Caution Urged in Using Multigene Panels to Assess Breast Cancer Risk
An international consortium of researchers writing in a special report published in the New England Journal of Medicine have urged caution in using multigene panels in assessing risk for breast cancer (NEJM 2015;372:2243–57). The authors said the clinical utility for many such tests has not been established and that without clinical validation there is potential for “substantial misuse of the technology.”
The 17 co-authors observed that advances in sequencing technology have made it easy to use multigene panels when assessing a woman’s risk of breast cancer. After the U.S. Supreme Court in 2013 invalidated Myriad Genetics’ patent claims on the DNA sequence for BRCA1 and BRCA2 genes—both strongly associated with breast cancer risk—many other companies have developed test panels. These panels test for more than 100 genes, only 21 of which have been specifically associated with breast cancer.
Of these 21, the panel identified just eight genes—BRCA1, BRCA2, CDH1, NF1, PALB2, ATM, CHEK2, and NBN—for which a strong association with breast cancer has been established. Absolute risk of breast cancer by age 80 in women who harbor these protein-truncating variants ranges from 76% with BRCA2 to 23% for NBN. Two other variants—PTEN and TP53—also confer increased risk, but the authors found that most published risk estimates involving them were subject to ascertainment bias due to the selected populations from which estimates were derived.
While multigene panels test for many of these genes, some also include common single-nucleotide polymorphisms associated with slight breast cancer risk, less than 1.5 times as high as in the general population.
“While there is clear evidence between some genetic variants and an increased risk of breast cancer—such as BRCA1 and BRCA2—for which there is an overwhelming body of evidence, risk estimates for many other mutations are far too imprecise at this stage and require further investigation,” said co-author Susan Domchek, MD, Basser professor of oncology at the University of Pennsylvania Perelman School of Medicine in Philadelphia. “Until we have a better understanding of associated risks of the range of genes found on these panels, individuals who are undergoing panel testing need to be informed of the potential for uncertainty regarding results.”
Whole-Genome Sequencing Deployed in Neonatal P. aeruginosa Outbreak
Researchers at the Royal Prince Alfred Hospital in Sydney, Australia reported using whole-genome sequencing (WGS) to quickly identify and subsequently contain an outbreak of Pseudomonas aeruginosa among babies in its neonatal intensive care unit (NICU) (Infect Control Hosp Epidemiol 2015; doi:10.1017/ice.2015.133). The findings demonstrate WGS as a “powerful tool in infection control.”
The authors deployed WGS after determining through routine surveillance that babies were colonized with P. aeruginosa. In the year before the outbreak, the incidence of P. aeruginosa colonization ranged between 0 and 2.3/1,000 patient days, but during the outbreak it rose to 11.4/1,000 patient days.
The authors screened all babies in NICU weekly and found 18 colonized with P. aeruginosa. They also took various environmental samples in the NICU and found P. aeruginosa in eight sinks, four sink drains, and five sink splashbacks. Samples from 23 babies and seven environmental samples were analyzed using WGS, and from this, the authors discovered that all but one baby was carrying the strain ST253, and that just one of the environmental samples had the same strain. This enabled them to focus their infection control activities on the sink in question.
“Bacteria, such as P. aeruginosa, have evolved into many strains and frequently contaminate the healthcare environment, which makes it difficult to determine the source of an outbreak and control it using traditional methods,” said lead author Rebecca Davis, MD, in a prepared statement. “[WGS] changes that. Our study found this technology allows us to implement rapid-response infection control protocols and stem the outbreak, which is critical for vulnerable patients, such as those in a neonatal intensive care unit.”
HDL-C Efflux Capacity Independently, Inversely Linked to Coronary Heart Disease
Cholesterol efflux capacity—the ability of high-density lipoprotein cholesterol (HDL-C) to promote transport of cholesterol out of the bloodstream and into the liver for excretion in bile—is significantly and inversely associated with coronary heart disease (CHD) events independent of traditional cardiovascular disease risk factors (Lancet Diabetes Endocrinol 2015; doi.org/10.1016/S2213-8587(15)00126-6). The findings suggest that HDL-C efflux capacity might be an effective target for future therapies aimed at reducing CHD.
HDL-C is inversely associated with CHD, but research has shown that interventions to raise HDL-C don’t necessarily lower CHD risk. This led the authors to explore the association between HDL-C efflux capacity and CHD. Their investigation involved case-control sample from the EPIC-Norfolk study. The researchers quantified HDL-C efflux capacity in 1,745 participants with incident CHD and in 1,749 controls.
In comparison to participants in the lowest tertile of HDL-C efflux capacity, those in the top tertile were at reduced risk for CHD, after adjustment for age and sex. Patients who developed CHD had significantly lower HDL-C efflux capacity. The strong, inverse association between HDL-C efflux capacity and CHD persisted even after adjusting for HDL-C or apolipoprotein A-1 concentrations.
“This is a definitive finding that HDL function…does predict later heart disease events, which implies that therapies that boost HDL function might reduce risk,” said senior author Daniel Rader, MD, chief of the division of translational medicine and human genetics and Seymour Gray professor of molecular medicine at the University of Pennsylvania Perelman School of Medicine in Philadelphia.
Already Available Pharmacogenomic Data Could Help Physicians Better Prescribe Cardiovascular Drugs
Through an extensive literature and pharmacogenomics database review, researchers at the University of Chicago and Stanford University have determined that clinically actionable pharmacogenomic information for cardiovascular drugs already exists (Mayo Clin Proce 2015;09:716–29). The authors suggest that this information be made available for implementation under research protocols to see if it affects physician decision-making and patient outcomes.
The investigators focused on 71 commonly prescribed cardiovascular drugs, searching the scientific literature for pharmacogenomics-related studies published about them between 2011 and 2013. They found 597 unique publications, involving 611 genetic variants and 884 drug-variant pairs. Overall, 51 of the 71 drugs had reported pharmacogenomic effects. The authors cross-referenced the reported associations with pharmacogenetic variant data in the Pharmacogenomics Knowledge Database, an online resource.
The researchers determined that there was clinically relevant data of high enough quality to justify clinical alerts for 92 drug-variant pairs involving 23 drugs. The combinations clopidogrel-CYP2C19, metoprolol-CYP2D6, simvastatin-rs4149056, dabigatran-rs2244613, hydralazine-rs1799983 and -rs1799998, and warfarin-CYP2C9 and -VKORC1 had the highest scores according to the Appraisal of Guidelines for Research and Evaluation II scoring instrument.
“Tens of thousands of patients have been studied and the connections between common medications and the genetic variants that can lead to adverse drug reactions or treatment non-response have been described, but few physicians track this information or even know where to find it,” said senior author Peter O’Donnell, MD, assistant professor of medicine at the University of Chicago. “One dose does not fit all, so we set out to boost awareness and simplify access. We assessed the quantity and quality of the literature, ranked the most relevant studies for clinicians, and condensed the data into a series of prescribing decision aids.”