By Monica Heger
This story was originally published on Aug. 30.
Researchers at Sweden's Uppsala University plan to sequence the exomes of 100 cancer patients before and after treatment to find biomarkers that predict drug responsiveness with the aim of eventually developing companion diagnostic tests.
The study, which will focus on patients with either colorectal cancer or chronic lymphatic leukemia, is being funded by SciLifeLab for SEK 4.5 million ($717,543) over the next two years, and will include researchers at the Royal Institute of Technology, the Swedish University of Agricultural Sciences, and Umeå University. SciLifeLab is a collaborative molecular bioscience and translational medicine institute that includes Stockholm University, the Karolinska Institute, the Royal Institute of Technology, and Uppsala University.
Researchers will use both the Illumina HiSeq 2000 and Life Technologies' SOLiD 5500 for the sequencing, and have not yet decided on a capture technique.
While the initial project will run over the next two years, Tobias Sjöblom, an associate professor of tumor biology and principal investigator of the project, said that he is also looking to partner with pharmaceutical companies in order to add more patients and eventually develop diagnostics. "The project will be a test run to scale this up," he told Clinical Sequencing News.
"So many of the drugs we use to treat cancer, we don't know why they're effective in the first place. We don't know their molecular targets," he said. As a result, researchers frequently also do not know why the drugs don't work, or what mutations are responsible for developing resistance.
Patients enrolled in the study will follow the standard of care treatment, receiving drugs currently on the market. Both CLL and colorectal cancer patients will receive chemotherapy, and CLL patients will also be treated with the targeted drug retuximab.
The team decided to use exome sequencing because the patients' tumor samples contain a mix of both cancer and normal cells, so exome sequencing will allow them "to achieve a sequencing depth of [tumor DNA] that is meaningful" and will better enable them to identify rare mutations, Sjöblom said.
Additionally, the study will serve as a model to help figure out how to implement sequencing in a clinical setting in Sweden.
"We want to set up a mechanism for bringing this information into the healthcare system, so [physicians] can act on it," Sjöblom said. However, "there is still work to do in figuring out how this is going to happen," he noted.
"The way I foresee it is the information will have to be validated within a conventional clinical genetics molecular pathology laboratory before therapy decisions are made," he added.
For the current study, the team has not yet figured out a framework for returning actionable results. "We want to return results, but we have to figure out how it should be done, and if it can be done within the consent" framework patients have already filled out. Sjöblom said that the patients have all consented to have their exomes sequenced, but returning actionable results would likely require additional consent. "It's not a simple yes or no answer," he said.
Sjöblom said that after the initial pilot, he would like to partner with pharmaceutical companies as well as institutes in Sweden's health system to make tumor profiling a routine part of care and also to develop companion diagnostic tests for different cancer drugs.
"The idea is to form a joint venture with pharma and the healthcare system," he said. While he stressed that's a long-term goal, he's already begun discussions with several undisclosed pharmaceutical companies.
Going forward, Sjöblom said the biggest challenges will be data analysis and collecting quality samples with the appropriate patient information.
Post-treatment sample collection is especially tricky because drugs can shrink tumors, so the proportion of cancer cells to normal cells is much smaller than it is in pre-treatment samples. Additionally, following treatment, patients have to be monitored continuously for signs of recurrence so that samples can be collected as soon as recurrence is detected.
Solid tumors, such as colorectal cancer tumors, tend to be more difficult to work with before treatment than liquid tumors like those found in CLL and other blood cancers because the tumors themselves are such a heterogeneous mixture of different cancer cells and normal cells, said Sjöblom. In blood cancers like CLL, around "90 percent of cells in the blood are tumor cells," though there are many fewer tumor cells in the blood post-therapy.
While data analysis for any next-gen sequencing project can be difficult, analyzing sequence data from patient samples following chemotherapy adds another layer of complexity. The drugs themselves can induce mutations, so figuring out which variants are enabling the tumors to build up treatment resistance is not just a matter of comparing the mutational profiles pre- and post-treatment.
Sjöblom said in order to "remove the noise from the signal," the team would integrate epigenomics tools, predictions from in vitro models, and systems biology approaches for the analysis.
Have topics you'd like to see covered by Clinical Sequencing News? Contact the editor at mheger [at] genomeweb [.] com.