Sheila Seal | bioRxiv Science

bioRxiv2016

Implementing rapid, robust, cost-effective, patient-centred, routine genetic testing in ovarian cancer patients

BackgroundAdvances in DNA sequencing have made gene testing fast and affordable, but adaptation of clinical services to capitalise on this for patient benefit has been slow. Ovarian cancer exemplifies limitations of current systems and potential benefits of increased gene testing. Approximately 15% of ovarian cancer patients have a germline mutation in BRCA1 or BRCA2 (collectively termed BRCA) and this has substantial implications for their personal management and that of their relatives. However, in most countries implementation of BRCA testing in ovarian cancer has been inconsistent and largely unsuccessful.\n\nMethodsWe developed a mainstream pathway in which BRCA testing was undertaken by cancer team members after 30 minutes online training. Patients with a mutation were sent a genetic appointment with their results. Cascade testing to relatives was performed via standard clinical genetic procedures.\n\nFindings207 women with ovarian cancer were offered gene testing through the mainstream pathway and all accepted. 33 (16%) had a BRCA mutation. The result informed management of 79% (121/154) women with active disease including 97% (32/33) women with a mutation. All mutation-positive women and ~3.5 relatives per family have been seen in genetics. Patient and clinician feedback was very positive. >95% found the pathway to be simple and effective. The pathway offers considerable reduction in time (~5-fold) and resource requirements (~13-fold) compared to the traditional genetic pathway. We estimate it would deliver {pound}2.6M NHS cost savings per year, and would allow implementation of national testing recommendations with existing infrastructure.\n\nInterpretationMainstream genetic testing is effective, efficient and patient-centred and offers a mechanism for large-scale implementation of BRCA gene testing in cancer patients. The principles could be applied in many other countries and to many other areas of genomic medicine.

Genetics↗

bioRxiv2015

CSN and CAVA: variant annotation tools for rapid, robust next-generation sequencing analysis in the clinic

BackgroundNext-generation sequencing (NGS) offers unprecedented opportunities to expand clinical genomics. It also presents challenges with respect to integration with data from other sequencing methods and historical data. Provision of consistent, clinically applicable variant annotation of NGS data has proved difficult, particularly of indels, an important variant class in clinical genomics. Annotation in relation to a reference genome sequence, the DNA strand of coding transcripts and potential alternative variant representations has not been well addressed. Here we present tools that address these challenges to provide rapid, standardized, clinically appropriate annotation of NGS data in line with existing clinical standards.\n\nMethodsWe developed a clinical sequencing nomenclature (CSN), a fixed variant annotation consistent with the principles of the Human Genome Variation Society (HGVS) guidelines, optimized for automated variant annotation of NGS data. To deliver high-throughput CSN annotation we created CAVA (Clinical Annotation of VAriants), a fast, lightweight tool designed for easy incorporation into NGS pipelines. CAVA allows transcript specification, appropriately accommodates the strand of a gene transcript and flags variants with alternative annotations to facilitate clinical interpretation and comparison with other datasets. We evaluated CAVA in exome data and a clinical BRCA1/BRCA2 gene testing pipeline.\n\nResultsCAVA generated CSN calls for 10,313,034 variants in the ExAC database in 13.44 hours, and annotated the ICR1000 exome series in 6.5 hours. Evaluation of 731 different indels from a single individual revealed 92% had alternative representations in left aligned and right aligned data. Annotation of left aligned data, as performed by many annotation tools, would thus give clinically discrepant annotation for the 339 (46%) indels in genes transcribed from the forward DNA strand. By contrast, CAVA provides the correct clinical annotation for all indels. CAVA also flagged the 370 indels with alternative representations of a different functional class, which may profoundly influence clinical interpretation. CAVA annotation of 50 BRCA1/BRCA2 gene mutations from a clinical pipeline gave 100% concordance with Sanger data; only 8/25 BRCA2 mutations were correctly clinically annotated by other tools.\n\nConclusionsCAVA is a freely available tool that provides rapid, robust, high-throughput clinical annotation of NGS data, using a standardized Clinical Sequencing Nomenclature.

Genomics↗