1705PD - Detection of Therapeutic Targets in Carcinomas of Unknown Primary

Date 10 September 2017
Event ESMO 2017 Congress
Session Basic science
Topics Cancer Biology
Carcinoma of Unknown Primary Site (CUP)
Basic Scientific Principles
Presenter Britt Clynick
Citation Annals of Oncology (2017) 28 (suppl_5): v595-v604. 10.1093/annonc/mdx391
Authors B. Clynick1, B. Dessauvagie2, G. Strerrett2, N. Harvey2, S. Subramaniam3, J. Herron4, R. Allcock1, B. Guo1, C. Saunder5, W. Erber1, K. Meehan1
  • 1Biomedical Sciences, University of Western Australia, 6009 - Perth/AU
  • 2Anatomical Pathology, PathWest, 6009 - Perth/AU
  • 3General Surgery, Royal Perth Hospital, 6000 - Perth/AU
  • 4Radiology, Sir Charles Gardiner Hospital, 6009 - Perth/AU
  • 5Surgery, Fiona Stanley Hospital, 6150 - Perth/AU



Current cancer treatment paradigms are based on features of the primary tumour and extent of disease. Carcinoma of unknown primary (CUP) is a histologically confirmed metastatic cancer in the absence of an identifiable primary tumour. As such it is difficult to determine the optimal treatment strategy and 5-year survival rates are less than 20%. This poor survival rate is due to both the unclassifiable metastatic malignancy and the use of empirical broad-spectrum chemotherapy. A shift towards personalising cancer management based on mutation profiling has been explored in many cancer types including CUP. Small studies have reported durable treatment responses in CUP patients receiving targeted therapies, specifically when EGFR and KIT variants are detected. The present study has explored whether biologically important oncogenic driver mutations and potentially actionable targets are present in CUP that have potential to better inform treatment decisions.


CUP cases (n = 32) diagnosed on histopathology criteria were selected for study. Sections were cut from paraffin blocks of formalin-fixed tissue and DNA isolated. Extracted DNA was amplified with the Ion AmpliSeq Cancer Hotspot panel and Oncomine Focus panel for the identification of biologically relevant and actionable mutations, respectively. Amplified DNA was sequenced using the Ion Torrent platform and data was processed using a stringent variant filtration pipeline.


Biologically relevant or therapeutically druggable variants were detected in 88% (n = 28) of cases. The most common variants were in TP53 (47%), KRAS (19%), MYC (6%), BRAF (6%) and CDKN2A (6%). There were potentially actionable targets in 14/32 (44%) cases, with the most common druggable variants being in the KRAS gene (exon 1) (6 cases) and MYC gene amplifications (2 cases).


This retrospective study successfully identified biologically relevant variants in 88% of CUP cases. 50% of these variants were potentially actionable with drugs currently approved for use in known primary cancer types or undergoing clinical trials. This would give a novel treatment option to patients with a currently incurable disease with poor survival. The data therefore supports the use of NGS at diagnosis to give biological insight into the drivers leading to the malignancy, and to allow consideration of novel targeted treatment options.

Clinical trial identification

Legal entity responsible for the study

Translational Cancer Pathology Laboratory




All authors have declared no conflicts of interest.