SABCS: MDR Testing Still Not Ready for Prime Time

Michael Smith
SAN ANTONIO, Dec. 15 -- Breast cancer patients who overexpress the multi-drug resistance (MDR) gene are less likely to have a good response to chemotherapy than others, researchers found.A meta-analysis of 84 studies, involving a total of 4,944 patients, showed that a higher level of expression of the gene is associated with at least a 40% increase in the risk of a poor response, Bruce Trock, Ph.D., of Johns Hopkins University, said at the San Antonio Breast Cancer Symposium.
Action Points --->
Explain to interested patients that researchers believe overexpression of the multi-drug resistance (MDR) gene predicts a poor response to chemotherapy for breast cancer.
Note that this meta-analysis, however, says published studies are still too imprecise to make testing for MDR overexpression a useful tool.
Note, too, that this study was published as an abstract and presented orally at a conference. The data and conclusions should be considered to be preliminary until published in a peer-reviewed publication.
But testing for MDR overexpression is still "not ready for prime time," Dr. Trock said, because research so far has been too imprecise to make it a useful tool.
The study was unable to say how the gene is linked to overall survival or disease-free survival, Dr. Trock said, because of varying study designs and differing endpoints.
The quality of the data "limits our ability to look at a stronger endpoint like survival," Dr. Trock said. He called for researchers studying multi-drug resistance to agree on how they should report their data and what information should be included.
The study comes 10 years after an earlier meta-analysis of MDR studies, also by Dr. Trock and colleagues, that showed the prevalence of overexpression of the gene was 41%.
That earlier study also showed that overexpression of the gene was associated with a tripling of the risk of a poor response to chemotherapy.
The current analysis found:
The prevalence of MDR overexpression ranged from 37.8% to 66.8% depending on when in the clinical course the measurement was made and what drugs were used in treatment.
The relative risk for poor response was 1.4 (with a 95% confidence interval from 1.1 to 1.7) if gene overexpression was seen before treatment.
The relative risk for poor response was 2.0 (with a 95% confidence interval from 1.5 to 2.6) if gene overexpression was seen after treatment.
Dr. Trock said the current study adds to the evidence that the MDR gene product is a drug pump. He said it's known that drugs derived from natural sources -- such as the anthracyclines and taxanes -- are most affected by the gene product.
In the study, he and colleagues found that overexpression before treatment with "MDR substrates" was 52%, compared with 67% for women with normal expression.
If MDR expression was measured after treatment with MDR substrate drugs, they found, the comparable figures were 48% and 72%.
The main problem with MDR as a clinical tool is that "it's not either-or" in that some women with MDR overexpression respond well to therapy, said Aman Buzdar, M.D., of M. D. Anderson Cancer Center, who was not involved in the study.
Dr. Buzdar said testing for MDR overexpression offers "some useful information" on a group basis but "on an individual basis it's not useful."
"You wouldn't do a test and make a decision based on the test at the present moment," he said.
The study was supported by Bristol-Myers Squibb. Dr. Trock reported financial links with Bristol-Myers Squibb.
Primary source: Breast Cancer Research and Treatment
Source reference:Trock B, et al "Multidrug resistance and breast cancer: a meta-analysis of MDR1 and its clinical significance" Breast Cancer Res Treat 2007; 106 (Supp1): Abstract 37.