A Study of Pre-Analytical Variables and Optimization of Extraction Method for Circulating Tumor DNA Measurements by Digital Droplet PCR

Published OnlineFirst March 1, 2019; DOI: 10.1158/1055-9965.EPI-18-0586

Research Article Cancer Epidemiology, Biomarkers A Study of Pre-Analytical Variables and & Prevention Optimization of Extraction Method for Circulating Tumor DNA Measurements by Digital Droplet PCR Luca Cavallone1, Mohammed Aldamry2, Josiane Laﬂeur1, Cathy Lan1, Pablo Gonzalez Ginestet3, Najmeh Alirezaie4, Cristiano Ferrario1, Adriana Aguilar-Mahecha1, and Mark Basik1

Abstract

Background: Circulating free DNA (cfDNA) is an exciting amplification on tumor and circulating tumor DNA novel method to diagnose, monitor, and predict resistance and (ctDNA) from patients with breast cancer. response to cancer therapies, with the potential to radically Results: Making use of a novel protocol for cfDNA extraction alter the management of cancer patients. To fulfill its potential, we increased cfDNA quantities, up to double that of commer- greater knowledge about preanalytical variables is required to cial kits. We found that a second centrifugation at 3,000 g optimize and standardize the collection process, and maxi- on frozen plasma is as efficient as a high-speed (16,000 g) mize the yield and utility of the small quantities of cfDNA centrifugation on fresh plasma and does not affect cfDNA extracted. levels. Methods: To this end, we have compared the cfDNA Conclusions: These results allow for the implementation of extraction efficiency of three different protocols, including protocols more suitable to the clinical setting. Finally, we a protocol developed in house (Jewish General Hospital). found that, unlike targeted gene amplification, whole genome We evaluated the impact on cfDNA levels of preanalytical amplification resulted in altered fractional abundance of variables including speed and timing of the second centri- selected ctDNA variants. fugation and the use of k-EDTA and CTAD blood collection Impact: Our study of the preanalytical variables affecting tubes. Finally, we analyzed the impact on fractional abun- cfDNA recovery and testing will significantly enhance the dance of targeted pre-amplification and whole genome quality and application of ctDNA testing in clinical oncology.

Introduction acquired resistance in cancer patients (4–7). More importantly, evidence suggests that ctDNA measurement may provide a ThepresenceofcirculatingtumorDNA(ctDNA)inblood more comprehensive picture of tumor heterogeneity than provides an accessible source of genetic material from solid tumor biopsies and allow the detection of genomic alterations tumors and holds great promise for the development of clini- in tumors present at different locations throughout the body cal tools for less invasive diagnostic testing in cancer pati- thus providing a more accurate prediction of patient's tumor ents (1–3).Owingtotherelativeeaseofaccessandthe burden (8, 9). The clinical applicability and utility of ctDNA is minimally invasive nature of blood collection, the detection highly promising, and the recent approval by the FDA of the of tumor genomic modifications in ctDNA with highly sensitive Cobas EGFR mutation test to guide therapy in patients with sequencing methods has demonstrated great potential to pro- non-small cell lung carcinoma opens a new era in clinical liquid vide sensitive biomarkers for non-invasive diagnosis, progno- biopsy testing (10, 11). Nonetheless, many challenges remain sis, prediction and monitoring of treatment response, and to be overcome for ctDNA testing to be adopted as a routine clinical tool (12, 13), and for now, the use of tissue biomarkers remains the gold standard to guide therapy for most cancer 1Department of Oncology, Lady Davis Institute, McGill University, Montreal, patients. The major limiting factor comes from the fact that Quebec, Canada. 2Department of Surgery, Jewish General Hospital, Montreal, ctDNA represents only a small fraction of the total amount of Quebec, Canada. 3Department of Epidemiology, Biostatistics and Occupational circulating cell free DNA (cfDNA; refs. 1, 5). Therefore, even 4 Health, Mcgill University, Montreal, Quebec, Canada. McGill University and though ctDNA contains genetic modifications suitable to be Genome Quebec Innovation Center, Montreal, Quebec, Canada. tested as potential cancer biomarkers its use and reliability is Note: Supplementary data for this article are available at Cancer Epidemiology, hampered by its relative scarcity. In fact, although some have Biomarkers & Prevention Online (http://cebp.aacrjournals.org/). reported difficulties in detecting tumor-derived mutations in Corresponding Author: Mark Basik, Lady Davis Institute and Jewish General plasma(14),othershavebeenabletoisolatectDNAinboth Hospital, 3755 Cote Ste Catherine, Montreal, Quebec, Canada H3T1E2. metastatic and nonmetastatic patients (14–16). These incon- Phone: 514-340-8222 ext. 24930; Fax: 514-340-8302; E-mail: sistencies are likely due not only to the inherent variability [email protected] of the fractional abundance of tumor-derived circulating doi: 10.1158/1055-9965.EPI-18-0586 DNA (from <0.01% to >90%), but also to the different sensi- 2019 American Association for Cancer Research. tivity of methodologies used for mutation detection and

www.aacrjournals.org 909

Cavallone et al.

