Evaluating the analytical validity of circulating tumor DNA sequencing assays for precision oncology.

Circulating tumor DNA (ctDNA) sequencing is being rapidly adopted in precision oncology, but the accuracy, sensitivity and reproducibility of ctDNA assays is poorly understood. Here we report the findings of a multi-site, cross-platform evaluation of the analytical performance of five industry-leading ctDNA assays. We evaluated each stage of the ctDNA sequencing workflow with simulations, synthetic DNA spike-in experiments and proficiency testing on standardized, cell-line-derived reference samples. Above 0.5% variant allele frequency, ctDNA mutations were detected with high sensitivity, precision and reproducibility by all five assays, whereas, below this limit, detection became unreliable and varied widely between assays, especially when input material was limited. Missed mutations (false negatives) were more common than erroneous candidates (false positives), indicating that the reliable sampling of rare ctDNA fragments is the key challenge for ctDNA assays. This comprehensive evaluation of the analytical performance of ctDNA assays serves to inform best practice guidelines and provides a resource for precision oncology.

Unique, dual-indexed sequencing adapters with UMIs effectively eliminate index cross-talk and significantly improve sensitivity of massively parallel sequencing.

BACKGROUND: Sample index cross-talk can result in false positive calls when massively parallel sequencing (MPS) is used for sensitive applications such as low-frequency somatic variant discovery, ancient DNA investigations, microbial detection in human samples, or circulating cell-free tumor DNA (ctDNA) variant detection. Therefore, the limit-of-detection of an MPS assay is directly related to the degree of index cross-talk. RESULTS: Cross-talk rates up to 0.29% were observed when using standard, combinatorial adapters, resulting in 110,180 (0.1% cross-talk rate) or 1,121,074 (0.29% cross-talk rate) misassigned reads per lane in non-patterned and patterned Illumina flow cells, respectively. Here, we demonstrate that using unique, dual-matched indexed adapters dramatically reduces index cross-talk to ≤1 misassigned reads per flow cell lane. While the current study was performed using dual-matched indices, using unique, dual-unrelated indices would also be an effective alternative. CONCLUSIONS: For sensitive downstream analyses, the use of combinatorial indices for multiplexed hybrid capture and sequencing is inappropriate, as it results in an unacceptable number of misassigned reads. Cross-talk can be virtually eliminated using dual-matched indexed adapters. These results suggest that use of such adapters is critical to reduce false positive rates in assays that aim to identify low allele frequency events, and strongly indicate that dual-matched adapters be implemented for all sensitive MPS applications.

Haplotyping germline and cancer genomes with high-throughput linked-read sequencing.

Haplotyping of human chromosomes is a prerequisite for cataloguing the full repertoire of genetic variation. We present a microfluidics-based, linked-read sequencing technology that can phase and haplotype germline and cancer genomes using nanograms of input DNA. This high-throughput platform prepares barcoded libraries for short-read sequencing and computationally reconstructs long-range haplotype and structural variant information. We generate haplotype blocks in a nuclear trio that are concordant with expected inheritance patterns and phase a set of structural variants. We also resolve the structure of the EML4-ALK gene fusion in the NCI-H2228 cancer cell line using phased exome sequencing. Finally, we assign genetic aberrations to specific megabase-scale haplotypes generated from whole-genome sequencing of a primary colorectal adenocarcinoma. This approach resolves haplotype information using up to 100 times less genomic DNA than some methods and enables the accurate detection of structural variants.

Emergence of constitutively active estrogen receptor-α mutations in pretreated advanced estrogen receptor-positive breast cancer.

