Printer Friendly

Rapid identification of plasma DNA samples with increased ctDNA levels by a modified FAST-SeqS approach.

Recent progress in the analysis of cell-free circulating tumor DNA (ctDNA) [6] now allows monitoring of tumor genomes by noninvasive means (1-7). Basically, 2 strategies to monitor ctDNA have emerged. First, targeted approaches, i.e., single or few tumor-specific somatic mutations present in the primary tumor, are used to monitor residual disease in the peripheral blood (6, 8, 9) (Fig. 1). Disadvantages of targeted approaches include the need for a detailed characterization of the primary tumor genome to identify suitable targets, which is not always possible (e.g., if only a biopsy was obtained). Furthermore, the analysis of only a single mutation or a few mutations limits the options to monitor the clonal evolution of a tumor genome. However, with highly sensitive approaches, mutations can be detected even if they are present at low allele frequencies (1, 10-13), which is important, as several recent publications demonstrated a high variability of ctDNA allele frequency even in advanced-stage metastatic disease (1, 6, 14-16).

Second, untargeted approaches establish genomewide patterns of copy number aberrations (CNAs) from ctDNA (2, 4, 15, 17) or assess the mutation spectrum by exome sequencing from plasma DNA (7) (Fig. 1). Advantages ofuntargeted strategies include independence of prior knowledge about characteristics of the primary tumor and easier identification of novel changes occurring during tumor evolution. However, a disadvantage is that a certain percentage of ctDNA in plasma DNA is required for the reliable reconstruction of tumor-specific copy number changes. According to our experience, this percentage is in the range of [greater than or equal to] 10% (4, 14, 15, 18). Owing to the high variability of ctDNA (from 0.01% to >90% (1, 6, 15, 16)), not all blood samples will therefore be suitable for genome-wide analyses (Fig. 1). We sought an untargeted strategy to identify plasma DNA samples that are suitable for a further workup with complex whole-genome sequencing approaches.

[FIGURE 1 OMITTED]

To this end, we used a modified Fast Aneuploidy Screening Test-Sequencing System (FAST-SeqS) method described by Kinde et al. (19) that was developed as a prenatal screening method to establish fetal chromosome status. We adapted this method to prescreen plasma DNA samples from cancer patients to estimate the fraction of ctDNA and its suitability for genome-wide analyses, such as plasma-Seq (4). We show that our modified FAST-SeqS (mFAST-SeqS) can be used to estimate the amount of ctDNA in plasma in a cost-effective and rapid manner without any prior knowledge of specific aberrations of the primary tumor (Fig. 1).

Methods

PATIENT SAMPLES AND CANCER CELL LINES

We analyzed 35 control samples (18 women, 17 men) without malignant disease and a set of samples from cancer patients with different tumor entities at different stages, which were available from previous studies (4, 14). We obtained 24 blood samples from 21 patients with advanced breast cancer (mean age 61 years; range 37-84), including hormone receptor-positive/negative and/or human epidermal growth factor receptor (HER2)-positive/negative tumors, from the Department of Obstetrics and Gynecology, Medical University of Graz. Follow-up samples taken at 13, 12, and 11 months were available from 3 patients (B1, B4, and B5). Prostate cancer patients (n = 56) (mean age 70.4 years; range 51-87) were recruited at the Department of Urology or the Division of Clinical Oncology, Department of Internal Medicine, Medical University of Graz. Four patients were under surveillance, 10 patients had a prostatectomy with localized cancer, and 42 patients had metastatic disease. Follow-up samples were available at 3, 10, 11, 9, and 6 months from 5 patients (P39, P40, P102, P111, and P147). All patients were of Caucasian origin. The study was approved by the ethics committee of the Medical University of Graz (approval numbers 21-227 ex 09/10, breast cancer; 21-228 ex 09/10, prostate cancer) and conducted according to the Declaration of Helsinki, and written informed consent was obtained from all patients. Additionally, this study includes cancer cell lines (HepG2, HT29, and MCF7) that were purchased from American Type Culture Collection and cultured according to the supplier's recommendations.

