Detection of EGFR mutations in non-small cell lung cancer by droplet digital PCR.

Activating mutations in EGFR predict benefit from tyrosine kinase inhibitor therapy for patients with advanced non-small cell lung cancer. Directing patients to appropriate therapy depends on accurate and timely EGFR assessment in the molecular pathology laboratory. This article describes the analytical design, performance characteristics, and clinical implementation of an assay for the rapid detection of EGFR L858R and exon 19 deletion mutations. A droplet digital polymerase chain reaction (ddPCR) assay was implemented with probe hydrolysis-dependent signal detection. A mutation-specific probe was used to detect EGFR L858R. A loss of signal design was used to detect EGFR exon 19 deletion mutations. Analytical sensitivity was dependent on DNA input and was as low as 0.01% variant allele fraction for the EGFR L858R assay and 0.1% variant allele fraction for the EGFR exon 19 deletion assay. Correlation of 20 clinical specimens tested by ddPCR and next generation sequencing showed 100% concordance. ddPCR showed 53% clinical sensitivity in the detection of EGFR mutations in plasma cell-free DNA from patients with lung cancer. The median clinical turnaround time was 5 days for ddPCR compared to 13 days for next generation sequencing. The findings show that ddPCR is an accurate and rapid method for detecting EGFR mutations in patients with non-small cell lung cancer.

Introduction

Lung cancer remains the second most common non-skin cancer of both men and women, and the most common overall, with an estimated 229,000 Americans diagnosed in 2020 [1]. Approximately 20% of patients with non-small cell lung cancer have mutations in the EGFR gene [2, 3], for which targeted tyrosine kinase inhibitor therapy provides significant treatment benefit [4].

Given the importance of matching patients with EGFR mutations to the appropriate targeted therapy, laboratories require testing approaches that are both timely and accurate. The most common mutations in EGFR that confer sensitivity are EGFR L858R (a point mutation in exon 21) and in frame deletions in exon 19. The gold standard for detecting these mutations has historically been EGFR sequencing, often by Sanger sequencing performed on formalin-fixed paraffin-embedded (FFPE) tissue [5, 6]. However, the availability of next generation sequencing (NGS) and new polymerase chain rection (PCR) technologies has opened the possibility for different approaches.

A recent technology that enables rapid mutational testing is droplet digital PCR (ddPCR). This assay is a modification of PCR wherein template molecules are distributed into many parallel reactions at limiting dilutions and each run to completion [7]. Using microfluidics, independent PCR reactions are suspended in aqueous droplets in oil, and sensitive optical systems can detect fluorescence from individual droplets in single or multiple channels [8, 9]. This setup provides a highly sensitive and specific system that is robust enough for use in the clinical laboratory, and such systems have been shown to be useful for a range of applications, from quantitation of DNA copy number [10] to analysis of circulating tumor DNA [11] and detection of SARS-CoV-2 [12].

Here we present our work validating a dual channel ddPCR system for EGFR L858R and exon 19 deletion detection in plasma and FFPE tissue. We describe the analytical validation for each assay, and we demonstrate that the assay has excellent performance characteristics and superior turnaround time compared to panel next generation sequencing.

Materials and methods

DNA isolation

DNA was isolated from formalin-fixed, paraffin-embedded tumor tissue. Tumor-enriched areas were macrodissected from ten 4-μm sections. DNA was isolated using QIAamp DNA mini kit (Qiagen; Hilden, Germany) and quantified using Qubit-based dsDNA detection (Qubit 3.0, Life Technologies; Carlsbad, CA). Cell-free genomic DNA (cfDNA) was isolated from plasma using the QIAamp Circulating Nucleic Acid Kit and then quantified using Qubit-based dsDNA detection. DNA was then diluted to a maximum concentration of 3 ng/μL.

Droplet digital PCR

DNA was mixed with commercially available master mix (Bio-Rad; Hercules, CA), PCR primers and fluorescently labeled probes (Applied Biosystems; Foster City, CA) for detection of the respective sequences (). A total volume of 20 μL was used in each reaction, comprised of 12.5 μL master mix, 0.625 μL of 40X Taqman probes, and 6.875 μL of RNase/DNase-free water, and 5 μL of sample DNA or control. For clinical testing, control samples were included with each run: water, a wild type control, as well as low (2% mutant DNA) and high (20% mutant DNA) controls for each mutation. Aqueous droplets containing individual PCR templates were generated in a water-oil emulsion-based droplet generator using 70 μL of oil per reaction. Using 50 μL of the resulting droplets, PCR for 40 cycles of alternating 94°C and 58°C was performed for detection of the respective sequences. These droplets were analyzed by flow cytometry (Bio-Rad QX200 ddPCR System), which detected probe-specific fluorescent signals. Flow cytometry event counts were used to detect wild type and mutant signals and quantify variant allele fraction.

Description of assay design

Independent ddPCR reactions were designed to detect the presence of EGFR c.2573T>G (L858R) and EGFR exon 19 deletion ().

For EGFR L858R detection, the assay contained a single set of primers and two competitive probes, one detecting the wild-type allele (VIC) and one detecting the mutant allele (FAM). Wild type events in the VIC channel ensured that EGFR DNA was present in the sample. Mutant events were represented as positive events in the FAM channel.

For exon 19 deletion mutations, a VIC-labeled probe hybridized to a shared reference sequence in exon 19, while a FAM-labeled probe only hybridized to the sequence spanning the deletion hotspot. Droplets with wild type DNA showed positive signal for both VIC and FAM. Droplets with deletion showed VIC signal only ().

Droplet digital PCR test for EGFR L858R for wild type DNA (A) and mutant EGFR L858R DNA at 1% variant allele fraction (B). EGFR exon 19 deletion analysis for wild type DNA (C) and mutant EGFR E746_A750del at 1% variant allele fraction (D).

Control EGFR experiment

Control EGFR L858R, exon 19 deletion (EGFR p.E746_A750del), and wild type DNA was obtained from Horizon Discovery (Horizon Discovery Ltd., Waterbeach, United Kingdom). Serial dilutions of mutant DNA were performed to generate DNA at 10%, 1%, 0.1%, 0.01%, and 0% variant allele fraction. Total EGFR DNA concentration was loaded at 250, 125, 62.5, 31.25 and 15.625 ng. Each combination of variant allele fraction and DNA concentration was run in duplicate.

DNA quantitation

DNA concentration could be calculated based on a Poisson distribution, where λ represented number of target DNA molecules and p was the ratio of positive events to all events. The principles of DNA quantitation in digital PCR systems have been previously described [13]. Both mutant DNA concentration and wild type DNA concentration could be derived to calculate the mutation variant allele fraction.

OncoPanel next generation sequencing

Orthogonal testing was performed with OncoPanel, a targeted next generation sequencing panel, as previously described [14]. Briefly, DNA was isolated from formalin-fixed, paraffin-embedded tissue with at least 20% tumor purity. Hybrid capture was performed to enrich for the coding regions of 447 cancer associated genes including EGFR using the Agilent SureSelect XT Fast Reagent Kit (Agilent Technologies Inc., Santa Clara, CA). Sequencing was performed on an Illumina HiSeq 2500 sequencer (Illumina Inc., San Diego, CA). Informatics was performed with a custom pipeline.

On validation, OncoPanel demonstrated 97.8% sensitivity and 100% specificity for single nucleotide variants and 100% sensitivity and 100% specificity for insertion and deletion variants excluding FLT3 and NPM1 variants. The limit of detection was set at 10% variant allele fraction for regions of at least 50x read depth, achieving 98.4% concordance in triplicate runs for variants that meet these criteria [14]. For clinical lung cancer specimens, EGFR mutational hotspots in exon 19 and exon 21 underwent additional manual inspection in Integrative Genomics Viewer.

Patient selection

Patient specimens tested by ddPCR and OncoPanel were identified by retrospective review of the medical record. For patients who had multiple cfDNA specimens tested by ddPCR, the result from the earliest specimen was used for analysis. This study was approved by the Mass General Brigham Human Research Committee. As the research was deemed no more than minimal risk, patient consent was waived by the ethics committee.

Results

Calculated effect of loading concentration on mutation detection

We calculated how DNA loading affected performance characteristics for the EGFR L858R and exon 19 deletion ddPCR assay designs. For the EGFR L858R assay, positive mutation events increased as DNA concentration increased (). In specimens with low variant allele fraction, the relationship of positive events increased linearly with DNA concentration.

Relationship between variant allele fraction (VAF), DNA loading concentration, and positive mutation events for EGFR L858R (A) and EGFR exon 19 deletion (B).

For the EGFR exon 19 deletion assay, VIC-only positive events increased with DNA concentration at low concentration. As the total DNA concentration increased past the optimal loading concentration, the proportion of positive events began to decrease (). As mutant variant allele fraction approached 0, the optimal loading concentration approached 1 event per droplet.

Analytical specificity

Wild type EGFR was tested across five DNA loading concentrations (250, 125, 62.5, 31.25, and 15.625 ng). Each concentration was run in duplicate for both the L858R and exon 19 deletion ddPCR assays. The 10 EGFR L858R wild type reactions generated a median of 15,282 droplets (range 7,954 to 18,141 droplets). Five reactions showed a single false positive event, and no reaction showed greater than one false positive event. The overall per droplet false positive error rate was 0.003% for the L858R reaction. In 10 EGFR exon 19 deletion reactions with wild type EGFR control DNA, no false positive mutant events were observed. Based on this observation, the assay was interpreted to be positive for EGFR mutation if 3 or more mutant events were present. With a positive threshold set at 3 events in duplicate reactions, the expected false positive rate in EGFR L858R analysis was 0.02%.

DNA quantification and limit of detection

Quantitation of DNA concentration based on Poisson approximation showed that calculated DNA concentration was linear with respect to DNA input for both the L858R and exon 19 deletion reactions (, R2 = 0.986 and R2 = 0.998, respectively).

Linear correlation between input DNA concentration and experimentally calculated concentration for EGFR L858R reaction (A) and EGFR exon 19 deletion reaction (B).

The limit of detection depended on input DNA concentration. Reactions were performed with mutant L858R or exon 19 deletion DNA at 0.01%, 0.1%, 1%, and 10% variant allele fraction across five DNA loading concentrations (250, 125, 62.5, 31.25, and 15.625 ng). Each reaction was performed in duplicate. A positive result was defined as generating at least 3 positive droplet events in both replicate reactions.

For the L858R reaction, the limit of detection was 0.01% variant allele fraction at 250 ng DNA input, 0.1% variant allele fraction at 125 and 62.5 ng input, and 1% at 31.25 and 15.625 ng input (). For the exon 19 deletion reaction, the limit of detection was 0.1% variant allele fraction at 250, 125, and 62.5 ng DNA input and 1% at 31.25 and 15.625 ng input ().

The observed difference in limit of detection between the L858R and exon 19 deletion assays was expected based on assay design principles. The analytical sensitivity decreased for the exon 19 deletion reaction at the high DNA concentration (see "Calculated Effect of Loading Concentration on Mutation Detection").

Accuracy and precision

For reactions above the limit of detection, a comparison of calculated variant allele fraction to true variant allele fraction was shown to be linear for both the L858R and exon 19 deletion reactions with R2 = 0.998 and 0.995, respectively.

Accuracy and precision of quantitation improved with increasing mutant events and variant allele fraction. For the L858R reaction, average percent error per reaction was 32.4%, 15.7%, 11.0% and 2.6% at variant allele fractions of 0.01%, 0.1%, 1%, and 10%, respectively. The coefficient of variation was 0.143, 0.172, 0.141 and 0.036.

For the exon 19 deletion reaction, average percent error per reaction was 53.7%, 11.7%, and 4.1% at variant allele fractions of 0.1%, 1%, and 10%, respectively. The coefficient of variation was 0.425, 0.131 and 0.056 ().

Experimental versus input variant allele fraction for specimens above limit of detection for EGFR L858R (A) and EGFR exon 19 deletion (B). Mean percent error decreases with increasing variant allele fraction for EGFR L858R (C) and EGFR exon 19 deletion (D). Coefficient of variation with respect to variant allele fraction for EGFR L858R (E) and EGFR exon 19 deletion (F).

Orthogonal validation and clinical sensitivity

We reviewed the application of the EGFR ddPCR assay in our clinical practice. Twenty patients had both ddPCR and OncoPanel targeted next generation sequencing performed on tissue specimens on the same pathological specimen. One cancer had EGFR L858R, five had EGFR exon 19 deletion, and 14 were wild type for EGFR. Overall concordance between ddPCR and OncoPanel in this cohort was 100%.

The EGFR ddPCR assay was also used to detect circulating cell free tumor DNA from patients’ plasma specimens. In total, 183 patients with advanced lung cancer had ddPCR performed on plasma as well as tissue evaluated by OncoPanel sequencing. Of 92 cancers with EGFR L858R or exon 19 deletion mutation detected by OncoPanel, 49 had mutations identified by ddPCR in plasma specimens. The overall clinical sensitivity of plasma cell free mutational analysis was 53%. An additional four plasma specimens from patients with EGFR-mutated lung cancers showed mutant events in replicate reactions but did not meet the criteria for reporting (at least 3 events in each replicate reaction). No false positive results were observed.

In this cohort of patients, the median age was 64 years (range 31 to 97 years) with 64 (70%) female patients. New diagnoses accounted for 28 of 92 (30%) patients with EGFR-mutated cancers with 82 (89%) being at stage IV and 89 (97%) being at stage III or IV. Most patients had received some form of treatment prior to their first plasma ddPCR assay, with 52 (57%) receiving a tyrosine kinase inhibitor, 21 (23%) receiving chemotherapy, and 36 (39%) receiving localized therapy to their primary tumor either in the form of surgery (20, 22%) or radiotherapy (16, 17%). Only 28 (30%) had received no pre-testing therapy for their primary tumor at all.

Sensitivity among patients with Stage IV disease was 59%, while for patients with Stage I-II disease, it was only 17%. Sensitivity also varied by treatment received, with sensitivity for patients receiving no treatment being 71%, for those receiving a tyrosine kinase inhibitor being 44%, and for those receiving any systemic therapy (tyrosine kinase inhibitor, chemotherapy, or checkpoint inhibitor) being 37%.

Turnaround time

Finally, we evaluated assay turnaround time, the number of days from when the specimen arrived in the laboratory to when the pathology report was generated, for both ddPCR and next generation sequencing in patients who had both assays performed. The median turnaround time for OncoPanel sequencing was 13 days (range 6 to 133 days). The median turnaround time for ddPCR was 6 days (range 1 to 10 days) for tissue specimens and 4 days (range 1 to 11 days) for plasma specimens. The difference in turnaround time between next generation sequencing and ddPCR was statistically significant (, p < 0.001).

Discussion

Here, we report the results of our validation for using a dual channel ddPCR system for detection of EGFR L858R and exon 19 deletion in FFPE and plasma. Our laboratory detects EGFR exon 19 deletions via a loss-of-signal testing strategy. Both loss-of-signal and gain-of-signal approaches have been previously described [15, 16]. The loss-of-signal approach allows for detection of a range of mutations without a priori knowledge of exact deletion breakpoints in exon 19. Concordance between NGS and ddPCR performed on the same tumor specimen was 100%, demonstrating that ddPCR is an accurate method for rapid assessment of EGFR mutations.

The low limit of detection for ddPCR makes the technology ideal for use in mutation detection in plasma that may be undetectable by other molecular genotyping methods. In our cohort, ddPCR performed on plasma cell free tumor DNA has a sensitivity of 53% with 100% specificity. The clinical sensitivity is lower compared to that reported in prior published studies which range from 63% to 82% [15-18]. The difference may reflect patient referral patterns in our practice and analytical limitations in clinical specimens. Our cohort has a slightly lower proportion of stage IV patients than most of those previously published. Additionally, more than half of our patients (56%) had already been treated with one or more TKIs at the time of testing which may reduce the ability of our assay to detect mutations, as evidenced by the markedly decreased sensitivity of our assay for patients who has received systemic therapy (37%) versus those who had not received any treatment at all (71%).

As NGS panels have become increasingly utilized for molecular characterization of advanced lung cancers, attempts have been made to find optimal testing strategies. NGS, while accurate and comprehensive, may take weeks to perform, with some patients with actionable mutations starting first-line chemotherapy before results are available [19]. Patients whose tumors harbor targetable mutations may benefit from rapid tests that are more limited in scope [20]. In our institutional experience, the median turnaround time for ddPCR is seven days faster than NGS. Early identification of actional EGFR alterations may aid in the timely selection of targeted therapy for patients who are acutely ill.

Plasma ddPCR can also be used as a tool for monitoring treatment response and tumor evolution. Historically, ddPCR has been used to monitor the EGFR T790M mutation, which confers resistance to early generation tyrosine kinase inhibitors erlotinib and gefitinib. Several studies have shown the utility of ddPCR for detecting this mutation over time [21-24]. This diagnostic testing strategy will have to be modified as osimertinib becomes a first line modality for EGFR-mutated lung cancers [25], leading to different mechanisms of therapeutic resistance [26].

Conclusions

This article describes our validation and clinical experience with a clinical ddPCR assay to test for common EGFR mutations. Experimentation confirms the analytical principles of DNA quantification by ddPCR. Clinical EGFR analysis by ddPCR is shown to be accurate with an added benefit of faster turnaround time compared to panel next generation sequencing. ddPCR is an accurate and rapid method for the detection of EGFR mutations with clinical utility for patients with advanced lung cancer.

(DOCX)

Click here for additional data file.

31 Aug 2021

PONE-D-21-21322

Detection of EGFR mutations in non-small cell lung cancer by droplet digital PCR

PLOS ONE

Dear Dr. Dong,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Oct 15 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see:  http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at  https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .

We look forward to receiving your revised manuscript.

Kind regards,

Alvaro Galli

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please provide additional details regarding participant consent. In the Methods section, please ensure that you have specified (1) whether consent was informed and (2) what type you obtained (for instance, written or verbal). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information.

3. Thank you for providing the following Funding Statement:

“The authors received no specific funding for this work.”

We note that one or more of the authors is affiliated with the funding organization, indicating the funder may have had some role in the design, data collection, analysis or preparation of your manuscript for publication; in other words, the funder played an indirect role through the participation of the co-authors.

If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study in the Author Contributions section of the online submission form. Please make any necessary amendments directly within this section of the online submission form. Please also update your Funding Statement to include the following statement: “The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

If the funding organization did have an additional role, please state and explain that role within your Funding Statement.

Please also provide an updated Competing Interests Statement declaring this commercial affiliation along with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc.

Within your Competing Interests Statement, please confirm that this commercial affiliation does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If this adherence statement is not accurate and there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared

4. Thank you for stating the following in the Competing Interests section:

“I have read the journal's policy and the authors of this manuscript have the following competing interests:

Geoffrey Oxnard is an employee of Foundation Medicine and holds equity in Roche. Lynette Sholl is a consultant for Genentech and Eli Lilly and receives research funding from Genentech.”

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

5. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: 1, the study was to generate more data to demonstrate potential value of ddPCR for clinical use in detecting tumor mutations especially with ctDNA. The manuscript is in good writen English.

2, more detailed technical information should be provided for readers to better understanding of the potential value and limitation of the ddPCR assays, including 1) brief procedure of the assays; 2) cfDNA/plasma input for the clinical validation; 3) key experimental and performance parameters of the reference NGS-panel assay, such as LOD, sensitivity, specificity.

3, there is a significant difference in terms of technical sensitivity or LOD between the ddPCR assay for L858R and that for E19 deletions, which needs an explanation or interpretation.

4, in the false positive tests, a few reactions reported single false positive event for L858R, and no false posive for E19 deletions, but the positive calling cutoff were set to be 3 positive droplets, which needs an explanation of rational. Considering the value of such assays to to detect ctDNA mutations which is at very limited concentration, maximization of clinical sensitivity with good control of specificity is important. 2 positive droplets from a single reaction or even duplicated reactions might be better.

5, there is a significant difference in terms of analytical and clinical sensitivity between the ddPCR assays and that reported previously such as ref. 15. This should be discussed with sufficiently deepth so that the readers can understand the reasons behind.

6, considering the broad application of NGS for ctDNA in clinic, it would be more informative if there is a head-to-head comparison between the ddPCR assays and a NGS assay for ctDNA detection.

Reviewer #2: The manuscript is aimed at validating a dual channel ddPCR system for EGFR L858R and exon 19 deletion

detection in plasma and FFPE tissue.

Despite the analysis have been conducted in a very rigorous way, the number of mutant samples analysed on FFPE is very low (6 out of 20) and the advantages in terms of specificity and turn around time of a ddPCR are well known.

Considering the low sensitivity (only 52%) using plasma samples, I would suggest the authors to accurately consider/discuss the clinical characteristics of patients, since this is one of the major challenge of liquid biopsy. Patients should try to set up their assay on an homogeneous population, to cut out any possible analytical bias to confirm their assay works.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Guanshan Zhu

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

28 Jan 2022

Dear reviewers,

Thank you for taking time to review our article and for your helpful comments. To comply with PLOS data availability guidelines, we have provided the minimal data set for all study results in a Supplemental File.

Please see the full response to each point of your suggestions below:

Reviewer #1:

1, the study was to generate more data to demonstrate potential value of ddPCR for clinical use in detecting tumor mutations especially with ctDNA. The manuscript is in good writen English.

Thank you for your comment.

2, more detailed technical information should be provided for readers to better understanding of the potential value and limitation of the ddPCR assays, including 1) brief procedure of the assays; 2) cfDNA/plasma input for the clinical validation; 3) key experimental and performance parameters of the reference NGS-panel assay, such as LOD, sensitivity, specificity.

We have updated methods to be more comprehensive. We have expanded the methods section to include more information from our protocol. We have specified the plasma input for cfDNA analysis. The method section for DNA isolation and ddPCR now reads as follows:

DNA Isolation

DNA was isolated from formalin-fixed, paraffin-embedded tumor tissue. Tumor-enriched areas were macrodissected from ten 4-μm sections. DNA was isolated using QIAamp DNA mini kit (Qiagen; Hilden, Germany) and quantified using Qubit-based dsDNA detection (Qubit 3.0, Life Technologies; Carlsbad, CA). Cell-free genomic DNA (cfDNA) was isolated from plasma using the QIAamp Circulating Nucleic Acid Kit and then quantified using Qubit-based dsDNA detection. DNA was then diluted to a maximum concentration of 3 ng/μL.

Droplet Digital PCR

DNA was mixed with commercially available master mix (Bio-Rad; Hercules, CA), PCR primers and fluorescently labeled probes (Applied Biosystems; Foster City, CA) for detection of the respective sequences (Table 1). A total volume of 20 μL was used in each reaction, comprised of 12.5 μL master mix, 0.625 μL of 40X Taqman probes, and 6.875 μL of RNase/DNase-free water, and 5 μL of sample DNA or control. For clinical testing, control samples were included with each run: water, a wild type control, as well as low (2% mutant DNA) and high (20% mutant DNA) controls for each mutation. Aqueous droplets containing individual PCR templates were generated in a water-oil emulsion-based droplet generator using 70 μL of oil per reaction. Using 50 μL of the resulting droplets, PCR for 40 cycles of alternating 94 °C and 58 °C was performed for detection of the respective sequences. These droplets were analyzed by flow cytometry (Bio-Rad QX200 ddPCR System), which detected probe-specific fluorescent signals. Flow cytometry event counts were used to detect wild type and mutant signals and quantify variant allele fraction.

We have added more details about the NGS panel assay validation, including LOD, sensitivity, and specificity. The following text has been added to the manuscript:

On validation, OncoPanel demonstrated 97.8% sensitivity and 100% specificity for single nucleotide variants and 100% sensitivity and 100% specificity for insertion and deletion variants excluding FLT3 and NPM1 variants. The limit of detection was set at 10% variant allele fraction for regions of at least 50x read depth, achieving 98.4% concordance in triplicate runs for variants that meet these criteria.[14]

3, there is a significant difference in terms of technical sensitivity or LOD between the ddPCR assay for L858R and that for E19 deletions, which needs an explanation or interpretation.

Thank you for your question. The difference in technical sensitivity/LOD between the L858R and E19 deletion assays is due to assay design (Figure 1 and Figure 2) and has to do with expected results as DNA concentration increases beyond limiting dilution. In the L858R assay, a droplet containing both wild type and mutant alleles will be double-positive for VIC and FAM and will contribute to the calculation for both wild type and mutant alleles. In the E19del assay, a droplet containing both wild type and mutant alleles will also be double-positive for VIC and FAM and will be indistinguishable from wild type only droplets. The end effect is that for the E19del assay, the analytical sensitivity decreases at high loading concentrations and is evident in the experimental data in Table 3.

This concept is explored in the section "Calculated Effect of Loading Concentration on Mutation Detection", including in Figure 3.

We have added the following text to the relevant section ("DNA Quantification and Limit of Detection") to help clarify this for the reader:

The observed difference in limit of detection between the L858R and exon 19 deletion assays was expected based on assay design principles. The analytical sensitivity decreased for the exon 19 deletion reaction at the highest DNA concentration (see "Calculated Effect of Loading Concentration on Mutation Detection").

4, in the false positive tests, a few reactions reported single false positive event for L858R, and no false posive for E19 deletions, but the positive calling cutoff were set to be 3 positive droplets, which needs an explanation of rational. Considering the value of such assays to to detect ctDNA mutations which is at very limited concentration, maximization of clinical sensitivity with good control of specificity is important. 2 positive droplets from a single reaction or even duplicated reactions might be better.

This is an interesting question about how to set thresholds for reporting based on data in analytical validation.

For the L858R assay, the overall per-event false positive rate is very low: 5/145019 = 0.003% of any droplet may show a false positive result. However, this risk is cumulative over 15,000 droplets per reaction. The expected number of false positive events is 15,000 x 0.003% = 0.517. This is consistent with the observed data, where 5/10 L858R reactions showed one false positive event each.

This data can be further modeled by a Poisson distribution with λ = 0.517. The probability of two or more positive event in a single reaction is 0.0955, and the probability of two replicates each having two or more positive events is 0.0955 x 0.0955 = 0.00911 (0.911%). A false positive rate of almost 1% is not acceptable for a clinical assay.

Changing the threshold from 2 to 3 events decreases the probability of false positive events (≥3) from a single reaction to 0.0157 and false positive in duplicate reactions to 0.000247 (0.02%).

For the exon 19 deletion assay, there were no false positives observed under experimental conditions. Applying a statistical rule of 3 (if an event does not occur in n observations, then the 95% confidence interval can be approximated by 3/n), changing the threshold from 2 to 3 events will decrease the false positive 95% confidence interval from 0.00304 (0.3%) to <0.001%.

You can rationalize setting the exon 19 threshold lower, but we have decided to keep it at 3 in our lab for convenience in clinical interpretation. There is value is having a single threshold for both assays to minimize risk of misinterpretation in a laboratory with many molecular pathologists rotating on service.

We have added the following line to the manuscript:

With a positive threshold set at 3 events in duplicate reactions, the expected false positive rate in EGFR L858R analysis was 0.02%.

5, there is a significant difference in terms of analytical and clinical sensitivity between the ddPCR assays and that reported previously such as ref. 15. This should be discussed with sufficiently deepth so that the readers can understand the reasons behind.

We performed additional analysis, reporting the demographic and clinical characteristics of patients in our cohort. These are addressed in detail in the response to reviewer #2. We noted that most patients in our cohort had been pretreated. Sensitivity among patients with Stage IV disease was 59% and for patients who had not received prior therapy was 71%.

During our reanalysis, we identified a data entry error. The final sensitivity for plasma ddPCR was 49/92 = 53%. It was erroneously reported as 48/93 in the prior version. All cases are shown in the minimal data set.

6, considering the broad application of NGS for ctDNA in clinic, it would be more informative if there is a head-to-head comparison between the ddPCR assays and a NGS assay for ctDNA detection.

We agree that a direct comparison between ddPCR assay and NGS assay for ctDNA detection would be informative. Unfortunately, our lab has not optimized our NGS assay to the depth of sequencing required for ctDNA detection, and this topic is beyond the scope of the current report.

Reviewer #2:

The manuscript is aimed at validating a dual channel ddPCR system for EGFR L858R and exon 19 deletion detection in plasma and FFPE tissue. Despite the analysis have been conducted in a very rigorous way, the number of mutant samples analysed on FFPE is very low (6 out of 20) and the advantages in terms of specificity and turn around time of a ddPCR are well known.

Thank you for your comment.

Considering the low sensitivity (only 52%) using plasma samples, I would suggest the authors to accurately consider/discuss the clinical characteristics of patients, since this is one of the major challenge of liquid biopsy. Patients should try to set up their assay on an homogeneous population, to cut out any possible analytical bias to confirm their assay works.

We have reviewed the clinical characteristics of our cohort and reexamined each case. During our reanalysis, we identified a data entry error. The final sensitivity for plasma ddPCR was 49/92 = 53%.

The clinical characteristics of the patients are included in the updated results and provided here:

In this cohort of patients, the median age was 64 years (range 31 to 97 years) with 64 (70%) female patients. New diagnoses accounted for 28 of 92 (30%) patients with EGFR-mutated cancers with 82 (89%) being at stage IV and 89 (97%) being at stage III or IV. Most patients had received some form of treatment prior to their first plasma ddPCR assay, with 52 (57%) receiving a tyrosine kinase inhibitor, 21 (23%) receiving chemotherapy, and 36 (39%) receiving localized therapy to their primary tumor either in the form of surgery (20, 22%) or radiotherapy (16, 17%). Only 28 (30%) had received no pre-testing therapy for their primary tumor at all.

Sensitivity among patients with Stage IV disease was 59%, while for patients with Stage I-II disease, it was only 17%. Sensitivity also varied by treatment received, with sensitivity for patients receiving no treatment being 71%, for those receiving a tyrosine kinase inhibitor being 44%, and for those receiving any systemic therapy (tyrosine kinase inhibitor, chemotherapy, or checkpoint inhibitor) being 37%.

We have also updated the discussion:

Our cohort has a slightly lower proportion of stage IV patients than most of those previously published. Additionally, more than half of our patients (56%) had already been treated with one or more tyrosine kinase inhibitors at the time of testing which may reduce the ability of our assay to detect mutations, as evidenced by the markedly decreased sensitivity of our assay for patients who has received systemic therapy (37%) versus those who had not received any treatment at all (71%).

Thank you for taking time to review our manuscript. We appreciate your input in improving the article.

Sincerely,

Fei Dong, M.D.

Brigham and Women's Hospital

Boston, MA

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

7 Feb 2022

Detection of EGFR mutations in non-small cell lung cancer by droplet digital PCR

PONE-D-21-21322R1

Dear Dr. Dong,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Alvaro Galli

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: All my previous comments have been well addressed. I have no further comments and am happy to see the manuscript to be accepted for publication on the journal.

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

15 Feb 2022

PONE-D-21-21322R1

Detection of EGFR mutations in non-small cell lung cancer by droplet digital PCR

Dear Dr. Dong:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Alvaro Galli

Academic Editor

PLOS ONE

Table 1

Primer and probe sequences for EGFR mutation detection.

	EGFR L858R	EGFR exon 19 deletion
Forward Primer	GCAGCATGTCAAGATCACAGATT	GTGAGAAAGTTAAAATTCCCGTC
Reverse Primer	CCTCCTTCTGCATGGTATTCTTTCT	CACACAGCAAAGCAGAAAC
Wild Type Probe	AGTTTGGCCAGCCCAA	ATCGAGGATTTCCTTGTTG
Mutant Probe	AGTTTGGCCCGCCCAA	AGGAATTAAGAGAAGCAACATC

Table 2

Number of positive mutant events for EGFR L858R reactions.

		Input Concentration (ng)
		15.625	31.25	62.5	125	250
Variant Allele Fraction	0%	0	1	0	1	1
		0	0	1	0	1
	0.01%	0	1	3	4	4
		0	2	1	2	6
	0.1%	0	2	11	17	39
		3	6	12	22	32
	1%	21	43	119	209	428
		31	33	88	219	330
	10%	267	592	1146	1951	2504
		249	503	422	1733	3426

Each combination of DNA input concentration and variant allele fraction is performed in duplicate. Positive results with at least three mutant events in each replicate are highlighted in gray.

Table 3

Number of positive mutant events for EGFR exon 19 deletion reactions.

		Input Concentration (ng)
		15.625	31.25	62.5	125	250
Variant Allele Fraction	0%	0	0	0	0	0
		0	0	0	0	0
	0.01%	0	0	1	1	0
		0	0	1	0	0
	0.1%	2	1	10	5	4
		2	9	5	13	9
	1%	28	36	42	72	43
		21	51	65	70	40
	10%	198	359	630	627	424
		245	412	553	643	434

Each combination of DNA input concentration and variant allele fraction is performed in duplicate. Positive results with at least three mutant events in each replicate are highlighted in gray.