Search

BackgroundThere are a number of equally efficacious chemotherapy options for the treatment of women with endometrial cancer, all of which work in only a subset of those women with this disease. An in vitro assay performed before therapy initiation to identify the drug(s) most likely to be effective for the individual patient would have clinical utility. Such an assay should yield response rates similar to those found in treated patient populations. The purpose of this investigation was to determine whether the patterns of in vitro tumor response rates as determined by ChemoFx are consistent with expected population response rates.MethodsNine hundred twenty-three tumor specimens from patients with high-risk early-stage, advanced stage, or recurrent endometrial cancer were sent for testing with the ChemoFx drug response marker from August 2, 2006, to August 31, 2009. Tumors were categorized as responsive (R), intermediately responsive (IR), or nonresponsive to each drug or combination tested. Response rates from clinical trials were identified and compared with the corresponding in vitro response rates.ResultsOf the 923 specimens received, 759 (82%) were successfully tested by ChemoFx. Of these, 755 were tested for at least 1 of 5 National Comprehensive Cancer Network-recommended endometrial cancer drugs. The response rates (R+IR) for these drugs were as follows: 66% carboplatin-paclitaxel, 48% carboplatin, 37% cisplatin, 23% doxorubicin, and 36% paclitaxel. Moreover, 20% of tumors were pan-sensitive (R or IR) to all 5 regimens tested, 27% were pan-resistant (nonresponsive), and 53% showed different degrees of response to different drugs.ConclusionsChemoFx in vitro response rates were consistent with published population response rates, and the ChemoFx drug response marker may provide clinically useful information to better optimize individual chemotherapy for treatment of women with endometrial cancer.

The study objective was to determine the effectiveness of a phenotypic chemoresponse assay in predicting response to chemotherapy measured by progression-free interval (PFI) in a retrospective series of ovarian cancer patients whose tumor specimens had been tested with the ChemoFx assay. A statistically significant correlation between assay prediction of response and PFI was observed in 256 cases with an exact or partial match between drug(s) assayed and received. In 135 cases with an exact match, the hazard ratio for progression of the resistant group was 2.9 (confidence interval [CI]: 1.4-6.3; P0.01) compared to the sensitive group and 1.7 (CI: 1.2-2.5) for the intermediate compared to the sensitive group. The median PFI for patients treated with drugs assayed as resistant was 9 months, 14 months for those with drugs assayed as intermediately sensitive, and PFI had not been achieved for those with drugs assayed as sensitive. These data indicate that the ChemoFx assay is predictive of PFI in ovarian cancer. As the majority of ovarian cancers display different degrees of response to different chemotherapy agents ex vivo, the incorporation of assay information into treatment selection has the potential to improve clinical outcomes in ovarian cancer patients.