Ebola is one of the nastiest, most frightening viruses known to man. Its victims suffer fevers, muscle weakness, and other symptoms that progress to severe bleeding, both internal and external, that eventually causes them to bleed to death. All known strains of Ebola virus have very high mortality rates, ranging from 25% to 90%, and there is no cure. Ebola virus was the subject of a dramatic and frightening 1995 disaster movie, Outbreak, starring Dustin Hoffman, Rene Russo, and Morgan Freeman.
Last week, in what may be the biggest medical breakthrough of its kind in years, a group of scientists published results in The Lancet describing a completely new type of anti-viral treatment that appears to cure Ebola. They report a 100% success rate, although admittedly the test group was very small, just 4 rhesus monkeys.
This is a breakthrough not only because it may give us a cure for an uncurable, incredibly nasty virus, but also because the same method might work for other viruses, and because we have woefully few effective antiviral treatments. We can treat bacterial infections with antibiotics, but for most viruses, we have either a vaccine or nothing. And a vaccine, wonderful as it is, doesn’t help you after you’re already infected.
The scientists, led by Thomas Geisbert at Boston University, used a relatively new genomics technique called RNA interference to defeat the virus. Here’s how it works. First, a little background: the Ebola virus is made of RNA, just like the influenza virus. And just like influenza, Ebola has very few genes - only 8. One of its genes, called L protein, is responsible for copying the virus itself. Two others, called VP24 and VP35, interfere with the human immune response, making it difficult for our immune system to defeat the virus.
Geisbert and his colleagues (including scientists from Tekmira Pharmaceuticals and USAMRIID) designed and synthesized RNA sequences that would stick to these 3 genes like glue. How did they do that? We know the Ebola genome’s sequence – it was sequenced way back in 1993. And we know that RNA sticks to itself using the same rules that DNA uses. This knowledge allowed Geisbert and colleagues to design a total of 10 pieces of RNA (called “small interfering RNA” or siRNA) that they knew would stick to the 3 Ebola genes. They also took care to make sure that their sticky RNA would not stick to any human genes, which might be harmful. They packaged these RNAs for delivery by inserting them into nanoparticles that were only 81-85 nanometers across.
In the key experiment, the scientists infected rhesus monkeys with a dose of Ebola that was 30,000 times greater than the normal fatal dose. They injected the siRNA treatments 30 minutes later, and again each day for 6 days. All the monkeys survived with no long-term effects.
When I read this story in The Lancet, my first reaction was: wow. A brand-new antiviral treatment, and against Ebola? In a commentary in The Lancet, Heinz Feldmann wrote that we are “in desperate need of approved countermeasures against Ebola-virus infections,” and called the new study a “milestone.” That’s putting it mildly.
Somehow, except for the UK Daily Mail, the major media outlets seem to have entirely missed this story, which I think is by far the biggest medical story of the week, if not the year. Ebola virus doesn’t get much attention (notwithstanding the 1995 movie about it), primarily because it occurs in central Africa. But air travel has made our world much smaller, and even a relatively isolated virus could easily get out of control.
There’s still a much work needed to turn the new Ebola treatment into an approved drug, but the prospect of a cure is suddenly very real, and the techniques used to create it are truly remarkable.
This is Steven Salzberg's blog on genomics, pseudoscience, medical breakthroughs, higher education, and other topics, including skepticism about unscientific medical practices. Here's where I can say what I really think about abuses and distortions of science, wherever I see them.
Alzheimer’s treatments don’t work, but you can buy them on the Internet
A new NIH-sponsored review of the scientific literature on Alzheimer’s disease has some disappointing news: nothing works. The panel's chair, Martha Daviglus, a professor of medicine at Northwestern University, said "We wish we could tell people that taking a pill or doing a puzzle every day would prevent this terrible disease, but current evidence doesn't support this."
Despite the lack of evidence, plenty of internet quacks are happy to recommend all sorts of supplements to treat Alzheimer’s, as we’ll see in below. First, the NIH panel’s main conclusion:
Not surprisingly, our old friend “Dr.” Joseph Mercola (he’s an osteopath, not an M.D., and I promise to write more about that one day) has plenty of recommendations to prevent or treat Alzheimer’s. He even has a web page showing his “top 5 tips to beat Alzheimer’s.” Let’s look at what he has to offer.