quantification (17–20). Furthermore, it has been shown that one aliquot was either centrifuged at low (3,000 g) or high theamountofisolatedcfDNAdependsontheefficiency of the (16,000 g) centrifugal force, respectively. Samples were cen- extraction method (21–23) and can be affected by pre-analyt- trifuged for 10 minutes at RT and 1 mL of supernatant from each ical variables including type of blood collection tube, centri- condition was collected for cfDNA extraction (Fig. 1B). fugation speed, and storage temperature (23–28). These preanalytical variables can also impact on the release of nonmutated Plasma processing in EDTA and CTAD tubes. Peripheral blood DNA from leucocytes that result in the dilution of the ctDNA from seven (n ¼ 7) healthy volunteers was collected by veni- fraction. In this work, we aimed at optimizing the yield of cfDNA puncture in 2 4mLEDTAtubesand2 4.5 mL BD Vacutainer and studied the effect of different pre-analytical variables on CTAD Tubes (referred to as CTAD Tubes in text; BD#367599; ctDNA detection using digital droplet PCR (ddPCR). We devel- Becton Dickinson). Tubes were inverted 10 times and imme- oped an improved protocol for cfDNA extraction from plasma, diately centrifuged at 1,500 g for 10 minutes at RT. The which formed the basis for the investigation of other variables plasma fraction was carefully collected and a second centrifu- such as the type of anticoagulant in blood collection tubes, and gation at 16,000 g was carried out at RT for 10 minutes. The centrifugation parameters. Finally, to obtain a larger amount of supernatant was collected and 1 mL aliquot was used for cfDNA DNA for downstream applications, we also tested if pre-amplifi- extraction (Fig. 1C). cation of cfDNA would affect the fractional abundance of ctDNA. Our findings contribute to the pre-analytical optimization of Plasma processing Q-CROC-03 patients. As part of the Q-CROC-03 cfDNA processing which is required for precise ctDNA results, trial, patients provided blood at different time points while thus strengthening the potential biomarker value of ctDNA mea- undergoing neoadjuvant chemotherapy for breast cancer. Blood surements in cancer patients. was collected in 2 4.5 mL CTAD tubes and centrifuged within 30 minutes of collection following standard operating pro- Materials and Methods cedures (SOP) submitted to the NCI-SOP database (https:// brd.nci.nih.gov/brd/sop/show/1961): samples were centrifuged Samples at 1,500 g for 15 minutes at RT and plasma immediately Cell lines. BT-20 and MDA-MB-436 from American Type Culture aliquoted and stored at 80C. For this study, aliquots were Collection (ATCC) were cultured according to ATCC recom- thawed and a second centrifugation at 16,000 g was performed mendation. Cells were authenticated using array comparative for 10 minutes at RT prior to analysis. genomic hybridization (aCGH) and mycoplasma testing con- firmed the cells were mycoplasma-free. Cells were grown to Breast cancer tumor DNA 70% to 80% confluence and harvested for DNA isolation using Patients in the Q-CROC-03 trial provided tissue samples QIAamp DNA Mini Kit (Qiagen; Catalog No. 51304). prior to or following standard neoadjuvant chemotherapy. DNA was extracted from tumor biopsies as per reported SOPs Human samples. Blood was collected from healthy control volun- (29). DNA from tumor samples and cell lines underwent whole teers, and plasma and tumor DNA were obtained from patients exome sequencing (WES) and mutation calling at the McGill with breast cancer taking part in the Q-CROC-03 (NCT01276899) University Genome Center (Supplementary methods). Data clinical trial. All participants provided informed consent and the have been deposited in the European Nucleotide Archive data- study was conducted in accordance with recognized ethical guide- base with study primary accession number PRJEB30048. lines (e.g., Declaration of Helsinki), and approved by the Jewish General Hospital Ethics Committee Review Board (JGH; proto- Generation of spiked DNA samples cols 05-006 and A04-M31-12B). We selected genes with single nucleotide variants (SNV) present in cell lines and one Q-CROC-03 patient: TP53 (BT20), Blood collection and plasma processing CDH5 and MAP1LC3B (MDA-MB-436), ROBO2 and PARK2 Plasma processing for cfDNA extraction optimization. Blood from a (patient Neo-05). Targeted pre-amplification of the genomic healthy volunteer was collected by venipuncture in 2 4mL regions containing specific SNVs in breast cancer cell lines or BD Vacutainer K2EDTA tubes (referred to as EDTA Tubes in text; patient tumor DNA was performed with primers listed in BD#366643; Becton Dickinson). Tubes were inverted ten times Supplementary Table S1. DNA was quantified with Qbit and immediately centrifuged at 1,500 g for 10 minutes at dsDNA HS Assay Kit (Catalog No. Q3285; Life Technologies) room temperature (RT). The plasma fraction was carefully and targeted pre-amplification of mutated DNA fragments collected and a second centrifugation at 16,000 g was carried (MUT DNA) was performed with Biorad SsoAdvanced PreAmp out at RT for 10 minutes. The supernatant was collected and Kit (Catalog No. 172-5160; Bio-Rad) as per the manufacturer's 1 mL aliquots were spiked with known amounts of reference instructions. DNA copies generated from the amplification DNA copies from cell lines or breast tumor to test extraction reaction were quantified by QX200 ddPCR with primers and protocols (Fig. 1A). probes (Integrated DNA Technologies) and known amounts of mutated DNA copies were spiked into 1 mL of control human Plasma processing at different centrifugation speeds and time points. plasma (Supplementary Table S1; Fig. 1A). Blood from nine (n ¼ 9) healthy volunteers was collected by venipuncture in 2 4 mL EDTA tubes. Tubes were inverted cfDNA extraction optimization 10 times and immediately centrifuged at 1,500 g for 10 minutes For the comparison of different circulating DNA extraction at RT. The plasma fraction was carefully collected and 1.5 mL methods, extraction was performed from 1 mL of plasma spiked aliquots were either immediately centrifuged or stored at 80 with known amounts of DNA copies as described above. We and centrifuged following 2 weeks of storage. At each time point, tested the QIAamp DSP Kit (Qiagen; Catalog No. 61504), the

