There’s a good rule of thumb about headlines (other than Betteridge’s Law of Headlines) that I use when perusing articles. It’s particularly useful for evaluating headlines about medical and science stories. Basically, if a headline says something like, “everything you know about X is/might be/could be wrong” or “everything scientists know about X is/might be/could be wrong,” it’s a highly reliable indication that much of what is in the article that follows is very likely to be unmitigated, grade-A bullshit. I realize that it might be confirmation bias on my part (I am, after all, a skeptic), but it’s a rule of thumb that has rarely let me down. It is, after all, a very lazy trope on the part of editors and journalists designed to pique the reader’s interest and imply that there’s been some sort of radical new finding that’s overturned much of what scientists thought about a topic before. Science, of course, doesn’t usually work that way. Findings that totally overturn a scientific paradigm are rare; most new findings are evolutionary, not revolutionary. This is good for science. It’s not so good for news stories and headline writers.
My rule definitely didn’t let me down yesterday, too, as I started seeing on my social medial an article in Newsweek by Jessica Wapner that definitely follows it. Oh, sure, there’s a paragraph or two out of a very long article presenting one scientist criticizing the crank idea presented, but for the most part the article is all credulous praise. Entitled, A New Theory on Cancer: What We Know About How It Starts Could All Be Wrong, the title fits the pattern, as does the article that follows. It is unmitigated, grade-A bullshit. Not only that, to the lazy trope of lionizing a “scientific maverick” (whose ideas are, of course, what show us that “what we know about how cancer starts could all be wrong”) who is most unworthy of lionization. Unsurprisingly, it is a theoretical physicist. It could be confirmation bias on my part again, but it sure does seem that physicists seem particularly prone to entering a new field, coming up with a new “insight” in it, and then wondering why all the scientists there hadn’t thought of the insight he’s had and indeed reject it. In any case, it’s a physicist we have met before, Paul Davies, who three years ago with his partner in crime Charles Lineweaver (another physicist, in this case an astrophysicist) made a splash with his idea that cancer is a throwback to an ancient genetic “sub-routine” where the mechanisms that usually instruct cells when to multiply and die malfunctions, thus forcing the cells to revert back to a default option that was programmed into their ancestors long ago, or, as it’s called, an atavism, a “reawakening” of ancient genetic subroutines. I wrote about it then. I described why his ideas are wrong and how they’ve inspired a quack. So did biologist P.Z. Myers (multiple times) Nothing’s changed since then, except that Wapner cranked up the worship of the Brave Maverick Scientist to a cringe-inducing 11.
Paul Davies knows what’s wrong with cancer research: too much cash and too little forethought. Despite billions of dollars invested in fighting this disease, it has remained an inscrutable foe. “There is this assumption that you can solve the problem by throwing money at it,” he says, “that you can spend your way to a solution.” Davies, a theoretical physicist at Arizona State University (ASU)—and therefore somewhat of an interloper in the field of cancer—claims he has a better idea. “I believe you have to think your way to a solution.”
Bloody hell. As a cancer surgeon and biologist, I couldn’t help but be monumentally pissed off when I read this paragraph. The hubris just oozes from every word of Davies’ quoted. So does the condescension and contempt for us poor, poor, oncologists, surgeons, and cancer biologists, who (or so Davies apparently thinks) are so deeply embedded into the existing paradigm that they can’t appreciate the brilliance of his insights. “I believe you have to think your way to a solution”? What does he think cancer biologists have been doing for well over a century now, dating back to the 1800s? Sitting around in their labs waiting for that sweet, sweet grant money to flow in, so that they can do whatever experiment that pops into their heads on any given day?
He also betrays the bias of his background as a theoretical physicist. Let me give him an education. To make progress against cancer takes a lot more than “thinking.” Yes, thinking is important, but it’s worthless without actual experimental evidence from the laboratory, which is then translated into the clinic and validated through clinical trials. This is a process that can take decades, of which “thinking” is only the first step. The hard work of cancer biology, like any science, is not necessarily coming up with ideas and insights. The real thinking that has to be done is to design strategies to assess which ideas lead to testable hypotheses that can result in a better understanding of cancer and, at least as importantly, identify mechanisms that can be targeted for therapy. The hubris that Davies demonstrates is breathtaking, and Wapner doesn’t call him out on it in the least.
