This will be a departure from my usual posts. Several announcements in the news and medical journals have caught my attention recently, and as I delved into the details, I thought I would share them with our SBM readers. Topics include AIDS cures, the continuing danger of polio, eating nuts for longevity, racial differences in vitamin D, and the use of pharmacogenetic testing to guide the dosage of anticoagulant drugs. They are all examples of science-based medicine in action.
Have patients been cured of AIDS?
I read that the HIV virus had returned in patients thought to have been cured by bone marrow transplants, and I mistakenly thought they were referring to the original claim of cure I had read about. Nope, that one still stands.
The CCR5-Δ32 mutation encodes a cell-surface receptor that allows the HIV virus to attach and enter a host cell. People with two copies of that gene are resistant to HIV infection, even considered immune. An AIDS patient in Germany, Timothy Ray Brown, developed leukemia and was intentionally given a bone marrow transplant from a person who had that mutation, and the HIV virus became undetectable in his blood. He was considered to be the first person “cured” of AIDS. Bone marrow transplants are risky and are not an option for treating AIDS in the absence of a malignancy. Subsequently, two other men with AIDS and lymphoma were treated with bone marrow transplants from donors without the mutation; it was hoped that the “graft vs. host” battle would eliminate the HIV virus from their bodies, and indeed their HIV levels became undetectable. After several months, they stopped taking their antiretroviral drugs, and doctors hoped they had been cured. Unfortunately, detectable levels of the virus returned after 12 weeks and 32 weeks off medication, respectively.
The CCR5-Δ32 mutation remains an intriguing avenue for further research into mechanisms, vaccines, and the possibility of a treatment based on the mutation that doesn’t involve the risks of bone marrow transplantation.
AIDS is one of the great triumphs of science-based medicine. It took only 2 years to identify the responsible retrovirus after the first clinical cases were reported. Big Pharma was quick to prove the value of a drug that was already available (zidovudine) and to develop several classes of new, effective drugs known for highly active antiretroviral therapy (HAART). It took little more than a decade to turn a death sentence into a chronically-manageable disease with a nearly normal life expectancy. And at the same time, AIDS denialists and superstitious CAMsters were responsible for around 171,000 AIDS cases and 343,000 deaths between 1999 and 2007 alone. I can’t think of a better example of why science-based medicine is important and why we need efforts like this blog to combat nonscientific thinking.
Has a baby been cured of AIDS?
A child whose mother was HIV positive was empirically started on antiretroviral treatment 30 hours after birth, and treatment was continued after tests on the infant met the diagnostic criteria for infection. On treatment, the infant’s HIV-1 RNA levels progressively dropped until they were undetectable at 29 days of age. Compliance was suspected to be poor, and the mother admitted she had stopped giving the child medication at age 15 months. The child is now 30 months old and neither the virus nor viral antibodies are detectable. This is only one case report, and it’s too early to say whether the child is cured, but it suggests the possibility that early antiretroviral treatment in newborns may interfere with the development of persistent reservoirs of virus. These patients may not need a lifetime of therapy.
Will eating nuts prolong your life?
A study published in The New England Journal of Medicine showed an association between the reported frequency of nut consumption and mortality. Headlines trumpeted “Go Nuts!” Previous studies had suggested a beneficial effect, and a randomized primary prevention trial had shown a reduction in cardiovascular events on a Mediterranean diet supplemented with walnuts, hazelnuts, or almonds. The new study was a high-quality trial showing a dose-response relationship: the more nuts a person ate, the less likely he was to die of all causes and of specific causes (cancer, heart disease, and respiratory disease). Separate analyses for tree nuts and peanuts showed similar results. The authors stressed that they had found a correlation and were not claiming causation. Unidentified variables could have been a cause, and reverse causality is possible: maybe people in poor health are less likely to eat nuts.
The FDA allows advertisers to claim that “eating 1.5 ounces per day of walnuts, as part of a low saturated fat and low cholesterol diet and not resulting in increased caloric intake, may reduce the risk of coronary heart disease.” Besides the high calorie density of nuts, another caveat is the possible risk of salmonella from nuts.
