Advertisement

AAPM 2016: Somatic Mutations and PET-Based Radiomic Features in Non–Small Cell Lung Cancer

Advertisement

Key Points

  • Using big data radiomics—extracting comprehensive information from PET images—researchers evaluated the associations between radiomic features of tumors and EGFR and KRAS mutations in about 350 patients with NSCLC.
  • Researchers found that radiomic features describing different aspects of the tumor, such as its shape and textures, appear to be associated with EGFR mutations.

A cutting-edge method of extracting big data from positron-emission tomography (PET) images can provide additional information to quantify lung tumors caused by a genetic mutation. This information could help guide the most effective treatment, suggest findings of a study of nearly 350 patients presented by Yip et al at the 58th Annual Meeting of the American Association of Physicists in Medicine (AAPM) (Presentation MO-DE-207B-1).

Advances in genomics have shown that non–small cell lung cancer (NSCLC), the most common type of lung cancer, often is caused by mutations in specific genes, including EGFR and KRAS. Approximately 15% of NSCLC patients have EGFR mutations and often benefit from tyrosine kinase inhibitor (TKI) therapies. Therefore, identification of these mutations is crucial for selecting the most effective treatment for these patients.  

PET imaging often is used to assess tumor glucose metabolism and is an essential tool for lung cancer management. Studies have shown that some of the biological and genetic variations within tumors potentially can be captured in PET images.

Study Findings

Using big data radiomics—extracting comprehensive information from PET images—researchers evaluated the associations between radiomic features of tumors and EGFR and KRAS mutations in about 350 patients with NSCLC. The mutations were confirmed by molecular testing based on biopsies of tumor tissues, the standard of care for mutation identification. Researchers found that radiomic features describing different aspects of the tumor, such as its shape and textures, appear to be associated with EGFR mutations. Their results suggest that different metabolic imaging patterns (or imaging phenotypes) that are quantified by radiomic features may be caused by EGFR mutations.

“Our long-term goal would be to use PET or another imaging technique to develop noninvasive imaging biomarkers that complement molecular tests,” said Hugo Aerts, PhD, Director of the Computational Imaging and Bioinformatics Laboratory at Dana-Farber Cancer Institute, Brigham and Women’s Hospital, and Harvard Medical School.

“Medical images are regularly acquired for every [patient with cancer] that comes into our clinic for treatment, which is the case at many other cancer centers as well,” said Stephen Yip, PhD, an instructor at Harvard Medical School, Dana-Farber, and Brigham and Women’s Hospital. “This early research suggests that standard-of-care PET imaging may help guide doctors in identifying patients with EGFR mutations, potentially providing valuable information for personalized lung cancer therapy.”

“More research needs to be done to further understand the complementary role of PET imaging to molecular testing,” Dr. Aerts said.

The researchers are now assessing combining PET-based radiomic features with features derived from other imaging tests—such as computed tomography or magnetic resonance imaging—to improve the accuracy of genetic mutation identification in lung and brain cancers.

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.


Advertisement

Advertisement



Advertisement