The Call Was Coming From Inside the House

Your body produces new cancer cells every day. Hundreds of billions of them. And every day, your body finds them and eliminates them.

Now to be more specific we're talking about cells that could become cancer if left unchecked: such as cells that are damaged, that divided incorrectly, or that got infected by pathogens. These early could-become-cancer cells are being produced by the billions every day our whole lives, and every day they are being cleared by our bodies. Billions of them. Every single day of your life. But when our bodies stop efficiently clearing these cells, they can start to accumulate. So then we have a lot more cells than we should have in a given area, and a lot of them have mutations and shouldn’t still be there. And now these accumulated cells are competing with each other, for what are now limited resources and space. When this starts happening, natural selection begins. Survival of the fittest: cell edition. Once this competition begins, the cells that can reproduce fastest, grab the now limited resources the most efficiently, or create environments that are harder for the other cells to tolerate, those are the ones that survive.

Because our cells reproduce so quickly, this natural selection process plays out at evolutionarily eye-watering speeds. Each time a cell mutates and doesn't get cleared, it’s made it past another round of selection, which means it's slightly better at surviving than the last version. Think about how antibiotics work. I’m sure you know you’re supposed to finish the entire course of antibiotics even if you start feeling better. Because if you stop early, the bacteria that survived the first doses are the strongest and most resistant ones, which means they’re the ones most likely to survive future rounds of antibiotics. The same thing happens with these cells. Each round of mutations that survives clearing is a little more resistant to your body's defenses. String enough of those mutations together, and that becomes what we call cancer. This is why the places with the highest cell turnover - the gut, the skin, the blood - have the highest cancer rates. More divisions means more mutations means more raw material for selection. Places with little turnover, like the heart, see almost no cancer.

This is all just basic natural selection, and mapped out this way it seems pretty obvious. But then, why does anyone get cancer in the first place, if our bodies are so good at clearing these aberrant mutation cells? And since this is clearly a clearing issue leading to natural selection on a cellular level, why is medicine looking at the cells themselves as the culprits? Why isn't medicine asking instead what caused this normally efficient clearing system to stop working?

Let's look back through the history of cancer research to see if we can figure out why it isn't being viewed through this lens.

In 1923 a researcher named Otto Warburg was studying cancer in his lab in Berlin, and he saw that cancer cells were using a process called glycolysis, which produced 2 units of energy instead of the usual 36 from the same amount of fuel. The understanding at the time was that cells only behaved like this to compensate when there wasn't enough oxygen, but the cancer cells he was observing were doing it with more than enough oxygen to go around. Why would they be doing something so seemingly wasteful?

Seeing this he thought he'd found THE answer to what cancer was. Here were cells that had switched to a completely different way of making energy; something normal cells never did unless they were oxygen-starved, and he reasoned that this must be the defining change that transformed normal cells into cancer cells. And he wasn’t alone in this belief. His discovery was seen as so momentous, and he was so celebrated for it, that the Rockefeller Foundation built him his own building specifically designed to his requirements: forty rooms, no offices or conference rooms, just labs and a library. This is where Warburg was doing his work, in Berlin, as a Jewish and almost certainly gay (though this was never publicly discussed) man, when in 1941, the Nazi regime came and dismissed him from his position. But then within weeks Hitler himself, during WWII, personally intervened and reinstated him. Hitler's mother had died of cancer, and he was so convinced Warburg was going to cure it, he personally let this gay Jewish man continue his work.

But unfortunately Warburg never did end up delivering the cure. Because while he'd found something real - many cancer cells are shifted to this emergency metabolism - he had the causation backwards. This wasn't how cells become cancer; it was what most cells do under certain conditions. The Warburg effect wasn't the cause but a consequence.

Then in 1992 a pediatric geneticist named Gregg Semenza discovered something called HIF-1, which is basically this master regulator in the cells. One of its jobs being to direct cells to shift to the glycolysis state. So now we knew about cancer cells switching to glycolysis, and we knew how cells got directed to make that switch. But the assumption at the time was still that generally speaking, HIF-1 sent that signal only when oxygen was low.

Then throughout the 1990s and 2000s, researchers started finding that sometimes HIF-1 was activating even when oxygen levels were perfectly normal. Inflammation could trigger it. Stress hormones could trigger it. Reactive oxygen species from overworked mitochondria could trigger it. By the mid-2000s, multiple studies showed that metabolic stress, inflammatory signaling, and oxidative stress could all activate HIF-1 and drive cells into Warburg metabolism without any oxygen deprivation at all. So Warburg had found that cancer cells were using glycolysis, and Semenza had found how cells got switched to glycolysis, and now we knew this could happen when the body was under various types of physiological stress.

