Jump to ContentJump to Main Navigation
Show Summary Details
More options …

Journal of Laboratory Medicine

Official Journal of the German Society of Clinical Chemistry and Laboratory Medicine

Editor-in-Chief: Schuff-Werner, Peter

Editorial Board: Ahmad-Nejad, Parviz / Bidlingmaier, Martin / Karsten, Conrad / Fraunberger, Peter / Ghebremedhin, Beniam / Holdenrieder, Stefan / Kiehntopf, Michael / Klein, Hanns-Georg / Klouche, Mariam / Kohse, Klaus P. / Kratzsch, Jürgen / Luppa, Peter B. / März, Winfried / Nebe, Carl Thomas / Orth, Matthias / Sack, Ulrich / Steimer, Werner / Weber, Bernard / Wieland, Eberhard / Zettl, Uwe K.

6 Issues per year


IMPACT FACTOR 2017: 0.216

CiteScore 2017: 0.22

SCImago Journal Rank (SJR) 2017: 0.158
Source Normalized Impact per Paper (SNIP) 2017: 0.082

Online
ISSN
2567-9449
See all formats and pricing
More options …
Volume 40, Issue 5

Issues

Potentials, challenges and limitations of a molecular characterization of circulating tumor DNA for the management of cancer patients

Potenzial, Herausforderungen und Grenzen einer molekularen Charakterisierung von zirkulierender Tumor-DNA

Peter Ulz / Armin Gerger
  • Division of Oncology, Medical University of Graz, Graz, Austria
  • Center for Biomarker Research in Medicine, Graz, Austria
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
/ Jelena Belic / Ellen Heitzer
Published Online: 2016-10-07 | DOI: https://doi.org/10.1515/labmed-2016-0049

Abstract:

A liquid profiling, i.e. the analysis of cell-free circulating tumor DNA (ctDNA), enables a continuous non-invasive monitoring of tumor-specific changes during the entire course of the disease with respect to early detection, identification of minimal residual disease, assessment of treatment response and monitoring tumor evolution. Technological improvements, advances in understanding the nature of ctDNA, the implementation of ctDNA analyses in clinical trials as well as efforts for the establishment of benchmarks, will bring an actual widespread clinic use within reach in the near future. However, despite this progress there are still hurdles that have to be overcome, which are discussed in this review. Moreover, present knowledge and new findings about the biology of ctDNA as well as selected potential clinical applications for metastatic cancer patients are pointed out.

Zusammenfassung:

Das sog. Liquid Profiling bezeichnet die Analyse von zell-freier zirkulierender DNA (ctDNA). Mit Hilfe eines Liquid Profilings ist es möglich Tumor-spezifische Veränderungen nicht-invasiv über den gesamten Krankheitsverlauf zu überwachen. Potentielle Anwendungen betreffen nicht nur das Überwachen des Therapieansprechens oder eine Detektion von minimaler Resterkrankung sondern auch die Verfolgung der Tumorevolution in Echtzeit und somit die Identifiziereng von neu aufkommenden Therapiezielen oder Resistenzmechanismen. Fortschritte, die unterschiedlichen Technologien für die Analyse von ctDNA betreffend, sowie Bemühungen Qualitätsstandards zu etablieren rücken eine tatsächliche Implementierung von ctDNA Analysen in die klinische Routine in greifbare Nähe. Aber auch trotz dieser Fortschritte gilt es noch einige Hürden zu überwinden, die unter anderem in diesem Übersichtsartikel diskutiert werden. Zudem werden das momentane Verständnis und neue Erkenntnisse in Bezug auf die Biologie der ctDNA, sowie ausgewählte klinische Anwendungen für Patienten mit metastasierter Tumorerkrankung präsentiert.

Reviewed publication

HoldenriederS.

Keywords: assessment of treatment response; cell-free circulating DNA; ctDNA; identification of actionable targets; monitoring tumor evolution; predictive and prognostic biomarker

Schlüsselwörter:: ctDNA; Identifizierung von Therapiezielen; prädiktive und prognostische Biomarker; Überwachung des Therapie Ansprechens; zell-freie zirkulierende Tumor DNA

Introduction

Real-time monitoring for the management of cancer patients by the use of liquid profiling, often also referred to as liquid biopsy, is close to becoming reality. It has been shown that the analysis of cell-free circulating tumor DNA (ctDNA) reflects the genomic landscape of a tumor and its metastases and therefore it can be used to assess molecular characteristics from a tumor in a non-invasive manner. Furthermore, ctDNA might indicate if a specific treatment, which was chosen based on a particular molecular target, remains relevant or if any resistance-conferring mechanisms emerge. Therefore, liquid profiling offers what tissue biopsies cannot – a continuous monitoring of tumor-specific changes during the entire course of the disease. The potential and the benefit of ctDNA analyses with respect to early detection, identification of minimal residual disease, assessment of treatment response and monitoring tumor evolution are extensively discussed elsewhere [1], [2], [3], [4], [5], [6], [7], [8]. Due to technological improvements the analysis of extremely rare alleles from plasma is now feasible and allows the use of ctDNA in many clinical scenarios. Although the analysis of ctDNA holds great promise its current use is, however, limited to the analysis of single or few hot spots, such as the detection of resistance conferring mutations in EGFR or KRAS under anti-EGFR treatment. For tissue specimens a variety of tests for predictive molecular hotspots are cleared by the FDA, while for plasma DNA only the cobas® EGFR Mutation Test v2 is approved (Table 1).

Table 1:

List of selected cleared or approved companion diagnostic tests.

For a widespread clinical use we lack a consensus on how to best use the available methods as well as benchmarks for pre-analytics and analytics of ctDNA. In addition, although ctDNA analyses are becoming more and more implemented in clinical trials, the actual benefit for patients in terms of survival has yet to be proven. Finally, we still have much to learn about the biology and dynamics of ctDNA. The purpose of this review is to point out potentials, challenges and limitations of ctDNA as a clinical biomarker. First, we give a brief overview about present knowledge and new findings about the biology of ctDNA. Second, we focus on potential applications of ctDNA for metastatic cancer patients. Finally, we point out what still has to be done to make the dream of an actual real-time monitoring of cancer patients come true.

