By David N. Leff
It didn't take a rocket scientist to point out the connection between lung cancer and cigarette smoking. Less of a no-brainer is the fact that tobacco also causes cancer of the urinary bladder.
Carcinoma of the lungs tops the charts of cancer deaths in the U.S., whereas bladder cancer - with 54,000 newly diagnosed cases in 1997 - ranks fourth among malignancies in men, and eighth among women. But caught in time, that well-hidden neoplasm isn't such a devastating killer.
So what else is new? Early diagnosis is the lifesaving secret in many cancers, but few of them tip their hand in timely fashion. One that does is cervical cancer, in which the borderline cells between cervix and uterus are easily scraped off by a Pap smear. By a similar token, DNA can be readily harvested from the human mouth by scraping off exfoliating skin cells from the cheeks.
No such luck for scanning the bladder's interior. That organ lies snugly out of sight in the body's pubic area. Scoping its walls for potential malignancy involves invasive procedures, reminiscent of transurethral biopsy or colonoscopy. (See BioWorld Today, Oct. 2, 1996, p. 1.)
Now a diagnostic test for bladder cancer, as simple and noninvasive as a Pap smear, is completing an initial 450-patient clinical trial in Europe and the U.S. It's based on the finding that as urine gushes through the bladder en route from kidney to excretion, it washes off cells shed from the bladder wall.
The test was developed by medical oncologist and otolaryngologist David Sidransky, at the Johns Hopkins University. It's based not on the oncogenes, such as p53, found in the nuclei of tumor cells, but rather in the mutated mitochondrial organelles that swarm in the cytoplasm of those cells.
"We are in the process of finishing a multi-institutional trial of that urinalysis test," Sidransky said, "on 250 bladder cancer patients and 200 controls. The samples have all been collected, the molecular assays are being run, and by late spring we should have the answers. Additional pilot studies done here and in Europe," he added, "have confirmed about a 90 percent sensitivity and, so far, 100 percent specificity on the test."
Bladder Cancer Scan Heads To Market
"Future bladder cancer scanning," he went on, "has been picked up by the Belgian company Virco, with offices in Mechelen, Belgium, the UK and U.S. Virco plans to launch this test commercially, we hope in the summer, once we finish the ongoing trial. And our mitochondrial research," he observed, "was supported by an NIH grant, so it's not in anybody's hands right now.
"One of the major areas that we wanted to work on," Sidransky continued, "is lung cancer. We've had tremendous difficulty, because of the needle-in-the-haystack problem. The cancer cells that come out in the lung secretions, whether they be from bronchoalveolar lavage or sputum, are tremendously diluted by normal nonneoplastic cells, compared to bladder tumors. In bladder it's a piece of cake because 90-plus percent of the material that comes out in the urine is cancerous. So it's much easier to make DNA diagnoses there."
He continued: "Not only did we find that most bladder tumors seemed to have mitochondrial mutations - which was not previously known - but also that kind of solved the needle-in-the-haystack problem, and allowed us to easily detect the cancer cells."
Sidransky cited estimates that "each mammalian cell contains several hundred to thousands of mitochondria, each of which carries one to 10 genomes." Reflecting their presumed evolutionary origin as free-living prokaryotic organisms that moved brazenly into eukaryotic life forms eons ago, each mitochondrial genome - 16.5 kilobases long - encodes 22 transfer RNAs, two ribosomal RNAs, and 13 respiratory chain (energy-supplying) subunits.
Besides analyzing DNA from mutated mitochondrial genomes in bladder and lung bodily fluids, Sidransky is also applying this technique to saliva from head-and-neck cancers. "These are prevalent in Southeast Asia," he pointed out, "but also in smokers in the U.S. Their tumors are in the oral cavity, the mouth, the back of the throat, and down into the voice box."
Sidransky is senior author of a progress report in the current issue of Science, dated Mar. 17, 2000. Its title: "Facile detection of mitochondrial DNA mutations in tumors and bodily fluids."
In experiments comparing 14 lung, 14 bladder and 13 head-and-neck tumors with their respective bodily fluids - lavage, urine and saliva - the co-authors found that the mutated mitochondrial DNA "was 19 to 220 times as abundant as nuclear p53 DNA. But our surprise," Sidransky told BioWorld Today, "was that most tumors seem to have these mutations - which is kind of a new finding."
From Specific Fluids To All-Purpose Blood Test
As his next step, he indicated, "We are now starting to explore the one area we haven't worked out. In addition to bodily fluids is the issue of simple blood tests. What we've learned is that a lot of cancer patients have circulating DNA in their blood, just as they have proteins. And just like the PSA [prostate-specific antigen] test, one can look at normal DNA changes in blood.
"We've documented this," he went on, "for many oncogene mutations in cell nuclei - p53, ras, microsatellites, methylation, things on the promoter. We know they occur, and we can detect them in blood. Our next step will be to look at the feasibility of a simple blood test to detect the mitochondrial mutations in the cytoplasm outside the nucleus. For all the normal places where you take fluid, we would be able to potentially use this simple blood test."
Sidransky envisions the use of mitochondrial DNA for cancer screening, much like mammography or Pap smears do now. "For example," he suggested, "a smoker at higher risk of developing lung cancer could provide a sputum sample. This would give a baseline description of the individual's mitochondrial genome. In subsequent visits over time, if new sputum samples reveal mutational changes, we would be in a better position to intervene while the cancer was still in an early and curable stage."