SERIAL DILUTIONS OF CELL LINES

To validate the reliability and repeatability, i.e., inter- and intraassay variability, we used a dilution series of HT29 DNA (100%, 50%, 10% 5%, 0%). The sensitivity of mFAST-SeqS was determined with serial dilutions of DNA from cancer cell lines (HepG2, HT29, MCF7) diluted in noncancerous DNA (Promega) in the following ratios: 50%, 25% 10%, 5%, and 1%.

EXTRACTION OF GENOMIC DNA AND PLASMA DNA

Genomic DNA was extracted with the QIAamp DNA Midi Kit (Qiagen) according to the manufacturer's protocol. Plasma DNA was prepared with the QIAamp DNA Blood Mini Kit (Qiagen) as previously described (15). For quantification of plasma DNA, we used the Qubit dsDNA HS Assay Kit (Life Technologies).

MODIFIED FAST-SeqS

Line-1 (L1) amplicon libraries were prepared on the basis of the protocol by Kinde et al. (19) with the following exceptions. We excluded the unique identifier in the primer sequences and adapted the amount of input DNA. For analyses of cell lines, we used 20 ng DNA. Depending on the concentration of plasma DNA, we used 5-10 [micro]L DNA corresponding to 0.1-5 ng total DNA. For samples with very high concentrations (>100 ng/mL plasma), a maximum of 5 ng was used for the assay.

Phusion HF Buffer (5X), 2 U Phusion Hot Start II Polymerase, 0.25 [micro]mol/L target-specific L1 primer, and 200 [micro]mol/L dNTPs were mixed and amplified with the following cycling conditions: denaturation at 98[degrees]C for 2 min, followed by 5 cycles of 98[degrees]C for 10 s, 57[degrees]C for 120 s, and 72[degrees]C for 120 s. PCR products were purified by incubation with 70 [micro]L AMPure Beads (Beckman Coulter) for 10 min. The supernatant was removed, and the bead pellet was washed twice with freshly prepared 70% ethanol. After drying, the pellet was resuspended in 12 [micro]L 1X Tris-EDTA-buffer. Purified PCR products (10 [micro]L) were directly used for a second PCR, in which sequencing adaptors and sample-specific indexes were added (for primer sequences, see Supplemental Table 1, which accompanies the online version of this article at http://www.clinchem.org/content/vol61/issue6). Reaction setup and cycling conditions remained the same but with increased numbers of cycles. For plasma DNA samples, we used 18 cycles; all other samples were amplified with 15 cycles. PCR products were quality checked and quantified on an Agilent Bioanalyzer with a 7500 DNA kit (Agilent).

SEQUENCING AND DATA ANALYSIS

L1 amplicon libraries were pooled in equimolar amounts and sequenced on a MiSeq (Illumina) generating 150-bp single reads. Because mFAST-SeqS libraries represent low-diversity libraries, we spiked either 10%-20% of a PhiX control library or mFAST-SeqS libraries into plasma-Seq libraries in different ratios depending on the number of samples to be sequenced. Sequence reads were aligned to the hg19 genome with Burrows-Wheeler alignment, version 0.7.4. Reads with a mapping quality >15 were counted with an in-house script. To correct for different sequencing yields, read counts were normalized to the total read count per sample. To test for over- and underrepresentation of each chromosome arm, we calculated z-scores by subtracting the mean and dividing by the SD of normalized read-counts for the respective chromosome arm from 35 controls. Because no reads aligned to the short arms of the acrocentric chromosomes 13p, 14p, 15p, 21p, 22p, and Y, these were excluded from the analysis. To get a general overview of aneuploidy, we introduced a genome-wide z-score. Therefore, normalized read counts per chromosome arm were squared and summed.

