Received April 15, 2004
Accepted June 29, 2004

Chemotherapy Sensitivity and Resistance Assays: A Systematic Review

David J. Samson *, Jerome Seidenfeld , Kathleen Ziegler , and Naomi Aronson

From the Blue Cross and Blue Shield Association Technology Evaluation Center, Washington, DC.

* To whom correspondence should be addressed. E-mail: david.samson@wro.bcbsa.com

Purpose: This systematic review evaluates evidence comparing therapy guided by chemotherapy sensitivity and resistance assays with empiric chemotherapy, emphasizing survival outcomes.

Methods: Prospective studies were sought comparing patients treated contemporaneously by assay-guided chemotherapy and empiric therapy. An initial MEDLINE search and a search performed by a Working Group of the American Society of Clinical Oncology were reviewed with attention to prespecified study selection criteria.

Results: This review identified 10 studies meeting selection criteria, plus one retrospective study, using seven different assays. Only two studies randomly assigned patients to assay-guided treatment or empiric treatment. Five of nine nonrandomized studies found significantly higher response rates for patients who received assay-guided therapy compared with those treated empirically. One of the two randomized trials found a significantly higher response rate in the assay-guided group. Four additional studies found response rates favoring assay-guided therapy, but comparisons did not achieve statistical significance. Two nonrandomized studies found overall survival to be significantly improved with assay-guided therapy. One randomized study used a cross-over design that made it difficult to determine whether survival differed between groups, while the other randomized trial found no difference in survival. Six studies provided no comparison of groups on baseline patient characteristics. Only one study reported adverse events data.

Conclusion: While higher response rates for assay-guided therapy have been observed, differences may be attributable to bias or confounding. Little evidence on survival is available. These results do not establish the relative effectiveness of assay-guided treatment and empiric treatment. Randomized trials are needed.