1. “Eat Healthy.” Mercola claims here that “Omega-3 fats, such as those from animal-based sources like krill oil, have also been found to help ward off Alzheimer’s and diabetes.” Wait, does Mercola sell krill oil? You bet! He even has a special webpage for his krill oil products, and ads for it appear all over his site. In direct contradiction to Mercola, the NIH report says “In a single randomized trial of omega-3 fatty acids with only 26 weeks of follow-up, there appeared to be no effect on cognitive functioning.”
2. “Exercise.” Although Mercola is wrong to claim that exercise reduces the risk for Alzheimer’s – the NIH report says it doesn't – at least this is good general health advice.
3. “Avoid Mercury.” Uh oh, now we’re getting deep into the woo. There’s no evidence that mercury causes Alzheimer’s, and Mercola goes well beyond just saying “avoid mercury.” He advises you to “remove dental amalgam fillings, which are one of the major sources of mercury.” This bit of unscientific advice has been around for decades, but it’s little more than pseudoscience. I recommend the excellent Quackwatch site for a detailed debunking of this one. It's so widespread that several official reports have investigated it, and the American Dental Association reported back in 1998 that amalgam fillings are safe and effective.
Mercola has a second bit of very bad advice here: he suggests that you avoid thimerosal-containing vaccines. As I’ve written before, Mercola is a major anti-vaccination scaremonger, and here he goes again – suggesting, without any evidence (and no citations to any scientific studies) that vaccines may cause Alzheimer’s.
4. “Avoid Aluminum.” According to Mercola, “aluminum has long been associated with Alzheimer's disease.” The NIH report, in contradiction to this, says no such association has been found. So who should you believe, an independent panel of distinguished experts on Alzheimer’s, or an Internet osteopath selling supplements based on unproven claims? Oh, and Mercola gets in another anti-vaccination claim here: he says that “some vaccines also contain aluminum.”
And don’t forget to “avoid aluminum cookware”! And hey... does Mercola sell cookware? You betcha! He calls it “the safe alternative to toxic cookware.”
5. “Exercise Your Mind.” The NIH report says this one isn’t proven either, but at least it doesn’t do any harm.
What we really know is that the main risk factors for Alzheimer's are things we have no control over: age and genetics, particularly the APOE protein (which I wrote about recently). The NIH panel recommends much more research, which is desperately needed. If people would donate their money to Alzheimer’s research rather than wasting it on ineffective treatments, maybe we’d find a cure just a little bit sooner.
Despite the lack of evidence, plenty of internet quacks are happy to recommend all sorts of supplements to treat Alzheimer’s, as we’ll see in below. First, the NIH panel’s main conclusion:
“There is currently no evidence of even moderate scientific quality supporting the association of any modifiable factor—dietary supplement intake, use of prescription or non-prescription drugs, diet, exercise, and social engagement—with reduced risk of Alzheimer's disease. The evidence surrounding risk reduction for cognitive decline is similarly limited.”In other words, we can’t cure it, and we can’t even delay it.
Not surprisingly, our old friend “Dr.” Joseph Mercola (he’s an osteopath, not an M.D., and I promise to write more about that one day) has plenty of recommendations to prevent or treat Alzheimer’s. He even has a web page showing his “top 5 tips to beat Alzheimer’s.” Let’s look at what he has to offer.
1. “Eat Healthy.” Mercola claims here that “Omega-3 fats, such as those from animal-based sources like krill oil, have also been found to help ward off Alzheimer’s and diabetes.” Wait, does Mercola sell krill oil? You bet! He even has a special webpage for his krill oil products, and ads for it appear all over his site. In direct contradiction to Mercola, the NIH report says “In a single randomized trial of omega-3 fatty acids with only 26 weeks of follow-up, there appeared to be no effect on cognitive functioning.”
2. “Exercise.” Although Mercola is wrong to claim that exercise reduces the risk for Alzheimer’s – the NIH report says it doesn't – at least this is good general health advice.