PLASMA-SEQ

The plasma-Seq method was described in detail previously (4). Briefly, shotgun libraries were prepared with the TruSeq DNA Nano Sample Preparation Kit (Illumina) with the following changes. We used 5-10 ng input DNA. On the basis of previous data, where we observed an enrichment of fragments in the range of 160-340 bp with the Agilent Bioanalyzer (14, 15), we omitted the fragmentation step. For selective amplification of the library fragments, we used 20-25 PCR cycles. The libraries were sequenced on a MiSeq (Illumina).

TARGETED RESEQUENCING

With Sure Select Custom DNA Kit (Agilent), we enriched 1.3 Mbp including exonic sequences of 55 cancer genes and 38 introns of 18 genes, where fusion breakpoints have been described. Analysis criteria have been previously described (4). Identified somatic mutations were validated with deep sequencing (18) to determine the mutant allele frequencies more accurately. We set the threshold for reliable detection of a sequence variation at 1%. Allelic fractions of <1% were considered sequencing errors.

ACCESSION NUMBERS

All sequencing raw data were deposited at the European Genome-Phenome Archive (http://www.ebi.ac.uk/ega/), which is hosted by the European Bioinformatics Institute, under the study accession number EGAS00001001133.

STATISTICS

To compare mFAST-SeqS with plasma-Seq, we used Pearson correlation. Comparisons were performed with genome-wide and chromosome arm-specific z-scores. Correlations were visualized with linear regression. ROC analysis was performed in R (20, 21).

Results

SEQUENCING OF CONTROL SAMPLES TO ESTABLISH REFERENCE VALUES

We modified the original FAST-SeqS protocol in 2 important aspects. First, whereas Kinde et al. performed read count and z-score analysis for whole chromosomes (19), we expanded the analysis to chromosome arms. Second, as our goal was the development of a prescreening tool that should perform fast at low cost, we used a lower amount of total reads than Kinde et al. (19). We analyzed plasma samples from controls to assess the consequences of these modifications (n = 35). After filtering for reads with a Phred-scaled mapping quality >15, we obtained a mean amount of 242190 reads (range 106205-925843) (see online Supplemental Table 2) that aligned to a mean of 22297 positions (see online Supplemental Table 2). Although the number of reads varied about 8.7-fold, the number of uniquely aligned positions varied only 2-fold (see online Supplemental Table 2). We applied read count analysis by counting the aligned reads to each chromosome arm and calculated a relative ratio for each chromosome arm. Although these ratios varied by a mean of 1.53-fold for smaller chromosomes (chromosomes 15-22), there was only a slight variability of 1.02-fold for larger chromosomes (chromosomes 1-14) (see online Supplemental Table 2). We calculated a z-score for each chromosome arm, with the mean ratios of control samples of the same sex. Altogether, we calculated 41 chromosome arm-specific z-scores for each control, accounting for a total of 1435 z-scores (range -3.35 to 3.85). Ten z-scores were >3 (0.07%) and 7 were less than -3 (0.05%) (see online Supplemental Table 3). Compared with data from Kinde et al., we observed a slightly higher variability, which was not unexpected since we used lower amounts of total reads. Moreover, we calculated a genome-wide z-score as a general measure for aneuploidy consisting of the squared sum of all chromosome-specific z-scores. This genome-wide z-score ranged from -1.50 to 2.51 for all healthy controls (n = 35) (see online Supplemental Table 3).

INTRA- AND INTERASSAY VARIABILITY OF mFAST-SeqS

We validated the reliability and repeatability of mFAST-SeqS by analyzing different dilutions of HT29 DNA (50%, 10% 5%, 0%) with 3-5 duplicates of each dilution in 5 independent runs. We obtained a good repeatability of the genome-wide z-score, with only small inter- and intraassay variations (Table 1) (see online Supplemental Fig. 1), although the number of reads varied 2-fold from run to run. Chromosome arm-specific z-scores were highly correlated within (r = 0.992) and between (r = 0.998) runs (see online Supplemental Fig. 1, A and B). Linear regression analysis revealed a good correlation between genome-wide z-scores and the dilution of HT29 cell lines (mean r = 0.974) (see online Supplemental Fig. 1C).