910 Cancer Epidemiol Biomarkers Prev; 28(5) May 2019 Cancer Epidemiology, Biomarkers & Prevention

Pre-Analytical Protocol Optimization for ctDNA Testing

QIAamp Circulating Nucleic Acid Kit (Qiagen; Catalog amplification (WGA) reactions using genomic DNA from tumor No. 55114), and a modified protocol developed in house com- tissue or cfDNA from plasma obtained from patients with breast bining components from both kits (referred to as modified cancer. It is based on phi29 DNA polymerase and was carried out hybrid JGH protocol). Both QIAamp kits were used in accordance in a 20 mL reaction volume according to manufacturers' protocols with the manufacturer's instructions. For the modified hybrid with a 3-hour incubation time at 30C. JGH protocol, we used the QIAamp Circulating Nucleic Acid Kit handbook instructions with the following modifications (see Development of ddPCR assays Supplementary Table S2). We designed a set of primer-probe combination for each gene We added 125 mL QIAGEN Proteinase K/mL of plasma and investigated (Supplementary Tables S1 and S3). Primers were 1 mL of buffer AL (from DSP Kit) containing carrier RNA (con- analyzed for specificity using the electronic PCR (In-silico PCR) centration of carrier RNA was established according to QIAGEN tool (http://genome.ucsc.edu/cgi-bin/hgPcr?command¼start). Circulating DNA kit handbook instructions) per 1 mL plasma. The ddPCR conditions were optimized to identify the optimal anneal- lysate mixture was vortexed for 30 seconds and incubated at ing temperature for each assay. 56C for 15 minutes with occasional agitation. 1.2 mL of ACB buffer/mL of plasma was added to the lysate mixture, vortexed for ddPCR analysis 30 seconds, and incubated for 5 minutes on ice. The lysate was ddPCR was performed on a QX200 ddPCR system (Bio-Rad). applied onto the QIAamp MinElute column provided in the Probes for mutated variants were labeled with FAM fluorescent QIAGEN DSP Kit and centrifugations were performed instead of dye and probes for wt variants were labeled with HEX fluorescent using a vacuum pump. The lysate was applied in several elutions dye. PCR reactions were prepared with ddPCR Supermix (Catalog of 650 mL each followed by centrifugation at 10,000 rpm for No. 1863024; Bio-Rad) and partitioned into droplets in a QX200 15 seconds, after applying the last 650 mL of lysate the column was droplet generator, emulsified PCR reactions were performed with centrifuged for 2 minutes. The flow-through from the collection C-1000 touch thermal cycler (Bio-Rad) according to the manu- tube was discarded and 600 mL of AW1 buffer (from DSP Kit) was facturer's instructions. Plates were read in the FAM and HEX added to the column, incubated for 5 minutes at RT and then channels on a Bio-Rad QX200 droplet reader using QuantaSoft centrifuged at 10,000 rpm for 1 minute. After discarding the AW1 software from Bio-Rad (version 1.7.4.0917) to assess the number eluate, 750 mL of AW2 buffer (from DSP Kit) was added to the of droplets positive for mutant (MUT) DNA, wt DNA, both, or column, incubated for 5 minutes at RT and then centrifuged at neither. At least two negative control wells with no DNA were 14,000 rpm for 2 minutes. After the flow-through was discarded, included in every run. 750 mL of EtOH (95%–100%) was applied to the column, For extraction efficiency, copies of spiked DNA recovered incubated for 5 minutes at RT and then spun at 14,000 rpm for from plasma were measured with ddPCR using probes labeled 2 minutes. After the centrifugation, the column was placed in a with FAM on at least three replicates for each gene. The recovery new collection tube, centrifuged for 3 minutes at 14,000 rpm and fraction was calculated as the number of MUT DNA copies incubated at 56C for 10 minutes to clear the column of inter- extracted from 1 mL of plasma over the number of MUT DNA fering substances and improve the quality of DNA to be eluted in copies spiked in 1 mL of plasma. cfDNA levels we measured the the next step. DNA elution consisted in the addition of 25 mLof number of wt TP53 DNA copies present in 1 mL of plasma AVE Buffer to the center of the membrane and after incubation using primers and probes against the wt region of TP53 labeled at RT for 3 minutes, the column was spun at 14,000 rpm for with HEX. 2 minutes to collect the eluate, which was transferred to a clean To measure the fractional abundance of mutated TP53 after 1.5 mL Eppendorf tube. The elution step was repeated twice. targeted or WGA, we used the formula: no. of copies of mutated DNA was stored at 20C for all the protocols tested for further TP53/no. of copies wtTP53 þ no. of copies mutated TP53. ddPCR analysis. Statistical analyses Targeted pre-amplification All statistical analyses reported were performed using R We generated a reference DNA sample using genomic DNA (R Development Core Team; ref. 30). from BT20 breast cancer cells (TP53 mutated) serially diluted (1/5) with genomic DNA from MDA-MB-436 cells [TP53 wild type (wt)] to obtain four fractions of reference DNA with a Results linear dilution of mutated TP53. Each fraction was split into Improved cfDNA extraction efficiency with a novel hybrid two samples of equal volume, one sample was subjected to extraction protocol targeted pre-amplification of the TP53-mutated variant (SsoAd- The quantity of cfDNA that can be extracted from plasma can vanced PreAmp Supermix; Bio-Rad; Catalog No. 1725160) vary depending on the efficiency of the extraction method using primers listed in Supplementary Table S1 and the other chosen (21, 22). For the first objective of this study, we aimed one used as control (non-pre-amplified). at optimizing the cfDNA extraction efficiency from plasma. We Genomic DNA from tumor tissue or cfDNA from plasma tested two commercially available DNA extraction kits, the obtained from breast cancer patients harboring TP53 mutations QIAGEN Circulating Nucleic Acid Kit (CNA) and the QIAGEN were used for targeted pre-amplification of TP53 variants using DSP Virus Kit (DSP) as well as a modified hybrid protocol primers listed in Supplementary Table S3. developed in house (JGH) and described above. To compare the three extraction protocols, we spiked 1 mL of human Whole genome amplification plasma with a known number of reference DNA copies with The Illustra GenomiPhi V2 DNA Amplification Kit (Catalog specific SNVs (Fig. 1A; Supplementary Table S1) and measured No. 25-6600-30; GE Healthcare) was used for whole genome cfDNA recovery by ddPCR. As shown in Fig. 2, the highest

