From Our 2011 Archives

Common Breast Cancer Gene Test May Be Flawed, Study Says

By Amanda Gardner
HealthDay Reporter

FRIDAY, Nov. 4 (HealthDay News) -- A widely used breast cancer test may not be accurate in identifying a gene that is critical in determining which life-saving treatment a woman should get.

The Oncotype DX, marketed by Genomic Health, results in a number of false-negatives for the HER2 gene, according to a study published Oct. 18 in the Journal of Clinical Oncology.

Women who test positive for the gene typically receive the targeted therapy Herceptin (trastuzumab) along with other medications to reduce their chances of recurrence and death.

"Women could be getting the completely wrong treatment," said study lead author Dr. David J. Dabbs, a professor and chief of pathology at Magee-Womens Hospital of the University of Pittsburgh Medical Center.

But, according to Dr. Lori J. Goldstein, director of the Breast Evaluation Center at Fox Chase Cancer Center in Philadelphia, the Oncotype DX test, which actually measures 21 different genes, was not designed to test for HER2, nor is it intended as a substitute for two other widely used and accepted lab tests.

"None of us would request Oncotype for the sole purpose of getting HER2," she said. "We're usually ordering it for other reasons and, as part of that test, we get HER2."

Oncotype has been shown to be reliable in identifying estrogen receptor-positive and progesterone-positive tumors, she added.

The only two valid assays for HER2 are immunohistochemistry (IHC) and fluorescent in situ hybridization (FISH). Both are approved by the U.S. Food and Drug Administration (FDA).

Genomic's Oncotype employs reverse transcription polymerase chain reaction (RT-PCR) to check for HER2.

HER2 is overactive in 15% to 20% of breast cancers and usually signals more aggressive disease, according to background information in the study.

To test the accuracy of Oncotype DX in assessing HER2 status, the researchers compared results from all three tests for 843 patients who had had samples tested at three laboratories, one in Pittsburgh and two in Ohio.

When results were negative with IHC and FISH, they were also generally negative with Oncotype.

But all 23 equivocal results as reported by IHC and FISH came out negative with Oncotype, the investigators found.

Only 28% of positive IHC and FISH results also came up positive with the Oncotype, while 33% came out as equivocal and 39% as negative.

"Anyone who has had this test performed needs to make sure that their hormone-receptor analyses were done by other FDA-approved methods and not rely upon this test alone," said Dabbs.

Overall, though, only a small proportion of tumors among the 843 women were HER2-positive, noted Goldstein, which could have biased the results.

And, in a statement provided to HealthDay, Genomic Health disagreed with the findings, proposing that the "conclusions appear to be one-sided" and require additional data.

The discordances between tests are "not uncommon" and because of those discrepancies the company began including RT-PCR measurement of HER2 in Oncotype "with the goal of providing added clarity in cases where HER2 results by IHC and FISH are uncertain or conflicting," according to the statement.

HER2 status "should be assessed in all patients by IHC and/or FISH," the company also stated.

A 2010 study funded by Genomic Health found that RT-PCR results actually did agree with FISH results.

MedicalNewsCopyright © 2011 HealthDay. All rights reserved.

SOURCES: David J. Dabbs, M.D., professor and chief, pathology, Magee-Womens Hospital, University of Pittsburgh Medical Center; Lori J. Goldstein, M.D., director, Breast Evaluation Center, Fox Chase Cancer Center, Philadelphia; Oct. 11, 2011, statement, Genomic Health; Oct. 18, 2011, Journal of Clinical Oncology