[FIGURE 2 OMITTED]

SPECIFICITY AND SENSITIVITY OF mFAST-SeqS

Next, we evaluated our mFAST-SeqS assay by analyzing 3 cancer cell lines, HepG2, HT29, and MCF7 (Fig. 2, A-C) (see online Supplemental Table 4). When we used raw data of our plasma-Seq to establish comparable chromosome arm-specific z-scores, we obtained correlation coefficients of 0.895, 0.883, and 0.857 for HepG2, HT29, and MCF7, respectively, indicating that copy number changes obtained with the mFAST-SeqS assay were highly concordant with those obtained from plasma-Seq (Table 2). Not surprisingly, high-level amplifications of specific parts of chromosomes resulted in high overall chromosome arm-specific z-scores. For instance, the high-level amplifications of chromosome 8q in both HT29 and MCF7 were reflected in 8q-specific z-scores of 42.2 and 15.0, respectively (Fig. 2D, first and third panel). We observed a close correlation between plasma-Seq and mFAST-SeqS if the entire chromosome arm was lost or gained, as exemplified by the loss of the short arm of chromosome 3 in HT29 (Fig. 2D, second panel) or gain of the entire chromosome 2 in HepG2 (Fig. 2D, fourth panel). In contrast, some chromosome arm-specific z-scores were below the threshold of 5, although CNAs were detected with plasma-Seq, which was the case when gains and losses co-occurred at the same chromosome arm (Fig. 2D, fifth panel).

To evaluate the sensitivity of mFAST-SeqS, we performed serial dilutions of 3 cancer cell lines. As expected, the genome-wide z-scores decreased with increasing dilution. Depending on the total amount and extent of aberrations, we observed a stronger decrease of genome-wide z-scores in the different dilutions (for details, see online Supplemental Fig. 2). Nevertheless, characteristic high-level amplifications such as the distal amplification of chromosome 8q in HT29 (see online Supplemental Fig. 2B) or the gain of chromosome 2q in HepG2 (see online Supplemental Fig. 2A) were detected when only 10% of cell line DNA was present. This suggests that a combined evaluation of genome-wide z-scores and chromosome arm-specific z-scores will detect the majority of plasma samples with a percentage of 10% ctDNA.

mFAST-SeqS FOR NONINVASIVE ANEUPLOIDY SCREENING IN CANCER PATIENTS

To evaluate whether mFAST-SeqS can be used as a prescreening tool for the presence of increased ctDNA levels, we analyzed 90 plasma samples from cancer patients. To determine z-score cutoffs for the identification of plasma samples where subsequent plasma-Seq would detect CNAs with a high probability, we used 24 plasma samples from 21 metastatic breast cancer patients for which copy number profiles established with plasma-Seq were available (mean plasma DNA concentration 64.4 ng/mL, range 5.1-440.8). The mean genome-wide z-score was 86.9 (range 2.2-349.6) (see online Supplemental Table 5). Genomewide z-scores were only weakly correlated with plasma DNA concentrations (r = 0.419). All samples with a genome-wide z-score >5 (n = 20) demonstrated CNAs with plasma-Seq, whereas samples with genome-wide z-score <5 (n = 4) showed balanced plasma-Seq profiles. Despite this relatively low sample size, we performed ROC analysis of genome-wide z-scores, which revealed a cutoff of 6 for the detection of CNAs with plasma-Seq. To not miss samples with potential CNAs, we chose a cutoff of 5, which revealed a sensitivity of 100% and specificity of 80% for the detection of copy number changes with plasma-Seq.

