Research institutions and companies are developing tests that can detect the presence of cancer using small fragments of DNA from tumors which have escaped into the bloodstream; so-called ‘liquid biopsies’. Illumina, the industry-leader in DNA sequencing, recently announced formation of a company to develop a blood test that can detect any kind of cancer at an early stage, and advocates of the technology say liquid biopsies could become a ubiquitous screening tool in doctors surgeries within five years. GENeS asked experts for their thoughts on the attention grabbing technology.
Dr. Dietrich Stephan, Chairman and Professor of the Department of Human Genetics, University of Pittsburgh Medical Center (webpage):
Expertise: Moving research findings on the causes of human diseases into diagnostic tests to identify those diseases early.
“Cancer is almost always caught when it has become visible by imaging, and is dangerous. Ideally we would have a test that identifies just a few cancerous cells which have evaded the normal immune system surveillance and are destined to become dangerous tumors, and liquid biopsies are the hope for gaining such an early detection test.
“Liquid biopsies are currently being used and developed to monitor patient response to cancer therapy, not for screening and diagnosis. Liquid biopsies are also used to decide whether to continue a therapy or switch to an alternative. These applications are first out of the gate because cancer patients have sufficient amounts of circulating tumor material for liquid biopsies to be able to detect that material, and the results can be confirmed by complementary tests.
“Developing liquid biopsies for early detection requires a very different study design. Because we have never been able to diagnose cancer at its earliest stages, we have no ‘gold standard’ against which to compare and refine a liquid biopsy test.
“There are technical hurdles to overcome in order to obtain the sensitivity needed to see very low amounts of material in the circulation. Additional stumbling blocks to development of a pan-cancer screening test include that different cancers shed DNA or tumor cells at different rates and the constitution of this material will differ from tumor to tumor, adding complexity to a single test.
“Very large, expensive, and complex prospective trials will need to be run to develop a pan-cancer screening test. Ideally, patients will be enrolled who are cancer free, tested routinely, and correlated with later subsequent cancer development. There are ways to streamline this development program, and in theory it is possible to do this within five years as some have claimed, but that is optimistic.”
Dr. Leonard Lichtenfeld, Deputy Chief Medical Officer, American Cancer Society (webpage):
Expertise: Dr. Lichtenfeld directs the ACS Cancer Control Science department which focuses on the prevention and early detection of cancer, as well as emerging science and trends in cancer.
“The science and the technology for detecting cancerous DNA circulating in the blood has progressed rapidly and there’s a substantial amount of very legitimate interest in applying the technology to the treatment of patients. It’s likely that within five or ten years, liquid biopsies will be used regularly in managing patients with cancer: they will be useful in helping to map the genome of a patient with cancer; they will be useful in predicting recurrence; and they will be useful in predicting the response to a cancer treatment.
“When we start talking about using liquid biopsies for the screening of people who have no evidence of cancer, we have to put the idea in the context of how effective other cancer screening tests have been and the lessons they have taught us. It took us several decades to find out that prostate cancer screening was not as successful as we initially thought. Although mammography and colorectal cancer screening have been successful in reducing deaths from cancer, we are continuing to learn that the biology of an individual’s cancer can also play a significant role in the outcome for the patient—even with early detection.
“In the early 1970s, the only way we could find most cancers was to physically feel a tumor or do an X-ray. But with mammography, for example, we’ve driven the technology so far that we can see cancers which are millimeters in size in the breast. Although there’s no question that mammography has helped many women, we’re also finding more breast cancers that may never have caused harm. As our detection techniques become more molecular, the chances are that the percentage of less harmful cancers that are detected is going to increase. So part of the responsibility in developing screening technology is to also develop an understanding of which cancers are truly going to cause harm.
“We possibly, possibly could have a liquid biopsy screening test that will tell us with some degree of accuracy if a patient has cancer available within the next five years. But whether we will have a test that can tell us accurately where that cancer is located, whether that patient actually needs to be treated, and what the best treatment is – that is highly uncertain. Inevitably we’re going to have a group of patients we help, but based on past history there’s likely to be a group of patients who will not benefit, and we need to have the wisdom to tell the difference.”
Dr. Carlos Moreno, Associate Professor, Department of Pathology and Laboratory Medicine & Department of Biomedical Informatics, Emory University School of Medicine (webpage):
Expertise: translational cancer bioinformatics and systems biology, to identify diagnostic markers for more effective and personalized therapies, to identify new potential therapeutic targets, and to better understand the biology of tumor progression.
“There is a significant difference between using a liquid biopsy for an existing cancer patient and for a screening test for early detection of cancer on a healthy population. The main concern in moving towards screening tests is the possibility of false positive test results that might trigger unnecessary invasive biopsies or medical treatments, which is the major concern with the PSA test for prostate cancer.
“It is critical that any potential screening test that might be used on healthy populations would have an extremely low false positive rate, as well has very high sensitivity. Consequently, it is really necessary to have these tests analyzed in clinical trials, published in scientific peer-reviewed journals, and approved by the FDA. And when the tests become available, it is equally important that the company that performs the test has the ability to generate extremely high quality DNA sequence data with the reliability that a clinical test requires.
“Developing blood tests for early detection of cancer is currently an area of very active research. There is a group in Hong Kong that is screening 20,000 healthy men for cancers, and another group at Johns Hopkins that is working on developing liquid biopsies for screening purposes. In addition, Illumina has recently announced that its new spin-off company Grail will enroll up to 30,000 people in its clinical trials.
“The technical hurdles which developers must overcome involve the ability to accurately detect mutations that might be present in one cell in 100,000 in the bloodstream (or less). Scientists are working on approaches that enrich the number of circulating tumor cells or cell-free tumor DNA that is released in the bloodstream when cancer cells die. As sequencing technology improves and gets cheaper, it is possible that such hurdles could be overcome. Given the pace of current projects and the ability to generate massive amounts of DNA sequence data, it is possible that a pan-cancer screening test could be available within 5 years, but it might take longer.”
Declared interests (see GENeS register of interests policy):
Dr. Dietrich Stephan: “I’ve been involved in liquid biopsies from the beginning, and disclose that I’m a shareholder and member of the Scientific Advisory Board of Guardant Health.”
No further interests declared