 21.4%), seven were from female patients (mutation frequency, 25.9%), and two were from male patients (mutation frequency, 13.3%). Mutant tumors were on average 11.4 mm in size, and patients with mutations were on average 43.9 years of age. The overall USP8 mutation frequency in our cohort was lower than in previously described cohorts, and we did not observe USP8 deletions that were frequent in other cohorts. We demonstrate that testing for USP8 variants can be performed from small amounts of FFPE tissue. NGS showed higher sensitivity for USP8 mutation detection than did Sanger sequencing. Assessment for USP8 mutations may complement histopathological diagnosis.

Novel somatic mutations in primary hyperaldosteronism are related to the clinical, radiological and pathological phenotype.

UNLABELLED: Aldosterone-producing adenomas (APAs) and bilateral adrenal hyperplasia are important causes of secondary hypertension. Somatic mutations in KCNJ5, CACNA1D, ATP1A1, ATP2B3 and CTNNB1 have been described in APAs. OBJECTIVE: To characterize clinical-pathological features in APAs and unilateral adrenal hyperplasia, and correlate them with genotypes. DESIGN: Retrospective study. SUBJECTS AND MEASUREMENTS: Clinical and pathological characteristics of 90 APAs and seven diffusely or focally hyperplastic adrenal glands were reviewed, and samples were examined for mutations in known disease genes by Sanger or exome sequencing. RESULTS: Mutation frequencies were as follows: KCNJ5, 37·1%; CACNA1D, 10·3%; ATP1A1, 8·2%; ATP2B3, 3·1%; and CTNNB1, 2·1%. Previously unidentified mutations included I157K, F154C and two insertions (I150_G151insM and I144_E145insAI) in KCNJ5, all close to the selectivity filter, V426G_V427Q_A428_L433del in ATP2B3 and A39Efs*3 in CTNNB1. Mutations in KCNJ5 were associated with female and other mutations with male gender (P = 0·007). On computed tomography, KCNJ5-mutant tumours displayed significantly greater diameter (P = 0·023), calculated area (P = 0·002) and lower precontrast Hounsfield units (P = 0·0002) vs tumours with mutations in other genes. Accordingly, KCNJ5-mutant tumours were predominantly comprised of lipid-rich fasciculata-like clear cells, whereas other tumours were heterogeneous (P = 5 × 10(-6) vs non-KCNJ5 mutant and P = 0·0003 vs wild-type tumours, respectively). CACNA1D mutations were present in two samples with hyperplasia without adenoma. CONCLUSIONS: KCNJ5-mutant tumours appear to be associated with fasciculata-like clear cell predominant histology and tend to be larger with a characteristic imaging phenotype. Novel somatic KCNJ5 variants likely cause adenomas by loss of potassium selectivity, similar to previously described mutations.

PRKACA mutations in cortisol-producing adenomas and adrenal hyperplasia: a single-center study of 60 cases.

OBJECTIVE: Cortisol excess due to adrenal adenomas or hyperplasia causes Cushing's syndrome. Recent genetic studies have identified a somatic PRKACA(L206R) mutation as a cause of cortisol-producing adenomas. We aimed to compare the clinical features of PRKACA-mutant lesions with those of CTNNB1 mutations, and to search for similar mutations in unilateral hyperplasia or tumors co-secreting aldosterone. DESIGN, PATIENTS, AND METHODS: In this study, 60 patients with cortisol excess who had adrenalectomies at our institution between 1992 and 2013 were assessed, and somatic mutations were determined by Sanger sequencing. A total of 36 patients had overt Cushing's syndrome, the remainder were subclinical: 59 cases were adenomas (three bilateral) and one was classified as hyperplasia. Four tumors had proven co-secretion of aldosterone. RESULTS: Among cortisol-secreting unilateral lesions without evidence of co-secretion (n=52), we identified somatic mutations in PRKACA (L206R) in 23.1%, CTNNB1 (S45P, S45F) in 23.1%, GNAS (R201C) in 5.8%, and CTNNB1+GNAS (S45P, R201H) in 1.9%. PRKACA and GNAS mutations were mutually exclusive. Of the co-secreting tumors, two (50%) had mutations in KCNJ5 (G151R and L168R). The hyperplastic gland showed a PRKACA(L206R) mutation, while patients with bilateral adenomas did not have known somatic mutations. PRKACA-mutant lesions were associated with younger age, overt Cushing's syndrome, and higher cortisol levels vs non-PRKACA-mutant or CTNNB1-mutant lesions. CTNNB1 mutations were more significantly associated with right than left lesions. CONCLUSIONS: PRKACA(L206R) is present not only in adenomas, but also in unilateral hyperplasia and is associated with more severe autonomous cortisol secretion. Bilateral adenomas may be caused by yet-unknown germline mutations.

Evaluation of chromosome 11p imbalances in aniridia and Wilms tumor patients.

Newborn sporadic aniridia patients with an 11p13 deletion including the WT1 gene have an increased risk to develop Wilms tumor. At present a risk for Wilms tumor cannot be estimated in patients with deletions not extending into, but ending close to WT1. Therefore, it is important to determine the distance of deletion endpoints from the WT1 gene and survey these patients for a longer follow-up time to obtain a more defined risk estimation. Using molecular methods, such as Multiplex Ligation-dependent Probe Amplification (MLPA), deletion endpoints can be mapped more accurately than with FISH. We describe here the analysis of six aniridia patients, in two of these the deletions extend close to the 3' end of WT1. At the ages of 3.8 and 4 years they have not developed a Wilms tumor, suggesting a low tumor risk in such patients. In addition we have studied 24 non-AN cases with a higher likelihood for WT1 alterations with MLPA and found no deletions. In conclusion newborns with aniridia should be studied with molecular methods that can determine deletion endpoints in 11p13 exactly. For a better Wilms tumor risk estimation cases with deletion endpoints close to WT1 should be followed for at least 4-5 years. Furthermore germ line intragenic deletions affecting WT1 in patients with a higher likelihood for a WT1 association, for example, bilateral tumors, genitourinary aberrations, or nephrotic syndrome, were not found in this study, suggesting that deletions are rare events.