Example heat maps showing chromosome arm-specific z-scores for various values of genome-wide z-scores and highly concordant mFAST-SeqS and plasma-Seq copy number profiles are displayed in Fig. 3. Chromosome-specific z-scores established from plasma-Seq data (n = 6) showed a strong correlation with those from mFAST-SeqS (mean r = 0.90) (Table 2). When we compared genome-wide z-scores from mFAST-SeqS with those of available plasma-Seq data (n = 18), we again observed a good concordance (r = 0.677) (Table 2). However, when we analyzed chromosome arm-specific z-scores, we noted 2 samples from the same patient (B5_3 and B5_4, genome-wide z-scores 4.0 and 3.8, respectively) with increased chromosome arm-specific z-scores for 8q of 6.4 and 7.0, respectively. In these 2 cases, plasma-Seq did not confirm the 8q CNAs (data not shown).

When we analyzed follow-up samples from the same patients collected 13 (B1) and 12 (B4) months after the first sample, mFAST-SeqS profiles were highly consistent (r = 0.97 for B1 and r = 0.99 for B4) (see online Supplemental Fig. 3, A and B), indicating that mFAST-SeqS delivers highly concordant results even when using biological replicates.

To validate these results, we analyzed a set of 66 samples from 56 prostate cancer patients at various disease stages (Fig. 4A) (see online Supplemental Table 6). Samples from patients with metastatic disease (n = 52) had significantly higher plasma DNA concentrations (mean 36.5 ng/mL; range 0.36-390) than patients in earlier stages (n = 14) (mean 2.4 ng/mL; range 0.5-5.5) (P < 0.0001). Twenty-eight samples (42.4%) had a genome-wide z-score <5. Not surprisingly, these patients included all patients under surveillance and 9 of 10 patients with localized cancer after prostatectomy, for which a low fraction of tumor-specific DNA was expected. Also, 15 samples from metastatic patients had a z-score <5, indicating low amounts of tumor-specific DNA, which was confirmed by a balanced profile after plasma-Seq of 3 samples (see online Supplemental Fig. 4). In 38 patients (57.6%), a genome-wide z-score >5 was observed, including 8 patients (12.1%) with a genome-wide score >100. Again, we did not observe a strong correlation between plasma DNA concentrations and z-scores (r = 0.304). However, when we correlated mutant allele frequencies (mAFs) of 11 samples identified by targeted resequencing of known cancer driver genes with the mFAST-SeqS z-score, we observed a strong correlation (r = 0.902) (see online Supplemental Fig. 5). On the basis of linear regression, a z-score of 5 would predict an mAF of 10.5%.

[FIGURE 3 OMITTED]

Furthermore, we observed a good concordance between mFAST-SeqS and plasma-Seq results as exemplified in Fig. 4B and Table 2 (r = 0.777 for 38 genome-wide z-score comparisons, and r = 0.945 for chromosome arm-specific z-scores). Analysis of follow-up samples of 3 patients (P40, P147, and P111) revealed strong correlations of r = 0.940, 0.992, and 0.988, respectively (see online Supplemental Fig. 6, A-C).

[FIGURE 4 OMITTED]

Discussion

Previous studies with plasma DNA from patients with cancer demonstrated highly variable allele frequencies of ctDNA (1, 16). In cases with low ctDNA allele frequency, targeted approaches allow high-sensitivity monitoring of ctDNA dynamics (Fig. 1) (10, 12, 22, 23). However, in cases with high ctDNA allele frequency, analyses can be extended to untargeted, genome-wide approaches to uncover novel changes occurring during tumor evolution. Hence, an important question is how to distinguish between plasma samples with low and high ctDNA content to decide on the optimal strategy for further processing of a plasma sample (Fig. 1). To address this, we adapted the FAST-SeqS screening method for fetal aneuploidy (19). Because our aim was to use mFAST-SeqS as a prescreening tool for an estimation of the ctDNA percentage, we used a lower total number of reads for FAST-SeqS than used by Kinde et al. (19). With cell line DNA and a set of plasma from breast cancer patients, we established a cutoff z-score of 5 for a subsequent detection of CNAs with our established plasma-Seq approach. We confirmed this cutoff with 11 plasma samples in which comparisons of mAF in established cancer driver genes such as BRCA1 (breast cancer 1, early onset), [7] TP53 (tumor protein p53), and CTNNB1 [catenin (cadherin-associated protein), beta 1, 88 kDa] with mFAST-SeqS z-scores showed a strong correlation.

