Inappropriate blood clotting can be a killer. Each year in the US, 300,000 to 900,000 people suffer a venous thromboembolism (VTE), which includes deep vein thrombosis (DVT) and pulmonary embolism (PE). African Americans have a 30–60% higher incidence of either or both than people of European ancestry.
The risk factors that come up on the usual health websites for VTE tend to be the environmental ones that apply to everyone. But mention of genetic factors tend to be those found primarily among people of European ancestry – even though African Americans are at higher risk.
A new study published in Molecular Genetics and Genomic Medicine identifies a strong genetic risk factor in African Americans, against the backdrop of the European risk factors being “nearly absent”. The findings are from the lab of Russ B. Altman, MD, PhD, from Stanford University, with lead author just-graduated Roxana Daneshjou, MD, PhD.
DVT and PE
DVT is the reason airline passengers are encouraged to move their feet, to keep blood from pooling and triggering clot formation. Symptoms, if any, are leg swelling, unexpected bruises, or a stabbing pain in an arm, leg, or the chest.
PE causes sudden death when a clot, perhaps dislodged from a sleeping foot, gets stuck in an artery in a lung. That’s what happened to NBC correspondent David Bloom on April 6, 2003, while squished in a vehicle for several days covering the war in Iraq. It also happened to the husband of a good friend of mine after a long flight.
VTE is devastating, but somewhat preventable with awareness of risk, movement, wearing compression stockings, and medication.
Environmental risk factors for DVT and PE include any situation that decreases blood flow, such as remaining still during travel or confined to bed following surgery. Extra weight, smoking, taking hormones for birth control or postmenopausal symptoms, certain cancer treatments, and pregnancy and recently giving birth are other risk factors that don’t evoke ancestry. Athletes aren’t immune and in fact may be at elevated risk if they travel to sporting events, or become injured or dehydrated and attribute symptoms to the event, not DVT.
A personal or family history of DVT or other cardiovascular condition also raises risk. It’s the identification of genetic risk factors that has been so Eurocentric. For them, mutation in the coagulation factor 5 gene hikes risk 3-5 fold, and in factor 2, 2-3 fold.
A Mother and Two Grown Daughters Inspire the Study
The new study grew out of a larger one. The researchers were sequencing exomes (the <2% of the genome that encodes protein) to find gene variants that affect how the body metabolizes the blood thinner warfarin. An African American mother and her two grown daughters, who’d each had a DVT or PE or both, were part of the study, but they didn’t have the European mutations. The researchers tested for all genes involved in blood clotting, particularly those behind DVT and PE in other populations, filtering for rare mutations that replace one amino acid type with another.
Bingo! The family had a variant in the gene that encodes protein S. (The mutation, V510M, changes a valine to a methionine at position 510 in the amino acid sequence). Protein S pairs with another protein to cut precursor forms of two coagulation factors.
Protein S deficiency has been known since the 1980s to cause DVT and PE. It affects only 0.16 to 0.21% of the general population. One mutation hikes risk at any age, while two copies of the mutation causes massive blood clotting in infants. However, since some relatives of affected individuals who also have the mutation don’t develop DVT or PE, other risk factors are at play. The official entry in Mendelian Inheritance in Man mentions families of Italian, Spanish, or Chinese background, but not African.
Gene in hand, the researchers searched for the mutation in data from two cohorts of African Americans already being studied: one with 306 cases and 370 controls, and the other 2203 individuals in the NHLBI GO Exome Sequencing Project.
First, past lab tests for the family from when they weren’t taking warfarin had indicated low levels of protein S. Then the cohort studies revealed the V510M protein S mutation in other African Americans, at a significant frequency compared to other groups. The NHLBI data found the mutation “nearly absent” in people of European descent (frequency 0.03%), but from 0.5%-1.42% in various populations with African ancestry. Protein S deficiency is extremely rare among Europeans but if there increases risk of DVT or PE five-fold. Tellingly, the rare European mutations in the protein S gene differ from the African version.
