Jeffrey Perkel has been a scientific writer and editor since 2000. He holds a PhD in Cell and Molecular Biology from the University of Pennsylvania, and did postdoctoral work at the University of Pennsylvania and at Harvard Medical School.
If physicians want to know what a tumor is doing, they have limited options. They can image the affected region, collect a tissue biopsy or monitor a blood biomarker such as PSA or CA 125. None of these is ideal: Imaging lacks resolution, biopsies are invasive and good molecular biomarkers are few and far between.
Ideally, researchers would be able to achieve the sensitivity and specificity of tissue biopsies with a simple blood test. Today, thanks to so-called “liquid biopsies,” they may have found a way to do just that.
Liquid biopsies
According to Max Diehn, assistant professor of radiation oncology at the Stanford University School of Medicine, a common problem with traditional cancer biomarkers is specificity. “The majority of biomarkers that have previously been explored are also produced to some extent by normal cells and therefore are not uniquely made by tumor cells. And that means that they are often not extremely specific.”
In theory, liquid biopsies could overcome that deficiency by searching for more specific tumor fingerprints in the bloodstream and other bodily fluids.
Those fingerprints come in two flavors. Circulating tumor cells (CTCs) are live cells that have sloughed off from a tumor mass, possibly representing the seeds of metastasis; circulating tumor DNA (ctDNA) is naked genetic material from dead and dying cancer cells, typically in the form of short snippets of DNA. In one recent study from Diehn’s lab, which described a new method for ctDNA detection in blood plasma of individuals with non-small cell lung cancer (NSCLC), fragments averaged 170 bases [1].
Of course, normal cells also exist in the bloodstream, and some normal cells die every day. Thus whether researchers are looking at CTCs or ctDNA, they must separate the wheat from the chaff. For CTCs, that typically is accomplished by cellular phenotype, for instance using the CELLSEARCH® assay from Janssen Diagnostics, which selects cells in blood based on the expression of epithelial cell markers.
For ctDNA, researchers must fish out fragments based solely on their sequence—and that’s a tall order, as non-mutant DNA from a cancerous cell is obviously uninformative. Thus, the technology is ideal for cases in which mutations already have been identified. Diehn’s study, for instance, began by identifying genomic regions commonly mutated in NSCLC patients to produce a so-called CAPP-Seq (CAncer Personalized Profiling by deep Sequencing) selector library—a collection of oligonucleotides for targeted sequencing representing 521 exons and 13 introns from 139 genes. The team used that library to harvest complementary DNA fragments from patient plasma (using the QIAamp Circulating Nucleic Acid Kit from QIAGEN), which they sequenced to 10,000x coverage to account for the relative paucity of ctDNA molecules compared with normal DNA.
“If one in a thousand DNA molecules is from cancer, at each position you care about, you have to go more than a thousand-fold deep to be able to reliably see it,” Diehn explains.
According to a 2014 study by researchers at the Johns Hopkins University School of Medicine, ctDNA can be found in the blood plasma or sera of patients with a wide diversity of tumors, although the assay’s sensitivity varies with tumor location and stage. “ctDNA was detectable in >75% of patients with advanced pancreatic, ovarian, colorectal, bladder, gastroesophageal, breast, melanoma, hepatocellular, and head and neck cancers,” the authors wrote, “but in less than 50% of primary brain, renal, prostate, or thyroid cancers” [2].
Nickolas Papadopoulos, professor of oncology at the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins University and a member of that team, says researchers can achieve greater sensitivity by getting “closer to the source”—for instance, by looking in the urine for possible genitourinary tumors or in saliva for some cancers of the head and neck. Indeed, the team recently reported improved detection of central nervous system (CNS) tumors in cerebrospinal fluid (CSF), finding CSF-ctDNA in 74% of CNS tumors overall and in 100% of “medulloblastomas, ependymomas, and high-grade gliomas that abutted a CSF space” [3].
As in many ctDNA studies, the Hopkins team detected ctDNA using targeted sequencing—in this case, a barcoding strategy called Safe-SeqS that can accurately report the number of ctDNA molecules in a sample. Researchers also can use digital PCR to quantify specific molecular changes, although as the authors of one study who used it note, a “[d]igital PCR assay provides high accuracy and sensitivity but requires the design of personalized assays, an expensive and rate-limiting step” [4].
