Novel Blood Test Combining DNA Sequencing and Machine Learning Shows Promise for Early Cancer Detection
Researchers at the Johns Hopkins Kimmel Cancer Center are developing a groundbreaking blood test that combines DNA sequencing and machine learning to detect cancer at an earlier stage. The innovative test, called GEMINI (Genome-wide Mutational Incidence for Non-Invasive detection of cancer), analyzes changes in DNA throughout the genome to identify the presence of tumors.
The process begins with a simple blood sample collected from individuals at risk of developing cancer. From this sample, cell-free DNA shed by tumors is extracted and sequenced using cost-efficient whole genome sequencing. Single molecules of DNA are then analyzed for sequence alterations to obtain mutation profiles across the genome. This data is then fed into a machine learning model that has been trained to identify changes in cancer and non-cancer mutation frequencies in different regions of the genome. The model generates a score, ranging from 0 to 1, with a higher score indicating a higher probability of cancer.
In lab tests, GEMINI, when combined with computerized tomography imaging, successfully detected over 90% of lung cancers, including those at stage I and II. The results of this proof-of-concept study were published in the journal Nature Genetics.
This study demonstrates that GEMINI, which incorporates genome-wide mutation profiles from cfDNA, can be used alongside other cancer detection methods for early cancer detection and monitoring during therapy, says Victor Velculescu, M.D., Ph.D., professor of oncology and co-director of the cancer genetics and epigenetics program at the Kimmel Cancer Center.
While the study primarily focused on detecting lung cancer in high-risk populations, the researchers also observed altered mutational profiles in cfDNA from patients with liver cancer, melanoma, and lymphoma. This suggests that the test may have broader applications beyond lung cancer.
To develop GEMINI, the researchers analyzed whole-genome sequences of cancers from 2,511 individuals across 25 different types of cancer. They identified distinct mutation frequencies across the genome for different tumor types, including an average of 52,209 somatic mutations per genome for lung cancer.
The researchers also identified genomic regions with the highest number of mutations, finding similarities in mutation frequency between tumor tissue and blood-derived cfDNA from patients with lung cancer, melanoma, or B cell non-Hodgkin lymphoma.
To evaluate the effectiveness of GEMINI in the detection of sequence alterations, the researchers tested cfDNA samples from 365 individuals in the Longitudinal Urban Cohort Aging Study (LUCAS), a cohort at high risk of developing lung cancer. The GEMINI scores were consistently higher in individuals with cancer compared to those without, indicating its accuracy in detecting cancer.
Furthermore, the researchers assessed the potential of combining GEMINI with DELFI (DNA evaluation of fragments for early interception), a previously developed test that detects changes in the size and distribution of cfDNA fragments across the genome. The combined approach improved the detection of early-stage lung cancer, detecting several cancer samples that GEMINI alone had missed. In the LUCAS cohort, GEMINI combined with DELFI correctly identified lung cancers 91% of the time.
The researchers also explored the use of GEMINI in other study samples, including individuals without detectable tumors at the time of blood collection. Remarkably, GEMINI detected abnormalities in cfDNA mutation profiles in these individuals, and they were later diagnosed with lung cancer between 231 and 1,868 days after sample collection. This evidence suggests that GEMINI can detect cancer years before standard diagnoses.
Additionally, GEMINI demonstrated the ability to distinguish between different subtypes of lung cancer and detect early-stage liver cancers. It also showed promise in monitoring lung cancer patients during therapy, as GEMINI scores decreased during the initial response to treatment.
The combined findings suggest that the integration of genome-wide GEMINI mutation analyses with DELFI fragmentation analyses of cfDNA could provide a cost-efficient and scalable approach for detecting cancer. However, larger clinical trials are necessary to validate the effectiveness of the GEMINI test before it can be widely implemented in clinical settings.
The development of this novel blood test for early cancer detection holds great promise for improving patient outcomes by enabling the timely diagnosis and treatment of cancer. As further research and testing are conducted, the potential impact of this innovative technology on cancer care is encouraging.
