Aug 3 2004
Scientists have developed a new screening technique to help them look for genes that change patients' responses to cancer drugs and other medications.
Researchers looking for such connections confront an enormous hunting ground of approximately 33,000 human genes. Normally their only options for mounting a search in such a vast field are either to rely on anecdotal reports of dramatically altered patient reactions, or to conduct extensive surveys of the genes for all the proteins known to interact with a given drug.
The new approach lets nature and a robotic screening system do the majority of the hunting for them. In their initial test, which will be described in the August 10 Proceedings of the National Academy of Sciences, investigators rapidly found potential connections between two chemotherapy drugs and two regions of human DNA that contain approximately 100 genes each. The study is currently available online.
"This isn't the answer to everything in terms of finding these links, but it's an important breakthrough," says senior investigator Howard L. McLeod, PharmD., associate professor of medicine, genetics and of molecular biology and pharmacology. "This approach is very likely to allow us to find links between pharmaceuticals and genes that we never would have been able to anticipate."
McLeod is an expert in pharmacogenetics, a new field where scientists are learning that a person's genes can dramatically influence the effectiveness of medications. These differences can change a drug that is a lifesaver for some patients into a toxin for others, or influence whether a medication provides little benefit or is a remarkably effective treatment. By identifying genetic factors that affect patients' responses to drugs, scientists hope someday to enable clinicians to customize treatment plans.
McLeod and colleagues in the Division of Biostatistics took advantage of cell lines established as part of the effort to map the human genome. Researchers at the Centre d'Etude du Polymorphisme Humain in Paris, France have created approximately 700 human cell lines from multiple generations of large families in Utah, France and elsewhere.
Washington University scientists exposed cells from more than 400 of the lines to varying doses of two chemotherapy drugs, 5-fluorouracil and docetaxel. The cells were non-cancerous, but chemotherapy can kill both cancerous and non-cancerous cells. Chemotherapy is given as a treatment for cancer because cancer cells are generally more sensitive to its effects, but many factors, including the genetics of the cells' non-cancerous precursors, can influence that sensitivity.
Scientists used a robotic screening system to look for cell lines with increased sensitivity to the drugs, demonstrated by higher numbers of cell deaths in response to low drug doses. The robot also highlighted cell lines with high resistance to the drugs where few or no cells were killed.
In the future, patients whose cells are particularly sensitive to chemotherapy may be able to be treated with relatively low doses, reducing side effects. Patients whose cells are particularly resistant may need special or added medications to assure a good outcome.
Scientists already know a great deal about inheritance of genetic markers among the cell lines. This enabled Washington University researchers to compare and contrast the genetics of a cell line with altered sensitivity to cell lines from other family members and from multiple generations of the same family. Children get a random mixture of genes from both parents, so both genetic markers and changes in sensitivity are sometimes passed from parent to child and sometimes aren't. When a particular genetic marker is consistently passed from parent to child at the same time as a change in sensitivity, that tells scientists they need to look near the marker for a gene that changes sensitivity.
The initial test of the new approach found connections between increased sensitivity to the drugs and areas on chromosomes 5 and 9.
"That part of chromosome 9 turned up in an earlier search we conducted for these genes," McLeod says. "Lightning's struck twice there now, so we're definitely going to be looking for a gene that affects sensitivity in this region."
McLeod's group already has applied the new screening technique to six more cancer drugs, but he says they've just begun to find ways to use the new approach.
"This is not a cancer research technique, it's a drug research technique," says James W. Watters, Ph.D., lead author of the study and instructor of medicine. "We want to find ways to look at new endpoints -- for example, how thoroughly does a drug hit its target of interest, or how much can it slow growth or other cellular processes? Then we'll be able to look at genetic effects on medications for a range of disorders."