If you don’t believe me that Davies is peddling grade-A bullshit, then check out this next passage:
Over the course of several years spent pondering cancer, Davies has come up with a radical approach for understanding it. He theorizes that cancer is a return to an earlier time in evolution, before complex organisms emerged. When a person develops cancer, he posits, their cells regress from their current sophisticated and complex state to become more like the single-celled life prevalent a billion years ago.
But while some researchers are intrigued by the theory that cancer is an evolutionary throwback, or atavism, plenty more think it’s silly. That theory suggests that our cells physically revert from their current form—a complex piece in the even more complex puzzle that makes a lung or a kidney or a brain—to a primitive state akin to algae or bacteria, a notion that seems preposterous to many scientists. Yet gradually, evidence is emerging that Davies could be right. If he is—if cancer really is a disease in which our cells act like their single-celled ancestors of eons ago—then the current approach to treatment could be all wrong.
“Gradually evidence is emerging that Davies could be right”? No. It’s not. Really, it’s not. Nothing Davies has thought of is remotely new. As I discussed a long time ago, it is not surprising that the hallmarks of cancer described by Douglas Hanahan and Robert Weinberg in 2000 and updated in 2011. The original six hallmarks were:
- sustaining proliferative signaling
- evading growth suppressors
- resisting cell death
- enabling replicative immortality
- inducing angiogenesis
- activating invasion and metastasis
Cancer biologists have known for a long time that the genes involved in developing cancer control cellular processes like these, and these processes are very ancient ones. Just because these processeas are ancient, however, does not mean that Davies is correct. His first line of argument is that cancer is found in virtually all multicellular organisms, which is true but basically irrelevant. Then he argues:
The evidence that cancer is an evolutionary regression goes beyond the ubiquity of the disease. Tumors, says Davies, act like single-celled organisms. Unlike mammalian cells, for example, cancer cells are not programmed to die, rendering them effectively immortal. Also, tumors can survive with very little oxygen. To Davies and his team, which includes Australian astrobiologist Charles Lineweaver and Kimberly Bussey, a bioinformatics specialist at ASU, that fact supports the idea that cancer emerged somewhere between 1 billion and 1 and a half billion years ago, when the amount of oxygen in the atmosphere was extremely low.
Tumors also metabolize differently from normal cells. They convert sugar into energy incredibly fast and produce lactic acid, a chemical normally resulting from metabolism that takes place in the absence of oxygen. In other words, cancer cells ferment, and scientists don’t know why. This phenomenon is known as the Warburg effect, named for Otto Warburg, a German biochemist who won a Nobel Prize in 1931 for his discoveries about oxygen and metabolism. Up to 80 percent of cancers display the Warburg effect. Researchers know that many cancers depend on the Warburg effect for their survival, but they don’t know why. To Davies, the strange way in which tumors metabolize also speaks of cancer’s ancient past: They are behaving as if there were no oxygen available.
Yes, tumor cells are not programmed to die (a process called apoptosis, or programmed cell death), an essential process in building multicellular organisms, particularly of keeping cells under control and from proliferating outside of control of the organism. However, as P.Z. Myers notes, there are normal cells that have the same ability, such as germ line cells. As a cancer biologist myself, I can’t help but note that another relevant cell type is the stem cell. Stem cells are cells that can differentiate into any cell type and have in essence unlimited replicative potential. It would make more sense to say that cancer cells in some ways resemble stem cells, and, yes, cancer stem cells are a hot (if controversial) area of research. Just ask Max Wicha, a pioneer in the field. Unlimited replicative ability is not an atavistic property.
As for surviving in very little oxygen, mammalian cells do the same thing. It is not an atavistic property. When oxygen tension is low, they revert to glycolysis, which doesn’t require oxygen. The disadvantage is that glycolysis produces much less energy per molecule of glucose than oxidative metabolism does. It is true that many tumor cell types do “prefer” glycolysis over aerobic metabolism. It’s something that Otto Warburg observed back in the late 1920s, and it’s the reason that PET scans work. PET scans use radiolabeled glucose Tumor cells, because of their dependence on glycolysis, use a lot of glucose, which means that areas with tumor cells “light up” on PET, allowing the detection of tiny tumor deposits. I’ve discussed the Warburg effect on multiple occasions. Again, there’s nothing “atavistic” here.
It’s also not as though we don’t understand some of these things as well. The Warburg effect is associated with the PKM2 isoform of a key enzyme in glycolysis, pyruvate dehydrogenase. Attempts to use inhibitors of aerobic glycolysis in the clinic or other drugs that target the Warburg effect have not been dramatically successful.
