Novel Machine Learning Blood Test May Allow Earlier Detection of Lung and Other Cancers

A unique blood testing technology currently under development that combines genome-wide sequencing of single molecules of DNA released by tumors with machine learning could enable earlier detection of lung and other cancers.

Researchers at the Johns Hopkins Kimmel Cancer Center (Baltimore, MD, USA) are developing GEMINI (Genome-wide Mutational Incidence for Non-Invasive Detection of Cancer), a test that searches for DNA alterations across the genome. After the collection of a blood sample from individuals at risk of developing cancer, cell-free DNA (cfDNA) released by tumors is extracted from the plasma and sequenced using cost-effective whole-genome sequencing. Individual DNA molecules are then analyzed for sequence alterations in order to obtain mutation profiles across the genome. Utilizing a machine learning model trained to distinguish cancerous from non-cancerous mutation frequencies in various genome regions, the system can identify individuals with cancer. The model generates a score between 0 to 1, with higher scores indicating a greater likelihood of cancer.

The development of GEMINI involved the examination of whole-genome sequences from 2,511 individuals with 25 different cancer types, as part of the Pan-Cancer Analysis of Whole Genomes study. This led to the identification of distinct mutation frequencies across the genome in various tumor types, such as lung cancers, which had an average of 52,209 somatic mutations per genome. High-frequency mutation regions were found to be comparable between tumor tissue and blood-derived cfDNA in patients with lung cancer, melanoma, or B cell non-Hodgkin lymphoma.

Laboratory testing with GEMINI demonstrated promising results, with the detection of over 90% of lung cancers, including early-stage I and II disease, when combined with computerized tomography imaging. Its potential was further demonstrated in a study with seven participants without detectable tumors at the time of blood collection, who later received a lung cancer diagnosis. The median GEMINI score of this group was 0.78, higher than individuals without cancer. Six participants who tested positive using GEMINI were later diagnosed with lung cancer from 231 to 1,868 days after samples were obtained, indicating that abnormalities in cfDNA mutation profiles can be detected years ahead of standard diagnoses. Further experiments also found GEMINI capable of differentiating between subtypes of lung cancers, detecting early liver cancers, and monitoring patients' response to lung cancer treatment.

“This study shows for the first time that a test like GEMINI, incorporating genome-wide mutation profiles from single molecules of cfDNA, in combination with other cancer detection approaches, may be used for early detection of cancers, as well as for monitoring patients during therapy,” said senior study author Victor Velculescu, M.D., Ph.D., professor of oncology and co-director of the cancer genetics and epigenetics program at the Kimmel Cancer Center.

Related Links:
Johns Hopkins Kimmel Cancer Center

Visit expo >

Gold Member

Pharmacogenetics Panel

VeriDose Core Panel v2.0

Verification Panels for Assay Development & QC
Seroconversion Panels

New

Silver Member

ACTH Assay

ACTH ELISA

New

HIV-1 Test

HIV-1 Real Time RT-PCR Kit