Cancer, a leading cause of mortality, is associated with mutated genes. Analysis of tumour associated genetic alterations is increasingly used for diagnostic, prognostic and treatment purposes. Precision or personalised medicine harnesses genomic knowledge banks to tailor individualised treatments based on patients’ or their tumours’ genetic signatures.
Somatic Mutations in Cancer
A high-quality genomic analysis is critical for personalised pharmacotherapy in patients with cancer. The advent of therapies targeting genomic alterations has improved the care of patients with certain types of cancer dramatically. The presence or absence of activated therapeutic driver mutations or gene targets (e.g., BRAF in melanoma, KRAS in colorectal cancer and EGFR mutation or ALK rearrangements in non-small cell lung cancers [NSCLC]) is currently employed to guide treatment decisions. Thus, elucidating the genetic profile of a given tumour is potentially useful in designing tailored treatment regimens that avoid unnecessary toxic therapy or overtreatment.
While molecular targets were initially detected in nucleic acid samples extracted from solid tumours, detection of circulating nucleic acids in blood has enabled the development of what has become known as “liquid biopsy”.
Certain fragments of DNA, cell-free DNA (cfDNA), have been shown to be elevated in the plasma of patients with cancer. This increase is the result of the rapid turnover of cells within the tumour, releasing DNA into the circulation (1; see Fig. 1). cfDNA has a rapid turnover with a short half-life allowing for evaluation of tumour changes in hours rather than weeks to months. The size of the DNA fragments is usually around 150-180 bp in length (26,30,35) with a higher prevalence of tumour associated mutations in the shorter fragments (31).
Recent technological advances have made it possible to analyse genetic aberrations of tumours via their released cfDNA. cfDNA analysis can complement and, in many instances, correlate to solid tissue biopsies (3,13,26) for real-time molecular monitoring of treatment, detection of recurrence, and tracking treatment resistance.
Liquid Biopsy is Less Invasive
The genetic profile of solid tumours is currently obtained from surgical or biopsy specimens. Tumour tissue genotyping is considered the standard choice for detection of potentially targetable alterations, however, performing biopsies, particularly in lung cancer, is not always possible in advanced disease stages and is associated with potential complications and adverse events. Also, information acquired from a single biopsy provides a limited snap-shot of a tumour and might fail to reflect its true genetic and cellular heterogeneity, as well as not guaranteeing enough material collected for accurate analysis in clinical practice. In the case of advanced or metastatic NSCLC as many as 31% of cases do not have accessible tissue (28). Due to tumour heterogeneity, biopsies often suffer from sample bias (29) and thus may yield false-negative results.
Circulating tumour DNA (ctDNA) ctDNA provides new insight into diagnosis, prognosis and patient follow-up compared to traditional tissue biopsy. Liquid biopsy provides a non-invasive alternative sample source, allowing the identification of genomic alterations that can be addressed by targeted therapy. This non-invasive type of liquid biopsy can be taken easily and repeatedly over the course of a patient’s treatment. ctDNA reflects the overall tumour information, can provide the genetic landscape of all cancerous lesions (primary and metastases) and is not biased by analysing only a small fraction of the tumour and is always accessible in contrast to the lung cancer tissue.
Testing of Circulating Tumour DNA (ctDNA) in Plasma
Various testing platforms described have the potential to add tremendous value to the care of cancer patients by offering comprehensive mutation profiling assay for clinical oncology patients. Most currently validated ctDNA assays have been compared against tissue genotyping. The variety of technologies emerging enables more precise and robust analysis of circulating tumour derived DNA extracted from blood with sufficient sensitivity and specificity to accurately detect cancer biomarkers. The specificities reported for these tests are consistently high, ranging between 90% and 100%, but their sensitivities are more variable, ranging between 70% and up to 100% (43-45) in stage IV disease, depending on the testing methodology (4,32,39). In principle, technologies can be divided into targeted approaches that aim to detect mutations in a set of pre-defined genes or untargeted approaches.
…the potential to add tremendous value to the care of cancer patients
Different validated technologies including Next Generation Sequencing (NGS), MassArray Agena Biosciences UltraSEEK and Droplet Digital PCR (ddPCR), are able to identify clinically relevant variants at a sensitivity down to 0.5% or less (44). Negative plasma genotyping result can be attributed to either an absence of the tumour’s mutations of interest, or the tumour is not shedding significant ctDNA into the blood. Thus, to rule out false negative results, ctDNA test should still be followed by a reflex tissue biopsy when feasible.
Circulating Tumour DNA (ctDNA) and Clinical Applications in Cancer
Prospective research cohort studies are ongoing to explore the possible clinical applications of ctDNA in cancer. The inclusion of ctDNA analysis into multiple clinical trials (ClinicalTrials.gov), signals the recent integration of this form of biomarker into routine clinical oncology.