Although PZ did a fine job deconstructing this nonsense, being a cancer physician, I found one part that really stood out to me:
Davies thinks the moneyed and narrow focus on targeted therapeutics is misguided. These new drugs tend to focus on attacking cancer’s strengths rather than its weaknesses; its muscle rather than its Achilles’ heel. For example, a medication might be designed to stop the abnormal protein that is allowing a cell to divide without stopping.
I laughed out loud when I read this. How stupid we are, we cancer biologists. We never would have thought of this without a genius like Davies to lead the way. Oh, wait. We did think of it. Long ago. Targeting mechanisms of that allow cancer cells to evade apoptosis is a major focus of cancer research. It’s even listed as one of Hanahan and Weinberg’s hallmarks of cancer! Bloody hell! The contempt Davies must have for the long history of cancer research is palpable. Small armies of scientists, postdocs, and graduate students have been working on just this problem for decades! Let’s put it this way. I started graduate school in 1990, and apoptosis was a major focus of research. I could show Prof. Davies my Cancer Biology class notes from then if he likes. (I’m pretty sure I still have them somewhere.) Davies makes it sound so easy, doesn’t he? Oh, and it’s not “the abnormal protein” that allows cells to divide without stopping. It’s proteins, as in many, and there are other forms of programmed cell death besides apoptosis.
It gets worse:
But, says Davies, for as long as cell division has existed, so have threats to it. “Life has had 4 billion years to evolve responses to those threats,” he says. Tumors are incredibly adept at circumventing the stress of a new drug by developing genetic abnormalities that preserve their ability to divide. Cancer patients know this strength all too well: Many once-potent therapies stop working because tumor cells become resistant, eventually exhausting all treatment options.
The atavistic theory portends new approaches. Drugging tumors with the lowest possible dose could prevent the evolution of therapy-resistant pathways that would otherwise enable the cancer to spread around the body. “You don’t have to get rid of it,” says Davies, “you just need to understand it and control it.”
Bloody hell times a thousand! Yes, tumor cells become resistant. It’s what they do. They’re genetically unstable and have a high rate of mutation, which leads to a lot of tumor heterogeneity, with the cells different parts of tumors often being quite different from each other, upon which selective pressures can work. Chemotherapy and other drugs are a powerful selective pressure that forces them to evolve resistance.
As for using low dose chemotherapy, Davies is behind the times, too. Has he never heard of metronomic chemotherapy? I have. Robert Kerbel first proposed it at least 15 years ago, and it is exactly what Davies describes, using the lowest possible dose of chemotherapy administered continuously or frequently. Of course, the goal here is not to keep resistance from developing. It’s actually based on Judah Folkman’s pioneering work on tumor angiogenesis (the ability of tumor cells to induce the ingrowth of blood vessels to supply its nutrient and oxygen needs. The idea is that low dose chemotherapy is antiangiogenic and targets the endothelial cells lining the blood vessels. There is a form of chemotherapy based on evolutionary considerations, though, known as adaptive chemotherapy. These ideas all long predate Davies’ leaping into the field of cancer research as the Great Physics Hope here to save us dumb physicians and biologists from our own stupidity.
Speaking of angiogenesis, if there’s a characteristic of tumor cells that is not the least bit atavistic, it’s the cancer cell’s ability to induce angiogenesis to feed itself. Single cell organisms have no need for angiogenesis. Nor do they have the need for the ability to evade an immune system. These are characteristics of multicellular organisms that cancer cells retain. Moreover, if you get into the weeds, you find that cancer cells interact with normal host cells and each other, and that that interaction is necessary for its spread and survival. There was just an interesting paper earlier this month that shows how cancer cells interact with host blood vessels and immune cells in order to invade the vasculature and spread via the bloodstream. When tumor cells arrive at other organs, it’s their ability to adhere to the surrounding collagen and cells and begin to grow that determines whether they turn into metastases. There’s a whole field known as tumor microenvironment that studies the interaction of cancer cells with the surrounding cells, structural proteins, and circulating signaling molecules of the host.
To shamelessly paraphrase Douglas Adams, cancer is complicated. Really complicated. You just won’t believe how vastly, hugely, mind-bogglingly complicated it is. I mean, you may think it’s a complicated to figure out your computer, but that’s just peanuts to cancer. Basically Davies cherry picks aspects of cancer that are shared with ancient single-celled organisms and ignores all the characteristics that are not. He dismisses the “existing paradigm” that cancer is primarily a genetic disease and ignores very old observations that certain tumor suppressors and oncogenes are directly associated with cancer. In some cancers, targeting these genes is an effective treatment. Gleevec, for instance, targets a single abnormal protein.