www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 28(5) May 2019 911

Cavallone et al.

Figure 1. Schematic of plasma processing conditions tested. cfDNA extraction protocol optimization with spiked in samples (A), evaluation of impact of different centrifugation conditions (B), and type of anti-coagulant (C) on cfDNA levels.

extraction efﬁciency was obtained with the JGH hybrid protocol yieldcomparedwiththeCNAprotocol.Whenwelookateach with an average percent recovery (all variants combined) of variant separately, we observe that recovery efﬁciency and 89% compared with 46.9% and 75.4% for the CNA and DSP variability vary for each gene tested independently of the protocols, respectively (P < 0.001), that is, a doubling of cfDNA protocol of extraction. Nevertheless, the hybrid JGH protocol resulted in more consistent recovery (>79% for all variants) and less variability compared with the other two protocols (Fig. 3). For all subsequent experiments, we adopted the hybrid JGH protocol for cfDNA extraction.

Impact of plasma processing on cfDNA levels One of the major limitations in ctDNA testing is the fact that ctDNA gets diluted in a pool of genomic cfDNA, which in turn masks the detection of tumor-specific mutated DNA. The source of nonmutated genomic DNA is for the most part circulating lymphocytes and its release can be triggered during blood collection and plasma processing procedures (31, 32). In general, protocols to process plasma recommend a first centrifugation of blood at low speed to enable the separation of plasma from blood cells, followed by a second high-speed centrifugation to minimize the contamination of plasma with nucleic acids from cellular debris (33, 34). However, performing the second high-speed centrifugation is not always suitable in the clinical setting due to the lack of ultra-high-speed centrifuges in hospitals and also to the increased burden that it represents for clinical research associ- ates and nurses. To evaluate the impact of the second centrifugation on plasma cfDNA levels, we tested if different speeds Figure 2. (3,000 g vs. 16,000 g) or time of centrifugation (immediately Median average percent recovery for all five genes measured (n ¼ 3–7 replicates) with each protocol tested. The JGH method shows the on fresh plasma or two weeks later on frozen plasma) could affect highest percent recovery of reference DNA compared with the other the amount of cfDNA detected by ddPCR (Fig. 1B). We measured two commercial protocols. Unpaired t test (, P < 0.001; , P < 0.01). the levels of cfDNA by counting the number of copies of wt TP53

912 Cancer Epidemiol Biomarkers Prev; 28(5) May 2019 Cancer Epidemiology, Biomarkers & Prevention

Pre-Analytical Protocol Optimization for ctDNA Testing

Figure 3. Percent reference DNA copies recovered (mean SD) for each gene after DNA extraction with different protocols. The JGH method shows less variability and higher recovery across the genes tested. Signiﬁcant unpaired t test value , P < 0.001; , P < 0.01; , P < 0.05.

