NEW YORK (GenomeWeb) – Researchers in the Netherlands have developed a diagnostic test to detect lung cancer by examining tumor RNA absorbed by circulating platelets called thrombocytes.
Published in Cancer Cell earlier today, the study investigated the potential and origin of spliced RNA profiles from tumor-educated platelets, or TEPs, for the noninvasive detection of early- and late-stage non-small cell lung cancer.
Platelets in patients that are cancer-free contain a different composition of RNA than TEPs, which are platelets that have interacted with a tumor. The TEP gene panel developed by the scientists allows vector machine-based classification of lung cancer.
The researchers examined more than 700 blood samples using a biomarker signature detection platform called ThromboSeq, including patients diagnosed with late-stage non-small cell lung cancer, a smaller group with early stage cancer, and a control group with no known cancer.
ThromboSeq enables clinical researchers to identify different cancer types by looking at TEP-derived RNA using next-generation sequencing.
ThromboSeq’s swarm intelligence algorithm scanned approximately 5,000 different RNA molecules found in the thrombocytes and recorded the small amount that indicate a cancerous tumor. The researchers then ran the blood samples through screenings to diagnose how accurately the thrombocytes identified cancer.
According to the researchers, ThromboSeq detected early-stage cancer with 81 percent accuracy and late-stage cancer with 88 percent accuracy. The algorithm yielded an accuracy up to 91 percent in a validation control group that matched for a patient's age, smoking status, and blood storage time.
The device's particle-swarm optimization (PSO) algorithms allowed efficient selection of RNA biomarker panels from platelet RNA-sequencing libraries. The gene panels then diagnosed cancer from TEPs, suggesting that swarm intelligence may also benefit the optimization of diagnostic readout of other liquid biopsy biosources.
"ThromboSeq might not only provide lung cancer diagnostics, but potentially any other tumor type as well, and may enable tumor-type stratification," VU University Medical Center researcher Myron Best said in a statement.
He and his colleagues based their method on swarming behavior that can be found in nature, where birds, insects, and fish swarm to defend themselves against predators or to search for food.
“Birds continuously adjust their location in the swarm relative to each other, thereby increasing the flock’s coverage, and thus, the efficiency of the food-searching process,” Best said. “We applied this natural phenomenon to our algorithms, which make use of the complex RNA repertoire present in platelets.”
Researchers concluded that the ThromboSeq platform allows for robust biomarker selection for blood-based cancer diagnostics, independent of bias introduced by factors such as age and smoking status. Best and his team plan to further optimize the algorithm with patients suspected of having undiagnosed lung cancer.
"Although the tumor-educated platelets blood test does not, so far, provide perfect predictions, it may complement alternative liquid biopsy bio-sources" added Best.