Researchers built a tool that can quickly predict which genes are responsible for cancer so that cancer treatments can be earlier suggested. It is among the most complete tools of its kind, and the first that contains a user-friendly web interface that needs little knowledge of bioinformatics.
The researcher noticed that iCAGES recognizes individual cancer “drivers” for 77 percent of the time when given a pair of randomly selected driver genes and non-driver genes, in comparison with about 51 percent for other computational tools.
Majority of cancers are caused by the accumulation of somatic versus inherited genetic mutations, or variants. A lot of variants involved in several types of cancer have been recognized with genetic sequencing studies of large numbers of patients. But, this information is not always clinically helpful on an individual level. Cancer “drivers” can differ from patient to patient, and there are no practical clinical tools available for predicting which variants in an individual’s genome are driving his or her disease and which are present but not responsible for the disease. “Even when the genes driving cancer are identified, clinicians don’t have an effective way of selecting among the hundreds of possible drug therapies,” mentioned the lead researcher of the study Kai Wang, PhD, who serves as the associate professor of biomedical informatics and the director of clinical informatics at the Institute for Genomic Medicine at CUMC.
How does iCAGES work?
To overcome all these issues, Dr. Wang and his team built a computational tool named integrated CAncer GEnome Score (iCAGES). First, iCAGES examines the patient’s entire genome, comparing it to the genomic sequence of the patient’s tumor to find possible cancer-causing variants. Then, iCAGES cross-references these variants to databases of known cancer-causing genes, with the help of statistical analyses and machine learning techniques to prioritize the most possible driver genes. In the end, iCAGES matches the prioritized variants to FDA-approved and investigational drug therapies that particularly address those variants or genes. The whole process requires only 30 minutes. In comparison, traditional approaches need multiple separate steps involving human input, taking more than many weeks.
In a test designed to demonstrate how the tool would be used in real practice, Dr. Wang tested iCAGES by means of detailed sequencing data from a patient with lung cancer. Out of 129 possible cancer drivers, iCAGES pointed out a gene called ARAF. iCAGES utilized the genomic sequencing data to choose sorafenib as the best drug candidate out of 122 possible treatments. The patient’s oncologists also had the same conclusions, but they used a much more difficult and time-consuming strategy, which requires expert knowledge during the decision-making process. “The patient was provided with sorafenib and had an outstanding clinical response,” noted Dr. Wang. “It is worth to notice that sorafenib is not FDA-approved. However, the result proposes that iCAGES may facilitate in identifying new treatment strategies and off-label use of available approved drugs.”
When tested on various cancer patient databases, iCAGES seemed to be better than other computational tools at indicating cancer drivers from personal genomes and at recognizing helpful treatment.
“We expect that iCAGES can aid clinicians take full gain of the huge amounts of data on genomic sequencing and cancer variants, and elucidate personalized cancer therapy,” remarked Dr. Wang.
Image credit: CUMC
For more fascinating research news, visit our website frequently. Subscribe to our newsletter; we bring the latest news to your inbox.