Search

Objectives To identify genomic variations in cell-free DNA (cfDNA) and evaluate their clinical utility in small cell lung cancer (SCLC). Materials and methods We performed whole genome sequencing using plasma cfDNAs derived from 24 SCLC patients for copy number variation (CNV) analysis, and targeted sequencing using 17 pairs of plasma cfDNA and their matched gDNA for mutation analysis. We defined somatic mutations by comparing cfDNA to its matched gDNA with 5% variant alleles as the cutoff for mutation calls. We applied Kaplan-Meier to correlate the genomic alterations with overall survival (OS) and progression-free survival (PFS). Results We observed widespread somatic copy-number alterations and mutations, including amplification of MYC at 8q24, FGF10 at 5p13, SOX2 at 3q26 and FGFR1 at 8p12, as well as deletion of TP53 at 17p13, RASSF1 at 3p21.3, RB1 at 13q14.2, FHIT at 3p14, and PTEN at 10q23. The most frequent mutations were genes involved in chromatin regulation ( KMT2D , ARID1A , SETBP1 and PBRM1) , PI3K/MTOR pathway( MTOR , PIK13G ), Notch1 signalling pathway ( NOTCH1 ), and DNA repair related gene ATRX . Kaplan-Meier analysis revealed poor OS and PFS in patients with somatic mutations in gene SETBP1 (P = 0.0061/0.0264, HR = 4.785/3.841, 95% CI = 2.014–28.25/1.286–16.58) and PBRM1 (P = 0.0276/0.0286, HR = 3.532/3.506, 95% CI = 1.275 to 25.34/1.26–24.87). Poor OS was also associated with somatic mutations in ATRX (P = 0.0099, HR = 4.024, 95% CI = 1.926–42.95), EP300 (P = 0.025/0.0622, HR = 3.382/2.891, 95% CI = 1.448–27.76/1.013-17.29), while poor PFS was associated with ATM mutation (P = 0.0038, HR = 4.604, 95% CI = 2.211–40.93). The mutation index produced by summing up the number of mutations in the five genes was significantly associated with the poor OS/PFS (P = 0.0185/0.0294) after adjusting the effect of the stage. Conclusions Our result supports blood plasma as a promising sample source for the genomic analysis in SCLC patients whose tumor tissues are scarcely available and demonstrates potential clinical utilities of cfDNA-based liquid biopsy for clinical management of this deadly disease. [ABSTRACT FROM AUTHOR]

Objectives Cell free tumor DNA (cfDNA) circulating in blood has a great potential as biomarker for cancer clinical management. The objective of this study is to evaluate if cfDNA in blood plasma is detectable in early stage lung cancer patients. Materials and methods We extracted cfDNAs and tumor tissue DNAs from 8 lung adenocarcinoma patients. We also extracted cfDNAs from 8 normal controls. To evaluate copy number variations (CNV) and identify potential mutations, we performed low pass whole genome sequencing and targeted sequencing of 50 cancer genes. To accurately reflect the tumor-associated genomic abnormality burden in plasma, we developed a new scoring algorithm, plasma genomic abnormality (PGA) score, by summarizing absolute log2 ratios in most variable genomic regions. We performed digital PCR and allele-specific PCR to validate mutations detected by targeted sequencing. Results and conclusions The median yield of cfDNA in 400ul plasma was 4.9ng (range 2.25–26.98 ng) in patients and 2.32 ng (range 1.30–2.81 ng) in controls ( p = 0.003). The whole genome sequencing generated approximately 20 million mappable sequence reads per subject and 5303 read counts per 1 Mb genomic region. Log2 ratio-based CNV analysis showed significant chromosomal abnormality in cancer tissue DNAs and subtle but detectable differences in cfDNAs between patients and controls. Genomic abnormality analysis showed that median PGA score was 9.28 (7.38–11.08) in the 8 controls and 19.50 (5.89–64.47) in the 8 patients ( p = 0.01). Targeted deep sequencing in tumor tissues derived from the 8 patients identified 14 mutations in 12 different genes. The PCR-based assay confirmed 3 of 6 selected mutations in cfDNAs. These results demonstrated that the PGA score and cfDNA mutational analysis could be useful tool for the early detection of lung cancer. These blood-based genomic and genetic assays are noninvasive and may sensitively distinguish early stage disease when combined with other existing screening strategies including low-dose CT scanning. [ABSTRACT FROM AUTHOR]