NCOA4-RET Fusion Research Articles

Drosophila Cancer Models Identify Functional Differences between Ret Fusions

We generated and compared Drosophila models of RET fusions CCDC6-RET and NCOA4-RET. Both RET fusions directed cells to migrate, delaminate, and undergo EMT, and both resulted in lethality when broadly expressed. In all phenotypes examined, NCOA4-RET was more severe than CCDC6-RET, mirroring their effects on patients. A functional screen against the Drosophila kinome and a library of cancer drugs found that CCDC6-RET and NCOA4-RET acted through different signaling networks and displayed distinct drug sensitivities. Combining data from the kinome and drug screens identified the WEE1 inhibitor AZD1775 plus the multi-kinase inhibitor sorafenib as a synergistic drug combination that is specific for NCOA4-RET. Our work emphasizes the importance of identifying and tailoring a patient's treatment to their specific RET fusion isoform and identifies a multi-targeted therapy that may prove effective against tumors containing the NCOA4-RET fusion.

Abstract 4268: An unbiased survey of cancer-related rearrangements in 5,917 solid tumors identifies therapeutically actionable fusions across multiple disease subtypes

Abstract The growing trend to make cancer treatment more precise depends on the accurate identification of clinically actionable genomic alterations within a patient's tumor. Fusion proteins that result from genomic rearrangements are attractive therapeutic targets because they are tumor-specific and cancer cells depend on their expression for survival. The goal of this study was to investigate the breadth of genomic rearrangements in 19 genes shown to be fused in at least one solid cancer across a collection of 5,917 tumors from a diverse variety of subtypes. All tumor samples were submitted to a CLIA-certified CAP-accredited laboratory (Foundation Medicine, Cambridge MA) for next-generation sequencing-based genomic profiling of 3,769 exons from 236 cancer related genes and 47 introns of 19 genes frequently rearranged in cancer. We observed 350 genomic rearrangements in 338 samples from 82 different solid tumor pathologies. Rearrangements involving kinases were observed in 176 samples (3% of total samples). We identified potentially druggable rearrangements of BRAF (n=13 unique partners; 9 pathologies), ALK (n=11 unique partners; 11 pathologies), FGFR2 (n=9 unique partners; 6 pathologies), ROS1 (n=7 unique partners ; 3 pathologies), RET (n=7 unique partners; 5 pathologies), FGFR3 (n=5 unique partners; 6 pathologies), NTRK1 (n=3 unique partners; 3 pathologies), ERBB2 (n=2 unique partners; 2 pathologies), and ERBB3 (n=2 unique partners; 2 pathologies). ALK, ROS1, RET, and NTRK1 fusions in lung cancer have all shown sensitivity to targeted agents in patients. The largest number of genomic rearrangements (n=87) was found in lung adenocarcinoma (12% of cases). Interestingly, we identified a KIAA1549-BRAF fusion in a triple negative breast cancer sample, a FIP1L1-PDGFRA fusion in glioblastoma, a KIF5B-RET fusion in ovarian cancer, and an NCOA4-RET fusion in colon adenocarcinoma. These fusions have been identified previously in pilocytic astrocytoma, myeloid malignancies, lung cancer and thyroid cancer, respectively, but have rarely been reported outside of these diseases; importantly, hypereosinophilic syndromes harboring FIP1L1-PDGFRA are sensitive to imatinib, and lung cancers with KIF5B-RET are sensitive to cabozantinib. Collectively, these data demonstrate that gene fusions exist across a multitude of tumor types and that known fusions are not always confined to the disease in which they appear most often. These rearrangements include clinically actionable fusions which may identify tumors that are sensitive to selective FDA-approved targeted agents. Citation Format: Juliann Chmielecki, Garrett Frampton, Doron Lipson, Jie He, Geoff Otto, Siraj Ali, Jeffrey S. Ross, Vincent A. Miller, Roman Yelensky, Philip J. Stephens. An unbiased survey of cancer-related rearrangements in 5,917 solid tumors identifies therapeutically actionable fusions across multiple disease subtypes. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4268. doi:10.1158/1538-7445.AM2014-4268

RET Fusions Define a Unique Molecular and Clinicopathologic Subtype of Non–Small-Cell Lung Cancer

The RET fusion gene has been recently described in a subset of non-small-cell lung cancers (NSCLCs). Because we have limited knowledge about these tumors, this study was aimed at determining the clinicopathologic characteristics of patients with NSCLC harboring the RET fusion gene. We examined the RET fusion gene in 936 patients with surgically resected NSCLC using a reverse transcriptase polymerase chain reaction (PCR) plus quantitative real-time PCR strategy, with validation using immunohistochemical and fluorescent in situ hybridization assays. A subset of 633 lung adenocarcinomas was also studied for EGFR, KRAS, HER2, and BRAF mutations, as well as ALK rearrangements. Patient characteristics, including age, sex, smoking history, stage, grade, International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society classification of subtypes of lung adenocarcinoma, and relapse-free survival, were collected. Of 936 patients with NSCLC, the RET fusion gene was exclusively detected in 13 patients (11 of 633 patients with adenocarcinomas and two of 24 patients with adenosquamous cell carcinomas). Of the 13 patients, nine patients had KIF5B-RET, three patients had CCDC6-RET, and one patient had a novel NCOA4-RET fusion. Patients with lung adenocarcinomas with RET fusion gene had more poorly differentiated tumors (63.6%; P = .029 for RET v ALK, P = .007 for RET v EGFR), with a tendency to be younger (≤ 60 years; 72.7%) and never-smokers (81.8%) and to have solid subtype (63.6%) and a smaller tumor (≤ 3 cm) with N2 disease (54.4%). The median relapse-free survival was 20.9 months. RET fusion occurs in 1.4% of NSCLCs and 1.7% of lung adenocarcinomas and has identifiable clinicopathologic characteristics, warranting further clinical consideration and targeted therapy investigation.