Pros and cons
Researchers can detect ctDNA in blood and other fluids from individuals with cancer, making the material a promising biomarker. Yet according to Papadopoulos, ctDNA is not merely a binary indicator; ctDNA levels also seemingly track with “tumor burden,” he says, with levels rising with disease severity and correlating with survival. And ctDNA can be detected even in some early-stage cancers, Papadopoulos notes, suggesting it may be possible to use ctDNA not only to monitor existing cancers but also as a screening tool.
Wyndham Wilson, chief of the lymphoid therapeutics section at the National Cancer Institute Center for Cancer Research, says ctDNA may also be used to predict tumor relapse. In a study of patients with diffuse large B-cell lymphoma [5], Wilson says, ctDNA “was able to pick up [tumor] recurrence months before a CT scan could. So you could start to treat that patient earlier” and hopefully improve his or her long-term prognosis.
That said, ctDNA poses several potential problems as a screening tool. For one thing, it isn’t yet clear how prevalent ctDNA is in different cancer types, stages and patients. Even when ctDNA is detected, it isn’t obvious where in the body it arose, or even if the detected mutation is necessarily associated with a tumor, meaning follow-up testing may be required.
CTC analysis, by contrast, offers live cells with physically contained and intact genomes. Researchers can thus isolate those cells, examine them microscopically and study their genomes and protein expression cell by cell. That, says Antonio Guia, vice president for research and development at Aviva Biosciences, a company that is developing a method for CTC isolation, enables “a little bit more broad-spectrum approach [than ctDNA analysis], so that you’re not looking for one specific fragment of one specific gene or one particular type of tumor. You can then target cells that have any number of different tumor types, as long as you can identify the cell as a tumor.”
One recent study by Harvard University’s Daniel Haber and colleagues, for instance, used single-cell RNA-Seq to identify signaling differences in individual prostate-tumor CTCs and to correlate those differences with response to treatment [6]. (The team isolated its CTCs using a custom microfluidic device called a CTC-iChip, which separates CTCs from other blood cells based on the fact that they are nucleated cells that express neither CD45 nor CD66.)
But when it comes to liquid biopsies, the question is not CTCs vs. ctDNA. Ultimately, says Jorge Villacian, chief medical officer at Janssen Diagnostics, a company that focuses on CTC diagnostics, the two biomarkers provide complementary information, and there’s room for both in the clinical toolkit.
“The more that [oncologists] embrace the concept of liquid biopsies, together with the regulatory authorities, it’s going to be a win-win for everybody—for the patients, for the physicians and for the system itself—because it will cause less morbidity, less complications, and potentially changes of treatment at an earlier stage that would benefit the patients with improved outcomes in the future.”
References
[1] Newman, AM, et al., “An ultrasensitive method for quantitating circulating tumor DNA with broad patient coverage,” Nat Med, 20:548-54, 2014. [PMID: 24705333]
[2] Bettegowda, C, et al., “Detection of circulating tumor DNA in early- and late-stage human malignancies,” Sci Transl Med, 6:224ra24, 2014. [PCMID: PMC4017867]
[3] Wang, Y, et al., “Detection of tumor-derived DNA in cerebrospinal fluid of patients with primary tumors of the brain and spinal cord,” Proc Natl Acad Sci, 112:9704-9, 2015. [PMID: 26195750]
[4] Dawson, S-J, et al., “Analysis of circulating tumor DNA to monitor metastatic breast cancer,” New Engl J Med, 368:1199-1209, 2013. [PMID: 23484797]
[5] Roschewski, M, et al., “Circulating tumour DNA and CT monitoring in patients with untreated diffuse large B-cell lymphoma: A correlative biomarker study," Lancet Oncology, 16:541-9, 2015. [PMID: 25842160]
[6] Miyamoto, DT, et al., “RNA-Seq of single prostate CTCs implicates noncanonical Wnt signaling in antiandrogen resistance,” Science, 349:1351-6, 2015. [PMID: 26383955]
Image: Shutterstock