Scientists have long known that the same type of cancer can play out in very different ways, from a genetic perspective, in one patient compared to another. But this new research shows that, even within the same patient—even within the same tumor—different samples of cancer cells have more genetic differences than they have similarities.

That's a very big deal. It means that cancer cells aren't just cells that grow uncontrollably. They also mutate. Which means that they evolve. That fact has serious implications for cancer treatment. Just like bacteria can evolve to become resistant to antibiotics, cancer cells can evolve resistance to the treatments we throw at them. At Not Exactly Rocket Science, Ed Yong explains how this discovery fits into the bigger picture of why curing cancer is so damned difficult:

For a start, cancer isn’t a single disease, so we can dispense with the idea of a single “cure”. There are over 200 different types, each with their own individual quirks. Even for a single type – say, breast cancer – there can be many different sub-types that demand different treatments. Even within a single subtype, one patient’s tumour can be very different from another’s. They could both have very different sets of mutated genes, which can affect their prognosis and which drugs they should take.

And now we know that's true within a tumor, as well. At the Cancer Research UK blog (where Ed used to work), Henry Scowcroft has a nice summary of how this one discovery explains three perplexing problems we've long had with cancer cells:

Firstly, cancer is very difficult to cure after it has spread. This is despite years of progress in chemotherapy and radiotherapy, two techniques that can offer respite to people with advanced cancer.

Secondly, most advanced cancers eventually become resistant to every type of drug used to treat them – both ‘traditional’ chemo and these newer agents. This is quite extraordinary: tumours can work out how to cope with chemicals that they’ve never ‘seen’ before – a biological superpower far beyond that of infectious diseases. Just consider how it’s taken ‘multidrug resistant’ bacteria like MRSA decades to evolve. Yet cancers can do this in a matter of months or even weeks. How?

And finally, researchers haven’t yet managed to develop tests to predict how a patient’s disease will progress, nor monitor their progress (a field called ‘biomarker’ research) – this is despite years of research, and a lot of tantalising pilot studies. Sometimes researchers detect a promising ‘signal’ by looking at samples from a handful of patients, only for this to disappear in larger numbers of people.

Read Ed Yong's full story on this research.

Read Henry Snowcroft's full story on this research.

H5N1, the famous bird flu, is deadly to humans. Of the 566 people who have contracted this form of influenza, 332 have died. But, so far, the people who have caught bird flu don't seem to have contracted the disease from other humans, or passed it on. Instead, they got it from birds, often farm animals with whom the victims were living in close contact. H5N1 was first identified 14 years ago, and there's never been a documented case of it being passed from person to person.

But that doesn't mean such a leap is impossible.

That's because of how the influenza virus works. Influenza is made up of eight pieces of RNA, containing 10 genes, and they all replicate independently of one another and there's no system for error correction*. That means you have more opportunity for mutations to arise that change what the virus does and who it can infect. Think of it like dice. Genetic replication is like putting a die in a jar, shaking it up and seeing what you get. Everybody does that. But influenza has eight die, not one. So it accumulates mutations faster. As a bonus, influenza viruses that infect the same host can share genes—essentially creating a baby virus that carries traits from different parents.

That's why, despite 14 years of relatively low-risk behavior, scientists are still concerned about what H5N1 might do in the future. All it would take, theoretically, is the right roll of the dice, and suddenly you have a flu virus with a 60% kill rate that can pass from person to person.

At least, theoretically. Could that actually happen? And, if so, how likely is it that the "right" bad combination of genes will come up? You can see why these are important questions to ask, and that brings us to the controversy.

Studying the genetics of H5N1 is nothing new. Its genome has been sequenced since 2004, for instance. But, until 2011, nobody had ever tested out a pretty fundamental idea in the control and management of H5N1: The theory that its genetics prevented it from simultaneously being both super deadly and passed from person to person.

That theory hinged on what we know about one of the proteins in H5N1. Specifically, the protein designated H5. Here's the LA Times:

Strains carrying the H5 type of a key influenza protein that helps the virus bind to cells in a host had never evolved to travel through the air from person to person. Even if H5N1 did evolve such an ability, some researchers reasoned that it might do so at the expense of its ability to take hold deep in the lung. And that would make it less lethal.