The nature of cfDNA and ctDNA

Although the presence of cell-free DNA (cfDNA) was already discovered in 1948 [9], it took almost 30 years before its use as a potential cancer biomarker was known. In 1977 Leon et al. for the first time reported elevated levels of cfDNA in the circulation of cancer patients. Moreover, after successful anticancer therapy even a decrease of cfDNA was observed [10]. More than 10 years later, using a method based upon the decreased strand stability of cancer cell DNA, Stroun et al. was able to show that DNA found in the circulation of cancer patients was derived from cancer cells [11]. Even back then, the authors suggested that the presence of increased amounts of cfDNA alone is not a sufficiently strong biomarker for the diagnosis of cancer and the way to use cfDNA for clinical purposes in oncology is the detection of tumor-specific changes. Sorensen et al. have been successful, for the first time, in providing evidence that parts of the cfDNA indeed originate from tumors [12]. The authors demonstrated the occurrence of mutant KRAS fragments in the plasma of patients with pancreatic adenocarcinoma. One year later it was already recognized that plasma DNA circulates mainly as mononucleosomes and it is likely to be associated with apoptosis. In the following years most groups focused on the detection of a variety of tumor-specific alterations such as microsatellite instability (MSI), loss of heterozygosity (LOH), point mutations, and aberrant methylation in plasma burden [13], [14], [15], [16], [17], [18], [19], [20], but hardly any data were published on the actual nature, the dynamics of release or the kinetics of clearance of cfDNA/ctDNA. Due to the fact that cfDNA is not only found in pathological conditions such as systemic lupus erythematosus (SLE), rheumatoid arthritis, trauma, myocardial infarction, stroke or cancer [1], [2], [21], [22], [23], [24], [25], [26], the release of DNA is in principle a physiological process. In healthy individuals cfDNA is degraded by peripheral blood DNase activity leading to relatively low levels in the circulation [27], [28]. It is thought that the fact that cancer patients show elevated levels of cfDNA might be attributed to lower DNase activity in blood plasma of cancer patients [27]. Another theory is that DNA of cancer patients could be resistant to DNase digestion [29], [30]. Under physiological conditions there is a balance between DNA release and clearance. However, as the volume of the tumors increases, the number of apoptotic and dead cells also increases resulting in elevated cellular turnover leading to an imbalance between cfDNA spread and clearance. Although all these data might be an explanation for elevated levels in cancer patients compared to healthy individuals it is still widely unclear why cancer patients show such a dramatic variability of absolute cfDNA levels and fractions of ctDNA. ctDNA fractions range from less than 1 to more than 90% even in metastatic stages. Possible reasons for this variability include tumor burden, stage, vascularity, proliferation state and cellular turnover, degree of vascularization and response to therapy [1], [31]. Moreover, circadian rhythms, inflammation, or other individual-specific factors may affect ctDNA release and clearance [32], [33].

In general, the release of ctDNA is thought to be a fluctuating, stochastic process rather than a continuous process. Due to the fact that cfDNA is cleared rapidly from the maternal plasma after delivery [34], [35], most data on stability and half-life of cfDNA come from fetal DNA studies. Recent data from Dennis Lo’s group suggests an initial rapid phase, with a mean half-life of 1 h and a subsequent slow phase, with a mean half-life of 13 h [35]. These data were confirmed in healthy subjects performing extensive exercise [36]. In cancer patients it is much more difficult to prove the clearance and dynamics of ctDNA due to the above mentioned imbalances regarding release and DNase activity, thus making these estimates not directly transferrable to ctDNA. Nevertheless, it could be shown that after a curative surgery or successful treatment, the amount of ctDNA in the circulation rapidly decreases indicating also a short half-life. This makes ctDNA a perfect biomarker for monitoring compared to conventional protein biomarkers, which often stay in the blood for weeks and therefore show delayed response.

It is well documented that cfDNA is double-stranded and forms a specific ladder pattern known from apoptotic cells ranging from 60 to >1000 bp with a predominant peak of around 166 bp, which corresponds to the length of DNA wrapped around a nucleosome including a linker region. In order to accurately evaluate fragment sizes in cancer patients the Lo group made use of aneuploid regions of the cancer genome, in which ctDNA is either enriched as for amplified regions or depleted as for deleted regions [37]. Massively parallel sequencing revealed in addition to the 166 bp peak a series of smaller peaks occurring at 10 bp periodicity at sizes of approximately 143 bp and shorter [37]. Size profiles were highly correlated with the amount of tumor DNA in the circulation, whereas the abundance of shorter fragments was associated with larger amounts of tumor DNA [37]. Similar observations were already made in 2011, when the Thierry group demonstrated the presence of a higher proportion of cfDNA fragments below 100 bp particularly in samples from cancer [38]. However, data from our group indicated that an inefficient clearance of nucleosomal derived DNA may lead to multiples of mono-nucleosomal DNA, which correlated with higher fractions of tumor DNA [33], [39]. Taken together these data suggest that besides the assessment of tumor-specific alterations from plasma, size profiling is an important accessible parameter and might also be used as a clinical application in the near future [40].

Concerning the cells of origin of cfDNA it has become clear that cfDNA constitutes a mixture of DNA released from cells from different tissues of the body with varying proportions in different conditions. Two studies from Lo’s group and Snyder and colleagues, respectively, contributed immensely to the present knowledge of cfDNA origin. In pregnant women, for example, it was shown that the placenta is the origin of fetal cfDNA (cffDNA) detectable in the maternal circulation whereas cffDNA ranged from 12.1 to 41.0% of overall cfDNA [35]. Moreover, these studies showed that cfDNA circulating in the blood of transplant recipients originated from the donor genome suggesting that cfDNA can be used to detect an organ-specific signature. Using whole genome bisulfite sequencing Lo’s groups established organ-specific DNA methylation signatures in order to trace back the origin of cfDNA fragments in pregnant women, patients with hepatocellular carcinoma and subjects following bone marrow and liver transplantation [41]. Consistent with previous reports in healthy individuals, cfDNA is primarily derived from apoptosis of normal cells of the hematopoietic lineage and other solid tissues contribute only to a small part of cfDNA. In contrast, in progressive cancer patients, a large part of the circulating DNA fragments originated from the tumor tissue [41]. Snyder et al. used a different approach in order to assess the epigenetic landscape of the tissue(s)-of-origin of cfDNA. The authors employed deep WGS (96- to 105-fold coverage) and generated maps of genome-wide in vivo nucleosome occupancy and observed features similar to micrococcal nuclease (MNase)-derived, nucleosome-associated fragments. Since nucleosome positioning varies between cell types, the authors reasoned that plasma derived nucleosome occupancy patterns could offer information regarding the cell type of origin. In healthy individuals a pattern seen in lymphoid and myeloid cells, consistent with hematopoietic cell death as the normal source of cfDNA, was observed.