Then he has the audacity to complain:
Many oncologists are skeptical that it ever will. Evolutionary biologist Chung-I Wu, at the University of Chicago, calls the atavistic theory “an extreme position.” Scientists have also criticized Davies’s reference to the discredited “recapitulation theory” that human embryos develop temporary vestigial organs—gills, a tail, a yolk sac—as support for the atavistic model. “I’ve been ridiculed by the biology community,” says Davies.
Yes, Davies has been ridiculed. I’m ridiculing him now. He richly deserves it. Like all good cranks, though, Davies views the ridicule as evidence that he’s right:
Davies is unfazed by the objections. “My feeling is, Who cares? The idea was to come in from the outside and lend a fresh perspective,” he says. Davies sees the criticism as largely rooted in territoriality and financial concerns. “Cancer is a multibillion-dollar industry that’s been running for decades. There’s a lot of vested interests out there.” After five years with the NCI program, Davies is now funded by NantWorks, a sprawling private health care company owned by scientist and billionaire investor Patrick Soon-Shiong (who made his fortune reworking the breast cancer drug paclitaxel to be more effective) to continue his work developing the atavistic model.
Ah, yes. The last refuge of the crank. When scientists ridicule your ideas, attack big pharma and the “territoriality” of cancer researchers. Now, don’t get me wrong. Scientists are human. They can be territorial. They sometimes protect their turf. They can have financial interests. But notice something. Davies had a National Cancer Institute grant to pursue his vision. The clear implication in the paragraph above is that he doesn’t have it anymore, which means that he likely failed to renew it, which means that the NCI found his reapplication wanting. Why might that be? In fairness, NIH grants are hard to get, but the implication is that Davies failed to produce enough promising data with his approach to convince the NCI to renew his grant. So now he’s in bed with big pharma himself! I wish he’d tell me: Why is it so horrible for scientists he dismisses as in the thrall of the existing paradigm and protecting their turf to be—dare I say—colluding with big pharma, but it’s perfectly fine for him to be funded by a billionaire’s health care company?
Of course, it’s not just Davies. There’s an oncologist, Mark Vincent, who should really know better but unfortunately seems to be trying to out-Davies Davies:
Vincent, who had his first atavism insight at around the same time as Davies, is also pursuing the theory. Vincent takes the single-celled phenomenon one step further, believing that cancer could be its own species. The stark difference between our healthy cells and cancerous ones looks more like a jump across the evolutionary tree rather than a hop to another branch. “It seems to me to be a different form of life,” he says. Vincent acknowledges that DNA mutations often cause cancer, but he sees the genetic paradigm as “very incomplete.”
Um, no. Cancerous cells are not their own “species.” For one thing, they can’t live on their own outside of the body. Yes, their genomes are screwed up. Real screwed up. And scientists would be the first to admit that the genetic “paradigm” is incomplete (although probably not “very incomplete”), because there are contributions to cancer from metabolism and the tumor microenvironment that we don’t yet understand. It is, however, interesting how many of the genes mutated in cancer are intimately involved with DNA repair.
Basically, what Davies proposes is nothing more than a 100 year old idea that he resurrected and tarted up for the 21st century. Because he doesn’t have a background in cancer, after finding this shiny new (to him) toy he became utterly fascinated with it and couldn’t understand why cancer biologists don’t find it as fascinating as he did. The reason, of course, is because it’s a 100 year old idea that was found long ago not to be consistent with the data. Pesky that!
Look, I don’t mind people from other disciplines becoming interested in cancer. Insights from physics, chemistry, and other sciences can be valuable and can shake things up. However, what makes cancer researchers grind their teeth when reading an article like this lionizing someone like Davies is being lectured by someone who didn’t have the humility to actually learn the history of the field he’s claiming to transform. If he had bothered to do that, maybe he wouldn’t have made such a fool of himself. Or maybe he would have. He was widely castigated by evolutionary biologists and cancer researchers in 2014 when he first published his ideas, and in the interim he apparently hasn’t developed any humility.
I might have to look into his tow or three most recent publications to see if his arguments have gotten any better, but PZ took care of one of those already. They haven’t.