Now let's jump back to the 1970s. When Macfarlane Burnet and Lewis Thomas proposed the concept that the immune system must be patrolling for and ridding the body of the lifelong accumulation of genetic changes. They posited that this was an evolutionary necessity.

But then in 1974, Osias Stutman at Memorial Sloan Kettering seemingly disproved this theory. He was actually trying to prove that immune surveillance was preventing cancer, by testing whether mice without immunity would develop more cancer than mice with it. But he ended up finding that the immune-deficient mice got cancer at the same rate as normal mice. And this discovery became a landmark paper, and the pace of research in understanding tumour/immune interactions slowed significantly in its wake. For many scientists, this closed the case: immune surveillance was wrong. And after this finding, Stutman would go on to become chair of the Department of Immunology at MSKCC and was an esteemed figure in the field.

But decades later, a researcher named Robert Schreiber saw a problem with Stutman's experiment. He figured out that Stutman's "immune-deficient" mice weren't actually immune-deficient. The mice in his study still had NK cells and some T cells, both of which contributed pretty significantly to immunity, so Schreiber redid the experiment with truly immune deficient mice. And the result? His immune deficient mice developed cancer much more readily than those with regular immunity. In the early 2000s, Schreiber was invited to present his findings at Memorial Sloan Kettering. "I'm sitting there getting ready for my talk and this elder gentleman walks in and sits in the front row." he recalls. The elderly man was Stutman - the man whose work he was disproving. "The first hand up was Stutman's," Schreiber recalled. "I'm just holding onto the lectern going, 'Oh, here it comes, he's gonna just lay into me.' And Stutman got up in front of everyone and said, 'You know, it is remarkable what you can do now at the turn of the century that we couldn't do in the 1970s.'"

Immune surveillance was hypothesized in the early '70s, seemingly disproven in '74, and then definitively reproven in the 2000s. And once they figured out that immune surveillance was in fact real, they took it a step further and documented that healthy people's immune systems could and were successfully clearing precancerous cells all the time. This was when the concept of your immune system patrolling your body, looking for damaged or abnormal cells, and eliminating them, became established science. They knew it worked. They could even see it working. But even with this knowledge, they didn't seem to be asking why it sometimes... stopped.

Don’t kill me flags

In 2002, in the wake of immune surveillance being reproven, researchers discovered that tumor cells were expressing a protein called PD-L1 on their surface. PD-L1 serves as a signal to the immune system patrollers - think of it like a flag you hang outside your door that says "don't kill me! I'm one of you!.” Healthy cells hang these flags so the immune system doesn’t accidentally kill them when it’s mounting a response, so when things like infection or inflammation are happening and the immune system is mounting a defense, cells are more likely to hang these flags. Tumor cells have been discovered to use these flags, which is why they are able to survive even with the immune system patrolling, which makes sense given what we talked about earlier - natural selection favors cells that can avoid being killed. If you are hanging these flags, your chances of surviving immune clearance go way up, so of course the tumor cells that survived are ones that still use this signal. But the medical literature framed this discovery of tumor cells using these flags as: tumors were actively suppressing the immune system, evading detection through the use of these flags. The medical literature was full of this kind of framing: tumors 'co-opt' immune checkpoints, tumors 'evade' immunosurveillance, tumors use 'key mechanisms' to escape immune destruction.

Once they discovered this, the solution seemed obvious: block those signals. If tumors were ‘actively’ suppressing immunity, just remove that suppression and unleash the immune system. So they tried it, and it worked. When researchers developed drugs that blocked PD-1 (the receptor the immune system uses to read the PD-L1 signal), tumors shrank, and some patients with previously fatal cancers who would have otherwise died, lived. In 2010, a clinical trial of a new treatment for patients with advanced skin cancer (melanoma) using the knowledge of these flags produced results at a level that medicine had never seen before. In several of the patients in the trial, signs of their cancer disappeared completely after treatment. These were people with metastatic disease, meaning their cancer was spreading to other body systems, who without treatment would have died within months. The drug that emerged from this research, ipilimumab, was very quickly approved by the FDA, in 2011. It was the first new treatment for advanced melanoma in over a decade, and following the wake of ipilimumab, the field exploded. Allison and Honjo, the people who developed this method, won the Nobel Prize in 2018. The word "revolution" appeared in paper after paper.

But there was a pretty big side effect to this treatment that wasn't getting as much air time as the solution: autoimmunity. And not just a little but a whole lot of it. Patients being treated with this new method developed things like inflammatory colitis, hepatitis, thyroiditis, and dermatitis. Their immune system, without the usual signals telling it not to, had started attacking healthy tissue throughout their bodies. Some of these autoimmune reactions were severe enough that patients had to stop the cancer treatment. This confirms what we just discussed: PD-L1 isn't a tumor trick. It's how ALL cells protect themselves during immune activity. Block it everywhere, and healthy tissue loses its protection too. When your immune system is attacking an infection or responding to injury, inflamed but regular tissue puts out the PD-L1 flags to prevent collateral damage—to say "yes, I know there's activity here, but don't attack ME, I'm just responding to the problem." Macrophages do it. Epithelial cells do it. Normal inflamed tissue does it all the time.

