College of American Pathologists
Printable Version

  Findings of an evidence-based assessment


CAP Today



May 2008
Feature Story

In an evidence-based assessment of the GeneSearch assay conducted for the Technology Evaluation Center (TEC) of the Blue Cross and Blue Shield Association, the main conclusion was, “At present there is not enough evidence to determine the impact of the assay on health outcomes,” says Barbara M. Rothenberg, PhD, a senior scientist at TEC. (The report can be accessed at Dr. Rothenberg was working with the one published study on GeneSearch and material presented to the Food and Drug Administration.

“There were a number of areas that FDA was concerned about,” Dr. Rothenberg told CAP TODAY. Two of those concerns formed the basis of the postmarketing studies the agency called for. One question is pragmatic and relates to timing: How long does it take to do the test in real life? For this purpose, the FDA required a study to determine whether GeneSearch extends operating time. The other issue: “Trying to understand whether false-positive results with GeneSearch are really false-positives or whether the assay is detecting metastases in parts of the lymph node that aren’t examined by histology,” Dr. Rothenberg says. “Would you get the same result if you did more extensive H&E sectioning?” In the assessment, she wrote, “While the researchers provide suggestive evidence, further research is needed to confirm or deny this hypothesis.”

“The real value of this assay,” Dr. Rothenberg says, “if it ends up performing well, is in avoiding second surgeries for women who have a positive assay but a negative frozen section and would have had to go back for second surgery. On the other hand,” she continues, “the risk is that if there are more false-positive results with the assay, there will be more axillary lymph node dissections [ALND] on patients who don’t really need them, which have some morbidity.” Unfortunately, Dr. Rothenberg adds, it is impossible to calculate this benefit-risk ratio because it is not clear how many apparent false-positive results with GeneSearch are truly false-positive. This issue may be addressed in part by one of the FDA-mandated postmarketing studies; additional high-quality studies would strengthen the evidence base.

Dr. Rothenberg identified another concern in the assessment: “If GeneSearch has higher sensitivity because it detects more micro­meta­stases (which is not clear), then the clinical significance of finding these patients may not balance the harms of ALND. Even though current guidelines recommend that ALND be performed if either [micro- or metastasis] is detected, they also state that the prognostic significance of micrometastases is unknown, as is the importance of ALND in these patients.” About 30 percent of GeneSearch’s additional sensitivity may be owing to micro­metastases.

Finally, Dr. Rothenberg notes that the FDA reviewer analyzed ROC (receiver operator characteristic) curves of frozen section histology and GeneSearch; ROC curves show the tradeoff between the true-positive and false-positive rate. “They concluded that both tests may lie along same ROC curve,” she says. “This may mean that in some sense there is no real difference between GeneSearch and frozen section.” Perhaps the apparent higher sensitivity of GeneSearch, as well as its lower specificity, is not an intrinsic feature of the assay, but simply a reflection of the chosen cutoff.

—William Check, PhD

Related Links Related Links