Are we Ready for DNA-based Population Screening? The Need for Large Collaborative Pilot Studies

December 10, 2019 by Michael Murray, Yale University; James Evans, University of North Carolina; Betsy Thompson and Muin J. Khoury, Centers for Disease Control and Prevention

a maginfying glass with DNA looking at a crowd

a maginfying glass with DNA looking at a crowd

This blog is a summary of a recent viewpoint published in the Journal of American Medical Association. There is increasing evidence that accepted approaches using medical history-based screening fail to identify the majority of individuals with genetic conditions associated with cancer and heart disease, including CDC tier 1 applications, such as hereditary breast and ovarian cancer, Lynch syndrome and familial hypercholesterolemia. For example, at one health care organization only 48 of 267 (18%) adults with pathogenic BRCA1/2 variants were aware that they carried these genetic variants prior to the DNA-based screening of over 50,000 adults. The failure to identify this risk in 82% of those with genetic risk is attributable to both limitations to the uptake of time-intensive medical history-based screening, as well as limitations in the sensitivity of the medical history-based screening method. This new evidence of significantly limited utility for medical-history-based screening to identify these patients has come to light in the setting of less expensive DNA-based testing, and greatly improved capacity for DNA variant interpretation compared to the 1990s.

Currently, it is estimated that over 1% of the population carries a pathogenic DNA variant associated with either FH (LDLR, APOB, PCSK9), HBOC (BRCA1, BRCA2) or LS (MLH1, MSH2, MSH6, PMS2). A move away from medical history-based screening and toward optimized DNA-based population screening for these genes has been proposed as an approach with the potential to offer short-term benefit for the estimated 3-4 million individuals in the US with one of these risks, and longer-term benefit to more people as the number of genes proposed for screening increases. However, many questions need to be answered prior to commencing with routine population-based screening.

Numerous important evidence gaps related to DNA-based population health screening are summarized by questions in the table below. Given the relatively low frequency of individuals with the target genetic risk in the population (i.e. aggregate 1-2%), evidence gathering pilot studies will require large collaborative cohorts to begin to address some these evidence gaps. Without large pilot studies, opportunities to evaluate evidence of clinical utility and economic feasibility will be lost or delayed.

12 Questions to be Addressed in Large Scale DNA-based Screening Pilots

How should screening be designed to offer inclusive benefits for the whole population? (with specific attention to the poor, as well as underrepresented racial and ethnic groups)
What are the appropriate population characteristics for screening? (e.g. age, gender)
What is the optimal testing strategy/technology? (e.g. exome sequencing, multi-gene panel, single nucleotide polymorphisms array)
What are the ideal lead institutions for carrying out DNA-based screening? (e.g. healthcare provider organizations, departments of public health, for profit companies)
Who should pay for DNA-based screening (primary screen)? (e.g. government funding, private insurance, self-pay)
Who should pay for clinical follow-up (secondary screen)? (e.g. government funding, private insurance, self-pay)
How often should data be re-analyzed? (e.g. compared to evolving databases like ClinVar to re-analyze data annually)
What strategy should be pursued for cascade testing? (e.g. should at-risk family members be automatically contacted by health system or health department)
What are the short-term clinical outcomes? (e.g. correcting diagnostic misattribution, pre-symptomatic diagnosis of cancer or heart disease)
What are the long-term clinical outcomes? (e.g. non-penetrance, overdiagnosis)
What are the best practices for reporting negative screening results? (it is critically important to avoid false reassurance)
What are the clinical workforce needs related to delivering DNA-based results and clinical follow-up at population scale? (i.e. How many medical geneticists, genetic counselors, specialists, others)

December 10, 2019 by Michael Murray, Yale University; James Evans, University of North Carolina; Betsy Thompson and Muin J. Khoury, Centers for Disease Control and Prevention