How Houston Doctors Plan To Fix Cancer Info Challenge
by: Jack Williams, December 23, 2013 2:12:00 pm
Here's the good news: researchers are getting a lot better at analyzing tumors at the molecular level and what they see are often different characteristics, even in tumors of the same type.
"We now recognize that tumors that we previously considered all the same type because they arose from the same organ, like the lung or the breast, are actually multiple different types and they have different characteristics at the molecular level and even those major types are divided into further sub-types depending on what kind of abnormalities there are in the tumor."
But here's the bad news: it's virtually impossible for a practicing clinician to sift through all that new information in a practical way.
"Information is being generated at a very rapid rate. So, for even a smart oncologist in practice that sees a lot of patients, it's very difficult to keep up to date with all this information. Thus there's a great need in putting together all the information that's being rapidly acquired so that decisions for each individual patient can be made better."
That's where Dr. Elmer Bernstam comes in. He's a professor of biomedical informatics at UT Health here in Houston. He's also part of new 3-year grant funded by the National Institutes of Health to solve the information problem.
The idea is to come up with a system for oncologists to easily access new information about tumors and therapies that might be better matched to specific molecular characteristics. He says 700,000 biomedical papers are published each year about cancer research.
"Let's say one out of a thousand is relevant to any particular specialty. That means that every day of their life, the clinician has to read two papers, Sundays, holidays, Christmas, whatever. Keeping up with the literature is no longer something that's practical for practicing clinicians."
Bernstam says the information delivery system could be in the form of an app, a website or a combination of a several technologies.
"What this really is is it's an information problem. It's a problem of taking all the data that are out there, all the text, all the databases that collect all sorts of information, and synthesizing that down into what is relevant and important, and those are two different things, for a particular patient at a particular time."
He says he hopes to have something helpful in a year or two, tools that could make it easier for doctors to match patients and personalized cancer treatments.
"In this grant, we're not really developing new treatments. We're not discovering new biology. What we're doing is we're making what is known actionable for individuals."
The nearly $1 million grant runs through August of 2016.