Various studies in breast, lung, melanoma and colorectal cancers have demonstrated the potential application of ctDNA analysis at each stage of clinical management: early diagnosis (13,26), molecular profiling (14), prognostication (12,15,16,26), detection of residual disease (7,17,26), monitoring response (36,37) and clonal evolution (7,18-20,34,35). Lastly, the recent approvals by the Federal Drug Administration (FDA) and the European Medicines Agency (EMA) for detection of the tyrosine kinase resistant clone EGFR p.T790M mutation in plasma as a companion diagnostic for second-line treatment of metastatic EGFR-mutant NSCLC (21, 25).
Fleischhacker and Schmidt (1) reviewed thirty four studies involving healthy subjects and patients with both malignant and non-malignant disease. While a trend toward the DNA concentration in the blood of cancer patients being much higher than in the blood of healthy controls and non-malignant patients is clear, the cfDNA concentration varies considerably (1). The variability of ctDNA levels in cancer patients likely associates with tumour burden, stage, vascularity, cellular turnover, and response to therapy. Several studies (13-16,26) have now demonstrated a high concordance between mutational profiles of oncogenes in matched tumour and plasma DNA samples from patients with various cancers. Using ctDNA to characterise a large population of patients with different tumour types, Bettegowda et al. found tumour-derived cfDNA in 82% of patients with solid tumours outside the brain, including more than 75% of patients with advanced ovarian, colorectal, bladder, gastroesophageal, pancreatic, breast, hepatocellular, and head and neck cancers, as well as melanomas. Less than 50% of patients with medulloblastomas or metastatic cancers of the kidney, prostate or thyroid, and less than 10% of patients with gliomas, had detectable ctDNA (26). Among patients with localised cancer of all types evaluated, only 55% had detectable ctDNA, but the proportion of patients with detectable ctDNA increased with tumour stage (26,17).
Liquid Biopsy and Early Detection of Relapse
Early detection is the holy grail of cancer management. The biggest advantage of liquid biopsy is the ability to detect the cancer biomarkers in blood earlier than conventional methods (22,23). ctDNA carries comprehensive, inherently specific, and highly sensitive information compared to protein biomarkers which have limited specificity and sensitivity (7,47). Computed tomography (CT) imaging improves detection of recurrence but is associated with radiation exposure and also has a high rate of false positivity (47).
Early detection is the holy grail of cancer management.
While correlating with imaging scans, it has been demonstrated that monitoring for tumour-derived DNA in the plasma can identify relapse or drug resistance well before clinical signs and symptoms appear, enabling earlier intervention and better outcomes (22,23). Detection of KRAS variants in ctDNA of patients receiving anti-EGFR therapies can identify relapse 10 months before radiographic documentation of disease progression (42). In the future, instead of extensive imaging and invasive tissue biopsies, employing ctDNA as liquid biopsies could be used to guide cancer treatment decisions and perhaps even screen for tumours that are not yet visible on imaging.
The Quality Choice for Monitoring Tumour Burden and Therapeutic Response
ctDNA could potentially be used to detect minimal residual disease and predicts recurrence after surgery or therapy. Over the past several years there have been multiple studies demonstrating the clinical utility of liquid biopsy ctDNA analysis following surgical resection of breast (17) and colorectal cancers (7,26). Studies demonstrate better outcomes when no tumour-derived DNA is found in patients following surgery, or chemotherapy in colorectal cancer patients (7,26) whereas those where tumour DNA is still present do better with the addition of more aggressive targeted treatment or chemotherapy.
Studies in melanoma (41), breast (15), ovarian (40), and colon (7) cancers have solidified the potential applications of ctDNA to monitor tumour burden dynamically and precisely during treatment process. Most studies, in particular BREAK-2 study (a phase II trial), showed that high basal ctDNA levels correlated with lower overall response rate (ORR) and lower progression free survival (PFS) to BRAF-inhibitor targeted therapy in melanoma patients (12,22,36,37). Progression free survival and overall survival (OS) were both significantly longer in the patients who had no detectable mutant EGFR alleles compared with those who did have detectable EGFR-mutant alleles after 3 cycles of treatment in advanced NSCLC33.
Bettegowda (26) and colleagues were able to detect circulating KRAS-mutant DNA fragments in the plasma of patients with KRAS-mutant tumours with high specificity and sensitivity, and high ctDNA levels were associated with decreased 2-year survival (26). Using digital polymerase chain reaction-based sequencing techniques and rearrangement analyses, in a group of 206 patients with metastatic colorectal cancer, the sensitivity of plasma ctDNA for detection of clinically relevant KRAS gene mutations was 87.2% and its specificity was 99.2% (26). In patients with metastatic colorectal cancer who developed resistance to EGFR antibodies, analysis of ctDNA identified the emergence of polyclonal KRAS, NRAS, BRAF, or EGFR mutations in 96% of panitumumab- or cetuximab refractory patients (26). Subsequently, Misale et al., were able to illustrate a way to use this information to overcome treatment resistance (27).