3. “Avoid Mercury.” Uh oh, now we’re getting deep into the woo. There’s no evidence that mercury causes Alzheimer’s, and Mercola goes well beyond just saying “avoid mercury.” He advises you to “remove dental amalgam fillings, which are one of the major sources of mercury.” This bit of unscientific advice has been around for decades, but it’s little more than pseudoscience. I recommend the excellent Quackwatch site for a detailed debunking of this one. It's so widespread that several official reports have investigated it, and the American Dental Association reported back in 1998 that amalgam fillings are safe and effective.
Mercola has a second bit of very bad advice here: he suggests that you avoid thimerosal-containing vaccines. As I’ve written before, Mercola is a major anti-vaccination scaremonger, and here he goes again – suggesting, without any evidence (and no citations to any scientific studies) that vaccines may cause Alzheimer’s.
4. “Avoid Aluminum.” According to Mercola, “aluminum has long been associated with Alzheimer's disease.” The NIH report, in contradiction to this, says no such association has been found. So who should you believe, an independent panel of distinguished experts on Alzheimer’s, or an Internet osteopath selling supplements based on unproven claims? Oh, and Mercola gets in another anti-vaccination claim here: he says that “some vaccines also contain aluminum.”
And don’t forget to “avoid aluminum cookware”! And hey... does Mercola sell cookware? You betcha! He calls it “the safe alternative to toxic cookware.”
5. “Exercise Your Mind.” The NIH report says this one isn’t proven either, but at least it doesn’t do any harm.
What we really know is that the main risk factors for Alzheimer's are things we have no control over: age and genetics, particularly the APOE protein (which I wrote about recently). The NIH panel recommends much more research, which is desperately needed. If people would donate their money to Alzheimer’s research rather than wasting it on ineffective treatments, maybe we’d find a cure just a little bit sooner.
Personal genetic testing, available soon at your local pharmacy
This past week, Pathway Genomics and Walgreens announced that they would start selling Pathway’s genetic testing kits at 6000 Walgreens stores. I thought this was good news – I, for one, would like to be able to run my own genetic tests. I’d like to know what risks my genes might carry, particularly if there was something I could do to reduce those risks. But what does Pathway’s test tell you, and is it worth it?
Pathway has been selling its testing kits since September, but selling them at a large chain store like Walgreens would undoubtedly reach many more customers. After their joint press release last Tuesday, though, someone at the FDA noticed, and stepped in to ask a few tough questions. In a letter on May 10 (read it here), the FDA told Pathway that they had to get FDA approval for their diagnostic kits, or else explain why they don’t need it. Pathway claims its kit is exempt from FDA approval, but that might be a tricky argument to prove. After the FDA sent its letter, Walgreens announced that it would hold off for now on selling the kits.
Meanwhile, the real question is, what will Pathway’s test tell you about your genes? Some biomedical scientists are saying that the science of genetic testing is too new to offer directly to consumers. They argue that the results of a genetic test are difficult to interpret, and that patients shouldn’t be given this complex information without expert guidance. Stanford University’s Hank Greely, quoted in the Washington Post, said "Information is powerful, but misunderstood information can be powerfully bad." Are doctors just trying to protect their business, or are they right? I decided to take a look.
Pathway promises to tell you about your risk for up to 70 diseases and conditions, if you pay $399 for the full suite of tests. (That's the price if you order directly from Pathway, which you can do right now. The Walgreens price was supposed to be lower.)
For example, if you are a carrier of the cystic fibrosis gene (which is included in the Pathway test), and if your spouse is also a carrier, then as a couple you have a 25% chance of having a child with cystic fibrosis. The genetics of this disease are well understood, and the test for the mutation should be highly reliable. This seems like a good value to me, and I see no reason why people shouldn’t be allowed to know if they’re a carrier for cystic fibrosis; it may be important to their future plans for having children.