Tumor cells hanging the PD-L1 flag in response to attacking T cells isn't "adaptive immune resistance" or a "clever evasion strategy." It's just... what cells do when they're in an inflammatory environment. It's the normal cellular response.

So this 'revolutionary' discovery is more evidence that tumor growth is just natural selection playing out. When immune clearance fails, cells accumulate and compete. The ones that happen to retain normal protective signals survive. Medicine was seeing the winners of a selection process and calling it a strategy. The discovery was: some tumor cells happened to retain or upregulate this normal protective signal, and the cells that did had survived immune surveillance. The ones that lost it got cleared. This proved that the immune system CAN clear tumors - the capacity exists, it's just being held back. When they removed the safety mechanisms with checkpoint inhibitors, tumors disappeared. But patients developed autoimmunity. Medicine had accidentally proven, again, that immune capacity matters. They just weren't looking at it that way.

Why does cancer spread

Then in 2015, researchers found that epithelial cells, which make up most of the cells that become cancer, have a built-in migration program called 'unjamming.' During wound healing, these cells shift from a solid-like 'jammed' state to a fluid-like 'unjammed' state, allowing them to migrate collectively to repair tissue. More specifically, these cells unjam when their neighbor goes missing. So to conceptualize this, let's think about a sheet of honeycomb for a minute. If you imagine all those little pockets as cells, you can see how they each have neighbors on every side. What triggers unjamming is if you were to cut a line in that sheet, all of the cells along that cut that had a neighbor a second ago and now don't would get the signal, "I lost my neighbor, we need to repair this gap that was just created." And what is particularly important here is that these cells move in sheets or clusters to repair these gaps, not individually. This is important because we know from the research that single rogue cancer cells rarely cause cancer to spread in the body; it’s almost always clusters of traveling cancer cells that are the catalyst for spread.

So let’s look at this one a bit more closely. Researchers know that cancer metastasis (when cancer spreads from its original location to other places in the body) involves cell clusters spreading to places they shouldn't be. They know most deadly cancers are epithelial. And a decade ago researchers figured out that epithelial cells naturally migrated using this unjamming program, and that this program creates moving clusters or sheets. But even now, ten years later, no one seems to have connected the dots.

And what this means is that cancer metastasis isn't some sophisticated evolutionary strategy that tumor cells developed. It's epithelial cells doing epithelial cell things; activating their normal wound-healing migration program. If there's tissue damage around growing tumors, that will create wound signals. Normal cells in the area respond appropriately, migrate a bit, repair, and stop. But tumor cells, which are just epithelial cells that survived selection for rapid growth when clearing failed, they also get these wound signals. They unjam and migrate too, but if they've lost the normal controls through mutation, without the normal stopping signals, they go places they shouldn't.

So if unjamming is how epithelial cells migrate (it is), and cancer cells are primarily epithelial cells (they are), and unjamming is triggered by edge creation (it is)... then what would we expect to see? And what would that mean for how we do things? The huge implication here is that surgery creates massive wound signals, it is literally creating wounds. And the data bears this out. Modern studies show right after surgery, within 6-12 months, there is a sharp peak of metastasis, which suggests a triggering event, not gradual progression. Radiation damages tissue, and even chemotherapy causes tissue damage that can cause these edges needed for unjamming to take place. So our front line treatments for cancer are triggering the migration program in cells that have already been selected for proliferation. We've known about unjamming for nearly ten years, and we're still using treatments that create the exact signals that trigger metastasis, then wondering why cancer spreads after treatment.

But maybe the biggest potential implication of all is biopsy. Biopsy is a front line diagnostic procedure. See anything out of the ordinary on your scans? Biopsy the potential mass to see if it's cancer. And interestingly, when someone actually looked, they found exactly what this framework predicts - more lymph node involvement, and specifically macrometastases (clusters), not individual displaced cells. In 2004 Hansen conducted a study of over 600 women who had breast cancer to see if there was a correlation with biopsy and metastasis, and she found that women who had needle biopsies were 50% more likely to have cancer in their sentinel lymph nodes than women who had their tumors surgically removed without prior biopsy. And not just cancer cells but macrometastases which means groups of cells clumped together, which is exactly what unjamming triggers. And cancer cells in your lymph nodes is categorically bad. Lymph node involvement is one of the primary criteria that determines how far along your cancer is and how aggressive your treatment should be, which means the biopsy may be increasing not just spread but also the intensity of subsequent treatment.