in plasma processed with the different conditions. As shown Targeted pre-amplification of cfDNA does not modify the in Fig. 4, cfDNA levels were not affected by any of the conditions fractional abundance of mutated variants in plasma tested (P ¼ 0.932), suggesting that the speed and the timing of the Because cfDNA quantities are often limited, it may be second centrifugation do not influence the amount of cfDNA advisable to perform a DNA pre-amplification step to increase purified from plasma and thus should not decrease the sensitivity DNA starting material for subsequent analysis. However, the of ctDNA detection. concern is that this pre-amplification step may introduce a bias We also analyzed two types of blood collection tubes, EDTA and alter the fractional abundance of ctDNA relative to wt tubes which are the standard of blood collection for biobank- DNA. The third part of our study thus aimed at evaluating if ing purposes and CTAD tubes, which contain buffered sodium the fractional abundance of mutated DNA could be affected citrate and additives (theophylline, adenosine, and dipyrida- by either "targeted" or "whole genome" DNA pre-amplification mol) that have been shown to inhibit ex vivo platelet activa- reactions. We first analyzed a reference DNA sample contain- tion (35). With increasing proteomics studies being performed ing a serial dilution of TP53 mutant DNA. Targeted pre- on clinical samples, CTAD tubes are being more frequently amplification was carried out and TP53 variant allele frequency utilized for biobanking so as to minimize platelet discharge of (VAF) was correlated with non-pre-amplified samples. The protein biomarkers during blood collection. We therefore test- linearity of the TP53 dilution was maintained with an ed whether there were differences on cfDNA detection by excellent correlation (R2 ¼ 0.9982, P value < 0.001) in VAF ddPCR in plasma collected in CTAD compared with EDTA for non-amplified versus pre-amplified samples (Supplemen- tubes. Despite a slight increase in average cfDNA levels in tary Fig. S2). Targeted pre-amplification was also tested on samples collected in EDTA tubes (mean ¼ 905.62 copies tissue DNA and serial plasma cfDNA samples from patients 468.49) compared with CTAD tubes (mean ¼ 805.83 copies with breast cancer harboring mutations in the TP53 gene (Sup- 413.32), the difference observed was not statistically significant plementary Table S3). A perfect correlation in VAF between (P ¼ 0.477; Supplementary Fig. S1). non-amplified and targeted pre-amplified samples was again

Figure 4. Box plots of cfDNA copies/mL of plasma detected by ddPCR in plasma processed with a second centrifugation at either low (3,000 g) or high (16,000 g) speed on the same day or after 2 weeks of storage at 80C. Median and SD of wt TP53 copies/mL in plasma for each condition tested are depicted (n ¼ 9). Statistical analysis was performed using Kruskal–Wallis chi-squared test (P ¼ 0.9532).

www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 28(5) May 2019 913

Cavallone et al.

Figure 5. Correlation of mutated TP53 VAF measured by ddPCR in tissue (n ¼ 4 patients; A) and plasma ctDNA (n ¼ 4 patients; B) after targeted pre-amplification and without pre-amplification. Between one and nine technical replicates were performed for each patient sample. Spearman's correlation coefficient was used to correlate the two variables.

obtained for both tissue (r2 ¼ 1) and plasma samples (r2 ¼ applications, we developed our own in-house hybrid protocol 0.993; Fig. 5A and B). These results confirm that allele fre- (JGH protocol), which produced significantly higher yields of quency is well preserved following targeted pre-amplification cfDNA than the original commercial kits. We were able to show reactions. that the protocol we developed was able to recover nearly the Finally, we performed WGA on DNA from tissue and plasma totality of spiked DNA in plasma (80%–100%) with minimal samples from the same patients as above. We found that WGA variability. As previously reported by others, our results further does not alter the fractional abundance of mutated TP53 in confirm that one of the causes of variability in cfDNA measure- tissue samples (r2 ¼ 0.9989), however, allele frequency was not ments comes from the different isolation methods used. preserved in plasma samples as shown with the poor correla- Currently, there is no consensus protocol for the pre- tion between the VAF of non-amplified versus amplified plasma analytical processing of blood for isolation of cfDNA and it DNA samples (r2 ¼ 0.4337; Fig. 6A and B). Therefore, WGA is known that different pre-analytical factors can influence its appears to introduce a bias when performed on cfDNA but not stability and levels. For instance, apoptosis or necrosis of on tissue DNA. peripheral blood mononuclear cells in the blood collection tube may lead to the release of genomic DNA, explaining the observed increase of cfDNA when the blood sample is stored Discussion for several hours before processing (23, 38). Ideally, the entire Currently, liquid biopsy shows considerable potential in cap- procedure from blood draw until plasma separation, should be turing the genomic profile of solid tumors. For instance, several performed within a couple of hours (33, 39). As rapid and groups have reported the reliable measurement of somatic muta- well-controlled blood processing is not always possible in the tions in KRAS and ERBB2 amplification in cfDNA (36, 37). One of clinical setting and blood samples may lie unprocessed for long the main limitations in using ctDNA as a biomarker in the clinical periods of time, blood collection tubes with stabilizing agents setting is the lack of analytical validation which starts with the that minimize the release of contaminating DNA in plasma standardization of sample processing methods, storage condi- have been developed. Streck cfDNA BCTs tubes are commonly tions, and protocols for both cfDNA extraction and quantifica- used and have been reported to prevent the lysis of white blood tion. Although most of the variation observed between different cells, and therefore any dilution of ctDNA with wt genomic individuals will be biological, there are many technical factors DNA (28, 31, 32). Although EDTA is the anticoagulant of that could cause a wide range of measurement inconsistencies. choice for most biological analyses, blood is also collected in Indeed, several reports have highlighted the lack of standardiza- CTAD tubes especially for proteomics studies. Until now, no tion (13, 33, 34) and the effect of different extraction methods on study had evaluated the effect of CTAD additives on cfDNA cfDNA yields (21–23). levels. Our results concluded that cfDNA concentration is not While assessing cfDNA extraction efficiency obtained using significantly altered when extracted from plasma collected in different components of commercial kits for downstream ddPCR both EDTA and CTAD anticoagulant tubes.

