Developing Ultrasensitive Library-Aliquot-Based Droplet Digital PCR

Subscriber access provided by Kaohsiung Medical University

Technical Note

Developing ultra-sensitive library aliquot-based droplet digital PCR for detecting T790M in plasma circulating tumor DNA of non-small-cell lung cancer patients Ke Yang, Jun Li, Jiuzhou Zhao, Pengfei Ren, Zhizhong Wang, Bing Wei, Bing Dong, Rui Sun, Xiaoyan Wang, H.J.M. (Harry) Groen, Jie Ma, and Yongjun Guo Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b01776 • Publication Date (Web): 29 Aug 2018 Downloaded from http://pubs.acs.org on August 30, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Developing ultra-sensitive library aliquot-based droplet digital PCR for detecting T790M in plasma circulating tumor DNA of non-small-cell lung cancer patients

Running Title: Ultrasensitive LAB-ddPCR for ctDNA EGFR T790M detection

Ke Yang1, Jun Li1, Jiuzhou Zhao 1, Pengfei Ren 1 , Zhizhong Wang 1, Bing Wei 1, Bing Dong 1, Rui Sun 1, Xiaoyan Wang 1, Groen, Harry J.M 2, Jie Ma1*, Yongjun Guo1*

1

Department of Molecular Pathology, The Affiliated Cancer Hospital of Zhengzhou

University, 127 Dongming Road, 2

Zhengzhou, 450003， China

Department of Pulmonary Diseases, University of Groningen and University

Medical Center Groningen, 9700 RB, The Netherland ASSOCIATED CONTENT Supporting Information

1

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 25

Author Yongjun Guo received funding from The Construction of the Big Data Platform for the Accurate Medical Treatment of Common Maligant Tumors Such As Lung Cancer (Major Science and Technology Projects of Henan 161100311500).

AUTHOR INFORMATION Corresponding Authors *

E-mail: [email protected] Phone:＋86 (0) 37165587506

*

E-mail: [email protected] Phone:＋86 (0) 37165587681

ORCID Ke Yang: 0000-0001-8085-8282 Notes The authors declare no potential conflicts of interest.

2


Page 3 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Abstract T790M secondary mutation in EGFR is the most well-established EGFR-TKIs resistant marker in non-small-cell lung cancer (NSCLC). The current methods to rapidly and accurately detect T790M in clinical practice are not satisfactory because of several obstacles, including unavailability of tumor tissue re-biopsies and low DNA copy number of T790M in circulating tumor DNA (ctDNA). Here, we develop a library aliquot-based droplet digital PCR (LAB-ddPCR) to increase detection sensitivity without affecting accuracy. This new LAB-ddPCR method is performed using the aliquots of ctDNA pre-capture Next-generation sequencing (NGS) library, in which the isolated ctDNA was amplified and enriched. We showed the LAB-ddPCR could precisely distinguish between T790M wild-type and mutation alleles without introducing extra false-positive signals. In a cohort of 70 post-TKIs NSCLC patients, the LAB-ddPCR identified 41 T790M-positive cases (sensitivity 58.57%), but ddPCR only detected T790M in 27 cases (sensitivity 38.57%). Taking the ARMS-PCR result from matched tumor re-biopsies into consideration, the LAB-ddPCR method is better than ddPCR. In conclusion, the LAB-ddPCR ctDNA test offers a feasible and flexible option for rapid and accurate detection of T790M secondary mutation, which is helpful in dynamically monitoring of drug response and disease progression throughout the therapeutic regimen. Keywords: EGFR T790M; droplet digital PCR; ctDNA; highly sensitive; NGS.


3


Page 4 of 25

Epidermal growth factor receptor tyrosine kinase inhibitors (EGFR -TKIs), e.g. gefitinib, erlotinib, and afatinib, have been recommended as the first-line therapy for non-small cell lung cancer (NSCLC) patients with EGFR sensitizing mutations 1. However, most post-TKIs patients eventually develop acquired resistance and become refractory in 9-13 months

2-4

. The occurrence of de novo EGFR T790M

secondary mutation is the most well-established biomarker for this resistance, and the third generation EGFR-TKIs have been developed to overcome this resistance, e.g. osimertinib, rociletinib, HM61713, and ASP8273, which showed optimistic clinic outcomes

5-7

. Therefore, fast and accurate detection of T790M is urgently

needed for dynamic monitoring of the disease progression and designing new treatment strategy opportunely for the patients. According to the NCCN guideline, it is suggested to investigate genetic aberrations in tumor tissues 1. However, this is not practical for EGFR-TKIs-resistant NSCLC patients, because the tumor re-biopsy is not always readily available for the test. Alternatively, circulating tumor DNA (ctDNA), as a liquid biopsy in the form of blood draws, is minimally invasive and feasible in the clinic practice to monitor tumor progression throughout the treatment 8. Previous studies using tumor tissues showed the secondary T790M mutation could be detected in about 60% of the EGFR-TKIs resistant NSCLC patients

9, 10

. However, the sensitivity of T790M

ctDNA test is only about 40%-50% 11-14, which is largely inhibited by the quantity of input material 15 Here we propose a method named library aliquots based ddPCR (LAB-ddPCR), in


4

Page 5 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


which the isolated ctDNA was first processed to NGS library preparation, followed by a ddPCR test on the aliquots of pre-capture library to identify T790M status. Since the ctDNA was amplified after NGS library preparation, we assume our LAB-ddPCR could largely increase the detection sensitivity of T790M ctDNA test.