PURPOSE: We undertook this study to determine the prevalence of estrogen receptor (ER) α (ESR1) mutations throughout the natural history of hormone-dependent breast cancer and to delineate the functional roles of the most commonly detected alterations. EXPERIMENTAL DESIGN: We studied a total of 249 tumor specimens from 208 patients. The specimens include 134 ER-positive (ER(+)/HER2(-)) and, as controls, 115 ER-negative (ER(-)) tumors. The ER(+) samples consist of 58 primary breast cancers and 76 metastatic samples. All tumors were sequenced to high unique coverage using next-generation sequencing targeting the coding sequence of the estrogen receptor and an additional 182 cancer-related genes. RESULTS: Recurring somatic mutations in codons 537 and 538 within the ligand-binding domain of ER were detected in ER(+) metastatic disease. Overall, the frequency of these mutations was 12% [9/76; 95% confidence interval (CI), 6%-21%] in metastatic tumors and in a subgroup of patients who received an average of 7 lines of treatment the frequency was 20% (5/25; 95% CI, 7%-41%). These mutations were not detected in primary or treatment-naïve ER(+) cancer or in any stage of ER(-) disease. Functional studies in cell line models demonstrate that these mutations render estrogen receptor constitutive activity and confer partial resistance to currently available endocrine treatments. CONCLUSIONS: In this study, we show evidence for the temporal selection of functional ESR1 mutations as potential drivers of endocrine resistance during the progression of ER(+) breast cancer.

Development and validation of a clinical cancer genomic profiling test based on massively parallel DNA sequencing.

As more clinically relevant cancer genes are identified, comprehensive diagnostic approaches are needed to match patients to therapies, raising the challenge of optimization and analytical validation of assays that interrogate millions of bases of cancer genomes altered by multiple mechanisms. Here we describe a test based on massively parallel DNA sequencing to characterize base substitutions, short insertions and deletions (indels), copy number alterations and selected fusions across 287 cancer-related genes from routine formalin-fixed and paraffin-embedded (FFPE) clinical specimens. We implemented a practical validation strategy with reference samples of pooled cell lines that model key determinants of accuracy, including mutant allele frequency, indel length and amplitude of copy change. Test sensitivity achieved was 95-99% across alteration types, with high specificity (positive predictive value >99%). We confirmed accuracy using 249 FFPE cancer specimens characterized by established assays. Application of the test to 2,221 clinical cases revealed clinically actionable alterations in 76% of tumors, three times the number of actionable alterations detected by current diagnostic tests.

Targeted next-generation sequencing of advanced prostate cancer identifies potential therapeutic targets and disease heterogeneity.

BACKGROUND: Most personalized cancer care strategies involving DNA sequencing are highly reliant on acquiring sufficient fresh or frozen tissue. It has been challenging to comprehensively evaluate the genome of advanced prostate cancer (PCa) because of limited access to metastatic tissue. OBJECTIVE: To demonstrate the feasibility of a novel next-generation sequencing (NGS)-based platform that can be used with archival formalin-fixed paraffin-embedded (FFPE) biopsy tissue to evaluate the spectrum of DNA alterations seen in advanced PCa. DESIGN, SETTING, AND PARTICIPANTS: FFPE samples (including archival prostatectomies and prostate needle biopsies) were obtained from 45 patients representing the spectrum of disease: localized PCa, metastatic hormone-naive PCa, and metastatic castration-resistant PCa (CRPC). We also assessed paired primaries and metastases to understand disease heterogeneity and disease progression. INTERVENTION: At least 50 ng of tumor DNA was extracted from FFPE samples and used for hybridization capture and NGS using the Illumina HiSeq 2000 platform. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: A total of 3320 exons of 182 cancer-associated genes and 37 introns of 14 commonly rearranged genes were evaluated for genomic alterations. RESULTS AND LIMITATIONS: We obtained an average sequencing depth of >900X. Overall, 44% of CRPCs harbored genomic alterations involving the androgen receptor gene (AR), including AR copy number gain (24% of CRPCs) or AR point mutation (20% of CRPCs). Other recurrent mutations included transmembrane protease, serine 2 gene (TMPRSS2):v-ets erythroblastosis virus E26 oncogene homolog (avian) gene (ERG) fusion (44%); phosphatase and tensin homolog gene (PTEN) loss (44%); tumor protein p53 gene (TP53) mutation (40%); retinoblastoma gene (RB) loss (28%); v-myc myelocytomatosis viral oncogene homolog (avian) gene (MYC) gain (12%); and phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit α gene (PIK3CA) mutation (4%). There was a high incidence of genomic alterations involving key genes important for DNA repair, including breast cancer 2, early onset gene (BRCA2) loss (12%) and ataxia telangiectasia mutated gene (ATM) mutations (8%); these alterations are potentially targetable with poly(adenosine diphosphate-ribose)polymerase inhibitors. A novel and actionable rearrangement involving the v-raf murine sarcoma viral oncogene homolog B1 gene (BRAF) was also detected. CONCLUSIONS: This first-in-principle study demonstrates the feasibility of performing in-depth DNA analyses using FFPE tissue and brings new insight toward understanding the genomic landscape within advanced PCa.