Targeted therapy remains the first-line treatment for melanoma tumours that harbour a BRAF mutation, particularly in Australia (11). Tracking of BRAF V600E ctDNA levels as patients undergo targeted treatment can potentially help identify the period when resistance to targeted therapy emerges. In a recent study of 634 patients with stage I to IV melanoma, Haselmann et al. (48), supported the routine use of ctDNA analysis to establish tumour genotype at diagnosis when treatment with targeted therapies is considered. With a high degree of concordance, ranging from 92.3% to 94.5%, of BRAF V600E mutation status between archival tumour tissue and plasma-based testing of cfDNA48, cfDNA provided reliable results for determining tumour genotype in early-stage melanoma patients, in whom ctDNA levels are often found to be very low (17,26,48). Furthermore, ctDNA can potentially be used as an immunotherapy predictive marker. Gray et al., and Lee et al., (22,24) showed that baseline ctDNA levels predict response to immunotherapy, PD-1 inhibitor, in melanoma patients, and that low basal ctDNA levels were significantly associated with long term clinical benefit. In this study, ctDNA levels at baseline and early during treatment provided an accurate prediction of tumour response, PFS and OS24. Nevertheless, prospective clinical studies of survival in a larger cohort of patients are underway to validate the predictive value of ctDNA in melanoma patients treated with immunotherapy.
Liquid Biopsy Can Guide Treatment Decisions
Because ctDNA can be used to monitor the effects of cancer treatment and give an early warning about possible recurrence, liquid biopsy can offer valuable insights into how best to treat cancer. The detection of resistant clones can offer clues to the reasons for treatment resistance (5,6,34; see Fig.2). Recently, a third-generation EGFR Tyrosine Kinase Inhibitor (TKI), that is effective in tumours harbouring the T790M EGFR mutation (~50-60% of lung cancer patients (5,6,10) ), was approved in Australia for patients with NSCLC harbouring the EGFR T790M mutation following progression on an EGFR TKI (9).
ctDNA can accurately reflect the tumour burden
ctDNA can accurately reflect the tumour burden. Serial analysis of ctDNA from the time of diagnosis throughout treatment can provide a dynamic picture of molecular disease changes, providing evidence that this non-invasive approach could also be used to monitor the development of secondary resistance and identify heterogeneous sub-clonal populations of tumour cells developing during the course of treatment (22,48). NCCN guidelines (8) recommend the use of liquid biopsy for lung cancer patients as an alternate for tissue in initial T790M EGFR testing (2,6). However, if the plasma is negative, then a tissue biopsy is recommended, if feasible. While the Cancer Council are looking at developing local guidelines for Australia, recently an Australian recommendation were made to test for resistance mutations using plasma ctDNA testing in NSCLC if available followed by a guided tissue biopsy if blood results are negative or indeterminate (2).
Challenges with Liquid Biopsy and Ongoing studies
Some studies are looking into offering liquid biopsy testing for patients with high risk such as hereditary cancer conditions. Indeed, the use of rapid systematic mapping approach found a broad range of relevant biomarkers allowing an evidence-based approach to identification of promising biomarkers for development of a blood-based cancer screening test in the general population. However, various prospective cohort studies are still in the research phase to determine the benefits of liquid biopsy testing as well as the positive predictive value of some cancer associated genes before it can be offered as a screening tool. It is of note that the detection of cancer-associated mutations on cfDNA in undiagnosed cancer patients might not indicate that the individual tested will develop cancer in her/his lifetime. Liquid biopsy should not be offered as a screening test as this stage as it has not yet been validated for that purpose.
Detailed experiences with early-stage cancer and low concentrations of ctDNA are required
While the American Society of Clinical Oncology (ASCO) and College of American Pathologists (CAP) encourage further prospective studies, the National Cancer Institute (NCI) is undertaking a validation study that can determine the role of liquid biopsy in detection of early-stage cancers, distinguishing cancer from benign conditions, and identifying fast- and slow-growing cancers. Most ctDNA studies focused on advanced-stage cancers with relatively high concentrations of ctDNA. Detailed experiences with early-stage cancer and low concentrations of ctDNA are required. The discovery of new targets for all types of tumours is underway which will help in the development of more effective drugs in future.
ctDNA holds great promise to fundamentally change how we approach patient treatment and clinical management
Overall, liquid biopsy offers a clear advantage to some cancer patients compared to conventional surgical methods. Previous studies highlight the suitability of ctDNA as a general monitoring tool for a clinical course of some cancers even for early stages, and for responses to targeted therapies. Clinical applications of liquid biopsies to inform molecular-based risk stratification and guide therapeutic intervention strategies may help reduce morbidity and overall costs, particularly for cancers where obtaining repeated tumour biopsies is challenging or unsafe. Liquid biopsies have value and may replace traditional tissue biopsies if a result is found. However, if a result is not found, a tissue biopsy is still recommended. Further research and clinical trials are ongoing to explore the clinical applicability for ctDNA in oncology. Indeed, ctDNA holds great promise to fundamentally change how we approach patient treatment and clinical management.