Figure 6. Correlation of mutated TP53 VAF measured by ddPCR in tissue (n ¼ 4 patients; A) and plasma ctDNA (n ¼ 4 patients; B) after whole genome pre-amplification and without pre- amplification. Between one and nine technical replicates were performed for each patient sample. Spearman's correlation coefficient was used to correlate the two variables.

914 Cancer Epidemiol Biomarkers Prev; 28(5) May 2019 Cancer Epidemiology, Biomarkers & Prevention

Pre-Analytical Protocol Optimization for ctDNA Testing

Published protocols for centrifugation of blood samples study allowed us to optimize cfDNA extraction efficiencies with typically use two sequential spins to separate plasma from blood anovelandsignificantly more efficient protocol and allowed us cells. Different groups recommend a second centrifugation at dif- to better understand the impact of pre-analytical factors on ferent speeds, with the highest speed reported at 16,000 g ctDNA analyses. Our results will be important to consider while (40, 41). Because most laboratories can centrifuge blood at designing future clinical studies. Development of reliable and low speed, we evaluated the influence of the second centri- robust non-invasive diagnostic platforms using standardized fugation at low (3,000 g) and high (16,000 g) speed. We protocols for ctDNA measurement will ensure that this bio- also evaluated the impact of delaying the second centrifugation marker fulfills its promise to add tremendous value to diag- following plasma storage at 80C for 2 weeks, or performing it nosis and monitoring of treatmentresponseofcancerpatients. immediately on fresh plasma. Our results confirm that proceeding with the second centrifugation at low or high speed and storing Disclosure of Potential Conflicts of Interest plasma samples for as long as two weeks at 80 C before the No potential conflicts of interest were disclosed. second spin does not alter cfDNA levels. These observations may facilitate the adoption of plasma processing protocols that are more accommodating to the clinical setting either by allowing Authors' Contributions a second centrifugation at low speed with standard centrifuges or Conception and design: L. Cavallone, M. Aldamry, A. Aguilar-Mahecha, M. Basik by allowing the shipment of frozen plasma samples to undergo Development of methodology: L. Cavallone, M. Aldamry high-speed centrifugation in the laboratory where ctDNA testing Acquisition of data (provided animals, acquired and managed patients, is performed. provided facilities, etc.): L. Cavallone, M. Aldamry, J. Lafleur, C. Lan, Finally, we also tested the reliability of pre-amplification C. Ferrario, A. Aguilar-Mahecha, M. Basik strategies, including targeted and WGA. Pre-amplification reac- Analysis and interpretation of data (e.g., statistical analysis, biostatistics, tions are a useful strategy when the amount of DNA is limited. computational analysis): L. Cavallone, M. Aldamry, P.G. Ginestet, N. Alirezaie, A. Aguilar-Mahecha WGA could theoretically enable the testing of different variants fi fi Writing, review, and/or revision of the manuscript: L. Cavallone, C. Ferrario, in the same sample whereas targeted gene-speci campli ca- A. Aguilar-Mahecha, M. Basik tion could enable technical replication of the assay in condi- Administrative, technical, or material support (i.e., reporting or organizing tions of limited cfDNA. First, we needed to ensure that the data, constructing databases): L. Cavallone, J. Lafleur, A. Aguilar-Mahecha process of pre-amplification did not alter allele frequency Study supervision: L. Cavallone, M. Basik ratios. We found that the fractional abundance of DNA was preserved when using targeted pre-amplification of known Acknowledgments mutated SNVs. Interestingly, these results were obtained when TheauthorswouldliketothanktheFRQSReseau Recherche Cancer for biobank support, Mr. Mark Sherman for hisgenerousdonationtoenable the fractional abundance of the mutated molecules was more the purchase of the ddPCR equipment, and Dr Eric Bareke for bioinfor- than 0.05%. We were not able to obtain consistent fractional matics support. M. Aldamry was supported with a scholarship from King abundances for samples with fractional ctDNA abundances Saud Bin Abdulaziz University for Health Sciences. This work was sup- under 0.05%. However, using WGA, we were not able to ported by a Quebec Breast Cancer Foundation grant to M. Basik, and a generate DNA with the same fractional abundance from ctDNA. Genome Quebec grant to M. Basik. This can be caused by the fact that cfDNA is composed of DNA fragments of different length which can affect the uniformity of The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked WGA in cfDNA samples (42). We therefore recommend caution advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate when applying WGA to ctDNA analysis. this fact. In summary, protocol harmonization and the use of consistent and reproducible extraction and processing methods are Received June 5, 2018; revised August 13, 2018; accepted February 25, 2019; required for ctDNA testing to enter the clinical setting. This published first March 1, 2019.

