By Nicholas Smith
We may well be on the threshold of a new hope for oncology. Shorthanded to ctDNA, circulating cell free tumor DNA is sloughed off from tumors. It can be detected in liquid biopsies of just a few milliliters of blood. This could revolutionize what oncology can achieve by diagnosing cancers earlier and more efficiently.
One of the presenters at the Molecular Diagnostics conference in Lisbon (April 2016) was Dr. Theresa Zhang, Vice President of Research Services at Personal Genome Diagnostics. She discussed how ctDNA analysis is poised to have a profound impact on how we treat cancer patients — in diagnostics, prognostics and drug monitoring. In the lab, automation will be key to achieving both cost-effective throughput and also the control over contamination that is critical to find the needle in the haystack: a few mutant DNA copies shed from a tumor into the blood, surrounded by thousands, or even tens of thousands of normal copies. We took the opportunity to ask her about the current status and future promise of ctDNA analysis in oncology.
Theresa is very enthusiastic about the potential for ctDNA in transforming oncology in practice. “Measuring ctDNA will help us with earlier detection in the future. It will be invaluable in studying minimal residual disease and reduce the level of overtreatment.”
“For many cancers, diagnosis is followed by surgery and then the tumor tissue is analyzed to decide on the risk of recurrence and the need for chemotherapy. That kind of risk assessment hasn’t been very accurate and the result is over-treatment or under-treatment. For example, in early stage breast cancer, many patients receive chemotherapy, even though large-scale clinical trials have shown that only 10-15% patients really benefit from chemotherapy.”
“ctDNA tests can be helpful in identifying those patients. ctDNA tests done on blood draws a few weeks after surgery can be used to determine whether the patient has residual tumor cells that are shedding ctDNA into the blood. If there is ctDNA in the plasma of patients after surgery, those patients will have a high likelihood of recurrence. The ability to detect early signs of recurrence could change the way we do risk assessment. Instead of statistical prediction, we are actually detecting recurrence.”
“Another aspect of diagnostics is directing targeted therapy in the late-stage patient. For example, patients who have specific mutations in epidermal growth factor receptor (EGFR) are eligible for specific treatment with EGFR inhibitors. A plasma-based version of the EGFR assay can be used as a diagnostic tool in some patients. Two of these assays, marketed by Roche and Qiagen have already been approved in Europe in 2015.”
Using ctDNA to monitor response to drug treatment is another hot area, as Theresa said. “Changes in circulating tumor DNA during treatment are very informative of response to treatment. The analysis of ctDNA, which is released from any lesion in the body, often detects response to treatment or relapse from treatment earlier than the conventional approach, such as imaging, which is directed at only a few lesions. Increased levels of ctDNA can be detected months before imaging can detect that the tumor has returned. It’s very exciting that we are able to detect a molecular relapse so early; this means that we know when the therapy has stopped working long before the imaging assay indicates there is a problem.”
“Being able to do a test non-invasively and get real-time information about the tumor is also valuable, because the tumor evolves during treatment or as the disease progresses. What we have from the diagnostic block, the archival tissue, is not informative enough to guide treatment decisions later on. Right now we are limited to archival samples, and real-time re-biopsies are not an option in most clinical settings. So ctDNA analysis offers an excellent alternative to serial re-biopsy, especially for late-stage cancer patients since their tumors tend to shed more ctDNA into blood which makes it easier to detect.”
Theresa acknowledges that there may be some disadvantages with ctDNA analysis, but they are definitely outweighed by the advantages. “With the current detection approach, not all the tumors shed detectable levels of ctDNA. And if the detection method is not sensitive enough, you will probably miss patients. But the biological background is very low for ctDNA – it only comes from cancer. The dogma is that if you find ctDNA in plasma then it means that you have cancer. We now know that some white blood cells may mutate clonally and shed mutated DNA into blood as we age, but these can be easily identified.”
After twelve years of experience at Merck, leading a molecular profiling group supporting oncology, Theresa joined Personal Genome Diagnostics two years ago, just when the ctDNA field was really starting to pick up. Early last year, the company launched, which analyzes circulating tumor DNA and interrogates sequence mutations, copy number changes and translocations in a large panel of 63 clinically actionable genes. Detection is based on Next Generation Sequencing, and the company is aiming at CLIA approval in the near future.
“PlasmaSelect-R has already been used by many pharmaceutical companies for retrospective analysis of their clinical trial samples. It helps them understand the genetic landscape of the patients enrolled in their clinical trials, which is often uncharacterized since patients have gone through many lines of treatment. Previously, pharmaceutical companies have had to rely on archival tumor block taken years prior to the trial but were missing the real-time status of the tumor. So where possible, they are now using liquid biopsies to assess the molecular landscape of the cancer they are treating.”
Analysis of ctDNA also gives insight into possible mechanisms of drug resistance by the tumor. “By comparing pre-treatment, baseline ctDNA results with the blood sample taken throughout the treatment, it is possible to determine if tumors have acquired new mutations and if these new mutations have contributed to resistance.”
Theresa has a clear view on the need for automation in the clinical laboratory when it comes to analyzing ctDNA. “Whenever a highly complex assay, such as a Next Generation Sequencing-based ctDNA assay is involved, automation is key because the assays are too complex to be scaled up in a reliable and controlled manner. Moreover, automation is critical for highly sensitive ctDNA tests. Indeed, we need to detect one mutant molecule out of 1,000, and in the future out of a background of 10,000 wild type molecules. So contamination becomes a real problem. While automation gives certainly cost effectiveness, it’s the key to controlling contamination.”