Racial differences in vitamin D
A recent study measured vitamin D-binding protein levels in white and black Americans. The Institute of Medicine recommends a blood level of 25-hydroxyvitamin D greater than 20 ng per milliliter to maximize bone health, but blacks can have lower levels without any decrease in bone mineral density. Vitamin D levels are lower in blacks than in whites; but since levels of vitamin D-binding protein are also lower due to genetic variants, their bioavailable vitamin D is the same. Blacks have higher calcium levels, bone density, and slightly higher parathyroid hormone levels than whites. Evolutionary explanations have been suggested for racial differences that developed as people with dark skin moved to climates with less sun. Blacks are being over-diagnosed as vitamin D deficient based on blood tests that do not reflect a true vitamin D deficiency. We need a better test to measure bioavailable vitamin D.
No country is safe from polio
In 1988 polio was endemic in 125 countries, with 350,000 new cases of paralysis annually. Thanks to an aggressive vaccination campaign, the incidence has been reduced by 99% and polio is now endemic in just 3 countries (Afghanistan, Nigeria, and Pakistan). Wild-type poliovirus type 3 is close to eradication, and type 2 is probably already eradicated, but type 1 remains in circulation and is a threat to every country including those that are currently polio-free. China’s last case of indigenous wild-type poliovirus infection was in 1994. A few cases of wild-type virus infections have been imported from Myanmar and India since then. China experienced a polio outbreak in 2011 in the Xinjiang Uygur Autonomous Region. The index case was a 16-month-old girl with no history of travel, but nucleotide sequencing confirmed that it was a strain imported from Pakistan. 21 confirmed cases and 23 clinically-compatible cases were found. The population was already close to herd-immunity thresholds. Of the ten affected children, three had not received oral polio vaccine (OPV), two had received one or two doses, and five had received three or more doses, indicating primary vaccination failure.
The Chinese Ministry of Health declared an emergency and responded vigorously. They were able to stop the spread of the disease within a month and a half, but they continued vaccinating until 43.7 million doses of oral polio vaccine had been given at a cost of $26 million. Adults were included, and OPV was given regardless of vaccination history. The oral vaccine has several advantages: low cost, ease of administration, can passively immunize non-recipients, and produces better intestinal immunity to reduce spread. It has the disadvantage that it can cause paralysis on rare occasions and can regain some of the properties of wild strains. Adverse event reporting was required during the vaccination campaign, but I wasn’t able to find the resulting data; it would be interesting to know, so we could directly compare benefits to risks. Eliminating type 2 from the trivalent vaccine is expected to reduce the risks. One shot of inactivated poliovirus can be used to maintain the protection against a possible recurrence of type 2.
Lessons to be learned: polio vaccines are not 100% effective, and every country in the world remains at risk until worldwide eradication is achieved (hopefully by 2014). Vaccine rejecters are a danger to worldwide public health.
Genetic testing not useful to determine warfarin dosage
Pharmacogenetics holds great promise for predicting individual responses to drugs based on genetic variants, but so far there are not many practical applications. One of the first proposals was to guide warfarin (Coumadin) dosage by genome testing. When I first heard this, my initial reaction was “Why bother?” Warfarin is an anticoagulant used to reduce the risk of blood clots in patients with atrial fibrillation, artificial heart valves, deep vein thrombosis, pulmonary embolism, and other conditions. Dosage is adjusted to keep the international normalized ratio (INR) of clotting factors within a prescribed range, usually between 2.0 and 3.0. This requires frequent blood tests and dosage adjustments (up to three times a week at first, and eventually decreasing to monthly in fortunate patients whose INR is stable). All kinds of medications, foods, and other factors affect the INR, so fluctuations are common. Overdosage can lead to serious bleeding complications and can be counteracted by vitamin K. Some patients require much higher doses than others; but that’s not a problem, since you can simply start with a low dose and titrate upwards as needed. Genetic variants of cytochrome enzymes affect the metabolism, requiring either higher or lower doses than usual to maintain the desired INR.
Pharmacogenetic enthusiasts predicted that genomic testing would be a more precise way to control dosage and would reduce the risk of clotting or bleeding complications. Recently three studies were simultaneously published in The New England Journal of Medicine to determine whether usual care or genome-guided care provided better control. The first found no difference, the second found a small (7%) advantage to genomic guidance, and the third found no difference. An accompanying editorial points out that the studies only covered the initiation phase without long-term follow-up, and concludes that pharmacogenetic testing is of marginal usefulness at best and represents a large expense. It suggests we should concentrate on making the current practice safer with better management.
This article was originally published in the Science- Based Medicine Blog.