Identification of new ALK and RET gene fusions from colorectal and lung cancer biopsies.

Applying a next-generation sequencing assay targeting 145 cancer-relevant genes in 40 colorectal cancer and 24 non-small cell lung cancer formalin-fixed paraffin-embedded tissue specimens identified at least one clinically relevant genomic alteration in 59% of the samples and revealed two gene fusions, C2orf44-ALK in a colorectal cancer sample and KIF5B-RET in a lung adenocarcinoma. Further screening of 561 lung adenocarcinomas identified 11 additional tumors with KIF5B-RET gene fusions (2.0%; 95% CI 0.8-3.1%). Cells expressing oncogenic KIF5B-RET are sensitive to multi-kinase inhibitors that inhibit RET.

Direct RNA sequencing.

Our understanding of human biology and disease is ultimately dependent on a complete understanding of the genome and its functions. The recent application of microarray and sequencing technologies to transcriptomics has changed the simplistic view of transcriptomes to a more complicated view of genome-wide transcription where a large fraction of transcripts emanates from unannotated parts of genomes, and underlined our limited knowledge of the dynamic state of transcription. Most of this broad body of knowledge was obtained indirectly because current transcriptome analysis methods typically require RNA to be converted to complementary DNA (cDNA) before measurements, even though the cDNA synthesis step introduces multiple biases and artefacts that interfere with both the proper characterization and quantification of transcripts. Furthermore, cDNA synthesis is not particularly suitable for the analysis of short, degraded and/or small quantity RNA samples. Here we report direct single molecule RNA sequencing without prior conversion of RNA to cDNA. We applied this technology to sequence femtomole quantities of poly(A)(+) Saccharomyces cerevisiae RNA using a surface coated with poly(dT) oligonucleotides to capture the RNAs at their natural poly(A) tails and initiate sequencing by synthesis. We observed transcript 3' end heterogeneity and polyadenylated small nucleolar RNAs. This study provides a path to high-throughput and low-cost direct RNA sequencing and achieving the ultimate goal of a comprehensive and bias-free understanding of transcriptomes.

Virtual terminator nucleotides for next-generation DNA sequencing.

We synthesized reversible terminators with tethered inhibitors for next-generation sequencing. These were efficiently incorporated with high fidelity while preventing incorporation of additional nucleotides, and we used them to sequence canine bacterial artificial chromosomes in a single-molecule system that provided even coverage for over 99% of the region sequenced. This single-molecule approach generated high-quality sequence data without the need for target amplification and thus avoided concomitant biases.

Single-molecule DNA sequencing of a viral genome.

The full promise of human genomics will be realized only when the genomes of thousands of individuals can be sequenced for comparative analysis. A reference sequence enables the use of short read length. We report an amplification-free method for determining the nucleotide sequence of more than 280,000 individual DNA molecules simultaneously. A DNA polymerase adds labeled nucleotides to surface-immobilized primer-template duplexes in stepwise fashion, and the asynchronous growth of individual DNA molecules was monitored by fluorescence imaging. Read lengths of >25 bases and equivalent phred software program quality scores approaching 30 were achieved. We used this method to sequence the M13 virus to an average depth of >150x and with 100% coverage; thus, we resequenced the M13 genome with high-sensitivity mutation detection. This demonstrates a strategy for high-throughput low-cost resequencing.