Materials and Methods Sections Patients and Specimen Seventy advanced NSCLC patients with acquired resistance to EGFR-TKIs were enrolled at the Henan Cancer Hospital from December 2016 to August 2017. The study inclusion criteria are as follows: (1) the patients were diagnosed as lung adenocarcinoma through tissue biopsies; (2) activating mutations 19 del or L858R were confirmed by NGS and/or ARMS-PCR in the tumor tissues; (3) according to the RECIST 1.1 (Response Evaluation Criteria in Solid Tumors 1.1) criteria, these patients showed progression of the disease (PD) after EGFR-TKIs treatment; (4) If available, the peripheral blood and tumor re-biopsies were collected simultaneously in advance to any further treatment. If not, only the peripheral blood was collected. The study was approved by the Henan Cancer Hospital research ethics committee. All patients signed an informed consent before the collection and use of their blood samples and tumor tissues. DNA extraction A total of 10 mL peripheral blood was collected and preserved into EDTA tube (SANLI，China) for ctDNA isolation. The isolation procedure was executed within 2 hours after collection. Briefly, the plasma was first centrifuged at 2000g for 10


5


Page 6 of 25

minutes at 4 , and then increased to 16000g for 10 minutes; cfDNA was captured by using the QIAamp® Circulating Nucleic Acid Kit (#55114, QIAGEN, Germany). Genomic DNA was extracted from formalin-fixed paraffin-embedded (FFPE) tumor tissues by using the QIAamp DNA FFPE Tissue kit (#56404, QIAGEN, Germany). The experiment was performed according to the protocol offered by the manufacturers. DNA concentration was measured by using the QubitTM dsDNA HS assay (#1204008450, Life Technologies, the U.S.). Extracted DNA was stored at -20 until analysis. NGS library preparation A total of 10ng-40ng ctDNA extracted from peripheral blood were used for NGS library preparation by using NextSeq 550AR Library Kit (#EP0301708, ANOROAD，China). The target panel includes 8 genes suggested by NCCN Guidelines (version 3.2018) Non-Small Cell lung cancer, including EGFR, ALK, ROS1, BRAF, ERBB2, MET, KRAS and BRAF. The sequencing region covers all the exons and the flanking splicing regions of these genes, as well as some introns (32, 33, 34 and 35) of ROS1, the intron 19 and 20 of ALK, and the intron 11 of RET. The experiment was performed according the protocols provided by the manufacture. Briefly, the isolated ctDNA was first subject to end-repairing, A-tailing and adapters ligation, then the eluted ctDNA was processed for PCR amplification (6-9 cycles) on Thermocycler (#IT041-0002, ExCell Biology，China). PCR library amplification procedure used is as follows, 94℃ for 5 minutes; 6-9 cycles of 94℃ for 30 seconds, 62℃ for 30 seconds, 72℃ for 30 seconds; 72℃ for 10 minutes, and 16℃ for 30


6

Page 7 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


minutes. After the clean-up, the ctDNA library was quantified by using Qubit dsDNA HS assay (#1204008450, Life Technologies, the U.S.). A total of 125~500ng pre-capture NGS library aliquots were used for target enrichment, and the captured targets were processed to LM-PCR with subsequent clean-up. The quality and quantity of the captured targets after clean-up is evaluated by Agilent 2100 Bioanalyzer (#DE13806001, Agilent Technologies, the U.S.). Genomic ddPCR test and LAB-ddPCR Droplet digital PCR assays were conducted on a QX200 digital PCR system (Bio-Rad, Hercules, CA). T790M was detected by using human EGFR T790M ddPCR detection kit (#CB240003, YuanQi, China) according to the manufacturer’s instructions. The PCR program is as follows: 95 for 15 seconds, 58

for 60 seconds, 98

for 10 minutes, 40 cycles of 94

for 10 minutes, and 4

reaction temperature is changed at a rate of 2.5

for 5 minutes. The

/sec. After thermal cycling, the

amplified samples were loaded into Bio-Rad reader (#771BR24449, Bio-Rad, Hercules, CA) to count the number of droplets for analysis. The experiment was performed according to the protocols of Mutation Detection Best Practices Guidelines provided by Bio-Rad (Bio-Rad, Hercules, CA). Genomic ddPCR was performed by using ctDNA directly as the input; alternatively, LAB-ddPCR was performed on the aliquots of pre-capture NGS library as described above. We optimized the amount of DNA input used for ddPCR assay from 20ng-800ng. An input of 300ng showed best balance in both high detection sensitivity and low false-positive signal, which is used in all the ddPCR tests. The detection limit and


7


Page 8 of 25

cutoff value of ddPCR were evaluated by using HDX Reference Standard DNA (#HD786, Horizon Discovery, Cambridge, UK) with serial variant T790M mutants ranging from 0.01% to 10%.

ARMS-PCR ARMS-PCR was performed according to the manufacturer’s instructions (catalog number，ACCB Biotech Ltd, China) on 15-24ng genomic DNA extracted from FFPE tumor biopsies by using Human EGFR T790M mutation quantitative detection kit (catalog number，ACCB Biotech Ltd, China) on a Stratagene Mx3000p platform (catalog number，Agilent Technologies, the U.S.).


8

Page 9 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Results and Discussion An input amount of 300ng used in LAB-ddPCR shows the best performance in T790M detection sensitivity and fidelity

First and foremost, we determined the optional input amount for LAB-ddPCR test on Reference Standard ctDNA containing 0.1% T790M. The LAB-ddPCR test was performed by using different amount of input (20ng, 75ng, 150ng, 300ng and 500ng), and the resulting total droplets (total events), occupancy droplets (occupancy events) and T790M VFs were evaluated. The result shows: a) the total events are comparable by using different amount of input (Figure1 A); b) the occupancy events increase with the input amount, but this increase is not that sharp from 300ng to 500ng (Figure1 B); 3) An input amount of 300ng shows the best detection accuracy, with a consistent variant frequency as known in the Reference Standard ctDNA (0.1%) (Figure1 C). These results suggest an input amount of 300ng appears to be best balanced in both detection sensitivity and false positivity.

LAB-ddPCR shows high fidelity in detecting ctDNA T790M mutation

Next, we investigated the detection fidelity of our LAB-ddPCR method, since the amplification in NGS library preparation may introduce false positive signals, which is significantly important in clinical testing. To this end, we first carried out the test in Reference Standard ctDNA with serial variant T790M mutants (5%, 1%, 0.5%, 0.1%, 0.05% and 0.025%.). Three independent experiments on Reference


9


Page 10 of 25

Standard ctDNA showed highly consistent results as expected (Figure2 A and B). When the T790M VFs is lower down to 0.025%, although 1 or 2 790M positive droplets could be observed, it’s still determined as negative, since the criteria for positive detection is more than 3 droplets. Importantly, we included 3 different types of negative controls, including water (nc-1), circulating free DNA (cfDNA) from healthy donors (nc-2) and ctDNA from NSCLC patients (nc-3) that the T790M status had been determined as negative by NGS; no positive result was identified (Figure2 A and B). Then we extended this test to 23 ctDNA samples isolated from NSCLC patients that the T790M status was determined by both NGS and LAB-ddPCR in parallel. As expected, LAB-ddPCR shows consistent detection of T790M as NGS; again, no false positive result was determined by LAB-ddPCR (Figure2 C). These data clearly demonstrates the fidelity of the LAB-ddPCR method, indicating it can detect T790M variant sensitively without introducing false positivity. Intra-operator variability of the experiment was also evaluated in both cfDNA from NSCLC patients and reference standard ctDNA, the results from different operators showed minimum difference (Figure2 D). Collectively, these results clearly demonstrate the LAB-ddPCR could perform a consistent and robust detection of T790M with a variant frequency (VF) over 0.05% from ctDNA without introducing false positivity. It’s noteworthy that the lower detection limit of NGS testing in our lab is also setted as 0.1% (unique reads 5000☓, and more than 5 unique variant allele reads), but the LAB-ddPCR could be finished only in 3 days, which is much shorter than


10

Page 11 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


NGS (usually 7 days). This could be significant important for patients.

LAB-ddPCR is more sensitive in detecting ctDNA T790M mutations than canonical ddPCR，， as compared to the result from FFPE tumor re-biopsies

Retrospective studies on FFPE tumor re-biopsies showed that the secondary T790M mutation occurred in about 60% of advanced NSCLC patients with acquired EGFR-TKIs resistance

9, 10

. However, studies performed by Wang Z et al

and Wang W (N=75) et al

17

16

(N=135)

could only detect T790M in 43% and 46.7% of the

EGFR-TKI resistant NSCLC patients from their ctDNA samples. This discrepancy may lead to some patients to miss the opportunity of receiving target drugs. The decreased detection sensitivity of canonical ddPCR is largely inhibited by low ctDNA input for the test. We summarized our recent 150 ctDNA isolations from 10mL whole blood of NSCLC patients, the yield (median=79.25, P25=53.25, P75=125.25) is far from the optimal input amount of 300ng (Supplementary Figure1). If the LAB-ddPCR could significantly increase the input amount, we assume this method could reveal the true T790M positivity from very low ctDNA yield. Hereby, we consecutively recruited 70 post-TKI resistant patients, and determined their T790M status by canonical ddPCR and LAB-ddPCR in parallel (Supplementary Table1).

Not surprisingly, the average input of LAB-ddPCR

significantly increased to 283.86ng (range, 34.80–690.00ng), as compared to 3.03ng (range, 0.72–238.80ng) in ddPCR (P

Developing Ultrasensitive Library-Aliquot-Based Droplet Digital PCR

Recommend Documents