References 1. Bettegowda C, Sausen M, Leary RJ, Kinde I, Wang Y, Agrawal N, et al. cancer by circulating tumor DNA profiling. Cancer Discov 2017;7: Detection of circulating tumor DNA in early- and late-stage human 1394–403. malignancies. Sci Transl Med 2014;6:224ra24. 8. De Mattos-Arruda L, Weigelt B, Cortes J, Won HH, Ng CK, Nuciforo P, 2. Heitzer E, Ulz P, Geigl JB. Circulating tumor DNA as a liquid biopsy for et al. Capturing intra-tumor genetic heterogeneity by de novo mutation cancer. Clin Chem 2015;61:112–23. profiling of circulating cell-free tumor DNA: a proof-of-principle. 3. Alix-Panabieres C, Pantel K. Clinical applications of circulating tumor cells Ann Oncol 2014;25:1729–35. and circulating tumor DNA as liquid biopsy. Cancer Discov 2016;6: 9. Chicard M, Colmet-Daage L, Clement N, Danzon A, Bohec M, Bernard V, 479–91. et al. Whole-exome sequencing of cell-free DNA reveals temporo-spatial 4. Russo M, Siravegna G, Blaszkowsky LS, Corti G, Crisafulli G, Ahronian LG, heterogeneity and identifies treatment-resistant clones in neuroblastoma. et al. Tumor heterogeneity and lesion-specific response to targeted therapy Clin Cancer Res 2018;24:939–49. in colorectal cancer. Cancer Discov 2016;6:147–53. 10. U.S. Food & Drug Adiministration. Cobas EGFR mutation test v2. 5. Siravegna G, Marsoni S, Siena S, Bardelli A. Integrating liquid biopsies into Available from: https://www.fda.gov/Drugs/InformationOnDrugs/ the management of cancer. Nat Rev Clin Oncol 2017;14:531–48. ApprovedDrugs/ucm504540.htm 6. Marzese DM, Hirose H, Hoon DS. Diagnostic and prognostic value of 11. Kwapisz D. The first liquid biopsy test approved. Is it a new era of circulating tumor-related DNA in cancer patients. Expert Rev Mol Diagn mutation testing for non-small cell lung cancer? Ann Transl Med 2017; 2013;13:827–44. 5:46–52. 7. Chaudhuri AA, Chabon JJ, Lovejoy AF, Newman AM, Stehr H, Azad TD, 12. Bardelli A, Pantel K. Liquid biopsies, what we do not know (yet). et al. Early detection of molecular residual disease in localized lung Cancer Cell 2017;31:172–9.

www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 28(5) May 2019 915

Cavallone et al.

