Liquid Biopsy and Precision Medicine in Oncology
Caroline C. Sigman
CCS Associates, Inc.
David R. Parkinson
ESSA Pharma
Gary J. Kelloff
National Cancer Institute
L

iquid biopsy-based biomarkers are increasingly used in precision medicine strategies to replace or complement tissue biopsy and imaging. They are used as predictive, response, resistance, and early detection biomarkers. Development of liquid biopsy-based biomarkers is challenging. They can be complex, being based on advanced ’omics and single-cell analysis; they need to reflect cancer biology; and they should be detectable at low levels and precisely enough to allow patients to be monitored effectively. The FDA and several international collaborative initiatives are now addressing the much-needed standards for developing and using liquid biopsy-based biomarkers. ctDNA is the most frequently studied liquid biopsy analyte, and the ctDNA research described in this article provides examples of the current and future status of liquid biopsy.

Liquid biopsies provide many opportunities for developing biomarkers to detect and monitor cancers and are potentially necessary for practicing precision oncology. They may be alternatives or complementary to tissue biopsy and anatomical and functional imaging (Table 1), and liquid biopsy-based biomarkers are increasingly incorporated in drug development and patient care. They are minimally invasive and potentially more sensitive than tissue-based biomarkers and so may help detect early-stage disease or disease onset and progression before there is clinical evidence and may be helpful for serial monitoring of patient treatment. They support precision medicine strategies with genomic, proteomic, immune status, and other information on a patient’s tumor and tumor microenvironment (TME). For example, many liquid biopsy studies have been published using next-generation sequencing (NGS) and other advanced technologies to analyze circulating DNA (circulating free DNA [cfDNA] or circulating tumor DNA [ctDNA]), RNA, microRNA, or immune system-related components and proteins free in blood or bone marrow or in circulating tumor cells (CTCs) and in exosomes from blood and other bodily fluids (Table 2).
Table 1. Liquid Biopsy, Tissue, and Imaging Biomarkers
Tissue Biomarker
Imaging Biomarker
  • Can probe many features
  • Probes 1−2 features
  • Single location, limited sampling
  • Tissue volume, complete tumor burden sampling
  • Variable cost
  • Often expensive
  • Invasive
  • Noninvasive
  • Serial assay challenging
  • Serial assay possible
  • Widely available
  • Less widely available ─ local assay
  • Liquid Biopsy Biomarkers can probe many features, detect complete tumor burden, are noninvasive, allow serial assay, are widely available, and may address assay limitations including sensitivity and heterogeneity
Table 1. Liquid Biopsy, Tissue, and Imaging Biomarkers
Tissue Biomarker
  • Can probe many features
  • Single location, limited sampling
  • Variable cost
  • Invasive
  • Serial assay challenging
  • Widely available
Imaging Biomarker
  • Probes 1−2 features
  • Tissue volume, complete tumor burden sampling
  • Often expensive
  • Noninvasive
  • Serial assay possible
  • Less widely available ─ local assay
Tissue and Imaging Biomarker
  • Liquid Biopsy Biomarkers can probe many features, detect complete tumor burden, are noninvasive, allow serial assay, are widely available, and may address assay limitations including sensitivity and heterogeneity
Table 2. Liquid Biopsy Analytes in Oncology
Circulating Tumor Cells (Enumeration, Physical Characteristics, DNA, RNA)
Exosomes (DNA, RNA, microRNA, Protein)
Whole Blood/Plasma (cfDNA, ctDNA, Immune System DNA, cfRNA, microRNA)
Abbreviations: cfDNA=circulating free DNA; ctDNA=circulating tumor DNA; cfRNA=circulating free RNA)

Critical Considerations in Development and Use of Liquid Biopsy Assays in Precision Oncology—ctDNA as Example

Two considerations for the use of ctDNA as an analyte can be applied generally in developing liquid biopsy assays:

  1. The analyte should be detectable at levels sufficient and precise enough to allow patients to be monitored effectively.
    This is the case for ctDNA in many clinical settings. The fraction of circulating DNA that is ctDNA in cancer patients is variable, but it is high enough to be detected and differentiated from nontumor DNA. Difficulties in detecting heterogeneous clonal variants or variants present only in single metastases and characterizing the biological levels of ctDNA complicate the reliability of ctDNA assays. However, within individuals, ctDNA levels vary with tumor burden and response to therapy, which can be useful in tracking patients’ disease. Analytically, it is essential to know the limit of detection (LOD) of the assay used to measure ctDNA, since it is likely that most assays will break at the low levels of ctDNA (e.g., <0.1% concentration) commonly encountered in patients. This characteristic has led to studies restricting the analysis of ctDNA to patients with concentrations above certain thresholds.
  2. The analyte should reflect cancer biology.
    Several studies have shown that ctDNA mutations correspond to primary tumor mutations. Besides somatic point mutations, NGS methods allow genome-wide detection of rearrangements and chromosome copy number changes. Nearly all tumors have rearranged DNA sequences that are not present in normal human plasma or tissue. One challenge is extracting the relatively few somatic structural alterations present in ctDNA from the much larger number of structural variants in germline DNA. Nonetheless, at least one study of cancer patients’ tissue and blood using an NGS ctDNA assay showed a high number of clinically actionable structural variants in both germline and somatic DNA, suggesting the relevance of this approach for evaluating patient treatment plans.

Some studies have shown that ctDNA assays, though highly sensitive (seeing approximately 80%-90% of mutations present in the primary tumor), do not detect all mutations identified in primary tumor tissue. Conversely, ctDNA assays may find mutations not seen in tumor tissue.

How ctDNA Liquid Biopsy Is Used in Precision Medicine in Oncology

There are four major potential applications of liquid biopsy in precision medicine strategies (see also Table 3):

  1. Predictive Biomarkers indicate whether a patient will benefit from a particular treatment. Companion or complementary diagnostics are predictive biomarkers. NGS-ctDNA-based liquid biopsy assays have been developed and cleared by the US Food and Drug Administration (FDA) as companion or complementary diagnostics across a broad array of drug targets and tumor types. These assays have also been used to assess blood tumor mutational burden (bTMB), microsatellite instability (MSI), and tumor fraction values, all of which may be useful in selecting treatments for patients in precision medicine settings. Specifically, bTMB measured by ctDNA is being studied as a predictive biomarker for response to immunotherapy. MSI evaluation has been shown to have high specificity, precision, and sensitivity, with a LOD of 0.1% ctDNA. For major cancer types, circulating DNA-measured MSI correlates with tissue-measured MSI. Another option, studied in multiple tumor types, is using ctDNA to detect drug targetable mutations in newly diagnosed or relapsing patients to direct therapy assignment.
  2. Response Biomarkers serve as indicators of drug efficacy and for monitoring patient treatments. Serial monitoring is a distinct advantage of using liquid biopsy as response biomarkers. Detection of minimal, measurable, or molecular residual disease (MRD) using complex NGS-derived or multiplexed immunohistochemistry biomarkers is an especially promising application for evaluating treatment efficacy and prognosis. The low LODs (0.1%) of some of the components of technologically advanced ctDNA assays make the detection of the “needle in the haystack,” which is needed to define MRD, a possibility. Retrospective studies in many tumor types have shown that ctDNA detects MRD before clinical progression is observed.
  3. Biomarkers of Emerging Resistance. ctDNA can detect “molecular” relapse or resistance before clinical or imaging evidence is apparent. This can be measured by increases in ctDNA amount or changes or increases in mutations in the ctDNA. Because serial analysis allows evaluation of clonal dynamics during treatment, ctDNA-based liquid biopsy can detect the multiple resistance pathways that may occur. It also detects decay of resistance mutations after a treatment has been stopped and the timing for restarting treatment with the previously effective drug, such as epidermal growth factor receptor (EGFR) inhibitors in colorectal cancer.
  4. Early detection biomarkers. The evaluation of noninvasive ctDNA-based biomarkers is of high interest for the early detection of cancer and cancer predisposition. Great strides have been made in developing tests for such biomarkers based on the NGS analysis of ctDNA combined with computational analysis. Many of these biomarkers are methylation pattern signatures, some having shown potential for detecting tumors by tissue of origin in undiagnosed subjects.
Table 3. Applications of ctDNA-Derived Biomarkers in Precision Oncology Studies
Type of Biomarker
Application
Predictive
Select patients for targeted or immunotherapy based on presence in ctDNA of specific mutations, MSI, or bTMB (bladder, breast, colon, NSCLC, prostate)
Response/MRD
Assign patients to post-treatment therapy or not based on presence or absence of ctDNA-derived MRD (colon)
Resistance
ctDNA monitoring during treatment to determine progression and timing for treatment change (bladder, breast, colon)
Early Detection
ctDNA methylation profiling with AI analysis (potential early detection of multiple cancer types, studied in breast, colon, NSCLC)
Abbreviations: AI=artificial intelligence; MRD=minimal, measurable, or molecular residual disease; MSI=microsatellite instability; NSCLC=non-small cell lung cancer; bTMB=blood tumor mutational burden
Table 3. Applications of ctDNA-Derived Biomarkers in Precision Oncology Studies
TYPE OF BIOMARKER
PREDICTIVE
Select patients for targeted or immunotherapy based on presence in ctDNA of specific mutations, MSI, or bTMB (bladder, breast, colon, NSCLC, prostate)
Response/MRD
Assign patients to post-treatment therapy or not based on presence or absence of ctDNA-derived MRD (colon)
RESISTANCE
ctDNA monitoring during treatment to determine progression and timing for treatment change (bladder, breast, colon)
EARLY DETECTION
ctDNA methylation profiling with AI analysis (potential early detection of multiple cancer types, studied in breast, colon, NSCLC)
Abbreviations: AI=artificial intelligence; MRD=minimal, measurable, or molecular residual disease; MSI=microsatellite instability; NSCLC=non-small cell lung cancer; bTMB=blood tumor mutational burden