As one scientist described it to the LA Times, this theory was, basically, "We've not seen this happen before so it can't happen." But that's not a particularly strong basis on which to pin all your fears about a global pandemic.

That's why researchers in Europe and the U.S. decided to try something risky—see whether they could prompt existing H5N1 viruses to mutate into the very thing everybody's been dreading. Nobody knows a lot about this research, but, at Slate.com, Carl Zimmer explains what is known:

They’ve carried out their experiments on ferrets, which respond to flu viruses much like humans do. What few details we know of the unpublished research comes from a talk Dutch virologist Ron Fouchier gave in August at a virology conference, along with subsequent news reports. Fouchier began the experiment by altering the H5N1 virus’s genes in two spots. Then he passed the virus from one ferret to another, allowing the virus to mutate and evolve on its own inside the animals. After several rounds, Fouchier ended up with an H5N1 virus that could spread through the air from one ferret to the other. If unleashed—and if proven capable of spreading from human to human with the same high mortality rate—it could make the deadly 1918 pandemic look like a pesky cold.

So that's one part of the controversy. Was this a responsible thing to do?

On the one hand, zomgwe'reallgonnadierunhide, right? On the other, this research has already taught us something really, really important. Not only can H5N1 make the leap to mammal-to-mammal transmission, but it did so faster and easier than the researchers had guessed. Knowing that matters, because it could help public health officials make better plans for where to use limited resources, and it could help other scientists figure out a way to fight a human transmissible H5N1 pandemic if it did happen in nature. But, if I may flip the waffle back over again, there are some legitimate scientists who don't think the benefits outweigh the risks of creating this thing. Carl Zimmer again:

Ian Lipkin, the director of the Center for Infection and Immunity at Columbia University, believes there’s no reason to assume that the mutations that arose in Fouchier’s experiments would be the ones that would arise out in the real world. “On the other hand,” Lipkin says, “publishing this information would give people a roadmap to creating Frankenstein viruses.”

And that brings us to the other part of the controversy: What to do with Fouchier's research.

This is where the government gets involved. These studies were funded by the National Institutes for Health. When NIH got the papers, they passed them on to the National Science Advisory Board for Biosecurity. On December 20, the NSABB recommended that Fouchier's study, and a similar one conducted by the University of Wisconsin's Yoshihiro Kawaoka, only be published once key data and details are removed, effectively rendering the studies un-reproducible.

The board can't technically force this. But the board is also a big deal and so Science, Nature, the NIH, and the paper's authors are all listening. That's why the papers haven't actually been published yet. The people involved are still figuring out how to handle them.

This matters a lot. Reproducibility—being able to read another scientist's research paper and independently test out their conclusions—is a key part of how science works. Remove that element, and it becomes harder to verify claims like this, not to mention much harder to actually get the benefits out of this risky research. The people involved are trying to work out a system under which qualified scientists could have access to the full data, but others say that isn't good enough. Especially considering the fact that H5N1 wouldn't make the best bioterrorism tool, anyway. Peter Christian Hall writing for Reuters:

[No one in the history of biological weapons] ever tried to weaponize a flu strain—for good reason.

Influenza in general is an equal-opportunity menace, particularly dangerous when a strain is so unfamiliar that humanity lacks immunity to it. This would put at great risk anyone trying to assemble a pandemic H5N1 to launch at “target” populations. Indeed, such an attack would unleash global contagion that would swiftly and inevitably incapacitate an aggressor’s own people. Influenza doesn’t respect borders.

Even arguably irrational terrorists like Aum Shinrikyo never got into anything near as notoriously unpredictable and uncontrollable as the flu, Hall writes. Of course, his argument is pretty similar to the one scientists used to use to reassure themselves that H5N1 couldn't be both deadly and human transmissible: We've not seen this happen before, so it won't.

Of course, it's also worth pointing out that these experiments were a lot more technologically complex than the short description here makes them sound. This isn't just about taking a bunch of ferrets and making them sick. It required some serious lab equipment that not just anybody has access to.

