AUSTIN, Texas (GenomeWeb) – A research team from the Sidney Kimmel Cancer Center at Johns Hopkins University is collaborating with Cepheid to adapt a cell-free methylated DNA assay for early diagnosis and monitoring of breast cancer to a cartridge format that could run on Cepheid's GeneXpert system.
Sara Sukumar, who is spearheading the Hopkins team, discussed the effort at a corporate-sponsored workshop at the Association for Molecular Pathology annual meeting held here this week.
In her presentation, Sukumar noted that there is an immediate need for a method to accurately assess tumor burden in metastatic breast cancer patients. A physical examination and radiological imaging help but do not always reflect tumor burden, she said.
As dozens of research groups and companies have come to realize, cell-free DNA has shown the most promise for early diagnosis and monitoring. However, several issues still make this approach challenging. For instance, it has been shown that a multi-marker panel is much more sensitive than a single marker, but creating such a panel takes a lot of time and trial and error.
In addition, DNA levels in patient serum or plasma is extremely low and variable; tumor DNA is just a fraction of the total circulating DNA; and "the size of the haystack" in which to find the proverbial tumor DNA needle keeps changing, as wild-type DNA often fluctuates in patients due to a number of lifestyle factors. For this reason, Sukumar's lab favors using methylated genes as biomarkers, as epigenetic alterations are among the most common genetic abnormalities.
As reported by GenomeWeb last year, Sukumar and colleagues previously developed a two-step multiplex methylation-specific PCR assay, called cMethDNA, as well as a panel of 10 genes that were found to be methylated in the serum of breast cancer patients, work that was published in Cancer Research. They also showed that the method performed well in training and test sets.
The researchers next embarked on a clinical utility study, called TBCRC005, to find if higher levels of tumor-related methylated DNA is associated with worse outcomes in metastatic breast cancer patients.
For this study, conducted between 2007 and 2009, the group collected serum samples from 141 patients diagnosed with metastatic breast cancer and found that there was an association between week-four cumulative methylation index (CMI) and progression-free survival. The assay was also able to predict response to certain therapies and predict the course of disease.
Sukumar noted that her lab has submitted this work — which she said is "the most comprehensive study to date of DNA methylation analysis in the blood of metastatic breast cancer patients" —for peer-reviewed publication. However, in the meantime, it had been seeking partners to simplify and commercialize the assay because, even though it performs well, the workflow is "not for the faint of heart."
Enter Cepheid, with which the Hopkins team has now established such a partnership. Sukumar said the partners have established feasibility and are now optimizing a Cepheid Xpert cartridge to run the test.
The GeneXpert methylation workflow, she noted, would comprise the steps of nucleic acid extraction, bisulfite conversion, desulphonation, target re-purification, and multiplexed methylation-specific PCR directly on an Xpert cartridge, with a turnaround time of three hours or less. The cartridge would be designed to detect methylated tumor DNA targets using both formalin-fixed paraffin-embedded tissue samples and plasma.
An initial comparison between the Johns Hopkins manual cMethDNA method and the prototype Cepheid cartridge has yielded a "very good correlation," Sukumar noted. She added that Cepheid's approach "looks like a promising alternative to more labor- and skill-intensive methods."
One major advantage of partnering with Cepheid, Sukumar said, is that the potential assay could be run on the more than 9,000 GeneXpert systems currently installed worldwide. Many of these are in resource-poor areas of the world through the company's "high-burden developing country" and tuberculosis testing program, which means that the breast cancer assay "might be able to be used worldwide to distinguish between malignant and benign tissue in third-world countries," where such a test is sorely needed.
Finally, Sukumar said, her group and Cepheid are already working on extending the assay to even earlier stages of breast cancer and have created a prototype for applications to other cancers.
Cepheid has been moving toward the oncology space for a few years now, and last year at the AMP meeting updated its pipeline in this area. At the time, a company executive noted that Cepheid was targeting four product applications in oncology — early detection, monitoring, therapy guidance, and biomarker discovery — and that it had tests in development for BCR-ABL monitoring, bladder cancer, and breast cancer.
Many of these tests will presumably be designed for the company's so-called "honeycomb" microwell configuration, which would confer a much higher level of multiplexing than the traditional Xpert cartridges in order to accommodate multi-marker gene panels.