The resolution of CNA (i.e., somatic copy number changes) detection with our assay cannot be given in absolute numbers, as it depends on the total amount of copy number changes, including CNAs and common copy number variations (i.e., germline copy number changes), and their amplitudes. As a consequence, a low overall amount of copy number changes might result in a lower z-score despite a tumor fraction of [greater than or equal to] 10%. On the other hand, few changes with high amplitudes, such as focal amplifications, may result in aberrant z-scores even if the ctDNA is <10%. Therefore, we suggest a combined evaluation of genome-wide and chromosome arm-specific z-scores. Furthermore, our assay may be influenced by very low plasma DNA concentrations, which would result in low input templates. However, these are likely to be cases unsuited for untargeted approaches to which highly-sensitive targeted techniques should be applied.

Advantages of this approach include that no prior knowledge about the genetic composition of tumor samples is necessary to estimate the amount of ctDNA, the speed of analysis (< 1 day: hands-on time 1 h, bioinformatic and statistical analysis < 1h), and the low cost per analysis (approximately 10 [euro] for consumables). The loss of resolution is, in our opinion, acceptable, as mFAST-SeqS merely serves as a decision support tool, to select the most appropriate in-depth/high-resolution strategy (Fig. 1).

It has previously been reported that ctDNA fragments might be smaller than 100 bp (24). However, this is in contrast to our own observations (14, 15) and reports by others (2). Indeed, the Qiagen Mini Kit for DNA extraction limits the isolation to fragment sizes of [greater than or equal to] 100 bp, and furthermore, the FAST-SeqS assay amplifies PCR products in the range of 124-142 bp. Smaller DNA fragments are omitted from our assay and mFAST-SeqS and plasma-Seq copy number profiles are closely correlated, which further suggests that a substantial number of tumor DNA fragments in the circulation must have a size of >100 bp.

With our plasma-Seq method that applies low-coverage whole-genome sequencing to establish copy number profiles, tumor-specific changes can be detected in the circulation at levels [greater than or equal to] 10% of circulating tumor DNA in patients with prostate cancer with a sensitivity of >80% and specificity of >80% (4). With plasma-Seq, we were able to identify CNAs in plasma samples of many metastasis patients (4, 14, 18); however, in some cases, we would not have performed plasma-Seq if we had had prior knowledge about the ctDNA fraction. Similar data about ctDNA variability were obtained by a recent study, which showed that approximately 80% of patients with metastatic disease had detectable levels of ctDNA, albeit with a high variability of mutant fragments (1). In another comprehensive study with exome sequencing, results could be achieved only in patients with a high systemic mutation burden in plasma > 5%-10% (7). Chan et al. showed that the fractional concentration of tumor DNA in plasma and the class of CNA strongly influence the detectability of such alterations in plasma (2).

Although sequencing costs have dropped rapidly in recent years, genome-wide, in-depth, high-coverage analysis is still expensive and time consuming, which hampers the introduction of ctDNA diagnostics into the clinic. mFAST-SeqS may contribute to a significant reduction of cost and increase of speed and could therefore serve as a valuable, untargeted prescreening tool to identify plasma DNA samples with high ctDNA content for decisions on further diagnostic steps (Fig. 1).

Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contribution to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.

Authors' Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest:

Employment or Leadership: None declared.

Consultant or Advisory Role: None declared.

Stock Ownership: None declared.

Honoraria: None declared.

Research Funding: M.R. Speicher, the Austrian Science Fund (FWF) (grants P20338, P23284, and W 1226-B18, DKplus Metabolic and Cardiovascular Disease), and the Oesterreichische Nationalbank (project 15093).

Expert Testimony: None declared.

Patents: None declared.

Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript.

References

(1.) 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 Transl Med 2014;6:224ra24.