Moreover, this isn't the first time scientists have made a deadly flu virus in the lab. Back in 2005, a team reverse-engineered the 1918 pandemic flu. After a lot of debate, their research was eventually published in full, reproducible form. Peter Palese was one of the scientists on that team, and he's written an essay on Nature about his experience, as part of a plea to publish the H5N1 research in full, too.

He makes a case both for the importance of risky research, and for why all science (even kind of scary science) needs to remain open source.

During our discussions with members of the NSABB, we explained the importance of bringing such a deadly pathogen back to life. Although these experiments may seem dangerously foolhardy, they are actually the exact opposite. They gave us the opportunity to make the world safer, allowing us to learn what makes the virus dangerous and how it can be disabled. Thankfully, the discussions were largely constructive — within a week, the NSABB recommended that we continue to study the virus under biocontainment conditions, and publish the results so that other scientists could participate in the research. After we published our full paper in 2005, researchers poured into the field who probably would not otherwise have done, leading to hundreds of papers about the 1918 virus. As a result, we now know that the virus is sensitive to the seasonal flu vaccine, as well as to the common flu drugs amantadine (Symmetrel) and oseltamivir (Tamiflu). Had we not reconstructed the virus and shared our results with the community, we would still be in fear that a nefarious scientist would recreate the Spanish flu and release it on an unprotected world. We now know such a worst-case scenario is no longer possible.

I make the same argument today that we made in 2005 — publishing those experiments without the details is akin to censorship, and counter to science, progress and public health. ... Giving the full details to vetted scientists is neither practical nor sufficient. Once 20–30 laboratories with postdoctoral fellows and students have such information available, it will be impossible to keep the details secret. Even more troublesome, however, is the question of who should decide which scientists are allowed to have the information. We need more people to study this potentially dangerous pathogen, but who will want to enter a field in which you can't publish your most scientifically interesting results?

Back in December, researchers at Caltech posted a research paper to arXiv that attempts to explain why the shape and structure of snowflakes change significantly depending on relatively small shifts in temperature.

In order to study this, they had to grow snowflakes in laboratory conditions. It was not an easy thing to figure out how to do. On his Snowcrystals page, physicist Kenneth G. Libbrecht show you how it's done.

There are many ways to grow snowflakes, but my favorite starts with something called a vapor diffusion chamber. This is essentially nothing more than an insulated box that is kept cold on the bottom (say -40C) and hot on the top (say +40C). A source of water is placed at the top, and water vapor diffuses down through the box, producing supersaturated air. The cold, supersatured air at the center of the chamber is ideal for growing ice crystals.

While working with this diffusion chamber, we rediscovered a wonderful technique for growing synthetic snow crystals that was first published in 1963 by meteorologist Basil Mason and collaborators [1]. One starts by putting a wire into the diffusion chamber from below, so that small ice crystals begin growing on the wire's tip. Then apply a high voltage to the wire, say +2000 volts, and voila -- slender ice needles begin growing from the wire.

Video Link

This is a great find by Not Exactly Rocket Science's Ed Yong. A tourist and a couple of researchers from the California Academy of Sciences have documented an instance of Pacific-dwelling jawfish hiding from predators by blending into the stripes of well-known camouflage guru, the mimic octopus.

This relationship is probably a rare occurrence. The black-marble jawfish is found throughout the Pacific from Japan to Australia, while the mimic octopus only hangs around Indonesia and Malaysia. For most of its range, the jawfish has no octopuses to hide against. Instead, Ross and Rocha think that this particular fish is engaging in “opportunistic mimicry”, taking advantage of a rare chance to share in an octopus’s protection.

Video Link

Thanks, Atvaark!

Eric Kostelich is one of the researchers who is trying to change that, by approaching the problem of brain cancer  from a new angle. Kostelich is a mathematician. In particular, he's interested in how we can use math to better predict the behavior of complex and chaotic systems. Right now, this mostly means that he studies the weather. In fact, he's part of a team that developed a new algorithm for weather prediction, called the Local Ensemble Transform Kalman Filter. But Kostelich thinks that the LETKF could have applications outside the nightly news.