Ivanov et al. further explored nucleosome occupancy in plasma DNA and with respect to gene expression patterns [42]. Moreover, our group also observed expression-specific nucleosome occupancy at promoters [43]. Using machine learning we were able to predict whether genes are expressed or not based on a nucleosome-depleted region, which is often observed at active promoter sites. Furthermore, we showed that whole genome sequencing data from plasma samples from cancer patients can deduce expressed cancer driver genes.

Taken together these novel findings based on methylation and nucleosome occupancy suggests that plasma DNA can reveal functional data and may aid in the development of novel biomarkers reflecting pathological changes in chromatin marks and the epigenetics landscape independent of genotypic differences between contributing cell types [42], [43], [44].

Potential applications of ctDNA in metastatic cancer patients

Although treatment of metastatic cancer has dramatically improved in recent years, mainly by the implementation of targeted therapies, metastases are the cause of 90% of human cancer deaths. Due to a massive heterogeneity and a continuous evolution of tumor cells metastases are difficult to treat. Treatment decisions are mostly based on the molecular characterization of the primary tumor, however, the molecular landscape of metastases can be fundamentally different from the primary lesion. And this is where the analysis of ctDNA enters the scene. As ctDNA is thought to be a surrogate for the entire tumor burden, it enables a comprehensive analysis of the tumor genome without the need for invasive procedures and thereby allowing assessment of tumor heterogeneity and the identification of novel occurring targets or resistance mechanisms. Moreover, it has been shown that changing levels of ctDNA reflect response to treatment, thus it may help clinicians to assess whether a chosen treatment is successful or fails long before a clinical response/progression is evident.

Monitoring treatment response and tumor evolution

A close monitoring of tumors during therapy is becoming increasingly important, whereas the classical tumor markers are becoming less important and are being replaced by “genetic” or “molecular” markers. A variety of studies demonstrated that a continuous monitoring of changing levels of ctDNA during initiation and maintenance of cancer therapy can be used to assess the patient’s response without burdening the patient, and more importantly progression can be detected before it is clinically obvious [45], [46], [47], [48].

A landmark study from Dawson et al. demonstrated that ctDNA levels showed a greater dynamic range, and greater correlation with changes in tumor burden in metastatic breast cancer patients, than did CA15-3 or circulating tumor cells. Moreover, it was shown that, ctDNA provided the earliest measure of treatment response in more than 50% of the patients [49]. The same group reported the first exome sequencing of plasma DNA from six patients with advanced cancer. In all patients resistance-conferring mutations could be identified indicating that the selective pressure of targeted therapy can drive an increase of resistant clones. Therefore, serial analyses of cancer genomes in plasma constitute a new paradigm for the study of clonal evolution in human cancers [50]. In a further study the authors showed that multifocal clonal evolution can be assessed using ctDNA and that serial changes in circulating levels of sub-clonal private mutations correlate with different treatment responses between metastatic sites [51]. Similar data come from Butler et al., who used whole-exome sequencing to investigate tumor evolution in breast cancer [52]. Although, identified mutations correlated well between metastases and cfDNA, an ESR1 mutation was not found in the primary tumor indicating that a subclonal change under the selective treatment pressure with anastrozole and herceptin occurred [52]. In a study from Frenel et al. it was shown that molecular changes at cfDNA mutation level were highly associated with time to disease progression by RECIST criteria. The authors tracked known tumor mutations monthly in advanced cancer patients, who completed at least two courses of investigational targeted therapy until disease progression and demonstrated potential treatment associated clonal responses during treatment [48].

Data from our group show similar results for metastatic prostate cancer patients. Decreasing levels of ctDNA were associated with good response to treatment with androgen-deprivation therapy (ADT) (Figure 1). On the contrary, consistent or elevated levels of ctDNA reflected treatment failure and the development of castration resistance. The molecular landscape underpinning response to the androgen receptor (AR) antagonists is not well defined, although AR gene aberrations are well-established resistance markers and can be identified noninvasively from plasma [53], [54], [55], [56]. Our approach for monitoring treatment response in plasma is two-fold (Figure 2). On the one hand, we established genomewide somatic copy number alterations (SCNAs) from low-coverage whole genome sequencing of cfDNA. Using the very same library we enrich for coding sequences of the most frequently mutated cancer driver genes, which in the case of prostate cancer also includes the entire transcribed region of ERG and TMPRSS2 in order to detect fusion breakpoints. With this approach we can comprehensively monitor tumor-specific changes on the genomic, the gene and the nucleotide level. Recently we further developed our previous analysis algorithms in order to identify focal amplifications [56]. Focal amplifications are highly relevant predictive biomarkers as they are frequently occurring under therapy and may therefore have important therapeutic implications as they may, on the one hand, explain the development of resistance against targeted therapies (e.g. AR amplification, KRAS amplification) or on the other hand, may contain genes for which targeted therapies exist (e.g. ERBB2 amplification, FLT3 amplification) [57]. With this approach we were able to identify novel therapy targets, to assess tumor evolution during the course of disease, and to correlate mutant allele frequencies with therapy response [39], [55], [56], [57], [58]. Our established plasma-Seq method can obtain clinically relevant information in a cost-effective and fast manner [39], [55], [58], [59], whereas many other comprehensive approaches are still too expensive and time-consuming for actual clinical use. One drawback is that our method only yields informative results if approximately 5%–10% of tumor DNA is present in the circulation. Due to the fact that even in highly metastasized patients there are clinical situations where ctDNA is present below optimal levels for the detection of genomewide alterations [31], [39], [55], [60], and approximately 20%–30% of analyses would not yield informative results, we now use our recently published mFAST-SeqS method in order to preselect samples with sufficient amounts of tumor DNA (Figure 2) [61]. As the calculated mFAST-SeqS-z-score highly correlates with mutant allele frequencies it can also be used as an untargeted approach for the establishment of changing ctDNA under a certain treatment.

