. .



Science & Technology
Cancer detection

Column
For the record
Profile
Malika Zartaj Baig
Polotics & Polocy
Pak-US relations
New plan for democracy
Financial indiscipline
Science & Technology
Online investing
Fruit fly genome
Cancer detection
Interview
Harvest Group
InformationTechnology
Organizational reforms for E-Commerce
Society
Managerial effectiveness

From Diana J. Choyce
Apr 03 - 09, 2000

One of the most important needs in treating cancer is in early detection. The sooner the disease is found, the more effective the treatment can be and the better chance of survival. Scientists at Johns Hopkins University have been developing a chemical testing method as part of a small pilot study with good results. If further research shows the method to be stable and effective, it could be on the market in 3 to 5 years.

The test is non-invasive and very simple. By testing a patient's saliva or urine or mutated DNA particles, cancer could be detected years before the onset of symptoms. Researchers are focusing on fluid samples taken from 20 patients suffering from cancers of the mouth, throat, voice box, lungs and bladder. "The test had 100 percent sensitivity. If the mutation was present in the tumor, we could detect it in the bodily fluid each time,'' said Prof. David Sidransky, lead author of a study which appeared in the journal Science. "The need to be able to detect cancer early is obvious. But current tests are predicated on people coming in with symptoms, which means the cancer is already at an advanced state.'' As an example of how effective the gene testing methods can be, the Johns Hopkins team tested a urine sample taken from the late Vice President Hubert Humphrey and found cancerous mutations in the DNA present. Humphrey gave the urine sample in 1967, nine years before he was diagnosed with bladder cancer. He died in 1978. "So, it's not unusual to be two, three, four or even more years ahead of time in doing the diagnosis. It just tells us how long the cancer process has been going on before we actually establish a diagnosis,'' Sidransky said.

Cancer will occur when the DNA in the nucleus of a cell mutates and causes a disruption in cell division. When a tumor dies, it then sends mutated DNA into the bloodstream. The "key" is to find the tiny particles of protein through molecular analysis. The Johns Hopkins team has already invented tests that look for mutated nucleic DNA. These are currently being developed by the European biotech company Virco. However these new "fluid" tests are 200 times more likely to detect cancer. Instead of searching for nucleic DNA, they search for the far more numerous DNA strands called mitochondria, which exist in the hundreds of thousands. Mitochondria are parts of human cells that provide energy for the cell. Like the cell nucleus, they contain DNA, the complex set of instructions for the cell. But unlike the nucleus, which gets DNA from each parent, mitochondria are inherited only from the mother. Researchers have known that mitochondria were important in cancer, as far as giving energy to the tumor, Sidransky said. But we never imagined that we'd find these mutations. "When you're looking for a needle in a haystack -- one cancerous cell among a lot of normal cells -- if it only contributes one or two mutant DNA copies, it's hard to find. But if the cell contributes hundreds of thousands, it's much easier. With additional research, we expect to be able to identify mitochondrial mutations through a simple blood test,'' Sidransky said. With this method, researchers said they can detect lung cancer in sputum, head and neck cancers in saliva, and bladder cancer in urine.

There are several possible applications for these new tests. "They're onto something useful ... an improved way of diagnosing cancer, said Dr. Garth Anderson, a senior researcher at Roswell Park Cancer Institute in Buffalo, N.Y. Anderson, who was not part of the research team, noted that the findings could help physicians treating cancer patients who want to see if the disease has returned after therapy. Between 1992 and 1996, the National Cancer Institute recorded 57.0 cases of lung cancer per 100,000 Americans, 16.8 cases of bladder cancer, 10.3 cases of oral cancer and 4.8 cancers of the larynx nose or trachea. Sidransky envisions a day when smokers, for example, get a baseline check on their lung fluids and then have it rechecked annually, looking for any changes that may mean cancer. And there is the possibility one day of kits for people to send in samples of body fluids to labs for testing, Sidransky said. He expects patients to begin asking their doctors for such tests within a couple of years, but fears it may take longer for insurance companies to begin paying for them. I hope that the tests will be cheap enough, in the $150 range to $200 range, that patients will get it anyway, he said.

The next thing, though, is to perfect the technology for use in the general population, which could take six months to a year, he said. The team also wants to extend its research to a much larger population. Its paper was based on tests in about 15 patients each with bladder, lung and head and neck cancer.This is obviously an important caveat, he said. We've only tested a few patients so we'd like to test a much larger number of patients. The work was funded by the Early Detection Research Network of the National Cancer Institute, which can help take the technology quickly to clinical trials, Sidransky said. The new test would not replace the need for biopsies and other invasive means of treating cancer, but would allow doctors to screen people for cancer and then direct treatment toward those who actually need it.