In a recent study, published December 21 in Biology Direct, he joined forces with cancer researchers, to see whether the statistical methods that make chaotic weather patterns more predictable could do the same thing for chaotic behavior in cancer cells. The results are promising. A couple of weeks ago, I spoke to Kostelich to find out more about the history of forecasting uncertainty, how algorithms like LETKF work, and what we might learn if we apply these systems to cancer. 

 Maggie Koerth-Baker: When you set out to apply the methods used to forecast the weather to cancer, why did you choose brain cancer? 

Eric Kostelich: Partly it's because I had a family member with a brain tumor. The scientist in me got interested because these really are terrible tumors. Getting to know a number of clinicians, it seemed to me that new ideas, anything that could help people live a little better would be welcome.

From a math standpoint these tumors are interesting in that they don't metastisize on the central nervous system. For instance, melanoma is a process that starts in a mole on surface of skin but spreads to the liver and other vital organs. You have to look at it throughout the entire body, and that's a daunting task. But a brain tumor tends to stay in the brain. That keeps it more mathematically attractive for our initial studies. And the brain also has some interesting geometry, with folds and so on, and with different functions in different places. The liver and kidneys are relatively heterogenous, but where a tumor is in the brain really affects what symptoms you see. My partner in this thought it would be really useful to be able to say, for an individual patient, based on studies of patients with similar tumors, there's a 60% chance that the tumor is more likely to grow in one direction, rather than another. They might be able to give a dose of radiation therapy in that one direction.

If you could help someone live a couple extra months, that would be a big advance for this type of cancer. Patients usually only live 12-14 months. Ted Kennedy suffered from this and his experience was typical. He died about 14 months from diagnosis. We've made a lot of strides in treating other kinds of cancer, but our approach to brain cancer hasn't changed much.

MKB: Why are brain cancers so difficult to treat?

EK: The brain isn't easily accessible. Between the brain and the outside, you've got the skull and accessing it involves drilling a hole. Plus, it's your brain. With a breast cancer you can remove a breast and a margin of tissue and still live. You can live without an arm or a breast. But you can't just remove a big chunk of the brain without crippling the patient. One of the objectives of treatment is to not make the patient's neurological condition any worse than tumor already has made it.

MKB: In this study, you used an algorithm called the Local Ensemble Transform Kalman Filter. What does all that mean? What does this algorithm do when applied to weather?

EK: Meteorologists never make just one prediction. They make many. In any of these models, you always have a grid of points on which you're trying to approximate behavior. To run that model on the computer you put in models for all the grid points, but you can't actually measure them all. What I mean is, you can measure the temperature and barimetric pressure and so forth at one point on the ground and plug it in. But the model also has another grid point above that one, high up in the air, and more points on up into the atmosphere. You can't measure the data at all of those. Another example, if you look at high impact weather like a major hurricane, they get started over remote tropical ocean where it's hard to get any measurements at all. Satellites are a big help, but there's always uncertainty. And because there's uncertainty, you get the idea of an ensemble forecast.

Ensemble forecasts go back to Edward Lorentz. [He's a pioneer in chaos theory—MKB] You run a number of forecasts with slightly different realizations of the numbers at all the different grid points. Forecast uncertainties depend not only on time, but also on space. You might have quite a bit of uncertainty about where a hurricane will end up in five days. But quite a bit of certainty about Phoenix being sunny for five days in June. If you look at many forecasts, you can get a handle of both where your uncertainties are and what magnitude they are. 

MKB: What makes it different from other algorithms used for making predictions about uncertain systems?

EK: You're trying to get mathematical combinations of forecasts that best fit observed conditions. Weather models are updated every six hours. Your uncertainties tend to grow with time. You can forecast tomorrow, but a week from now is more iffy. So the models have to be updated on a regular basis. By doing things locally you get better computational efficiency. We thought, "Well, if we can do this for the weather ..."