Monitoring treatment response to androgen depriviation therapy (ADT) of metastatic prostate cancer patients. (A+B) Patients, who responded well to the treatment, showed a massive decrease in ctDNA levels. After 2 months of therapy neither SCNAs nor the mutations in CTNNB1 and TP53, respectively could be observed. Moreover, the genomewide z-scores established with mFAST-SeqS, which correlated with ctDNA levels, dropped dramatically. (C) This patient did not respond to the therapy. In contrast elevated levels of ctDNA were observed after 6 months of treatment, which is reflected in higher log-2 ratios of the SCNAs and an increase of a somatic BRCA1 mutation from 47.4% to 79.5%. Treatment failure could be attributed to a high level amplification of the AR gene, which is known to be associated with resistance to ADT and the development of castration resistant prostate cancer.
Figure 1:

Monitoring treatment response to androgen depriviation therapy (ADT) of metastatic prostate cancer patients.

(A+B) Patients, who responded well to the treatment, showed a massive decrease in ctDNA levels. After 2 months of therapy neither SCNAs nor the mutations in CTNNB1 and TP53, respectively could be observed. Moreover, the genomewide z-scores established with mFAST-SeqS, which correlated with ctDNA levels, dropped dramatically. (C) This patient did not respond to the therapy. In contrast elevated levels of ctDNA were observed after 6 months of treatment, which is reflected in higher log-2 ratios of the SCNAs and an increase of a somatic BRCA1 mutation from 47.4% to 79.5%. Treatment failure could be attributed to a high level amplification of the AR gene, which is known to be associated with resistance to ADT and the development of castration resistant prostate cancer.

Workflow of plasma-Seq for monitoring treatment response in metastatic cancer patients. After plasma DNA extraction, cfDNA is quality checked on an Agilent Bioanalyzer. An enrichment of fragments with a near mode of 166 bp indicates high quality of cfDNA. In order to select samples with sufficient fractions of tumor DNA (>5%–10%), which are suitable for a comprehensive, genomewide analysis a pre-screening using mFAST-SeqS is performed. Samples with genomewide z-scores above 5 are further analyzed with plasma-Seq. To this end a library preparation is performed using a slightly modified protocol of the Illumina TruSeq Nano Library Prep kit. One part of the library is directly subjected to low-coverage whole genome sequencing on an Illumina MiSeq or a NextSeq. The other part of the library is enriched for the most frequently mutated cancer driver genes and sequenced in a separate run. With this approach we can comprehensively monitor tumor-specific changes on the genomic, the gene and the nucleotide level. For samples with genomewide z-scores below 5, high resolution methods are needed in order to obtain informative results.
Figure 2:

Workflow of plasma-Seq for monitoring treatment response in metastatic cancer patients.

After plasma DNA extraction, cfDNA is quality checked on an Agilent Bioanalyzer. An enrichment of fragments with a near mode of 166 bp indicates high quality of cfDNA. In order to select samples with sufficient fractions of tumor DNA (>5%–10%), which are suitable for a comprehensive, genomewide analysis a pre-screening using mFAST-SeqS is performed. Samples with genomewide z-scores above 5 are further analyzed with plasma-Seq. To this end a library preparation is performed using a slightly modified protocol of the Illumina TruSeq Nano Library Prep kit. One part of the library is directly subjected to low-coverage whole genome sequencing on an Illumina MiSeq or a NextSeq. The other part of the library is enriched for the most frequently mutated cancer driver genes and sequenced in a separate run. With this approach we can comprehensively monitor tumor-specific changes on the genomic, the gene and the nucleotide level. For samples with genomewide z-scores below 5, high resolution methods are needed in order to obtain informative results.

In our very recent work we demonstrated that prostate cancer genomes show a high plasticity with newly occurring focal amplifications as a driving force in progression [56]. Furthermore, subclonal diversification of the tumor was correlated with clonal pattern changes in the tumor genome. Similar data come from Carreira et al. who identified multiple independent clones in metastatic disease, which are differentially represented in tissue and the circulation [54]. Their data suggest complex dynamics of tumors with temporal and spatial heterogeneity. These distinct mechanisms of resistance at different sites which emerged and regressed depending on treatment selection pressure again highlight the need for sequential monitoring of advanced prostate cancer.

Identification of actionable targets

Due to the increasing number of targeted therapies, which are administered on the basis of tumor-specific changes or aberrant signal transduction, there is a paradigm shift in cancer therapy. Molecular stratification of patients to targeted therapies is currently based on a molecular characterization of tumor tissues. However, the analysis of tumor tissue comes with several limitations such as bad DNA quality due to paraffin embedding, the fact that archival biopsies may not reflect the current and complete genetic profile of a tumor, and that a biopsy bears some risks and burdens both the patients and the health care system. Liquid profiling, i.e. the analysis of ctDNA, is a promising alternative to tissue biopsies since it is minimally-invasive, it can be immediately analyzed and most likely provides a more complete molecular profile of tumor heterogeneity than a single biopsy achieves. Although, there is a great potential for ctDNA as a predictive marker, there are hardly any published data proving that ctDNA molecular profiling can indeed identify actionable targets, help guide treatment selection and that patients would indeed benefit from such analyses.

