ORLANDO, Fla. (GenomeWeb News) – Researchers are not only coming up with strategies for integrating genomic strategies into existing clinical trials, but are also using these analyses to inspire new trials, Washington University Genome Center Co-Director Elaine Mardis reported at the American Association for Cancer Research annual meeting here yesterday.
Speaking during a session on recent findings from genomic analyses of tumors, Mardis discussed progress that she and her collaborators are making on this front — and described a potential plan for a clinical trial to evaluate the use of genome sequencing for identifying relevant mutations in breast cancer, predicting drug response, and guiding treatment.
As part of an American College of Surgeons Oncology Group clinical trial, Mardis and other Washington University researchers have already sequenced the genomes of 50 estrogen receptor positive breast cancers and matched normal blood samples — findings that Washington University oncologist Matthew Ellis presented at the AACR meeting earlier this week.
The tumor genomes represent DNA from core biopsy samples taken prior to aromatase inhibitor treatment, while DNA from blood samples was used to generate matched normal genomes for each patient. Both tumor and normal genomes were sequenced via paired-end sequencing with the Illumina platform.
The genome data is being used to learn more about breast cancer biology, Mardis explained, but those involved in the study are also using it to look for genomic profiles that can be used to predict patient response to aromatase inhibitors — which are typically used to treat ER-positive breast cancer prior to surgery.
Of the 50 women tested, 26 had tumors that responded to aromatase inhibitor treatment and 24 had unresponsive tumors, as measured by a Ki67 protein-based assay used to gauge proliferation index in the tumors following aromatase inhibitor treatment.
Along with genome sequence data, the team also has SNP array data for the tumor-normal samples, as well as Agilent array CGH data, information on surgical results, and more, Mardis added.
Overall, she noted, researchers are finding fewer mutations genome-wide in the tumors that respond to aromatase inhibitor treatment, while the aromatase inhibitor resistant tumors are more complex and contain more mutations.
For instance, researchers have tracked down more than 800 affected genes in the treatment resistant group so far, along with hundreds of somatic deletions and translocations and some 3,100 structural variants.
In the future, the team plans to continue comparing the genomic signatures in the aromatase inhibitor sensitive and resistant genomes, with a focus on impacted pathways and an eye towards designing a predictive assay and algorithm for drug response in the ER-positive breast cancer patients, Mardis said.
Eventually, they hope to take a crack at clinical trials to test the use of genomic profiling to guide treatment for women with estrogen receptor positive breast cancer, she added.
At the moment, the design for such a trial would likely involve doing whole genome sequencing and analyses of normal tissue and pre-treatment tumor samples over the span of four weeks to find genomic information for predicting aromatase inhibitor response and come up with targeted therapies based on druggable gene mutations detected in tumors genomes, if applicable.
Based on a sequencing output of 300 billion base pairs per flow cell over a 10 day sequencing run, Mardis noted, it should be possible to get 60 times whole tumor genome coverage and 30 times matched normal genome coverage for each clinical case, as well as one sequencing lane of data each for tumor and normal exomes and tumor RNA.
In the meantime, Mardis said, she and here colleagues are also doing genome sequencing as part of another American College of Surgeons Oncology Group clinical trial looking at trastuzumab treatment for HER2-positive breast cancer.