For other diseases, though, the Pathway test provides much more ambiguous information. Let’s look at late-onset Alzheimer’s disease, something that almost everyone worries about as they get older. The Pathway site doesn’t specify which gene they test, but there is no known “Alzheimer’s gene.” Research on Alzheimer’s has demonstrated that some genetic mutations are associated with an increased risk, but the picture is far from clear. For example, mutations in the Apolipoprotein E (APOE) gene on chromosome 19 appear to increase the risk for Alzheimer’s after age 60. According to the National Institute on Aging, the e4 (epsilon-4) variant of this gene:
Yes, emphatically yes!
It’s my genome, and I should be able to look at it if I want to. So what if the information is complex? If someone is curious enough to pay for this test, then when they get their results back, perhaps they’ll be motivated to learn more about genetics so that they can understand their own risk. Giving people information about their own genome seems like a great way to educate the public, and to make people more aware of the power of genetic information. Of course there are risks: charlatans might offer to sell bogus treatments for all kinds of genetic conditions, for example. People might misinterpret the results. But that doesn’t mean we shouldn’t allow people to discover more about themselves.
In a very real way, genetic information is no different from other medical facts. We have easy access to our blood pressure numbers, and no one is suggesting that we shouldn’t, despite the fact that interpretation of blood pressure numbers can be complicated. (For example, should you reduce salt intake, or take statins to lower your blood pressure, or change your lifestyle in other ways?)
I hope the FDA allows Pathway to sell its DNA test at Walgreens. Personally, I think the value of the tests they’re offering today is mostly to satisfy curiosity – we don’t have treatments for most of the genetic diseases they can detect, so there's not much you can do about them. Some of the results will be useful to some people, though, and over time we’ll develop new treatments that will make more genetic tests valuable. Now that Pandora’s box is open, we can’t close it again. And we shouldn’t.
Pathway has been selling its testing kits since September, but selling them at a large chain store like Walgreens would undoubtedly reach many more customers. After their joint press release last Tuesday, though, someone at the FDA noticed, and stepped in to ask a few tough questions. In a letter on May 10 (read it here), the FDA told Pathway that they had to get FDA approval for their diagnostic kits, or else explain why they don’t need it. Pathway claims its kit is exempt from FDA approval, but that might be a tricky argument to prove. After the FDA sent its letter, Walgreens announced that it would hold off for now on selling the kits.
Meanwhile, the real question is, what will Pathway’s test tell you about your genes? Some biomedical scientists are saying that the science of genetic testing is too new to offer directly to consumers. They argue that the results of a genetic test are difficult to interpret, and that patients shouldn’t be given this complex information without expert guidance. Stanford University’s Hank Greely, quoted in the Washington Post, said "Information is powerful, but misunderstood information can be powerfully bad." Are doctors just trying to protect their business, or are they right? I decided to take a look.
Pathway promises to tell you about your risk for up to 70 diseases and conditions, if you pay $399 for the full suite of tests. (That's the price if you order directly from Pathway, which you can do right now. The Walgreens price was supposed to be lower.)
For example, if you are a carrier of the cystic fibrosis gene (which is included in the Pathway test), and if your spouse is also a carrier, then as a couple you have a 25% chance of having a child with cystic fibrosis. The genetics of this disease are well understood, and the test for the mutation should be highly reliable. This seems like a good value to me, and I see no reason why people shouldn’t be allowed to know if they’re a carrier for cystic fibrosis; it may be important to their future plans for having children.
For other diseases, though, the Pathway test provides much more ambiguous information. Let’s look at late-onset Alzheimer’s disease, something that almost everyone worries about as they get older. The Pathway site doesn’t specify which gene they test, but there is no known “Alzheimer’s gene.” Research on Alzheimer’s has demonstrated that some genetic mutations are associated with an increased risk, but the picture is far from clear. For example, mutations in the Apolipoprotein E (APOE) gene on chromosome 19 appear to increase the risk for Alzheimer’s after age 60. According to the National Institute on Aging, the e4 (epsilon-4) variant of this gene:
“occurs in about 40 percent of all people who develop late-onset AD and is present in about 25 to 30 percent of the population. People with AD are more likely to have an APOE e4 allele than people who do not develop AD. However, many people with AD do not have an APOE e4 allele.”Complicated, no? And this is the clearest association between a genetic mutation and Alzheimer’s — a half-dozen other genes are associated with slightly increased risk, and research is ongoing. So should anyone be allowed to test his/her genome for this mutation?
