Single‐tube two‐pronged approach using both cell‐free DNA and RNA for multimodal biomarker tests at the time of biopsy

The emergence of noninvasive liquid biopsy procedures as an alternative to surgical biopsies has fueled an intensive research effort and investment into the detection of cell‐free disease biomarkers. By pairing next‐generation sequencing (NGS) and RT‐qPCR technologies on cell‐free DNA (cfDNA) and RNA (cfRNA), respectively, we have established and validated a high‐throughput and quick‐turnaround workflow for highly sensitive detection of genomic alterations using a single tube of blood. This emphasizes the importance of our cfDNA/cfRNA recovery and enrichment knowhow to maximize the chance to capture relevant rare biomarkers. On the cfDNA‐NGS axis, sensitivity and specificity of 95% and 98% were recorded for somatic variants of ≥1% VAF, with high concordance observed between the in‐house and orthogonal assays. Overall accuracy for the cfRNA and RT‐qPCR portion of the workflow was >95%. Real‐world data of patients underwent the two‐pronged testing scheme revealed dramatic beneficial outcomes guided by genomic‐matched treatments. When implemented at the time of biopsy to expedite treatment decision‐making in the earliest possible way, the single‐tube two‐pronged approach can effectively reduce time to treatment initiation by about 70%. The cfDNA–cfRNA paired test is suitable for implementation in routine clinical decision‐making to maximize the benefit without delay.

alterations is currently suffering from poor extraction and recovery efficiency of low abundance and fragmented cfDNA and cfRNA, 6,7 requiring large sample input (usually more than 20 mL of blood) just to compensate the sample loss during extraction. 8Today, it is impractical to co-extract cfDNA and cfRNA from the same tube of blood, most clinical laboratories thus offered NGS-based large gene panel utilizing only cfDNA for detection of both mutations and fusions.
While a cornerstone of liquid biopsy component, it appears evident today that the mutation detection via cfDNA alone cannot provide the full genotypic information to meet increasing clinical demand.
Let us first recall the stakes.Circulating cfDNA is a current mainstream in the liquid biopsy field.With the FDA has already approved PCR-and NGS-based IVD to detect genetic alterations in plasma cfDNA that matched to specific molecularly targeted therapies for cancer, 9 and the evidence-based recommendations from International Association for the Study of Lung Cancer (IASLC), 10 it marked a tipping point for the widespread use of cfDNA tests in the clinic, and mostly in patients with advanced-stage cancer.What is more, the cfDNA-only testing faced several challenges: low abundance (only two copies per cell) and the dominance of noncoding sequences ($98% of total DNA), information limited to tumor cells (missing valuable pathophysiological insight such as changes in tumor microenvironment, tumor-immune cell interactions, and blood vessel epithelium functions, etc.).In this regard, in addition to tumor-informed cfDNA tests, cfRNA-based diagnostics should be implemented in clinic whenever possible to maximize benefits of cancer patients.
Although NGS technology is superior for the comprehensive assessment of various genetic alterations, it is not without limitations; lower sensitivity in detecting fusions and difficulty to detect and interpret low-frequency somatic mutations (<0.5%) continue to plague NGS approach.Deployment of high-sensitivity qPCR on cfRNA to detect actionable fusions and to quantitate gene expression levels (e.g., PD-L1 for immunotherapy) have the potential to overcome NGS limitations.Quantitative PCR methods can significantly reduce cost, turnaround time while increasing assay sensitivity and throughput, which in turn enhances our ability to detect novel rare transcripts, novel alternative splice isoforms and direct measurement of transcript abundance.Among hundred thousands of genomic aberrations present in cancers, only a small number contributes to the development and progression of the disease.Separating these so-called driver mutations from those that are inconsequential in the diseasepassenger mutations-is vital to filter-in and detect clinically relevant biomarkers for targeted therapy.Currently, roughly 70% of cancers have known, detectable driver mutations.Quantitative PCR is a fitfor-purpose multiplex platform for detecting known driver mutations and fusions at higher sensitivity.Together, the two-pronged approach combining NGS and qPCR-based platforms on cfDNA and cfRNA, respectively, provides an unprecedented opportunity in a clinical setting to enable sensitive and accurate detection of full spectrum of actionable and informative genomic alterations all from a simple blood draw, and now possibly from a single tube of blood.
We have established a streamlined and standardized NGS and qPCR testing workflow based on cfDNA and cfRNA sample inputs to bridge the cfDNA gap in liquid biopsy.The goal of this study was to demonstrate the great advantage of integrating dual cfDNA-cfRNA workflow in term of sensitivity, throughput, cost and clinical outcome in a CLIA/CAP clinical laboratory setting.Our efforts will contribute to the validation and decision of clinical liquid biopsy applications.
Recommendations to circumvent the challenges and improve the protocol for future genomic and transcriptomic profiling research will also be documented, which we hope will aid efforts to identify mutation as well as gene expression signatures that could be used to guide patient treatments and predict patient treatment responses.

| Patients and samples
Peripheral blood samples from patients with various types of cancer were collected between October 2019 and November 2021 from multiple clinics in California.Healthy donor samples were collected from volunteers without a history of cancer.All participants provided signed informed consent.The information of clinicopathological characteristics of these patients was obtained from chart review.Studies involving human subjects were performed under IRB approved protocol.Ten milliliter of whole blood was collected into Cell-Free DNA Blood Collection Tube ® (Streck, La Vista, NE, USA), transported and stored at ambient temperature for no longer than 7-days according to the manufacturer's instruction. 11

| Cell-free DNA/RNA extraction
Plasma was prepared from whole blood via a two-step centrifugation method as described previously. 12Plasma samples were stored at À80 C until batch processing.DNA extraction was performed using the QIAamp Circulating Nucleic Acid Kit (Qiagen, Hilden, Germany) using 4 mL of plasma.In parallel, RNA was extracted from 500 μL of plasma by MagMAX™ Viral RNA Isolation Kit (ThermoFisher Scientific, Waltham, MA, USA).Nucleic acid concentration was quantified using the Qubit dsDNA and RNA High-Sensitivity Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA).Quality control was assessed on Agilent's 4200 Tape Station system using Genomic DNA and High Sensitivity RNA Screen Tape (Agilent, Santa Clara, CA, USA) to obtain average fragment length and DNA/RNA integrity numbers.Contrived samples with known SNV, CNV or fusions at know VAF were purchased from SeraCare Life Sciences (Milford, MA, USA) and Horizon Discovery (Waterbeach, UK).

| Panel design
Cancer associated and actionable genes were selected based on publicly available mutation databases.Eighty-eight frequently mutated genes of clinical relevance for various cancer were selected to increase the likelihood of detecting an aberration while also keeping manageable genomic size of the panel.The 68-gene panel was designed using the SureDesign software (Agilent).A pool of 7260 120-mer biotinylated RNA oligonucleotide "baits" was designed to target the coding regions of the 68 genes, along with 1542 single nucleotide polymorphism (SNP) loci.The SNP targets were evenly distributed across the genome to improve whole-chromosomal coverage for copy number event identification, as well as serving as a patientspecific DNA "fingerprint" to allow confirmation of individuals across longitudinal sampling.Exonic regions had 2Â bait tiling, while SNP regions had 1Â, resulting in a total library size of $850 kb.

| Next generation sequencing
Thirty nanograms of cfDNA were used as input for library preparation using Twist Library Preparation EF Kit 2.0 (Twist Bioscience, San Francisco, USA).Briefly, samples were end repaired, A-tailed, and adapters were added to the ends by ligation.Both adapter-ligated DNA fragments were amplified by index PCR using DNA-specific index primers (using unique and combinatorial index primer mixes) to add index sequences for sample multiplexing.The samples were run through two rounds of hybridization and capture to bind targeted regions and optimally enrich the libraries.The enriched libraries were then subjected to PCR amplification, cleanup with sample purification beads, and quantification by Qubit and analyzed on Tape station using D1000 Screen Tape.Bead-based normalization of the enriched libraries was performed to ensure a uniform library representation in the pooled libraries before sequencing.Sequencing of the pooled libraries was performed on NextSeq (Illumina, San Diego, IL, USA) on a highoutput kit (300 cycles) following manufacturer's instructions.

| Bioinformatic pipelines
All genomic positions and variants denoted in this study use the Genome Reference Consortium Human Build 38 (GRCh38) hg19 coordinates.Fastq files were generated from the NextSeq run using bcl2fastq2 Conversion Software v2.20 set to trim adapters, and to include UMIs in the read header.4][15] Multisample vcf (variant call format) files were then split into individual sample vcf files.These variant and bam files were lifted from hg38 coordinates to hg19 coordinates to enable variant comparisons.Variants had to be called by two out of three variant callers in order to be considered "true" variants.
Somatic variant identification was performed in paired sample mode using standard filter settings, with the exception of minimum variant frequency (reduced to 0.001%).Somatic and germline variants (small indels and SNV) were filtered and identified from paired vcf files.
Only missense variants with a Phred score ≥ 20 and a minimum read depth of 5000Â and at least three supporting consensus reads were considered for further manual curation.Manual curation was performed by visualizing VCF files using Integrative Genomics Viewer (Broad Institute, MA, USA) in order to remove any remaining sequencing artifacts.
Copy number variants (CNVs) were identified using VarSeq CNV caller.Coverage log ratios were calculated against a median reference generated using germline SNPs that underwent the same library preparation and sequencing process.Only references with an overall divergence from the cfDNA sample of <30% were used for CNV calling.
Copy number gains were required to have a p-value of <0.0001, a z-score of >0.25, and a minimum target depth of 50Â with a cutoff value of 1.9.

| RT-qPCR analyses
Plasma cell-free RNA (cfRNA) was reverse transcribed to doublestranded cDNA using ProtoScript ® II First Strand cDNA Synthesis Kit and NEBNext ® Ultra™ II Non-Directional RNA Second Strand Synthesis Module (New England Biolabs, Ipswich, MA, USA).The RNA fusion analysis of all samples was carried out by using TaqMan Gene Fusion Assays (ThermoFisher Scientific, Waltham, MA, USA) according to the manufacturer's protocols.The fusion detection real-time PCR assays can detect 41 clinically relevant fusions in genes of ALK, ROS1, NTRK1/2/3 and RET in a single run.

| Statistical analysis
For PFS and OS, the survival curves were estimated using Kaplan-Meier method and differences between groups were tested using the log-rank test.The univariable Cox proportional hazard model was performed to estimate the hazard ratio for each 1-unit increase in the Heng risk factors and other risk factors.p < 0.05 was considered statistically significant (two-sided).Graphic analysis was performed using GraphPad Prism 8.0 software (GraphPad Software, Boston, MA, USA).

| RESULTS
The single-tube two-pronged testing workflow is illustrated in Figure 1.The assay applied a dual approach coupling two sample inputs on two platforms to interrogate various genomic alterations at high accuracy with faster turnaround (5 days).Following cfDNA and cfRNA extraction, the samples were subjected to NGS and RT-qPCR analysis, respectively.The result endpoints were integrated and summarized on a customized report with the detection of genomic alterations tailored to specific FDA-approved targeted therapies and clinical trials.This combined and streamlined workflow has shown superior advantage over NGS-only process in terms of sensitivity, labor, cost, and time, and has over 5000 cancer patients worldwide benefited from our service since 2019.
The NGS-based 68-gene targeted panel test was developed for the detection of somatic alterations in cfDNA samples.DNA was sequenced to detect variants in the full exonic coding region of 68 genes covering single nucleotide variant (SNV) and insertions/ deletions (Indels), as well as copy number variation (CNV) in 10 genes.
In parallel, plasma cfRNA was extracted, enriched and used to detect ALK, ROS1, NTRK1/2/3 and RET fusions by RT-qPCR method.Our two-pronged testing approaches were performed as a CLIA/CAP laboratory developed test using cfDNA and cfRNA extracted from a single tube of plasma sample.The assay employed two automatic nucleic acid extraction workflows for higher throughput and production capacity.Using the Illumina NextSeq platform, hybrid captureselected libraries were amplified and sequenced to high uniform depth (targeting >10 000Â median coverage with >90% of exons at coverage >5000Â), and the sequence data was analyzed to detect genomic variants and signatures.The integrated personalized report is intended to provide accurate, actionable and the earliest possible information in accordance with medical guidelines for management of cancer patients.
Performance characteristics were established using cfDNA and cfRNA derived from a wide range of contrived and clinical specimens harboring variants with strong and potential clinical significance (tier 1 hotspot) and those resided in nonhotspot regions (Table 1).For genomic profiling, each performance study included representative variant types from each alteration class (SNV, Indels, CNV and fusions/splice variants), in various genomic contexts across a broad selection of genes.Sequencing analytical studies were conducted to confirm key standard measurements including positive percent agreement (PPA) and negative percent agreement (NPA).
Analytical PPA and NPA for the 68-gene NGS panel with SNV and Indels of ≥1% and ≥5% variant allele frequency (VAF) were 95% and 98%, and 100% and 100%, respectively.In CNV analysis, PPA and NPA reached 100% and 97% if the cutoff value was set at 1.4; when using a higher stringent cutoff of ≥1.9, the test PPA was 93% and NPA was 100%.Plasma cfRNA-based RT-qPCR assays showed PPA of >95% and NPA of >99% for ALK, ROS1, NTRK1/2/3 and RET fusions (Table 1).Overall, both cfDNA/NGS and cfRNA/RT-qPCR platforms showed excellent test performance which was required for clinical applications.There is an urgent clinical need to further improve current time to treatment initiation (TTI) by accelerating biomarker test reporting while increasing uptake of biomarker testing for all cancer patients.
The critical steps require us, as a diagnostic laboratory, to address sample acquisition, handling, testing and reporting.We proposed here that biomarker testing should be initiated even before a pathological diagnosis of cancer is confirmed (Figure 3).Empowering surgical oncologists to initiate biomarker testing via cfDNA and cfRNA with a simple blood draw at time of biopsy (i.e., "Plasma First") bypasses the Although digital PCR (dPCR), the modern form of PCR, might excel in sensitivity, qPCR is still the best choice for many other applications.We recommend to use RT-qPCR in cfRNA gene fusion assays because of its wide dynamic range; its ability to quantify different expression levels, discriminate splice variants and multiplex; and its capability for high-throughput applications and automation.Indeed, it is better known in laboratories and in a regulatory context, and widely used in screening for genes, health conditions, or infectious diseases.
With constant innovation and its widespread use, qPCR applications will only broaden.but not in tissue samples could be attributed to spatial and temporal tumor heterogeneity, for example, clonal evolution, especially older tissue biopsy material. 16r study also demonstrated the effectiveness of simultaneous cfDNA-cfRNA multimodal biomarker test at the time of biopsy in a clinical laboratory setting and implemented under real world conditions.The results of this study showed that the median TTI was significantly reduced $70% with deployment of this two-pronged approach at the time of biopsy compared to the current tissue biopsy or postdiagnosis blood biopsy procedure (10 days vs. 30-40 days).The reduction in TTI was due to an elimination in the time to acquire tumor tissue and pathology report, and implementation of the "plasma first" workup strategy.The possibility of rapidly reflexing to tissuebased analysis is reassuring, notably when there is no actionable alteration detected from plasma.

| CONCLUSIONS
Our study here showed the feasibility of circulating cfDNA-cfRNA analyses as a tumor biopsy surrogate in clinical practice.The combination of NGS-driven cfDNA technology and RT-qPCR-based cfRNA assays is suitable for implementation in routine clinical decision-making.NGS on cfDNA allowed for broad, deep, highly specific analysis of genomic regions of interest, which can be further enhanced by the cfRNA-based RT-qPCR as a perfect diagnostic pair for the selection of patients who could derive benefit from targeted and immune checkpoint inhibitor therapies.Even more, this single-tube two-pronged approach could potentially be applied to other stages of cancer progression such as early detection, treatment follow-up and monitoring of minimal residual disease for recurrence.

F
I G U R E 1 Workflow for the two-pronged parallel cfDNA and cfRNA approach for multimodal biomarker profiling.T A B L E 1 Key performance characteristics of the 68-gene cell-free DNA (cfDNA) next-generation sequencing (NGS) panel and cell-free RNA (cfRNA)-based RT-qPCR multiplex tests.

F I G U R E 2
Standard curves for absolute quantification of ALK, ROS1, and NTRK1/3 fusions.Linear regression analysis revealed the assay performance has good linearity, reproducibility, and amplification efficiency, making these multiplex RT-qPCR assays quantitatively robust.number.This approach has been used extensively in clinical situations for diagnostic detection of infectious diseases such as COVID-19.Absolute quantification of ALK, ROS1, NTRK1/2/3 and RET fusions were determined by constructing corresponding standard curves.We applied plasmid vectors which carried single copy of each gene, one plasmid molecule thus represented one copy of each target.The log of each known copy number in the dilution series was plotted against the Ct value to produce standard curves for each biomarker as shown in Figure2.The standard curves showed excellent amplification efficiency, linear dynamic range, and reproducibility of the assays, demonstrating these multiplex RT-qPCR tests' robustness.Limit of detection for NTRK1, NTRK3, ALK and ROS1 and assays were determined to be 1.0 copy, 11.2 copies, 17.5 copies and 28.0 copies, respectively.In routine diagnostic workflow, our cfRNA-derived RT-qPCR assays were highly automated, highly multiplexed and high throughput incorporating pre-loaded 384-well plate format (1536-well plate if needed), making them the most cost-, time-and labor-efficient platforms.
time delay for diagnosis, oncology consultation, tissue acquisition and pathology report, thus significantly shortening $70% of TTI from 30 to 40 days down to around 10 days.Awaiting tissue-based biomarker test results can significantly increase the time to treatment decisions for cancer patients and can result in unnecessary initiation of chemotherapy with worse outcomes.As newer targeted therapies with corresponding predictive biomarkers become available, knowledge translation in this area has demonstrated vital in specialist understanding about blood-based molecular testing.Overall, collaboration and frequent communication between oncologists, pathologists and laboratories are essential to successful and timely biomarker testing.Clinical validation of the paired cfDNA-cfRNA testing entailed an analysis of clinical outcomes from real-world cancer patients.Two representative cases were shown here.Figure 4 illustrated a CT scan for the change of target lesions in one patient with advanced NSCLC who was 68-gene NGS panel tested positive on EGFR mutation before and after EGFR TKI (tyrosine kinase inhibitor) administration.The cfDNA-informed test result was able to guide and expedite treatment decision leading to drastic improvement in clinical outcome and the lung mass, fluid, lung collapse declined significantly following 14-day of targeted therapy.The evaluation of tumor volume response by conventional imaging techniques using RECIST (Response Evaluation Criteria in Solid Tumors) has its limitations in the detection of early therapy response, especially in the case of targeted treatment.PET/CT provides rapid, noninvasive, in vivo assessment and quantification of glucose metabolism and could be a powerful tool for measurement of treatment response.Changes in tumor glucose metabolism precede changes in tumor size and can possibly reflect drug effects at a cellular level, resulting in a potential advantage over morphologic imaging.Shown in Figure 5 is the PET/CT image of a NSCLC patient who was tested ROS1 fusion positive by cfRNA RT-qPCR (but negative by FFPE-based ROS1 test) before and after receiving crizotinib.Following ROS1-targeted treatment, tumor size diminished dramatically to undetectable range as a result of tumor cell death and loss, indicating a "complete" patient response to the therapy.Furthermore, these data confirmed the superior sensitivity of cfRNA-based RT-qPCR assay over tissue-and NGS-based tests.Collectively, the present study highlighted real-world evidence regarding the accuracy and clinical utility of the paired cfDNA-cfRNA platforms as the first-choice patient eligibility test for targeted medicine.4 | DISCUSSION The performance of the 68-gene cfDNA NGS panel coupled with cfRNA-derived RT-qPCR tests has demonstrated its advantage over other cfDNA-only NGS genomic testing platforms.First, since a F I G U R E 3 Comparison between tissue-only diagnosis and treatment scheme (top) and high-sensitivity dual cfDNA-cfRNA profiling at the time of biopsy (bottom).Plasma-first testing strategy demonstrated $70% reduction in time to treatment initiation.singlegene can encode hundreds of copies of transcripts compared to only two copies of DNA sequences in each cell, it provides us with much higher chance for the detection of cfRNA over cfDNA in a given blood sample, thereby significantly increasing detection sensitivity and reducing false negativity.Second, the RT-qPCR platform enabled higher sensitivity in detecting actionable fusions such as ALK, ROS1, NTRK1/2/3 and RET, an analytic sensitivity that is multiple folds greater than cfDNA-only NGS while maintaining near-perfect specificity.In the case of NTRK1, our RT-qPCR technique even reached single molecule sensitivity.All of these outstanding outcomes were attributed to our proprietary "cfRNA pool multiplier" technology and the fit-for-purpose design of primer/ probe sets using "specific" information derived from the secondary/tertiary structures as well as the degradation patterns of target transcripts.Third, there are $98% noncoding, repetitive sequences existed in every cfDNA sample (i.e., only $2% of the genome are gene-rich regions), further compounding the difficulty in sequencing cfDNA for those informative and actionable fusions at low VAF.Finally, sequencing tumor-specific cfRNA, while necessary to discern unknown fusions and splicing variants that arise, is costly and can take several weeks, whereas actionable fusions and splicing variants can be successfully identified by RT-qPCR in a couple of days for a fraction of the cost of NGS.Adding to the complexity of sequencing, only a fraction of the clinical samples have sufficiently detectible RNA for quality results.Additionally, sequencing runs are performed in batches, sometimes delaying timely reporting of results.The RT-qPCR approach, which uses minimal hardware and software, can complement and augment comprehensive NGS analysis, helping focus sequencing efforts on cfDNA samples representing actionable and informative variants.
The 68-gene NGS panel covers $847 200 base pairs (850 kb), providing complete coverage for all exons in 68 genes and the critical exons in 10 genes.The current NGS data revealed high concordance with 20 contrived samples (100% at 5% VAF; and 97% at 2% VAF)and with an orthogonal method on 39 clinical samples (99%), establishing an accurate, automated and accessible cfDNA testing platform for broad clinical applications.Collectively, the cfRNA-derived test sensitivity and specificity, and the combined breadth and depth of SNV, indels and CNV from NGS, together with the minimal requirement of sample volume and fast turnaround are the primary F I G U R E 4 Targeted therapy of a NSCLC patient who was tested positive for actionable EGFR mutation by cfDNA NGS receiving TKI demonstrated a significantly improved clinical outcome (A: before; B: after).F I G U R E 5 Targeted therapy of a NSCLC patient who was tested positive for ROS1 fusion by cfRNA RT-qPCR receiving crizotinib displayed a dramatic improvement (A: before; B: after).Most importantly, FFPE-based ROS1 fusion result was negative in this patient thus missing the golden opportunity for treatment.differentiators of our two-pronged approach versus other cfDNAalone NGS methodologies.The cfDNA/NGS and cfRNA/RT-qPCR dual pairs could theoretically be applied and adapted to different scenarios depending on required sensitivity, such as mutation and gene expression signatures for early detection, minimal residual disease, and cell and gene therapy.A significant paradigm shift toward precision and personalized medicine is occurring in cancer management.The hallmark is to identify actionable driver mutations from individual patient and match to targeted therapies that can improve patient outcomes while decreasing exposure to collateral toxicity.ROS1 fusion is the well-established actionable biomarkers patients with NSCLC.In this study, the clinical sensitivity for plasma-based cfRNA via RT-qPCR was demonstrated to be greater than that of tissue-based DNA NGS sequencing in the detection of ROS1 fusion.The false negative result with the gold standard tissue gDNA NGS in the advanced NSCLC patient highlighted an important clinical limitation: a negative finding in tumor tissue did not rule out the presence of a potential genomic alteration.Overall, our cfRNA RT-qPCR method displayed high analytic sensitivity and thus could fulfill the gap in tissue biopsy.The genomic profiling analysis solely by tumor tissue is imperfect, presenting a molecular snapshot of a small tumor section rather than a complete representation of genetic alterations in a patient.It has been well documented that the discrepancy between genomic alterations detected in plasma Test accuracy of the 68-gene next-generation sequencing panel.
Test accuracy study was performed using commercially available reference standards.Of the 20 contrived samples from SeraCare with known SNV, Indels and CNV at various allele frequencies, the NGS panel demonstrated overall concordance of 97% at 2% VAF with 10 ng DNA input.The test accuracy increased to 100% for all alterations at or above 5% VAF using 10 ng of DNA (Table2).The detection of genomic variants by our 68-gene panel was further compared to results of another validated NGS assay, the SureSelect Agilent All-In-One Solid Tumor 98-Gene Panel, to assess concordance with orthogonal method (Table2).Of the 39 clinical samples tested by both assays, the concordance was 99% across all variant types and well above the set acceptance criteria of 90%.Overall, our 68-gene NGS panel enables for broad, rapid, highly specific analysis of genomic regions of interest, and these data support the use of the 68-gene panel as a biomarker test to guide treatment selection and follow-ups.Quantitative PCR is a high-sensitivity and commodity technique in clinical diagnostics to quantify the amount of target gene by introducing fluorescent or intercalating dyes to detect PCR product as it accumulates in real time during PCR cycles.Particularly, it can be used for absolute quantification of RNA transcript down to single copy T A B L E 2