13. Merker JD, Oxnard GR, Compton C, Diehn M, Hurley P, Lazar AJ, et al. 28. Kang Q, Henry NL, Paoletti C, Jiang H, Vats P, Chinnaiyan AM, et al. Circulating tumor DNA analysis in patients with cancer: American Society Comparative analysis of circulating tumor DNA stability in K3EDTA, of Clinical Oncology and College of American Pathologists joint review. Streck, and CellSave blood collection tubes. Clin Biochem 2016;49: J Clin Oncol 2018;36:1631–41. 1354–60. 14. Brock G. Liquid biopsy for cancer screening, patient stratification and 29. Aguilar-Mahecha A, Lafleur J, Pelmus M, Seguin C, Lan C, Discepola F, et al. monitoring. Transl Cancer Res 2015;4:280–90. The identification of challenges in tissue collection for biomarker studies: 15. Diehl F, Schmidt K, Choti MA, Romans K, Goodman S, Li M, et al. the Q-CROC-03 neoadjuvant breast cancer translational trial experience. Circulating mutant DNA to assess tumor dynamics. Nat Med 2008;14: Mod Pathol 2017;30:1567–76. 985–90. 30. R Development Core Team 2008. R: A language and environment for 16. Dobrzycka B, Terlikowski SJ, Mazurek A, Kowalczuk O, Niklinska W, statistical computing. R Foundation for Statistical Computing, Vienna, Chyczewski L, et al. Circulating free DNA, p53 antibody and muta- Austria. Available from: http://www.R-project.org. tions of KRAS gene in endometrial cancer. Int J Cancer 2010;127: 31. Toro PV, Erlanger B, Beaver JA, Cochran RL, VanDenBerg DA, Yakim E, et al. 612–21. Comparison of cell stabilizing blood collection tubes for circulating 17. Diehl F, Li M, He Y, Kinzler KW, Vogelstein B, Dressman D. BEAMing: plasma tumor DNA. Clin Biochem 2015;48:993–8. single-molecule PCR on microparticles in water-in-oil emulsions. 32. Norton SE, Lechner JM, Williams T, Fernando MR. A stabilizing reagent Nat Methods 2006;3:551–9. prevents cell-free DNA contamination by cellular DNA in plasma during 18. Diaz LA Jr, Bardelli A. Liquid biopsies: genotyping circulating tumor DNA. blood sample storage and shipping as determined by digital PCR. J Clin Oncol 2014;32:579–86. Clin Biochem 2013;46:1561–5. 19. Leary RJ, Sausen M, Kinde I, Papadopoulos N, Carpten JD, Craig D, et al. 33. El Messaoudi S, Rolet F, Mouliere F, Thierry AR. Circulating cell free DNA: Detection of chromosomal alterations in the circulation of cancer preanalytical considerations. Clin Chim Acta 2013;424:222–30. patients with whole-genome sequencing. Sci Transl Med 2012;4: 34. Chang Y, Tolani B, Nie X, Zhi X, Hu M, He B. Review of the clinical 162ra154. applications and technological advances of circulating tumor DNA in 20. Devonshire AS, Whale AS, Gutteridge A, Jones G, Cowen S, Foy CA, et al. cancer monitoring. Ther Clin Risk Manag 2017;13:1363–74. Towards standardisation of cell-free DNA measurement in plasma: con- 35. Ahnadi CE, Sabrinah Chapman E, Lepine M, Okrongly D, Pujol-Moix trols for extraction efficiency, fragment size bias and quantification. N, Hernandez A, et al. Assessment of platelet activation in several Anal Bioanal Chem 2014;406:6499–512. different anticoagulants by the Advia 120 Hematology System, fluo- 21. Fong SL, Zhang JT, Lim CK, Eu KW, Liu Y. Comparison of 7 methods for rescence flow cytometry, and electron microscopy. Thromb Haemost extracting cell-free DNA from serum samples of colorectal cancer patients. 2003;90:940–8. Clin Chem 2009;55:587–9. 36. Misale S, Yaeger R, Hobor S, Scala E, Janakiraman M, Liska D, et al. 22. Fleischhacker M, Schmidt B, Weickmann S, Fersching DM, Leszinski GS, Emergence of KRAS mutations and acquired resistance to anti-EGFR Siegele B, et al. Methods for isolation of cell-free plasma DNA strongly therapy in colorectal cancer. Nature 2012;486:532–6. affect DNA yield. Clin Chim Acta 2011;412:2085–88. 37. Gevensleben H, Garcia-Murillas I, Graeser MK, Schiavon G, Osin P, Parton 23. Page K, Guttery DS, Zahra N, Primrose L, Elshaw SR, Pringle JH, et al. M, et al. Noninvasive detection of HER2 amplifi cation with plasma DNA Influence of plasma processing on recovery and analysis of circulating digital PCR. Clin Cancer Res 2013;19, 3276–84. nucleic acids. PLoS One 2013;8:e77963. 38. Xue X, Teare MD, Holen I, Zhu YM, Woll PJ. Optimizing the yield and 24. Hidestrand M, Stokowski R, Song K, Oliphant A, Deavers J, Goetsch M, et al. utility of circulating cell-free DNA from plasma and serum. Clin Chim Acta Influence of temperature during transportation on cell-free DNA analysis. 2009;404:100–4. Fetal Diagn Ther 2012;31:122–8. 39. Elshimali YI, Khaddour H, Sarkissyan M, Wu Y, Vadgama JV. The clinical 25. Garcia J, Dusserre E, Cheynet V, Bringuier PP, Brengle-Pesce K, Wozny AS, utilization of circulating cell free DNA (CCFDNA) in blood of cancer et al. Evaluation of pre-analytical conditions and comparison of the patients. Int J Mol Sci 2013;14:18925–58. performance of several digital PCR assays for the detection of major EGFR 40. Riva F, Bidard FC, Houy A, Saliou A, Madic J, Rampanou A, et al. Patient- mutations in circulating DNA from non-small cell lung cancers: the specific circulating tumor DNA detection during neoadjuvant chemother- CIRCAN_0 study. Oncotarget 2017;8:87980–96. apy in triple-negative breast cancer. Clin Chem 2017;63:691–9. 26. Parpart-Li S, Bartlett B, Popoli M, Adleff V, Tucker L, Steinberg R, et al. The 41. Guo N, Lou F, Ma Y, Li J, Yang B, Chen W, et al. Circulating tumor DNA effect of preservative and temperature on the analysis of circulating tumor detection in lung cancer patients before and after surgery. Sci Rep 2016;6: DNA. Clin Cancer Res 2017;23:2471–7. 33519. 27. Chan KC, Yeung SW, Lui WB, Rainer TH, Lo YM. Effects of preanalytical 42. Underhill HR, Kitzman JO, Hellwig S, Welker NC, Daza R, Baker DN, et al. factors on the molecular size of cell-free DNA in blood. Clin Chem 2005; Fragment length of circulating tumor DNA. PLoS Genet 2016;12: 51:781–4. e1006162.

916 Cancer Epidemiol Biomarkers Prev; 28(5) May 2019 Cancer Epidemiology, Biomarkers & Prevention

A Study of Pre-Analytical Variables and Optimization of Extraction Method for Circulating Tumor DNA Measurements by Digital Droplet PCR

Luca Cavallone, Mohammed Aldamry, Josiane Lafleur, et al.

Cancer Epidemiol Biomarkers Prev 2019;28:909-916. Published OnlineFirst March 1, 2019.

Updated version Access the most recent version of this article at: doi:10.1158/1055-9965.EPI-18-0586

Supplementary Access the most recent supplemental material at: Material http://cebp.aacrjournals.org/content/suppl/2019/03/01/1055-9965.EPI-18-0586.DC1

Cited articles This article cites 40 articles, 13 of which you can access for free at: http://cebp.aacrjournals.org/content/28/5/909.full#ref-list-1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department Subscriptions at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cebp.aacrjournals.org/content/28/5/909. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.