Ancestry, Not Race
Skin color is not a very good surrogate for drug response, for many reasons. A study I’ve mentioned in my genetics textbook for so many editions that I don’t recall the source (but it may be Sarah Tishkoff’s group at Penn) cataloged 23 markers for genes that control drug metabolism in 354 people representing blacks (Bantus, Ethiopians, and Afro-Caribbeans), whites (Norwegians, Armenians, and Ashkenazi Jews), and Asians (Chinese and New Guineans). The genetic markers fell into four very distinct groups that predicted which of several blood thinners, chemotherapies, and painkillers would work— and these response groups didn’t at all match the groups based on skin color.
The fallacy of “race-based prescribing” has made headlines for 15 years, and thanks to human genome sequencing and annotation, may finally be on the way out. I tackled the issue back in 2002 for The Scientist in Race and the Clinic: Good-Science? In 2006 the issue catapulted into the headlines when the FDA approved marketing of a heart failure drug combination called BiDil to African Americans on shaky evidence, a move that likely resulted in misprescribing both ways – to black patients in whom the drug was ineffective yet not to white patients in whom it would have worked.
Now that everyone is officially talking about precision medicine, Vence Bonham, JD, Shawneequa Callier, JD, and Charmaine D. Royal, PhD, revisited the oversimplified link of race to drug response in a recent New England Journal of Medicine article. Ancestry, not race, is the critical consideration, because it is the transmission of genes we’re talking about for establishing inherited risk, not how much melanin infiltrates the uppermost skin layer. We all have about the same number of melanin-producing cells.
Steel Syndrome in the Puerto Rican Community of East Harlem
I recently wrote about a compelling example of how considering ancestry in subpopulations can actually prevent medical harm, for Rare Disease Report.
Steel syndrome is an orthopedic condition that is much more common among people of Puerto Rican ancestry than among others in East Harlem. I learned about it from Eimear Kenny, PhD, an assistant professor in Genetics and Genomic Sciences at the Institute of Personalized Medicine at the Icahn School of Medicine at Mount Sinai and her graduate student Gillian Belbin, MS, at the American Society of Human Genetics annual conference last fall. Their findings are already leading to expanded genetic testing that can save children from hip surgery that can actually worsen symptoms.
While looking for “hidden patterns of relatedness in genetic data of New York City populations to infer demography and ancestry,” Kenny and Belbin found a genetic signature so distinctive that it could reveal Puerto Rican ancestry even among individuals who were adopted and have no clue where they came from. At the same time, the researchers noticed a constellation of musculoskeletal symptoms in several people in the Puerto Rican community — short stature along with joint pain, joint replacements at young ages, pinching of the spinal cord in the neck, and characteristic facial features. But they didn’t know what it was.
The third piece of evidence came when they read a report about a mutation in a collagen gene (COL27A1) that causes Steel syndrome, described by orthopedist Howard H. Steel in 1993 about 23 children from Puerto Rico. His report included other symptoms – high arches, soft bone tips, dislocated elbows and hips, fused finger and toe bones, and scoliosis. Most importantly, Dr. Steel advised against hip surgery, which might otherwise be done, because the nature of the problem is different than treating hips dislocated by injury.
A report from 2010 reiterated the warning against surgery, and added facial features: a large head with a long face, broad nose, small low ears rotated slightly backward, and a bulging forehead.
Kenny and Belbin searched their data for the collagen mutation and found it: 6 cases among the 8,000 or so people of Puerto Rican ancestry living in the area. “If you’re looking at Hispanics as a whole you may not find it, because it is much less frequent in that population than if you understand the finer scale structure of the population,” Dr. Kenny concluded.
Correctly diagnosing Steel syndrome can prevent further harm to an affected individual, and depends upon a health care practitioner being aware of the clustering of the condition in a certain slice of the population. In the case of DVT and PE, ancestry awareness – not race-based prescribing – can prevent death.
As the Precision Medicine Initiative moves ahead and stratifies complex populations by ancestry and identifies the threads of mixed ancestries, diagnostics and therapeutics will become increasingly tailored, and presumably more effective.
Thanks to Jennie Dusheck, friend and writer at Stanford Medicine, for alerting me to the new paper. See her version here.