Yes, emphatically yes!
It’s my genome, and I should be able to look at it if I want to. So what if the information is complex? If someone is curious enough to pay for this test, then when they get their results back, perhaps they’ll be motivated to learn more about genetics so that they can understand their own risk. Giving people information about their own genome seems like a great way to educate the public, and to make people more aware of the power of genetic information. Of course there are risks: charlatans might offer to sell bogus treatments for all kinds of genetic conditions, for example. People might misinterpret the results. But that doesn’t mean we shouldn’t allow people to discover more about themselves.
In a very real way, genetic information is no different from other medical facts. We have easy access to our blood pressure numbers, and no one is suggesting that we shouldn’t, despite the fact that interpretation of blood pressure numbers can be complicated. (For example, should you reduce salt intake, or take statins to lower your blood pressure, or change your lifestyle in other ways?)
I hope the FDA allows Pathway to sell its DNA test at Walgreens. Personally, I think the value of the tests they’re offering today is mostly to satisfy curiosity – we don’t have treatments for most of the genetic diseases they can detect, so there's not much you can do about them. Some of the results will be useful to some people, though, and over time we’ll develop new treatments that will make more genetic tests valuable. Now that Pandora’s box is open, we can’t close it again. And we shouldn’t.
The first therapeutic cancer vaccine
Time for some good news. This past week, the FDA approved the first-ever vaccine to treat cancer. The treatment is called Provenge, developed by a small company named Dendreon , and it’s designed for advanced prostate cancer. The idea of developing a vaccine to treat cancer has been around for a long time, but never before has a treatment made it all the way to the clinic. This is very good news, especially if it heralds many more new vaccine-like cancer treatments.
So what is Provenge? Unlike most vaccines, it’s not a simple shot. A typical vaccine contains some form of the infectious agent, usually a virus or a bacterium, which we administer as a shot (or a jab, as our friends in the UK put it). The virus is usually either dead or modified so that it cannot cause disease, but enough of it remains to “train” your immune system. The immune system reacts to the dead virus and, though the amazing mechanism of acquired immunity, is then primed to attack and eliminate a live virus (or bacterium) if it ever shows up again.
A cancer vaccine is not nearly so simple. Unfortunately, we can't just grind up a bit of cancerous tissue, treat it somehow, and then inject it. The reason this new treatment is called a vaccine is that, like conventional vaccines, it trains a person’s immune system to recognize and attack the unwanted invaders – in this case, the cancerous cells. But this presents a big problem, because cancer cells are the body’s own cells. How do we get the immune system to attack them without triggering a much broader, possibly damaging, autoimmune response? As Bruce Goldman explained in a 2002 article, the trick is to find some feature on the surface of cancer cells that marks them as abnormal, and then to get the immune system to target that feature (or antigen).
Luckily, because cancer cells have gone haywire, they tend to have lots of abnormal surface features. But every cancer tends to be a little bit different, which means that these antigens are different in each patient. That means that the vaccine has to be customized for each patient, an exquisitely difficult process. Cancer vaccines use another cell type, called dendritic cells, to create a vaccine that is customized for each patient: first a bit of the patient’s tumor has be collected, and then grown with dendritic cells, and then re-injected, which alerts the immune system to attack just the tumor cells. This is the process used in Provenge.
(Note that Provenge is the first therapeutic vaccine because it is used to treat cancer. There is already a preventive cancer vaccine available: Gardasil, which prevents some types of cervical cancer. But that’s another topic.)
Three years ago, the FDA rejected Dendreon’s first application, asking for more evidence that it worked. The company then conducted a new placebo-controlled, double-blind trial and found that Provenge (sipuleucel-T) extended patients' median survival time by 4.1 months. That might not sound like much, but for men with advanced prostate cancer, it’s better than anything else available. And this is only the additional survival time. Five years ago, Dendreon reported on an earlier study in which “34 percent of patients receiving Provenge were alive at 36 months compared to 11 percent of patients receiving placebo.” In that earlier study, median survival time increased by 4.5 months, essentially the same result as the newer study.