In a proof of principle study De Mattos-Arrunda et al. identified two additional mutations of potential therapeutic interest in ctDNA, which could not be detected in the corresponding tumor samples in a patient with a ER+/Her2- carcinoma [62]. More and more groups now start prospective studies in order to assess the potential of ctDNA as a predictive marker in real clinical settings. The Division of Clinical Oncology at our university, for example, conducts in cooperation with our group the ICT (Individualized Cancer Treatment) study. In this study the success of a targeted therapy based on a molecular-genetic tumor profiling will be evaluated by the progression-free survival (PFS) of the targeted therapy compared to the PFS of the last evidence-based drug therapy. Molecular profiling will be performed from ctDNA and if the patient consents for a freshly taken biopsy from a metastasis. While tumor samples are analyzed at the Institute of Pathology, blood samples are immediately analyzed at our institute. The analysis of these samples is two-fold, on the one hand, we perform our plasma-Seq approach to identify potential actionable SCNAs, particularly focusing on focal amplifications, as these mostly harbor driver genes that might be used as actionable targets, such as CDKs, FLT3, HER2, or MET. As the current number of targets with associated therapeutic options is limited, we additionally perform mutational analysis using a cancer hotspot panel that covers only known clinically relevant genes and hot spots. Publicly available databases such as My Cancer Genome [63], TARGET [64] or Therapeutic Targets Database (TTD) [65] are used to derive clinically relevant conclusions. Whether or not a targeted therapy can be administered and whether an off-label use is indicated, will be decided by a molecular tumor board including a geneticist, a pathologist and oncologists.

A similar, but more comprehensive study is performed at Cancer research UK in Manchester.

The Tumour chARacterisation to Guide Experimental Targeted Therapy Trial (TARGET) tests the hypothesis that ctDNA molecular profiling can be used to help guide selection of experimental medicines and monitors tumor response for patients that receive a matched targeted therapy [66]. To this end, in blood, archival tumor samples and optional fresh biopsies from eligible patients a total of 654 genes will be analyzed. In a first recruiting phase 100 patients will be included in order to optimize analysis algorithms and reporting strategies. In a second phase 250 patients will be recruited with the aim to report and discuss results in a molecular tumor board within 14 days. Outcomes will include the number of patients for which ctDNA analysis guided trial selection and response rates/PFS to matched targeted drugs.

Hurdles that have to be overcome

Although liquid profiling represents a promising tool for cancer therapy, there are many hurdles that have to be overcome for a widespread clinical implementation starting from the as yet unspecified regulations with respect to implementation and validation of blood-based tests as well as the current lack of benchmarks and a consensus of quality requirements to the question whether or whoever will reimburse such investigations and at what level of evidence reimbursement should be carried out.

Clinical validation is also essential for implementing such tests. Such validation involves assessment of the sensitivity, specificity, cut-offs, and other parameters of a test [67], [68]. The first large scale quality control scheme external quality assessment (EQA) (SPIDIADNAplas, http://www.spidia.eu/) which among other things investigated the role of pre-analytical variables on cfDNA quality and quantity parameters (integrity and quantity) was reported in 2015 [69]. Fifty-six laboratories throughout Europe received the same plasma samples and extracted plasma DNA using their own established method. Isolated cfDNA was sent to the SPIDIA facility where cfDNA quantity and quality, expressed as recovery and integrity, respectively, was assessed [69]. Consistent with previous reports, results of this study indicated that confounding parameters like different storage conditions and extraction procedures are in fact sources of the wide variation in the quantity and integrity of cfDNA, further highlighting the need for harmonization and the establishment of standard operating procedures (SOPs). In order to meet these requirements the EU-funded IMI-project CANCER-ID consortium was founded in 2014 (http://www.cancer-id.eu/the-project/) and aims to establish standard protocols for and clinical validation of blood-based biomarkers.

Another unsolved question is who should do the testing? Usually clinical testing should be performed in accredited laboratories in order to ensure the quality of testing systems and the reproducibility of results. With this respect EQA should be carried out to evaluate both pre- and post-analytical activities.

Moreover, it is not clear yet, to what extent plasma DNA should be analyzed. Although several tests for the analysis of hotspot mutations for EGFR or KRAS are already in clinical use, these tests are without a doubt not the only ones that will find application in daily clinical routine. Due to the reduced costs and increased efficiency of sequencing, hundreds of genes can be analyzed all at the same time. Instead of just testing for a single target, in the near future multiple types of genetic alterations that might drive decisions for therapy will be tested. However, at the same this means that data interpretation will become more difficult and we do not know how and what kind of data to report to the clinic. Even for markers that are clearly associated with treatment response variable levels of evidence exist. Joint efforts are currently being made by members of Illumina, Memorial Sloan Kettering, MD Anderson, Dana-Farber and Fred Hutchinson, which founded the “Actionable Genome Consortium” in order to accurately define what constitutes a “cancer actionable genome,” or the genomic changes that define an individual patient’s tumor, so that oncologists and pathologists can determine optimal therapies and testing strategies to improve patient outcomes [70].

Another challenge in the context of big data is to decide whether and how to report incidental findings to patients. These issues can only be solved by validation and sophisticated bioinformatic algorithms to make the data more useful. Roadmaps are needed of how patients are referred, matched, and enrolled onto different types of early-phase trials depending on their molecular profiles regardless of whether the profiling was done in tumor tissue or non-invasively from plasma.

Usually the intended use determines the rules for a diagnostic test. A test for the identification of gene variants also carries the risk of incidental findings that are not actually relevant to the diagnostic indication or affect the germline. Importantly, in contrast to tumor tissue, in which tumor cells are usually enriched by a dissection, cfDNA is mostly derived from the hematopoietic system and tumor-specific fragments are often underrepresented. The exact origin of a mutation, i.e. whether it is somatic or germline, is still barely determined. One should keep in mind that the detection of germline mutations in known cancer-predisposing genes does not only affect the patient himself and his/her treatment, but has also tremendous implication for his/her family members. If a germline mutation is detected, genetic counseling of patients and their families is essential. Preliminary data indicated that those affected from such issues want to be informed and educated about such incidental findings and related actions. Most institutions lack both capacity and expertise for the interpretation of germline variants and genetic counseling. Correct clarification or advice for such families can only be ensured by trained medical and clinical laboratory geneticists. As most of these issues also apply to molecular pathologists and tissue based analyses, a close cooperation between clinicians, pathologists and human geneticists is absolutely essential in order to finally bring liquid profiling into clinic.

