In a new development, a team of researchers at John Hoplins Kimmel Cancer Center has developed a new AI-based blood testing technology. The technology is found to detect more than 90% cases of lung cancer in samples investigated of almost 800 individuals with or without cancer.
The test called DNA evaluation of fragments for early interception identifies unique patterns in the DNA fragments shed from cancer cells that circulate in the bloodstream. The use of this technology to study blood samples of 796 individuals in Netherlands, Denmark, and the U.S. revealed that it accurately differentiated patients with and without cancer.
The test when combined with analysis of clinically-related risk factors, a protein biomarker, and followed by computed tomography imaging helped in the detection of 94% patients with cancer at various stages and of subtypes. This percentage included 91% patients with first-stage or less invasive stage cancer and 96% with advanced stage cancers.
Clinically, lung cancer is the most common type associated with cancer deaths. It claims almost 2 million lives each year globally. However, in the U.S., less than 6% of individuals with risk of lung cancer undergo low-dose computed tomography scanning that is recommended. This is despite projections that tens of thousands of fatalities could be avoided. The screening statistics for risk of lung cancer is even fewer worldwide.
The reasons for fewer screening are several. This includes concerns of potential harm related to investigation of false positive result, exposure to radiation, and worries about complications of invasive procedures.
Clearly, there is an urgent, unmet clinical need for the development of alternative, noninvasive methods. This is to improve scanning for cancer among high-risk individuals, and ultimately the whole population.
