Affymetrix, based in Santa Clara, Calif., has created a p53 chip that measures slightly less than 13 millimeters square-bigger than a thumbtack, but smaller than the standard issue 32-cent stamp. Engineers at Affymetrix have subdivided this real estate into a checkerboard of 20,000 “probe cells,” each bristling with a uniform carpet of millions of identical DNA probes measuring 18 base pairs long. Using fluorescently labeled reagents, prepared DNA from a tumor can be washed over the chip, and extremely sensitive scanners are programmed to detect minuscule variations in intensity in the checkerboard pattern-which arise from slight genetic changes in p53. Once the DNA has been prepared, the test can be done in four hours.
Although p53 is only one of numerous genes implicated in the evolution of a tumor, it has already been thrust into a prominent role in experimental cancer treatments. In April of 1996, for example, Oncormed began using the Affymetrix chip to perform “molecular staging”-that is, assessing the status of a tumor-in patients with head and neck cancers prior to clinical testing of an experimental form of gene therapy developed by Onyx Pharmaceuticals. The therapy is in Phase II testing, and some patients have responded favorably in preliminary results.
The ability to do molecular profiling, according to Leslie Alexandre, a vice president at Oncormed, allows oncologists to determine both the virulence of a tumor and the extent of its metastatic spread. “If there is a p53 mutation in the tumor,” she says, “then you would look at the lymph nodes and look for the fingerprint to see how far the tumor has spread.” Oncormed has reached agreements with Rhone-Poulenc Rorer and Schering-Plough to do p53 testing associated with gene therapy trials.
The potential significance of such testing goes well beyond staging individual tumors. Drug companies are intensely interested in ways of predicting which patients are likely to respond to chemotherapy, and there is some evidence that p53-and molecular profiling like it-may help identify patients likelier to respond. Not only will this allow drug companies to achieve higher response rates in drugs being tested, but it may, Alexandre says, provide a way to “resurrect” drugs that fail in Phase III trials by identifying a small group for whom the drug is very effective.
As with many new technologies, there may be a social cost to molecular profiling. The most precious resource available to any cancer patient is not money, but hope. Tests based on the p53 chip and related chips may turn out to be one more sophisticated, definitive, and molecular way of telling a patient there is no hope.