- Fleischhacker M et al., (2007) Biochem Biophys Acta 1775:181-232.
- John T et al., (2017) Asia-Pacific J of Clin Oncol 13:296-303.
- Thierry A et al., (2014) Nat Med 20:430-435.
- Huang T et al., (2013) Biomark Res 1: 3.
- Thress K et al., (2015) Lung Cancer 90:509-515.
- Wu Y et al., (2017) Br J cancer 116:175-185.
- Tie J et al., (2016) Sci Transl Med 8:346ra92.
- National Comprehensive Cancer Network. Clinical Practice Guidelines in Oncology: Non-small cell lung cancer (7 April 2017)
- AstraZeneca. TAGRISSO (osimertinib mesilate) product information; 2016.
- Yu H et al., (2013) Clin Cancer Res 19:2240–2247.
- Spagnolo F et al., (2015) Onco Targets Ther 8:157-168.
- Ascierto P et al., (2013) J Clin Oncol 31 (26):3205-3211.
- Newman A. M et al., (2014) Nat Med 20 (5):548-554.
- Murtaza M et al., (2015) Nat Commun 6:8760.
- Dawson S et al., (2013) N Engl J Med 368 (13):1199-1209.
- Forshew T et al., (2012) Sci Transl Med 4 (136):136ra68-136ra68.
- Garcia-Murillas I et al., (2015) Sci Transl Med 7 (302):302ra133-302ra133.
- Murtaza M et al., (2013) Nature 497 (7447):108-112.
- Wang W et al., (2017) Cancer Med 6 (1):154-162.
- Van Emburgh B et al., (2014) Mol Oncol 8 (6):1084-1094.
- Premarket approval P150044 d Cobas EGFR MUTATION TEST V2. Cited 2016; Available from: https://www.accessdata.fda.gov/ scripts/cdrh/cfdocs/cfpma/ pma.cfm?id¼P150044
- Gray E et al., (2015) Oncotarget 6 (39):42008-42018.
- Girotti M et al., (2016) Cancer Discov 6 (3):286-299.
- Lee J et al., (2017), Ann Oncol.
- Calapre L et al., (2017) Cancer Letters 404:62-69.
- Bettegowda C et al., (2014) Sci Transl Med 6:224ra24.
- Misale S et al., (2014) Sci Transl Med 6:224ra26.
- Kim E et al., (2008) Lancet 372:1809-1818.
- Bedard PL et al., (2013) Nature 501:355-364.
- Jiang P et al., (2015) Proc Natl Acad Sci U S A 112: E1317-1325.
- Diehl F et al., (2005) Proc Natl Acad Sci U S A102:16368-16373.
- Chen Q et al., (2007) J Mol Diag 9: 272-76.
- Mok T et al., (2015) Clin Cancer Res 21:3196-3203.
- Frenel J et al., (2015) Clin Cancer Res 21:4586-4596.
- Canzoniero J & Park B (2016) Cancer 2 (1865): 266-274.
- Sanmamed M et al., (2015) Clin Chem 61 (1):297-304.
- Santiago-Walker A et al., (2016) Clin Cancer Res 22 (3):567-574.
- Pavlakis N et al., (2017) Cancer Council Australia Lung Cancer Guidelines Working Party. What is the Optimal First-line Maintenance Therapy for Treatment of Stage IV Inoperable NSCLC?
- Shi J et al., (2007) Hum Mutat 28: 131-36.
- Forshew T et al., (2012) Sci Transl Med 4(136):136ra68.
- Bidard FC et al., (2014) Int J Cancer 134(5):1207–1213.
- Misale S et al., (2012) Nature 486(7404):532–536.
- Diehl F et al., (2008) Nat Med 14:985-990.
- Sherwood JL et al (2017) Esmoopen.bmj.com/contents/2/4e/e000235.
- Higgins MJ et al., (2012) Clin Cancer Res 18:3462-3469.
- Diaz et al., (2012) Nature 486:537-540.
- Chao M & Gibbs P (2009) J Clin Oncol 27:e279–e280.
- Haselmann V et al., (2018) Clinical Chemistry 64 (5):830–842