Prediction of clinical benefit from androgen deprivation therapy in salivary duct carcinoma patients.

Androgen deprivation therapy (ADT) is first-line palliative treatment in androgen receptor-positive (AR+) salivary duct carcinoma (SDC), and response rates are 17.6-50.0%. We investigated potential primary ADT resistance mechanisms for their predictive value of clinical benefit from ADT in a cohort of recurrent/metastatic SDC patients receiving palliative ADT (n = 30). We examined mRNA expression of androgen receptor (AR), AR splice variant-7, intratumoral androgen synthesis enzyme-encoding genes AKR1C3, CYP17A1, SRD5A1 and SRD5A2, AR protein expression, ERBB2 (HER2) gene amplification and DNA mutations in driver genes. Furthermore, functional AR pathway activity was determined using a previously reported Bayesian model which infers pathway activity from AR target gene expression levels. SRD5A1 expression levels and AR pathway activity scores were significantly higher in patients with clinical benefit from ADT compared to those without benefit. Survival analysis showed a trend toward a longer median progression-free survival for patients with high SRD5A1 expression levels and high AR pathway activity scores. The AR pathway activity analysis, and not SRD5A1 expression, also showed a trend toward better disease-free survival in an independent cohort of locally advanced SDC patients receiving adjuvant ADT (n = 14) after surgical tumor resection, and in most cases a neck dissection (13/14 patients) and postoperative radiotherapy (13/14 patients). In conclusion, we are the first to describe that AR pathway activity may predict clinical benefit from ADT in SDC patients, but validation in a prospective study is needed.

SSTR2A expression in medullary thyroid carcinoma is correlated with longer survival.

PURPOSE: Medullary thyroid carcinoma (MTC) derives from the parafollicular C-cells of the thyroid gland. Somatostatin receptors (SSTRs) are expressed in various neuroendocrine tumours including MTC. The aim of this study was to evaluate SSTR2A as a prognostic factor for MTC, to study distribution of SSTR2A expression within tumours and to compare expression of SSTR2A between primary tumours and corresponding lymph node metastases. METHODS: Patients who underwent surgery between 1988 and 2014 for MTC from five tertiary referral centres in The Netherlands were included. In total, primary tumours of 114 patients and lymph node metastases of 34 patients were analysed for expression of SSTR2A using a tissue microarray, and correlated with clinicopathological variables and survival. RESULTS: The mean age of patients was 45.5 years (SD 16.2), 55 patients were male (49.5%). Primary tumours of 58 patients (50.9%) showed SSTR2A expression. In multivariate Cox-regression analysis, SSTR2A positivity correlated independently with better overall survival (OS) (HR 0.3; 95% CI 0.1-1.0). In stage IV MTC patients, 10-year survival rates for SSTR2A-negative and positive patients were 43% and 96%, respectively. In 53.9% of patients with lymph node metastases, expression in primary tumour and lymph node metastases differed. CONCLUSION: SSTR2A expression is correlated with longer OS in MTC, especially for stage IV patients, suggesting that SSTR2A expression might be a useful prognostic factor in MTC. The SSTR2A status of the primary MTC does not predict expression in lymph node metastases.

Mutations in g protein encoding genes and chromosomal alterations in primary leptomeningeal melanocytic neoplasms.

Limited data is available on the genetic features of primary leptomeningeal melanocytic neoplasms (LMNs). Similarities with uveal melanoma were recently suggested as both entities harbor oncogenic mutations in GNAQ and GNA11. Whether primary LMNs share additional genetic alterations with uveal melanoma including copy number variations is unknown. Twenty primary LMNs ranging from benign and intermediate-grade melanocytomas to melanomas were tested by direct sequencing for hotspot mutations in the genes GNA11, GNAQ, BRAF, NRAS and HRAS. Furthermore, the lesions were tested for copy number variations of chromosomes frequently present in uveal melanoma (1p, 3, 6 and 8q) by multiplex ligation-dependent probe amplification (MLPA). Genome-wide analyses of copy number alterations of two leptomeningeal melanocytic neoplasms were performed using the OncoScan SNP-array. GNAQ(Q209) mutations were present in eleven LMNs, while two of 20 cases carried a GNA11(Q209) mutation. No BRAF, HRAS or NRAS hotspot mutations were detected. Monosomy 3 and gain of 8q were present in one leptomeningeal melanoma, and one intermediate-grade melanocytoma harbored a gain of chromosome 6. With MLPA, the melanocytomas did not show any further gross chromosomal variations. Our data shows that primary LMNs, like uveal melanoma, harbor oncogenic mutations in GNAQ and GNA11 but lack mutations in BRAF, NRAS and HRAS. This finding may help in the differential diagnosis between a primary LMN and a metastasis from a cutaneous melanoma to the central nervous system. Copy number variations in some aggressive LMNs resemble those present in uveal melanoma but their prognostic significance is unclear.