(2.) Chan KC, Jiang P, Zheng YW, Liao GJ, Sun H, Wong J, et al. Cancer genome scanning in plasma: detection of tumor-associated copy number aberrations, single nucleotide variants, and tumoral heterogeneity by massively parallel sequencing. Clin Chem 2013;59:211-24.

(3.) Heitzer E, Auer M, Ulz P, Geigl JB, Speicher MR. Circulating tumor cells and DNA as liquid biopsies. Genome Med 2013;5:73.

(4.) 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.

(5.) Heitzer E, Ulz P, Geigl JB. Circulating tumor DNA as a liquid biopsy for cancer. Clin Chem 2015;61:112-23.

(6.) 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 Transl Med 2012;4:162ra154.

(7.) 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.

(8.) Diehl F, Li M, Dressman D, He Y, Shen D, Szabo S, et al. Detection and quantification of mutations in the plasma of patients with colorectal tumors. Proc Natl Acad Sci USA 2005;102:16368-73.

(9.) McBride DJ, Orpana AK, Sotiriou C, Joensuu H, Stephens PJ, Mudie LJ, et al. Use of cancer-specific genomic rearrangements to quantify disease burden in plasma from patients with solid tumors. Genes Chromosomes Cancer 2010;49:1062-9.

(10.) Diehl F, Schmidt K, Choti MA, Romans K, Goodman S, Li M, et al. Circulating mutant DNA to assess tumor dynamics. Nat Med 2008;14:985-90.

(11.) Forshew T, Murtaza M, Parkinson C, Gale D, Tsui DW, Kaper F, et al. Noninvasive identification and monitoring of cancer mutations by targeted deep sequencing ofplasma DNA.Sci Transl Med 2012;4:136ra68.

(12.) Taly V, Pekin D, Benhaim L, Kotsopoulos SK, Le Corre D, Li X, et al. Multiplex picodroplet digital PCR to detect KRAS mutations in circulating DNA from the plasma of colorectal cancer patients. Clin Chem 2013;59:1722-31.

(13.) Newman AM, Bratman SV, To J, Wynne JF, Eclov NC, Modlin LA, et al. An ultrasensitive method for quantitating circulating tumor DNA with broad patient coverage. Nat Med 2014;20:548-54.

(14.) 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.

(15.) 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.

(16.) Thierry AR, Mouliere F, El Messaoudi S, Mollevi C, Lopez-Crapez E, Rolet F, et al. Clinical validation of the detection of KRAS and BRAF mutationsfrom circulating tumor DNA. Nat Med 2014;20:430-5.

(17.) Chan KC, Jiang P, Chan CW, Sun K, Wong J, Hui EP, et al. Noninvasive detection of cancer-associated genome-wide hypomethylation and copy number aberrations by plasma DNA bisulfite sequencing. Proc Natl Acad Sci USA 2013;110:18761-8.

(18.) Mohan S, Heitzer E, Ulz P, Lafer I, LaxS, 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.

(19.) Kinde I, Papadopoulos N, Kinzler KW, Vogelstein B. FAST-SeqS: a simple and efficient method for the detection of aneuploidy by massively parallel sequencing. PloS One 2012;7:e41162.

(20.) Zweig MH, Campbell G. Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clin Chem 1993;39:561-77.

(21.) R Project. R: a language and environment for statistical computing. http://wwwRprojectorg. Accessed April 2015.

(22.) Kinde I, Wu J, Papadopoulos N, Kinzler KW, Vogelstein B. Detection and quantification of rare mutations with massively parallel sequencing. Proc Natl Acad Sci USA 2011;108:9530-5.

(23.) Spindler KL, Pallisgaard N, Vogelius I, Jakobsen A. Quantitative cell-free DNA, KRAS, and BRAF mutations in plasma from patients with metastatic colorectal cancer during treatment with cetuximab and irinotecan. Clin Cancer Res 2012;18:1177-85.

(24.) Mouliere F, Robert B, Arnau Peyrotte E, Del Rio M, Ychou M, Molina F, et al. High fragmentation characterizes tumour-derived circulating DNA. PloS One 2011;6: e23418.

