Introduction to the DELFI Approach

Recent advancements in artificial intelligence (AI) technology have opened new doors for early lung cancer detection. Researchers from the Johns Hopkins Kimmel Cancer Center and various other institutions have developed and validated a liquid biopsy that could potentially identify lung cancer at earlier stages.

Study Overview and Findings

A study published on June 3 in Cancer Discovery established that AI technology can identify individuals more likely to have lung cancer based on specific DNA fragment patterns in their blood. This study involved approximately 1,000 participants with varying cancer statuses across 47 centers in 23 U.S. states.

Addressing Low Screening Rates

Despite the U.S. Preventive Services Task Force recommending yearly CT screenings for 15 million high-risk individuals, only 6%–10% follow through. The DELFI approach aims to boost these screening rates by providing a simple blood test that can be conducted in a doctor’s office.

AI’s Role in Lung Cancer Detection

According to corresponding author Victor E. Velculescu, M.D., Ph.D, this blood test will allow doctors to determine if a patient has potential signs of lung cancer and whether follow-up via CT scan is advisable.

Understanding DNA Fragmentation

The innovative test developed by Velculescu and his team exploits the contrasting ways DNA is organized in normal versus cancer cells. In healthy cells, DNA is tightly packaged, akin to a neatly rolled yarn, while in cancer cells, it appears more chaotic and disheveled.

Study Methodology

The team trained AI software to recognize unique DNA fragment patterns in a cohort of 576 individuals, later validating the findings in a separate group of 382 subjects. The test demonstrated an impressive negative predictive value of 99.8%.

Impact on Screening and Detection Rates

The group’s computer simulations suggested that if screening rates increased to 50% over five years due to this test, the number of detected lung cancers could quadruple and the early detection rate might increase by about 10%, potentially preventing around 14,000 deaths from lung cancer.

Accessibility and Cost-Effectiveness

Velculescu emphasizes the potential of this inexpensive test to be administered at a large scale, revolutionizing lung cancer screening accessibility and helping more people to get the necessary screenings in time.

Current Status and Future Directions

The DELFI test is currently available as a laboratory-based solution under the Clinical Laboratory Improvement Amendments, with intentions to seek approval from the U.S. Food and Drug Administration for lung cancer screening.

Research Collaborators and Support

Robert B. Scharpf from Johns Hopkins co-authored the study alongside colleagues from a variety of esteemed institutions. The research was supported by multiple organizations, including DELFI Diagnostics and the National Institutes of Health.

Intellectual Property and Conflict of Interest

Velculescu and Scharpf are involved with various patent applications related to cell-free DNA detection, which have been licensed to notable companies like DELFI Diagnostics and LabCorp. Their financial relationships with these companies adhere to Johns Hopkins’ conflict of interest policies.

Conclusion

The DELFI approach represents a promising step towards enhancing early lung cancer detection through accessible, non-invasive technology. By integrating AI with liquid biopsy, researchers hope to significantly improve screening rates and outcomes for lung cancer patients.