Editors' ChoiceCancer

An Emerging Lung Cancer Biomarker

See allHide authors and affiliations

Science Translational Medicine  02 Apr 2014:
Vol. 6, Issue 230, pp. 230ec58
DOI: 10.1126/scitranslmed.3009040

Biomarker development takes two principal forms. In recent years, numerous large-scale genomic studies resulted in a deluge of data. Identifying the genetic alterations that drive and maintain tumorigenesis (“driver” mutations) from these data sets requires complex bioinformatics methods. Even after a candidate driver is identified, however, it may not be amenable to pharmacologic targeting. An alternative approach to biomarker identification is to begin in the clinic. A decade ago, after observing that a subset of patients with advanced lung cancer responded dramatically to epidermal growth factor receptor (EGFR) inhibitors, investigators sequenced the EGFR gene in biopsies of responding patients. From this work came the recognition that sensitizing EGFR mutations result in heightened kinase activity, oncogene addiction, and exquisite sensitivity to EGFR inhibitor therapy.

Imielinski and colleagues used this second approach to determine the molecular underpinnings of outlier responses to sorafenib in a lung cancer clinical trial. Sorafenib, a multitargeted kinase inhibitor, has modest activity in lung cancer, and the Eastern Cooperative Oncology Group (ECOG) 2501 trial was no exception. Among the 306 evaluable patients, there were only nine responders, one of whom achieved ongoing disease control for 5 years. The investigators obtained biospecimens from this case and performed massively parallel DNA sequencing of tumor tissue and peripheral blood. Initially, this yielded 25,150 somatic mutations. These included 101 nonsynonymous mutations affecting the coding of 99 genes, 15 of which were consistently detected in RNA sequencing. Within this subset, based on pathway biology, the investigators surmised that the most likely candidate oncogenic driver was an ARAF S214C mutation.

ARAF encodes a serine-threonine kinase in the Raf protein family. Similar to other Raf family members, ARAF transduces mitogen-activated protein kinase (MAPK) pathway signaling from Ras to MAPK kinase (MEK) and extracellular signal–regulated kinase (ERK) and is a sorafenib target. However, it had not been previously implicated in tumorigenesis. To investigate the broader importance of their findings, Imielinski and colleagues queried The Cancer Genome Atlas (TCGA) and identified three ARAF codon 214 somatic mutations and three mutations in the related gene RAF1, representing ~1% of lung cancers. ARAF mutations were also identified in rare cases of colorectal cancer (2%), gastric cancer (1%), and melanoma (1%). The functional impact of ARAF mutations was demonstrated by ectopic expression of ARAF S214 variants in immortalized human airway epithelial cells, which increased soft agar colony formation and phosphoMEK. This transformation was inhibited by clinically achievable concentrations of sorafenib and the MEK inhibitor trametinib.

ARAF mutations are not ready for targeting in clinical practice, but this initial work paves the way for confirmatory preclinical studies and testing of other clinical specimens. If ARAF mutations are ultimately validated as a predictive biomarker, the present work will serve as a key reminder that clinical observation of outlier patient responses remains a powerful and efficient approach to driver mutation discovery.

M. Imielinski et al., Oncogenic and sorafenib-sensitive ARAF mutations in lung adenocarcinoma. J. Clin. Invest. 10.1172/JCI72763 (2014). [Full Text]

Navigate This Article