Standards and Incentives Are Needed

Accomplishing precision medicine research goals requires consistent data from multiple studies and assays across numerous clinical sites. Thus, standards are an overarching requirement (e.g., for sample collection, analytical and clinical validation, and analysis) to keep abreast of advancing technologies and make the data created useful across laboratories and studies. In addition to these protocol standards, precompetitive and intellectual property issues related to the assays should be managed to allow wide application across the research and clinical communities.

The numbers and complexity of liquid biopsy assays have proliferated in the wake of rapidly advancing technologies in ’omics, molecular targeting, and single-cell analysis. The assay targets are rare, so reliable evaluation of results requires many samples, sometimes not available in one study. Many assays are not yet analytically or clinically validated or may be run according to diverse protocols or with slightly varying technologies in different laboratories, yielding noncomparable results.

An expert panel from the American Society of Clinical Oncology (ASCO) and the College of American Pathologists (CAP) tackled the problems associated with clinical use of liquid biopsy and lack of standards, focusing on ctDNA assays. Their review produced in 2018 included guidelines for future research and called for tumor tissue profiling to confirm ctDNA tests, concluding that there was no evidence of clinical utility and little evidence of clinical validity of blood ctDNA assays except in clinical trial settings. The panel recommended developing tools and guidance for the use of ctDNA assays in clinical practice. Testing protocols including standard procedures for processing and analyzing liquid specimens are a must, as are reliable control materials against which assay results and methods can be validated.

Several collaborative initiatives are now addressing standards for liquid biopsy (Table 4). A few of these efforts are profiled briefly below:

  • Blood Profiling Atlas in Cancer (BloodPAC) Consortium was formed in 2016 in concert with the Cancer Moonshot initiative to compile and disseminate data and provide standard procedures and methods of analysis for liquid biopsies used in precision medicine and includes members from academia, private foundations, industry, and the FDA and the National Cancer Institute (NCI).
  • Foundation for the National Institutes of Health (FNIH) ctDNA Quality Control Materials (QCM) project is a collaboration involving private sector, NCI, FDA, National Institute of Standards and Technology, and academic and not-for-profit partners. Its objective is to develop QCMs and demonstrate their comparable performance to ctDNA from clinical settings. The QCM could then establish performance characteristics for assays within and across laboratories, clinics, and research studies.
  • Friends of Cancer Research ctMoniTR. Friends of Cancer Research convenes public-private partnerships such as ctMoniTR for research and policymaking to improve cancer drug development and treatment. The project has gathered data from retrospective and prospective clinical trials to determine if ctDNA correlates with cancer therapy efficacy and has defined standard practices for sample collection, data processing, and analysis that can be incorporated into clinical trials.
  • International Liquid Biopsy Standardization Alliance (ILSA) is comprised of 10 organizations, including those listed above along with CANCER-ID, European Liquid Biopsy Society, International Society of Liquid Biopsy, Japanese bio Measurement & Analysis Consortium, Medical Device Innovation Consortium, and the National Institute for Biological Standards and Control to implement liquid biopsy in oncology clinical practice and drug development. FDA recognizes the alliance as a Collaborative Community in which the FDA can participate and provide regulatory perspective.
Table 4. Collaborations on Standards, Guidelines, and Scientific Advances in Liquid Biopsy
  • Advances in Circulating Tumor Cells (Annual International Workshop)
  • CANCER-ID
  • BloodPAC
  • ASCO/CAP Joint Review on ctDNA Testing
  • National Institute for Biological Standards and Control, UK (NIBSC)
  • European Liquid Biopsy Society (ELBS)
  • FNIH ctDNA Quality Control Materials Project
  • FDA/EMA
  • FDA/AACR Workshops on Liquid Biopsies in Oncology Drug and Device Development
  • Friends of Cancer Research ctDNA Pilot Project (ctMONiTR)
  • International Society of Liquid Biopsy (ILSB)
  • Japanese bio Measurement and Analysis Consortium
  • NCI, DoD, VA APOLLO
  • National Cancer Institute, Division of Cancer Prevention Liquid Biopsy Consortium
  • Medical Device Innovation Consortium (MDIC)
  • International Liquid Biopsy Standardization Alliance (ILSA)

Regulatory Guidance

Besides providing advice on the levels and types of evidence necessary to use liquid biopsy assays, regulatory guidance helps clarify reimbursements and can offer pathways to easing concerns on the preservation of intellectual property. In 2018, the FDA granted marketing approval to six solid tissue-based NGS tests or test systems, published guidance on the regulatory approach that will be taken for these tests that provides recommendations for designing, developing, and validating NGS-based tests, and announced their commitment to work with NGS test developers to use the least burdensome approach for reviewing their tests. A second guidance describes a process by which test developers may rely on clinical evidence from FDA-recognized public databases (e.g., NIH’s ClinGen) and peer-reviewed literature to support their tests and help assure the accurate clinical evaluation of the genomic test results. Concomitantly with the FDA marketing approvals, the Centers for Medicare & Medicaid Services (CMS) confirmed payer support for these assays by issuing guidance recommending reimbursement for companion diagnostic NGS testing and reimbursement for evidence development for other NGS testing. Further, in 2020 as described above, two liquid biopsy ctDNA-based NGS assays were approved by FDA as companion diagnostics agnostic to tumor type and drug.

Challenges and Research Approaches

For comprehensive and insightful discussion on the current and possible future role of liquid biopsy in precision oncology, we refer readers to excellent recent reviews (see Supplemental Resources). As evidenced by the high volume of recent publications, liquid biopsy has exceptional promise as a noninvasive adjunct or alternative to tissue biopsy and imaging in developing precision medicine strategies in oncology, but there are challenges to overcome.
The first is cancer biology, including tumor heterogeneity and clonal evolution and selection, as well as normal genetic variation. Understanding the role of the TME, particularly immune status, is also critical. Researchers are aggressively tackling these issues by developing and applying the NGS methods highlighted above and other high content technologies involving RNA sequencing, immune profiling, proteomics, systems biology, multiplexed immunohistochemistry molecular imaging, and extensive computational analysis. Newer efficient clinical study designs, such as multi-arm platform trials, expansion cohorts, use of standard protocols across collaborative networks, and targeted evaluation of drug combinations, should also benefit the development of liquid biopsy.

The second challenge is keeping up with advancing technology in developing standards for assays’ analytical and clinical validation. This issue is also being aggressively addressed through organizations and public-private collaborations dedicated to setting standards, collecting data, and providing methods for analyzing data from multiple sources. A significant effort is to promote multisector precompetitive partnerships with intellectual property rights resolved to accomplish these goals.

A final challenge is embedding liquid biopsy into the research and clinical care communities. This requires establishing clinical utility (benefit to patients) of liquid biopsy testing strategies and promoting insurance coverage for the use of these strategies while their utility is defined (i.e., coverage with evidence development). The FDA and CMS are engaged in this effort, as shown in the approvals of NGS liquid biopsy assays and the FDA guidances on the development of NGS assays.