Welcome to PeerView

SECTION IV: Biological Markers, Genetic Expression

Abstract 1700
Gene Expression Profiles for Docetaxel Chemosensitivity
Jenny C. Chang, MD, Baylor College of Medicine, Houston, Texas

Systemic chemotherapy is known to improve survival in patients with operable breast cancer but there are no available methods to identify patients who will benefit from chemotherapy versus those who will not. A group of investigators from Baylor College of Medicine hypothesized that patterns of gene expression could identify breast cancers that are sensitive or resistant to specific chemotherapy.

"If this hypothesis is valid, then patterns of gene expression could be used to select appropriate chemotherapy for women with breast cancer," noted Dr. Jenny Chang, who presented the results at the 38^th annual ASCO meeting.

The study had three aims:

To identify gene profiles that might predict sensitivity or resistance to Taxotere (docetaxel).
To confirm differential expression of selected genes by reverse transcription polymerase chain reaction (RT PCR) and immunohistochemistry.
To validate the expression profiles in an independent cohort of patients.

Dr. Chang presented the results of the first two aims.

Trial Design
Six core biopsies were obtained from 40 patients with locally advanced breast cancer before treatment with neoadjuvant Taxotere (100 mg/m² every three weeks for four cycles). Dr. Chang assessed patient responses by comparing clinical bi-dimensional tumor measurements before and after treatment.

"Biopsies were performed in the outpatient clinic. No complications were observed, and patient acceptance was high," Dr. Chang reported.

Response to Taxotere neoadjuvant therapy was evaluated in 21 patients. Four (19%) had clinical complete responses and eight (38%) had partial responses, for an overall response rate of 57%. Eight (38%) had stable disease and one (5%) had progressive disease.

Dr. Chang reported genetic data for the first 19 patients. Of the 2,332 genes selected for further analysis, 823 underwent unsupervised clustering, which Dr. Chang defined as an unbiased sorting of samples based on similarities of genes within each cluster. "Unsupervised clustering is important in identifying clusters of patients who share similar patterns of expression. Only then can patterns be observed that predict a favorable response to Taxotere," she explained.

Two main groups of genes were identified based on gene expression patterns and labeled as: Cluster A (eight tumors) and Cluster B (11 tumors).

Results
Unblinded results showed that almost all the tumors in Cluster A were sensitive to Taxotere, while most of those in Cluster B were resistant, Dr. Chang reported. Sensitive tumors were defined as those with less than 25% residual disease and resistant tumors were defined as those with greater than 25% residual disease. Two patients in Cluster B were sensitive; one of them was misclassified and the other was actually found to be resistant.

"Unsupervised clustering, without any consideration for treatment response, accurately divided patients into two groups — one sensitive to Taxotere and the other less responsive or completely resistant," she noted.

Dr. Chang and colleagues showed that elevated expression of beta tubulin is a marker for some, but not all resistant tumors. Non-responders also had elevated levels of inflammatory response genes, as well as elevated proliferation and oncogene expression levels.

"Breast cancer is heterogeneous and it appears that mechanisms other than beta tubulin expression are associated with Taxotere resistance. This profile of Taxotere resistance differs from expected profiles of general chemoresistance," she said.

Low levels of beta tubulin expression were observed in sensitive tumors. These tumors also had a higher expression of mitochondrial proteins and motility-related microfilament proteins. This was consistent with an apoptosis-inducing mode of action for Taxotere.

Currently, Dr. Chang's team is confirming array results by RT PCR and immunohistochemistry and they have found good correlation for the first 15 genes analyzed.

"Gene expression arrays can identify profiles that predict sensitivity and resistance to Taxotere chemotherapy. Breast cancer is heterogeneous, and different genetic properties may lead to resistance or sensitivity. Gene profiles that predict response may be drug-specific, so that establishing and validating these profiles could enable selection of chemotherapy to optimize treatment and minimize toxicity for women with breast cancer," Dr. Chang concluded.

Key Points In Focus:

While estrogen and progesterone receptor status can be used to identify patients who will respond to hormonal cancer therapies, no predictors of response have been identified for systemic chemotherapy.
Molecular profiles of gene expression for tumors that are resistant or sensitive to Taxotere can be used to predict which patients will respond to this agent.
Identifying patterns of gene expression specific for a particular chemotherapy may be able to reduce unnecessary treatment for women with breast cancer, as well as associated toxicity and costs.

--[ TOP ]--