Evolving Algorithms Take Aim At Prostate Cancer
Science Daily — Researchers are sending forth strings of binary digits to go forthand multiply-all in the name of human health.
Using strands of 0s and 1s to stock a digital community, completewith mating, offspring, and the occasional mutation, is nothing new toscientists who create genetic algorithms. These mathematical formulas aregoverned by Darwin's "survival of the fittest" mantra and rely on theprinciples of evolution to create solutions in a variety of applications,including financial analyses, computer-assisted scheduling, and targetdetection for the military.
Now scientists at the University of Rochester Medical Center areexploring use of the formulas to treat prostate cancer by improving howradiation is delivered. The formulas have been developed by medicalphysicist Yan Yu, Ph.D., a well-known expert on radiation treatment planningwho heads a task force on the subject for the American Association ofPhysicists in Medicine. His algorithms are at the heart of a new start-upcompany aimed at improving a common prostate cancer procedure calledbrachytherapy, where tiny radioactive seeds are implanted into the prostateand destroy cancer cells in the organ over several weeks. It's anincreasingly popular procedure for treating the nearly 200,000 men each yearwho are diagnosed with cancer of the prostate, which is an organ about thesize of a small peach between the rectum and bladder that contributes fluidsto semen.
Medical physicists like Yu, an associate professor of radiationoncology, play an often unseen role in cancer treatment, ultimately decidinghow to deliver radiation to best eradicate cancerous tissue or organswithout hurting healthy tissue. In cases of prostate cancer, the stakes arevery high: the bladder, rectum, urethra, and nerves that control sexualfunction are all packed together near the organ, making it an especiallychallenging disease to treat. Brachytherapy is a little bit like baking asingle muffin from the inside out, with physicists and physicians trying toensure that every bit of the "muffin" is cooking at the exact sametemperature while the other muffins nearby stay cool.
Choosing the right pattern for the seeds, which are radioactiveparticles about the size of a grain of rice, is daunting. Physicianscommonly turn to commercial programs to help them decide how to place theparticles. The recommended treatment plan is put in the hands of a surgeon,who actually inserts several dozen seeds into the prostate during a one- ortwo-hour surgical procedure.
Yu's work, dubbed PIPER for Prostate Implant Planning Engine forRadiotherapy, uses artificial-intelligence technology to recommend aradiation treatment plan. Based on an ultrasound scan of a patient'sprostate and other pelvic organs, PIPER sets up a competition between thepossible configurations. Yu and colleagues create a digital community with64 "members" whose binary codes each represent a different radiationpattern. Community members compete to pass on their "genetic material"-bitsof binary code-to the next generation. Each pattern's viability isdetermined by mathematical criteria which favor radiation plans thatirradiate the prostate efficiently and knock out cancerous cells whilesparing vital organs. In a game of virtual natural selection, binary codethat embodies these qualities survives and multiplies, while poor codeperishes.
Over the course of 200 generations the community evolves, with codescoming together randomly, combining their genetic material, and evenmutating occasionally. In this way, the genetic algorithm creates a hugerange of potential solutions that it constantly sifts to find the bestcandidate for a treatment plan. "A genetic algorithm can look at a muchgreater range of options than we otherwise could. There might be certaincombinations that would never occur to a physicist to try," says Yu.
In just two minutes, PIPER presents to physicists and physicians the"winner"-the plan that the program decides is most likely to work best. Thespeed of PIPER, whose underlying technology has been patented by theUniversity, may make it possible to do the radiation planning right in theoperating room immediately before surgery, instead of several weeksbeforehand as is now standard. That's a tremendous advantage, says surgeonEdward Messing, M.D., chair of the Department of Urology, who has performedscores of brachytherapy procedures at the University's Strong MemorialHospital.
"The prostate you see in the operating room is never the same oneyou saw three weeks previously in your office," says Messing. "Hormonetherapy before surgery can shrink the prostate, for instance, and evenanesthesia can change the positioning of the pelvis the day of surgery. Thisoftentimes makes deviations from the now-dated plan necessary duringsurgery." Yu's goal is to make the process more precise. With a radiationtreatment plan compiled just minutes before the operation, the plan is morelikely to match what physicians actually confront in the operating room.
To commercialize the technology, the University has joined with RealTime Enterprises, a Rochester software engineering firm, to create a newcompany, RTek Medical Systems LLC. RTek will focus on the development of newradiation treatment planning systems based on Yu's algorithms. Any suchsystem must be tested rigorously before any application for marketingapproval will be submitted to the U.S. Food and Drug Administration, and theFDA must approve any product before it would become available.
At the University, the FDA has approved a clinical trial of thecurrent system as an investigational device on about 30 patients. Messing,Yu and their colleagues will study the radiation treatment plans recommendedby the system, along with the effects on tumor control, quality of life, andcomplications in patients who get the PIPER treatment compared to patientswho receive a commercially available treatment. The clinical study willbuild on more than five years of basic research that has been funded by avariety of sources, including the National Cancer Institute and the WhitakerFoundation.
New product development and support is part of the expertise thatReal Time Enterprises (RTE) brings to the partnership. The companyspecializes in software for medical equipment and devices like bloodpressure monitors, vision screeners, and other diagnostic equipment usedaround the world.
"This line of research really is compelling," says RTE PresidentRobert Ruppenthal. "People who treat cancer are visibly excited andimpressed by the potential of this technology, because it's so differentfrom what's out there today. Current treatment planning tools are trial anderror: They will tell a physician what dose a given seed distribution willdeliver, but they will not tell the doctor where to put the seeds."
RTek is the result of an accelerated effort by the University tocommercialize technology developed in its laboratories. "There are manyexciting technologies being developed in the Medical Center and across theentire University," says Jay Stein, M.D., senior vice president of theMedical Center and vice provost for health affairs. "This is one of thefirst of what we expect to be an ongoing series of technologies we plan tocommercialize."
Note: This story has been adapted from a news release issued by University Of Rochester.