Conclusions

The analysis of ctDNA for liquid profiling has not without reason attracted a great deal of attention in recent years as its utility for early detection, monitoring response to treatment, and predicting survival of cancer patients has been proven in numerous studies. Liquid profiling offers real-time monitoring of cancer patients during the entire course of the disease as blood can easily be obtained at any time before, during and after a therapy. Recent studies have even shown that also functional data, such as tissue-of-origin or expression patterns, can be acquired by the analysis of cfDNA. Yet, although progress was being made in clarifying the biology and nature of ctDNA, we still lack knowledge about the exact dynamics and clearance mechanisms. Moreover, despite many promising studies ctDNA is not fully implemented in daily patient care yet. The major hurdle for a routine use is probably the fact that we lack benchmarks, a consensus on what to analyze and how to report these results and that most methods are not routinely applicable considering time, costs, technical equipment and the training requirements of the personnel. From an analytical perspective many methods for the analysis of ctDNA can be automated which is of advantage for a broad clinical application. Nevertheless, regulatory issues with respect to reimbursement and clearance for clinically applicable tests need to be solved. The field is rapidly developing and improving and it is only a matter of time until ultimately the patients benefit from these developments.

References

  • 1.

    Heitzer E, Ulz P, Geigl JB. Circulating tumor DNA as a liquid biopsy for cancer. Clin Chem 2015;61:112–23. Google Scholar

  • 2.

    Diaz LA Jr, Bardelli A. Liquid biopsies: genotyping circulating tumor DNA. J Clin Oncol 2014;32:579–86. Google Scholar

  • 3.

    Alix-Panabieres C, Schwarzenbach H, Pantel K. Circulating tumor cells and circulating tumor DNA. Ann Rev Med 2012;63:199–215. Google Scholar

  • 4.

    Schwarzenbach H, Hoon DS, Pantel K. Cell-free nucleic acids as biomarkers in cancer patients. Nat Rev Cancer 2011;11:426–37. Google Scholar

  • 5.

    Schwarzenbach H, Pantel K. Circulating DNA as biomarker in breast cancer. Breast Cancer Res 2015;17:136–44. Google Scholar

  • 6.

    Karampini E, McCaughan F. Circulating DNA in solid organ cancers-analysis and clinical application. QJM 2016;109:223–7. Google Scholar

  • 7.

    Esposito A, Bardelli A, Criscitiello C, Colombo N, Gelao L, Fumagalli L, et al. Monitoring tumor-derived cell-free DNA in patients with solid tumors: Clinical perspectives and research opportunities. Cancer Treat Rev 2014;40:648–55. Google Scholar

  • 8.

    Anker P, Stroun M. Circulating DNA in plasma or serum. Medicina 2000;60:699–702. Google Scholar

  • 9.

    P MPaM. Les acides nucleiques du plasma sanguin chez l’homme [in French]. Seances Soc Biol Fil 1948;142:241–3. Google Scholar

  • 10.

    Leon SA, Shapiro B, Sklaroff DM, Yaros MJ. Free DNA in the serum of cancer patients and the effect of therapy. Cancer Res 1977;37:646–50. Google Scholar

  • 11.

    Stroun M, Anker P, Maurice P, Lyautey J, Lederrey C, Beljanski M. Neoplastic characteristics of the DNA found in the plasma of cancer patients. Oncology 1989;46:318–22. Google Scholar

  • 12.

    Sorenson GD, Pribish DM, Valone FH, Memoli VA, Bzik DJ, Yao SL. Soluble normal and mutated DNA sequences from single-copy genes in human blood. Cancer Edipem Biomar 1994;3:67–71. Google Scholar

  • 13.

    Chen XQ. Microsatellite alterations in plasma DNA of small cell lung cancer patients. Nat Med 1996;2:1033–5. Google Scholar

  • 14.

    de Kok JB, van Solinge WW, Ruers TJ, Roelofs RW, van Muijen GN, Willems JL, et al. Detection of tumour DNA in serum of colorectal cancer patients. Scand J Clin Lab Inv 1997;57:601–4. Google Scholar

  • 15.

    Kopreski MS, Benko FA, Kwee C, Leitzel KE, Eskander E, Lipton A, et al. Detection of mutant K-ras DNA in plasma or serum of patients with colorectal cancer. Br J Cancer 1997;76:1293–9. Google Scholar

  • 16.

    Yamada T. Detection of K-ras gene mutations in plasma DNA of patients with pancreatic adenocarcinoma: correlation with clinicopathological features. Clin Cancer Res 1998;4:1527–32. Google Scholar

  • 17.

    Castells A. K-ras mutations in DNA extracted from the plasma of patients with pancreatic carcinoma: diagnostic utility and prognostic significance. J Clin Oncol 1999;17:578–84. Google Scholar

  • 18.

    Chen X, Bonnefoi H, Diebold-Berger S, Lyautey J, Lederrey C, Faltin-Traub E, et al. Detecting tumor-related alterations in plasma or serum DNA of patients diagnosed with breast cancer. Clin Cancer Research 1999;5:2297–303. Google Scholar

  • 19.

    Esteller M, Sanchez-Cespedes M, Rosell R, Sidransky D, Baylin SB, Herman JG. Detection of aberrant promoter hypermethylation of tumor suppressor genes in serum DNA from non-small cell lung cancer patients. Cancer Res 1999;59:67–70. Google Scholar

  • 20.

    Vasioukhin V, Anker P, Maurice P, Lyautey J, Lederrey C, Stroun M. Point mutations of the N-ras gene in the blood plasma DNA of patients with myelodysplastic syndrome or acute myelogenous leukaemia. Brit J Haematol 1994;86:774–9. Google Scholar

  • 21.

    Amoura Z, Piette JC, Chabre H, Cacoub P, Papo T, Wechsler B, et al. Circulating plasma levels of nucleosomes in patients with systemic lupus erythematosus: correlation with serum antinucleosome antibody titers and absence of clear association with disease activity. Arthritis Rheum 1997;40:2217–25. Google Scholar

  • 22.

    Rodrigues EM, Simon D, Ikuta N, Klovan C, Dannebrock FA, de Oliveira CO, et al. Elevated cell-free plasma DNA level as an independent predictor of mortality in patients with severe traumatic brain injury. J Neurotraum 2014;31:1639–46. Google Scholar

  • 23.

