With the annual clinical genetics meeting of the American College of Medical Genetics and Genomics taking place in Nashville this week, the QIAGEN Silicon Valley blog staff chatted with Sean Scott who is leading our Ingenuity Clinical initiative. Sean spends most of his time talking to clinical geneticists and lab directors to understand their unmet needs and requirements for NGS-based test interpretation and reporting — so we asked him to tell us about the key trends that he sees. Here’s what he had to say.
Scale Challenges on the Horizon:
As labs migrate from single-gene Sanger sequencing tests to multi-gene panels and exome or whole genome NGS-based tests, labs are inevitably going to encounter increased complexities in tests, phenotypes, and variants that will constrain their ability to effectively interpret and report out on these tests.
Clinical geneticists and lab directors can spend two to three hours, on average, classifying an observed variant in a case. Particularly tricky variants can take upwards of five to seven hours to assess. Standard practice is to score and classify observed variants based upon available clinical evidence in scientific literature — combined to some degree with various prediction algorithms — and labs spend a significant amount of time searching for relevant literature and non-published clinical evidence in databases and other sources that are often missing relevant articles and information.
This is an incredibly time-consuming, human-intensive, and error-prone approach. It’s not difficult to extrapolate out on the scale challenges when you run exome- or genome-based tests with today’s mostly manual approach. Labs will not be able to hire enough clinical geneticists to handle this increased workload. Next-generation sequencing has tremendous potential in the clinical environment; however, the market for NGS-based clinical testing will not develop quickly unless labs can streamline and scale their test interpretation and reporting workflow from variant annotation through scoring, classification, and reporting. Our goal is to greatly reduce the amount of time it takes an average lab to classify variants as well as to increase confidence in clinical assessment of the observed variant.
There are three basic components to our clinical decision support software:
- Content manually curated from literature
- Computational engine for the automated scoring and classification of variants
- Workflow support for the review, assessment, and reporting of clinically relevant variants
Here at QIAGEN, our mission is to manually curate all relevant human mutation-related clinical evidence and then structure and model it to make it computable for automated variant scoring and classification. With Ingenuity Clinical, we’ll provide support screening and diagnostic applications as well as hereditary /germline and somatic cancer test indications, where labs want to link clinically relevant somatic variants to approved drug indications and available clinical trials. We aim to help labs bring more effective tests to market faster, reduce the costs associated with test interpretation, foster professional association standards, and better demonstrate the clinical utility of their lab-developed tests to ensure reasonable levels of reimbursement.
Internal Knowledge Is Key:
Comprehensive coverage of literature and the clinical evidence embedded within it is a must-have for effective test interpretation. While 80 percent of variant classification is based on scientific literature, the remaining 20 percent is based on domain expertise within the lab and the case history data these labs have accumulated. That knowledge may not be represented in the scientific literature, but it is just as useful and important as published information. A lab’s ability to develop an internal case database can be a clear competitive advantage, improving its ability to make assessments and setting it apart from other clinical labs. Ingenuity Clinical will enable labs to capture this internal information and incorporate it into the underlying clinical evidence to improve their interpretation capabilities.
ACMG Leads in Defining Guidelines:
ACMG has been very proactive at defining guidelines for variant scoring and classification; they want consistency, quality metrics, and standards across labs. We’ve been working with ACMG to explore how we can help with their educational curricula and how we can help support their guidelines around variant interpretation. We are able to translate those guidelines into bioinformatics terms and implement them in our clinical decision support tool. This will be really important in ensuring standardization and compliance across labs. ACMG is very progressive for trying to drive standards for NGS-based testing, and we are glad to have a good working relationship with the association.