One disappointing part of the recent news is the price: Dendreon announced that it will charge $93,000 per patient. That’s an awful lot of money, and it’s not clear yet if insurance will cover this. No doubt desperate patients will be glad to pay for the few extra months of life. In fact, Dendreon said that its price is based on the assumption that people will pay $23,000 per extra month of life, based on the cost of other cancer drugs. We can only hope the price will drop dramatically as the company improves its manufacturing process, but (unfortunately) lower production costs don’t always get passed on to patients.
These are not miraculous results: Provenge is not a cure. And unlike conventional vaccines, it doesn’t prevent the disease; instead, it treats people who are already sick. But it’s a different type of treatment, and we can be optimistic that this opens the door for many improvements in cancer vaccines, not just for prostate cancer but for many other types of cancer. Trials are already under way for several other types of cancer, and the future looks very encouraging.
So what is Provenge? Unlike most vaccines, it’s not a simple shot. A typical vaccine contains some form of the infectious agent, usually a virus or a bacterium, which we administer as a shot (or a jab, as our friends in the UK put it). The virus is usually either dead or modified so that it cannot cause disease, but enough of it remains to “train” your immune system. The immune system reacts to the dead virus and, though the amazing mechanism of acquired immunity, is then primed to attack and eliminate a live virus (or bacterium) if it ever shows up again.
A cancer vaccine is not nearly so simple. Unfortunately, we can't just grind up a bit of cancerous tissue, treat it somehow, and then inject it. The reason this new treatment is called a vaccine is that, like conventional vaccines, it trains a person’s immune system to recognize and attack the unwanted invaders – in this case, the cancerous cells. But this presents a big problem, because cancer cells are the body’s own cells. How do we get the immune system to attack them without triggering a much broader, possibly damaging, autoimmune response? As Bruce Goldman explained in a 2002 article, the trick is to find some feature on the surface of cancer cells that marks them as abnormal, and then to get the immune system to target that feature (or antigen).
Luckily, because cancer cells have gone haywire, they tend to have lots of abnormal surface features. But every cancer tends to be a little bit different, which means that these antigens are different in each patient. That means that the vaccine has to be customized for each patient, an exquisitely difficult process. Cancer vaccines use another cell type, called dendritic cells, to create a vaccine that is customized for each patient: first a bit of the patient’s tumor has be collected, and then grown with dendritic cells, and then re-injected, which alerts the immune system to attack just the tumor cells. This is the process used in Provenge.
(Note that Provenge is the first therapeutic vaccine because it is used to treat cancer. There is already a preventive cancer vaccine available: Gardasil, which prevents some types of cervical cancer. But that’s another topic.)
Three years ago, the FDA rejected Dendreon’s first application, asking for more evidence that it worked. The company then conducted a new placebo-controlled, double-blind trial and found that Provenge (sipuleucel-T) extended patients' median survival time by 4.1 months. That might not sound like much, but for men with advanced prostate cancer, it’s better than anything else available. And this is only the additional survival time. Five years ago, Dendreon reported on an earlier study in which “34 percent of patients receiving Provenge were alive at 36 months compared to 11 percent of patients receiving placebo.” In that earlier study, median survival time increased by 4.5 months, essentially the same result as the newer study.
One disappointing part of the recent news is the price: Dendreon announced that it will charge $93,000 per patient. That’s an awful lot of money, and it’s not clear yet if insurance will cover this. No doubt desperate patients will be glad to pay for the few extra months of life. In fact, Dendreon said that its price is based on the assumption that people will pay $23,000 per extra month of life, based on the cost of other cancer drugs. We can only hope the price will drop dramatically as the company improves its manufacturing process, but (unfortunately) lower production costs don’t always get passed on to patients.
These are not miraculous results: Provenge is not a cure. And unlike conventional vaccines, it doesn’t prevent the disease; instead, it treats people who are already sick. But it’s a different type of treatment, and we can be optimistic that this opens the door for many improvements in cancer vaccines, not just for prostate cancer but for many other types of cancer. Trials are already under way for several other types of cancer, and the future looks very encouraging.