    Zhong XY, von Muhlenen I, Li Y, Kang A, Gupta AK, Tyndall A, et al. Increased concentrations of antibody-bound circulatory cell-free DNA in rheumatoid arthritis. Clin Chem 2007;53:1609–14. Google Scholar

  • 24.

    Rainer TH, Wong LK, Lam W, Yuen E, Lam NY, Metreweli C, et al. Prognostic use of circulating plasma nucleic acid concentrations in patients with acute stroke. Clin Chem 2003;49:562–9. Google Scholar

  • 25.

    Antonatos D, Patsilinakos S, Spanodimos S, Korkonikitas P, Tsigas D. Cell-free DNA levels as a prognostic marker in acute myocardial infarction. Ann N Y Acad Sci 2006;1075:278–81. Google Scholar

  • 26.

    Crowley E, Di Nicolantonio F, Loupakis F, Bardelli A. Liquid biopsy: monitoring cancer-genetics in the blood. Nat Rev Clin Oncol 2013;10:472–84. Google Scholar

  • 27.

    Tamkovich SN, Cherepanova AV, Kolesnikova EV, Rykova EY, Pyshnyi DV, Vlassov VV, et al. Circulating DNA and DNase activity in human blood. Ann N YAcad Sci 2006;1075:191–6. Google Scholar

  • 28.

    Hawes MC, Wen F, Elquza E. Extracellular DNA: A bridge to cancer. Cancer Res 2015;75:4260–4. Google Scholar

  • 29.

    Cox RA, Gokcen M. Comparison of serum DNA, native DNA-binding and deoxyribonuclease levels in ten animal species and man. Life Sci 1976;19:1609–14. Google Scholar

  • 30.

    Reitz M, Lober G, Kleemann P, Dick W. Secretion of neutral and acid DNases in cultivated human lymphocytes after incubation with DNA; possible consequences for inhalation anesthesia. Z Naturforsch C 1995;50:419–24. Google Scholar

  • 31.

    Bettegowda C, Sausen M, Leary RJ, Kinde I, Wang Y, Agrawal N, et al. Detection of circulating tumor DNA in early- and late-stage human malignancies. Sci Tansl Med 2014;6:224ra24. Google Scholar

  • 32.

    Garcia-Olmo DC, Picazo MG, Toboso I, Asensio AI, Garcia-Olmo D. Quantitation of cell-free DNA and RNA in plasma during tumor progression in rats. Mol Cancer 2013;12:8. Google Scholar

  • 33.

    Heitzer E, Auer M, Hoffmann EM, Pichler M, Gasch C, Ulz P, et al. Establishment of tumor-specific copy number alterations from plasma DNA of patients with cancer. Int J Cancer 2013;133:346–56. Google Scholar

  • 34.

    Lo YM, Zhang J, Leung TN, Lau TK, Chang AM, Hjelm NM. Rapid clearance of fetal DNA from maternal plasma. Am J Hum Genet 1999;64:218–24. Google Scholar

  • 35.

    Yu SC, Lee SW, Jiang P, Leung TY, Chan KC, Chiu RW, et al. High-resolution profiling of fetal DNA clearance from maternal plasma by massively parallel sequencing. Clin Chem 2013;59:1228–37. Google Scholar

  • 36.

    Beiter T, Fragasso A, Hudemann J, Niess AM, Simon P. Short-term treadmill running as a model for studying cell-free DNA kinetics in vivo. Clin Chem 2011;57:633–6. Google Scholar

  • 37.

    Jiang P, Chan CW, Chan KC, Cheng SH, Wong J, Wong VW, et al. Lengthening and shortening of plasma DNA in hepatocellular carcinoma patients. Proc Natl Acad Sci USA 2015;112:E1317–25. Google Scholar

  • 38.

    Mouliere F. High fragmentation characterizes tumour-derived circulating DNA. PLoS One 2011;6:e23418. CrossrefGoogle Scholar

  • 39.

    Heidary M, Auer M, Ulz P, Heitzer E, Petru E, Gasch C, et al. The dynamic range of circulating tumor DNA in metastatic breast cancer. Breast Cancer Res 2014;16:421. Google Scholar

  • 40.

    Jiang P, Lo YM. The long and short of circulating cell-free DNA and the ins and outs of molecular diagnostics. Trends Genet 2016;32:360–71. Google Scholar

  • 41.

    Sun K, Jiang P, Chan KC, Wong J, Cheng YK, Liang RH, et al. Plasma DNA tissue mapping by genome-wide methylation sequencing for noninvasive prenatal, cancer, and transplantation assessments. Proc Natl Acad Sci USA 2015;112:E5503–12. Google Scholar

  • 42.

    Ivanov M, Baranova A, Butler T, Spellman P, Mileyko V. Non-random fragmentation patterns in circulating cell-free DNA reflect epigenetic regulation. BMC genomics 2015;16(Suppl 13):S1. Google Scholar

  • 43.

    Ulz P, Thallinger GG, Auer M, Graf R, Kashofer K, Jahn SW, et al. Inferring expressed genes by whole-genome sequencing of plasma DNA. Nat Genet 2016. Aug 29. Google Scholar

  • 44.

    Snyder MW, Kircher M, Hill AJ, Daza RM, Shendure J. Cell-free DNA comprises an in vivo nucleosome footprint that informs its tissues-of-origin. Cell 2016;164:57–68. Google Scholar

  • 45.

    Imamura F, Uchida J, Kukita Y, Kumagai T, Nishino K, Inoue T, et al. Monitoring of treatment responses and clonal evolution of tumor cells by circulating tumor DNA of heterogeneous mutant EGFR genes in lung cancer. Lung Cancer 2016;94:68–73. Google Scholar

  • 46.

    Ball MW, Gorin MA, Guner G, Pierorazio PM, Netto G, Paller CJ, et al. Circulating tumor DNA as a marker of therapeutic response in patients with renal cell carcinoma: a pilot study. Clin Genitourin Can 2016. Mar 24. Google Scholar

  • 47.

    Montagut C, Siravegna G, Bardelli A. Liquid biopsies to evaluate early therapeutic response in colorectal cancer. Ann Oncol 2015;26:1525–7. Google Scholar

  • 48.