Our system outperforms the Weather Service's system by quite a bit. The Brazilian government is actually using our approach for their next gen weather prediction systems. What we try to exploit in our approach is, in a mathematical sense, the uncertainties in a forecast. Uncertainties tend to lie in certain directions. All these mathematical models are partial differential equations in dynamical systems. In math, we have a concept we call a phase space. That's "phase" as in phases of the moon. It's a math abstraction and it's where we think of all the math action taking place. In the case of the weather, it appears that the uncertainties in weather models lie primarily in certain directions of the abstract phase space. Our approach takes advantage of that in a clever way. Because we know where the uncertainties are more likely to be, we can pay more attention to those places.

MKB: The ideas here—combining new observations with prior forecasts, and paying the most attention to where you know the most uncertainty is likely to be—these are things that can seem like a bit of an obvious thing to laypeople. How new are these ideas really? Is there something that makes this more special and surprising than it seems on the surface?

EK: The notion of data assimilation, combining observations and prior forecasts, has been an integral part of weather forecasting for several decades now. Computers were used to do this on a regular basis, starting around the mid 1960s. By that point they were powerful enough that you could build a realistic enough grid to say something about the weather. You and I take weather forecasting for granted. But the reality is that this is one of the great triumphs of modern science.

Think about Hurricane Katrina. It was known three days in advance that this hurricane would come close to New Orleans. Thirty years ago, we couldn't have been able to tell you that. Today we can evacuate a few hundred miles of coastline instead of telling the entire Gulf region, "This is coming and we don't know where." That's a huge advance.

There's some really interesting history on this. The term forecast goes back to Robert Fitzroy, he was the captain of the Beagle, the ship that Darwin traveled on to the Americas in the 1830s and 40s. He was very intersted in questions of weather and storms because he'd lost several crewmen in a gale. The ship basically tipped over. He saved the ship, but lost several lives. When he was back in England he was appointed meterological statist. He was appointed to keep all the records for the crown. So, by the late 1850s, the telegraph had come in and Fitzroy had people at all the ports telegraph in the weather information to him. He looked at patterns of temperature and pressure rising and falling and was able to combine those patterns with new observations and say, "There's a storm coming in. Stay at port." This was the beginning of forecasting. Shipwrecks fell by half in a few years. 

MKB: So where do you make the connection between all of this, and something like brain cancer? 

EK: No forecast is perfect. In Fitzroy's day, if the storm didn't materialize, then there were complaints. People lost money by staying in port and Fitzroy actually got into political trouble. They stopped doing forecasting for a while until a fisherman's lobby brought him and his methods back.

Now fast forward 150 years. Say you have a brain tumor. What you'd like to know is, "What is going to happen to me?" But doctors are really very much where Fitzroy was 150 years ago before he invented forecasting. It's like, stick your finger up in the air and make a guess. Nowadays, for weather, we have mathematical models that, combined with satellites, can predict hurricanes before they materialize. We can tell several days in advance where the storm will go. Doctors can't do anything near that with cancer. All they can do is look at a scan and some bloodwork and say we'll see you in a couple months. But we're on the verge of being able to gather lots more information about what's going on in the human body. 

What we'd like to do with this oncoming data deluge, what we're trying to do, is devise math tools that will help clinicians make sense of all the new data and associate probabilties with that data. Then they can tell people something more useful. Not perfect. But useful.

MKB: I think most lay people operate under the assumption that weather prediction like this isn't very accurate, beyond a day or so ahead of time. You're wanting to do cancer prediction over 60-day cycles. Why would that kind of time frame be reliably accurate enough to matter?  

EK: Weather service looks hard and long at that question. One way in which you can assess the goodness or lack thereof of the forecast is to say, "I'm going to predict the weather two, three, four days out. Then you go out and measure on those days and compare the reality to the forecast. The bigger the difference, the worse the forecast. By that measure, forecasts today made 3-4 days out are as accurate as a 36-hour forecast was 30 years ago. So the weather forecasts are more accurate than people give them credit for. It's just that when you blow it that's what people remember.