Primary melanoma of the CNS in children is driven by congenital expression of oncogenic NRAS in melanocytes.

UNLABELLED: NRAS mutations are common in human melanoma. To produce a mouse model of NRAS-driven melanoma, we expressed oncogenic NRAS (NRAS(G12D)) in mouse melanocytes. When NRAS(G12D) was expressed in the melanocytes of developing embryos, it induced melanocyte proliferation and congenital melanocytic lesions reminiscent of human blue nevi but did not induce cutaneous melanoma. Unexpectedly, however, it did induce early-onset primary melanoma of the central nervous system (CNS). The tumors were rapidly proliferating and caused neurologic symptoms, rapid health deterioration, and death. NRAS is not a common driver oncogene of primary melanoma of the CNS in adults, but we report two cases of primary melanoma of the CNS in children, both of which carried oncogenic mutations in NRAS. We conclude that acquisition of somatic mutations in NRAS in CNS melanocytes is a predisposing risk factor for primary melanoma of the CNS in children, and we present a mouse model of this disease. SIGNIFICANCE: We show that the acquisition of NRAS mutations in melanocytes during embryogenesis is a risk factor for early-onset melanoma of the CNS. We have developed a powerful mouse model to study this rare but devastating childhood disease, and to develop therapeutic approaches for its treatment.

Improved discrimination of melanotic schwannoma from melanocytic lesions by combined morphological and GNAQ mutational analysis.

The histological differential diagnosis between melanotic schwannoma, primary leptomeningeal melanocytic lesions and cellular blue nevus can be challenging. Correct diagnosis of melanotic schwannoma is important to select patients who need clinical evaluation for possible association with Carney complex. Recently, we described the presence of activating codon 209 mutations in the GNAQ gene in primary leptomeningeal melanocytic lesions. Identical codon 209 mutations have been described in blue nevi. The aims of the present study were to (1) perform a histological review of a series of lesions (initially) diagnosed as melanotic schwannoma and analyze them for GNAQ mutations, and (2) test the diagnostic value of GNAQ mutational analysis in the differential diagnosis with leptomeningeal melanocytic lesions. We retrieved 25 cases that were initially diagnosed as melanotic schwannoma. All cases were reviewed using established criteria and analyzed for GNAQ codon 209 mutations. After review, nine cases were classified as melanotic schwannoma. GNAQ mutations were absent in these nine cases. The remaining cases were reclassified as conventional schwannoma (n = 9), melanocytoma (n = 4), blue nevus (n = 1) and lesions that could not be classified with certainty as melanotic schwannoma or melanocytoma (n = 2). GNAQ codon 209 mutations were present in 3/4 melanocytomas and the blue nevus. Including results from our previous study in leptomeningeal melanocytic lesions, GNAQ mutations were highly specific (100%) for leptomeningeal melanocytic lesions compared to melanotic schwannoma (sensitivity 43%). We conclude that a detailed analysis of morphology combined with GNAQ mutational analysis can aid in the differential diagnosis of melanotic schwannoma with leptomeningeal melanocytic lesions.

Activating mutations of the GNAQ gene: a frequent event in primary melanocytic neoplasms of the central nervous system.

Primary melanocytic neoplasms of the central nervous system (CNS) are uncommon neoplasms derived from melanocytes that normally can be found in the leptomeninges. They cover a spectrum of malignancy grades ranging from low-grade melanocytomas to lesions of intermediate malignancy and overtly malignant melanomas. Characteristic genetic alterations in this group of neoplasms have not yet been identified. Using direct sequencing, we investigated 19 primary melanocytic lesions of the CNS (12 melanocytomas, 3 intermediate-grade melanocytomas, and 4 melanomas) for hotspot oncogenic mutations commonly found in melanocytic tumors of the skin (BRAF, NRAS, and HRAS genes) and uvea (GNAQ gene). Somatic mutations in the GNAQ gene at codon 209, resulting in constitutive activation of GNAQ, were detected in 7/19 (37%) tumors, including 6/12 melanocytomas, 0/3 intermediate-grade melanocytomas, and 1/4 melanomas. These GNAQ-mutated tumors were predominantly located around the spinal cord (6/7). One melanoma carried a BRAF point mutation that is frequently found in cutaneous melanomas (c.1799 T>A, p.V600E), raising the question whether this is a metastatic rather than a primary tumor. No HRAS or NRAS mutations were detected. We conclude that somatic mutations in the GNAQ gene at codon 209 are a frequent event in primary melanocytic neoplasms of the CNS. This finding provides new insight in the pathogenesis of these lesions and suggests that GNAQ-dependent mitogen-activated kinase signaling is a promising therapeutic target in these tumors. The prognostic and predictive value of GNAQ mutations in primary melanocytic lesions of the CNS needs to be determined in future studies.