    Frenel JS, Carreira S, Goodall J, Roda D, Perez-Lopez R, Tunariu N, et al. Serial next-generation sequencing of circulating cell-free DNA evaluating tumor clone response to molecularly targeted drug administration. Clin Cancer Res 2015;21:4586–96. Google Scholar

  • 49.

    Dawson SJ. Analysis of circulating tumor DNA to monitor metastatic breast cancer. N Engl J Med 2013;368:1199–209. Google Scholar

  • 50.

    Murtaza M, Dawson SJ, Tsui DW, Gale D, Forshew T, Piskorz AM, et al. Non-invasive analysis of acquired resistance to cancer therapy by sequencing of plasma DNA. Nature 2013;497:108–12. Google Scholar

  • 51.

    Murtaza M, Dawson SJ, Pogrebniak K, Rueda OM, Provenzano E, Grant J, et al. Multifocal clonal evolution characterized using circulating tumour DNA in a case of metastatic breast cancer. Nature Comm 2015;6:8760. Google Scholar

  • 52.

    Butler TM, Johnson-Camacho K, Peto M, Wang NJ, Macey TA, Korkola JE, et al. Exome sequencing of cell-free DNA from metastatic cancer patients identifies clinically actionable mutations distinct from primary disease. PLoS One 2015;10:e0136407. Google Scholar

  • 53.

    Azad AA, Volik SV, Wyatt AW, Haegert A, Le Bihan S, Bell RH, et al. Androgen receptor gene aberrations in circulating cell-free DNA: biomarkers of therapeutic resistance in castration-resistant prostate cancer. Clin Cancer Res 2015;21:2315–24. Google Scholar

  • 54.

    Carreira S, Romanel A, Goodall J, Grist E, Ferraldeschi R, Miranda S, et al. Tumor clone dynamics in lethal prostate cancer. Sci Tansl Med 2014;6:254ra125. Google Scholar

  • 55.

    Heitzer E, Ulz P, Belic J, Gutschi S, Quehenberger F, Fischereder K, et al. Tumor-associated copy number changes in the circulation of patients with prostate cancer identified through whole-genome sequencing. Genome Med 2013;5:30. Google Scholar

  • 56.

    Ulz P, Belic J, Graf R, Auer M, Lafer I, Fischereder K, et al. Whole-genome plasma sequencing reveals focal amplifications as driving force in metastatic prostate cancer. Nature Comm 2016 22;7:12008. Google Scholar

  • 57.

    Heitzer E, Ulz P, Geigl JB, Speicher MR. Non-invasive detection of genome-wide somatic copy number alterations by liquid biopsies. Mol Oncol 2016;10:494–502. Google Scholar

  • 58.

    Mohan S, Heitzer E, Ulz P, Lafer I, Lax S, Auer M, et al. Changes in colorectal carcinoma genomes under anti-EGFR therapy identified by whole-genome plasma DNA sequencing. PLoS Genet 2014;10:e1004271. Google Scholar

  • 59.

    Ulz P, Auer M, Heitzer E. Detection of circulating tumor DNA in the blood of cancer patients: an important tool in cancer chemoprevention. Methods Mol Biol 2016;1379:45–68. Google Scholar

  • 60.

    Leary RJ, Sausen M, Kinde I, Papadopoulos N, Carpten JD, Craig D, et al. Detection of chromosomal alterations in the circulation of cancer patients with whole-genome sequencing. Sci Tansl Med 2012;4:162ra54. Google Scholar

  • 61.

    Belic J, Koch M, Ulz P, Auer M, Gerhalter T, Mohan S, et al. Rapid identification of plasma DNA samples with increased ctDNA levels by a modified FAST-SeqS approach. Clin Chem 2015;61:838–49. Google Scholar

  • 62.

    De Mattos-Arruda L, Weigelt B, Cortes J, Won HH, Ng CK, Nuciforo P, et al. Capturing intra-tumor genetic heterogeneity by de novo mutation profiling of circulating cell-free tumor DNA: a proof-of-principle. Ann Oncol 2014;25:1729–35. Google Scholar

  • 63.

    My Cancer Genome. https://www.mycancergenome.org/

  • 64.

    TARGET. https://www.broadinstitute.org/cancer/cga/target

  • 65.

    Therapeutic Targets Database (TTD). http://bidd.nus.edu.sg/BIDD-Databases/TTD/TTD.asp

  • 66.

    The Tumour chARacterisation to Guide Experimental Targeted Therapy Trial TARGET. Google Scholar

  • 67.

    Febbo PG, Ladanyi M, Aldape KD, De Marzo AM, Hammond ME, Hayes DF, et al. NCCN task force report: evaluating the clinical utility of tumor markers in oncology. J Natl Compr Canc Netw 2011;9(Suppl 5):S1–32; quiz S3. Google Scholar

  • 68.

    Woodcock J. Assessing the clinical utility of diagnostics used in drug therapy. Clin Pharmacol Ther 2010;88:765–73. Google Scholar

  • 69.

    Malentacchi F, Pizzamiglio S, Verderio P, Pazzagli M, Orlando C, Ciniselli CM, et al. Influence of storage conditions and extraction methods on the quantity and quality of circulating cell-free DNA (ccfDNA): the SPIDIA-DNAplas external quality assessment experience. Clin Chem Lab Med 2015;53:1935–42. Google Scholar

  • 70.

    ACG. Actionable Genome Consortium to guide NGS in cancer. Nat Biotech 2014;32:965. Google Scholar

About the article

Received: 2016-07-04

Accepted: 2016-09-08

Published Online: 2016-10-07

Published in Print: 2016-10-01


Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.

Research funding: None declared.

Employment or leadership: None declared.

Honorarium: None declared.

Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.


Citation Information: LaboratoriumsMedizin, Volume 40, Issue 5, Pages 323–334, ISSN (Online) 1439-0477, ISSN (Print) 0342-3026, DOI: https://doi.org/10.1515/labmed-2016-0049.

Export Citation

©2016 Walter de Gruyter GmbH, Berlin/Boston.Get Permission

Comments (0)

Please log in or register to comment.
Log in