A famous case was Veterans' Day a few years ago in Washington DC.  The Weather Service forecasted a dusting of snow, but there ended up being something like 14 inches. Basically, the snow was 100 miles off from where they thought it would be. A 100-mile error, 24 hours out, that isn't so bad really in a global perspective, but the local impact is very great. People remember that. On the flip side, though, in 1900 a hurricane hit Galveston, Texas. The Cubans had telegraphed DC to tell us that there was a storm heading West. But we blew the Cubans off and there were no warnings for Galveston. 8000 people drowned. We still have bad hurricanes today, but we don't have 8000 drowning because they don't know it's coming. In that respect, we're doing pretty good. But it's still not perfect.

Our basic approach here, in thinking about cancer in general and brain cancer specifically, is could we adapt the accuracy of a 3-4 day weather forecast for a month or two or three. So that the models show what is likely to happen to an individual patient's tumor. It's a fair question about whether we can move that timeline up. But it works because of differences in what you're forecasting. For weather, the uncertainties tend to double every couple of days. As far as we understand cancer, the uncertainties you have in the state of a tumor, they don't double every two days. They double over a month or two. It's a different mathematical beast than the weather. A couple of months for cancer is like a couple of days for the weather. Now, they can vary quite a bit in how aggressive they are. There are cases in the literature where in some patients the tumors double in size in a couple weeks. But more typically the doubling times are on the order of a month or two.

On the other hand, forecasting cancer can be harder than forecasting the weather. In the case of the weather, air is a fluid, and there's a couple hundred years of physics that go into understanding how fluids move in laboratory conditions. If you're going to write a weather model there's no doubt what equations you start with. In the case of cancer, we don't know how glioblastoma cells really work very well. It's a much greater challenge to write that mathematical model because our understanding is much less complete. We're trying to take into account that whatever model we write down is likely to be off. Possibly by quite a bit. But the question is, "Can data assimiliation system make a clinically useful forecast?" It doesn't have to be perfect to be useful.

MKB: The brain cancers you're looking aren't really treatable. Like you say, most people die from them within 14 months. What's the benefit, then, of having a more accurate prediction of how they will spread? If you still can't treat the cancer, what does it help to know how it will behave? 

EK:  My understanding, and from personal experience with a family member, is that you're right, this isn't going to cure cancer in general or glioblastoma specifically. But one of the real goals of treatment is to help patients live as well as possible for as long as possible. The age of highest incident for the type of brain cancer we studied is between 40 and 65. If this result allows you to live two months longer than you otherwise would maybe that makes the difference between seeing your daughter get married or not. We can't prevent the inevitable, but we might help them live better or longer. If we can develop good enough mathematical models and be able to tell patients that going through another round of chemo isn't likely to help, then they can decide to spend that time with family instead of in the hospital. That's beneficial in it's own way.

Eric Kostelich's research is available to read, for free, online. 

A recently published study found a correlation between higher rates of coffee drinking in women and decreased risk of depression. Naturally, that finding made headlines. But blogger Scicurious has a really nice analysis of the paper that picked up a significant flaw in the way the data is being interpreted. There was a correlation between drinking more coffee and a lowered risk of depression. But that wasn't the only correlation the researchers found—just the only correlation they made a big deal of in their conclusions.

On her blog, Scicurious lists the other correlations and explains why it's hard to draw any solid conclusion from this data set:

1) Smoking. The interaction between depression risk, smoking, and coffee consumption was “marginally” significant (p=0.06), but they dismiss it as being due to chance because it was “unexpected”. Um. Wait. Nicotine is a STIMULANT. It is known to have antidepressant like effects in animal models (though the withdrawal is no fun). This is not unexpected.

2) Drinking: heavy coffee drinkers drink more. But note that they don’t say that drinking coffee puts you at risk for drinking alcohol.

3) Obesity: heavy coffee drinkers are, on average, thinner, but not more physically active. They do not conclude that coffee drinking prevents obesity.

4) Church going: heavy coffee drinkers are less likely to go to church. Less likely to go to church, less likely to develop depression…heck, forget depression, maybe coffee prevents religion now! Now THAT would be a heck of a finding.

Here’s the thing. I do believe that high coffee consumption correlates with decreased risk of depression. But a lot of other things do as well. I am not convinced that the high coffee consumption wasn’t part of a lifestyle that correlated with decreased risk of depression, maybe they have stronger support networks or less incidence of depression in the family. It could be many other things.