Jelena Belic, [1] Marina Koch, [1] Peter Ulz, [1] Martina Auer, [1] Teresa Gerhalter, [2] Sumitra Mohan, [1] Katja Fischereder, [3] Edgar Petru, [4] Thomas Bauernhofer, [5] Jochen B. Geigl, [1] Michael R. Speicher, [1] and Ellen Heitzer [1] *

[1] Institute of Human Genetics, [3] Department of Urology, [4] Department of Obstetrics and Gynecology, and [5] Division of Oncology, Medical University of Graz, Graz, Austria; [2] Institute of Molecular Biotechnology, Technical University of Graz, Graz, Austria.

* Address correspondence to this author at: Institute of Human Genetics, Medical University of Graz, Harrachgasse 21/8, 8010 Graz, Austria. Fax +43-316-380-9605; e-mail ellen.heitzer@medunigraz.at.

Received October 2,2014; accepted March 20,2015.

Previously published online at DOI: 10.1373/clinchem.2014.234286

[6] Nonstandard abbreviations: ctDNA, cell-free circulating tumor DNA; CNA, copy number aberration; mFAST-SeqS, modified Fast Aneuploidy Screening Test-Sequencing System; HER2, human epidermal growth factor receptor; L1, line-1; mAF, mutant allele frequency.

[7] Human genes: BRCA1, breast cancer 1, early onset; TP53, tumor protein p53; CTNNB1, catenin (cadherin-associated protein), beta 1,88 kDa.
Table 1. Intra-and interassay variability of
genome-wide z-scores calculated from HT29 dilutions.

HT29      Dilution I   Dilution II   Dilution III
DNA, %

         Mean    SD    Mean    SD    Mean    SD

100      131.7   7.2   126.1   3.3   127.4   3.2
50       49.7    2.2   49.6    2.1    48     2.3
10       16.9    1.2   17.3     2    16.5    0.4
5        15.2    0.8   14.6    1.3   14.4    1.4
0         2.4    0.6    2.4    0.6   -0.1    0.5

HT29     Dilution IV    Dilution V       All
DNA, %

         Mean    SD    Mean    SD    Mean   SD

100      126.9   3.1   128.2   1.4   128    3.7
50       48.2    2.1   48.9    0.6   48.9   1.9
10       15.9    2.1   17.3    1.6   16.8   1.5
5        14.1    0.5   15.2    1.2   14.7    1
0        -0.2    0.2    2.4    0.6   1.4    0.5

Table 2. Comparison of mFAST-SeqS and plasma-Seq data.

Sample    Correlation of   Genome-wide z-score
            chromosome
           arm-specific
             z-scores

                           mFAST-SeqS   plasma-Seq

HepG2         0.895           84.2         26.3
HT29          0.883          128.5         61.0
MCF7          0.857           93.8        260.6
B4_1          0.907          213.4        191.5
B40_1         0.870           54.1         79.2
B41_1         0.885          131.01        37.5
B4_2          0.906          259.9        187.5
B1_2          0.902           94.9         31.3
B80_1         0.920          349.6         75.0
P40_1         0.907           47.0        75.67
P40_2         0.925           26.6        10.95
P110_1        0.967          175.3         76.7
P110_3        0.952          353.5        183.33
P111_1        0.956           41.9         9.72
P111_4        0.962           94.1        39.05
COPYRIGHT 2015 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Belic, Jelena; Koch, Marina; Ulz, Peter; Auer, Martina; Gerhalter, Teresa; Mohan, Sumitra; Fischered
Publication:Clinical Chemistry
Article Type:Report
Date:Jun 1, 2015
Words:5152
Previous Article:Noninvasive prenatal diagnosis of Duchenne muscular dystrophy: comprehensive genetic diagnosis in carrier, proband, and fetus.
Next Article:Controlled cannabis vaporizer administration: blood and plasma cannabinoids with and without alcohol.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters