Ultrasensitive Detection of Circulating Tumor DNA in Non-Small Cell Lung Cancer by Deep Sequencing: Translational Research

Abstract

Purpose/Objective(s)There are no blood-based biomarkers in routine clinical use for patients with non-small cell lung cancer (NSCLC). Circulating tumor DNA (ctDNA) represents a promising analyte for non-invasively detecting disease burden in NSCLC patients while simultaneously providing access to tumor genomes. However, existing methods for ctDNA detection lack the necessary analytic sensitivity and/or require patient-specific optimization that prevents wide clinical applicability. We therefore developed a deep sequencing-based approach for quantifying ctDNA that is specifically optimized for NSCLC and does not require patient-specific optimization.Materials/MethodsOur approach, called Cancer Personalized Profiling by Deep Sequencing (CAPP-Seq), incorporates improved technical and bioinformatic methods for ultrasensitively and specifically detecting ctDNA. We validated CAPP-Seq using spiking experiments performed with clinically relevant amounts of cell free DNA. We then applied our approach to serial plasma samples from patients with stage I through IV NSCLC who received a variety of treatments.ResultsCAPP-Seq detected at least 1 mutation in ∼95% of NSCLC patients, achieved similar analytic sensitivity to digital PCR (∼0.01%), and simultaneously detected single nucleotide variants, indels, rearrangements, and copy number alterations. Analysis of serial plasma samples from 30 patients with early and advanced NSCLC revealed that absolute concentrations of ctDNA were highly correlated with tumor volume both before and after treatment. CAPP-Seq allowed quantitation of ctDNA in stage I patients, facilitated detection of minimal residual disease, detected subclonal resistance mutations in plasma, and helped distinguish posttreatment normal tissue changes from residual cancer. Finally, we demonstrated the ability of CAPP-Seq to be used for noninvasive genotyping and biopsy-free tumor detection.ConclusionsCAPP-Seq allows highly sensitive and non-invasive detection of ctDNA in patients with NSCLC. Our approach provides a personalized biomarker for nearly every patient with NSCLC without the need for patient-specific optimization. It could therefore be routinely applied clinically and has the potential for accelerating the personalized detection, monitoring, and treatment of NSCLC. Purpose/Objective(s)There are no blood-based biomarkers in routine clinical use for patients with non-small cell lung cancer (NSCLC). Circulating tumor DNA (ctDNA) represents a promising analyte for non-invasively detecting disease burden in NSCLC patients while simultaneously providing access to tumor genomes. However, existing methods for ctDNA detection lack the necessary analytic sensitivity and/or require patient-specific optimization that prevents wide clinical applicability. We therefore developed a deep sequencing-based approach for quantifying ctDNA that is specifically optimized for NSCLC and does not require patient-specific optimization. There are no blood-based biomarkers in routine clinical use for patients with non-small cell lung cancer (NSCLC). Circulating tumor DNA (ctDNA) represents a promising analyte for non-invasively detecting disease burden in NSCLC patients while simultaneously providing access to tumor genomes. However, existing methods for ctDNA detection lack the necessary analytic sensitivity and/or require patient-specific optimization that prevents wide clinical applicability. We therefore developed a deep sequencing-based approach for quantifying ctDNA that is specifically optimized for NSCLC and does not require patient-specific optimization. Materials/MethodsOur approach, called Cancer Personalized Profiling by Deep Sequencing (CAPP-Seq), incorporates improved technical and bioinformatic methods for ultrasensitively and specifically detecting ctDNA. We validated CAPP-Seq using spiking experiments performed with clinically relevant amounts of cell free DNA. We then applied our approach to serial plasma samples from patients with stage I through IV NSCLC who received a variety of treatments. Our approach, called Cancer Personalized Profiling by Deep Sequencing (CAPP-Seq), incorporates improved technical and bioinformatic methods for ultrasensitively and specifically detecting ctDNA. We validated CAPP-Seq using spiking experiments performed with clinically relevant amounts of cell free DNA. We then applied our approach to serial plasma samples from patients with stage I through IV NSCLC who received a variety of treatments. ResultsCAPP-Seq detected at least 1 mutation in ∼95% of NSCLC patients, achieved similar analytic sensitivity to digital PCR (∼0.01%), and simultaneously detected single nucleotide variants, indels, rearrangements, and copy number alterations. Analysis of serial plasma samples from 30 patients with early and advanced NSCLC revealed that absolute concentrations of ctDNA were highly correlated with tumor volume both before and after treatment. CAPP-Seq allowed quantitation of ctDNA in stage I patients, facilitated detection of minimal residual disease, detected subclonal resistance mutations in plasma, and helped distinguish posttreatment normal tissue changes from residual cancer. Finally, we demonstrated the ability of CAPP-Seq to be used for noninvasive genotyping and biopsy-free tumor detection. CAPP-Seq detected at least 1 mutation in ∼95% of NSCLC patients, achieved similar analytic sensitivity to digital PCR (∼0.01%), and simultaneously detected single nucleotide variants, indels, rearrangements, and copy number alterations. Analysis of serial plasma samples from 30 patients with early and advanced NSCLC revealed that absolute concentrations of ctDNA were highly correlated with tumor volume both before and after treatment. CAPP-Seq allowed quantitation of ctDNA in stage I patients, facilitated detection of minimal residual disease, detected subclonal resistance mutations in plasma, and helped distinguish posttreatment normal tissue changes from residual cancer. Finally, we demonstrated the ability of CAPP-Seq to be used for noninvasive genotyping and biopsy-free tumor detection. ConclusionsCAPP-Seq allows highly sensitive and non-invasive detection of ctDNA in patients with NSCLC. Our approach provides a personalized biomarker for nearly every patient with NSCLC without the need for patient-specific optimization. It could therefore be routinely applied clinically and has the potential for accelerating the personalized detection, monitoring, and treatment of NSCLC. CAPP-Seq allows highly sensitive and non-invasive detection of ctDNA in patients with NSCLC. Our approach provides a personalized biomarker for nearly every patient with NSCLC without the need for patient-specific optimization. It could therefore be routinely applied clinically and has the potential for accelerating the personalized detection, monitoring, and treatment of NSCLC.