A larger study by Peters-Engl and colleagues found the same association Hansen did - a 37% increased risk - but after statistical adjustments, concluded the finding wasn't significant. They corrected away what they saw like we saw the researchers do in the placebo Cochrane study. A third study by Moore in 2004 conducted at Memorial Sloan Kettering showed the direct correlation: no biopsy (1.2%) → FNA (small needle) (3.0%) → core needle (bigger needle) (3.8%) → surgical biopsy (4.6%). With a p=0.002 which basically means the chances of this happening by chance are extremely unlikely. So what we are clearly seeing is: more manipulation = more metastasis.

The medical response has been to say those cancers must have already spread. But now we have the mechanism. Biopsy creates wound edges. Wound edges trigger unjamming. And tumor cells that unjam go into the lymphatics and bloodstream through normal routes, and this happens because our front line diagnostic procedures are telling them to migrate.

So while individual doctors or medicine as a whole may know this, sort of, if they truly understood how these connect—if they'd integrated this into their framework—everything would look different. If they understood and acknowledged the potential implications, biopsy wouldn't be the reflexive first step every time a scan shows something suspicious. We would be doing more studies looking at whether all this cancer screening is really helping anyone. Miller et al. did a 25-year randomized controlled trial of nearly 90,000 women that showed women randomized to screening vs no screening for breast cancer had the exact same death rates. So it’s not a given that all this screening is providing more benefit than harm. And to be clear, none of this means we shouldn’t look for cancer at all, or that we shouldn’t treat it when we find it. It just means that we should know what we’re doing when we do it. Know the risks and make calculated judgement calls, instead of looking for cancer, cutting it and triggering spread, and then acting surprised when it moves around the body. But they aren’t thinking about it this way. Which means that while medicine may 'know' the individual facts, they clearly haven't put them together into a framework that actually changes how they think about, research, or treat cancer.

One guy saw it…sort of

There was one researcher who did see it differently. Robert Gatenby, a radiologist at Moffitt Cancer Center, had been thinking about cancer through an evolutionary lens since the 1990s. He remembered being taught in medical school that "cancers grow because cancer cells grow faster than normal cells", which he knew was wrong on multiple levels. Cancer cells don't grow faster because a mutation flipped a switch. They grow faster because, in a competitive environment where damaged cells are accumulating instead of being cleared, fast proliferation is a winning strategy. We're seeing the result of selection, not the cause of the problem.

In 2009, Gatenby published a paper proposing that we stop trying to eradicate all cancer cells in the way we have always done, and instead try using treatments strategically; use evolution against itself. He called it "adaptive therapy." He figured that maximum-dose chemotherapy kills sensitive cells, creating space and resources for resistant cells to proliferate. That's why cancer so often comes back resistant—we're selecting for resistance with each treatment. But what if we cycled treatment on and off, maintaining a population of sensitive cells to suppress resistant ones?

In 2017, Gatenby published results from a 11-patient prostate cancer trial. He was able to demonstrate that using this approach could double patient survival time, while cutting drug use in half. The paper appeared in Nature Communications, which is one of the most prestigious journals in science. Other centers started testing variations of his protocols, and he got follow-up funding for larger trials. By 2022, five years after publishing those initial results, multiple adaptive therapy trials were ongoing or planned in different cancers. But in 2025, sixteen years after his original concept paper, eight years after proving he could double survival time, adaptive therapy was still described as "promising" and "gaining traction." Sixteen years. Dramatic results. Proper channels. Published in a prestigious journal. And it's still "promising."

And even Gatenby, who understands cancer cells are responding to selection pressures and is using that understanding to save lives, even he is still asking: "How do we manage the tumor population better?" rather than looking upstream to: "Why did the clearing system fail in the first place?" His innovation is still, ultimately: how can we best kill cancer cells directly. He's figured out how to get really good at bailing water out of the boat. But he's not asking why the boat is sinking.

So what we’re seeing here is that cancer isn't that mysterious, it’s really just normal epithelial cells doing two normal epithelial cell things - proliferating and migrating - without normal supervision. When normal immune clearance fails or depletes, natural selection favors the fastest replicators. When wound signals trigger migration, these selected cells spread chaotically. Every 'hallmark of cancer' is just normal cell behavior without normal regulation. And we didn't need to spend decades in laboratories or billions in research funding to understand this. We didn't need to become oncologists or molecular biologists. We just needed to recognize the pattern: when immune clearance fails, natural selection happens at the cellular level. The winners become what we call cancer.

This understanding predicts cancer should develop specifically in epithelial tissues (which can unjam and migrate) where clearance has failed. Cancer distribution data confirms this: malignancies cluster in epithelial tissues (lung, GI tract, breast), while benign tumors that don't metastasize occur in tissues without this migration program.

The research community has meticulously documented every piece of this puzzle - the Warburg effect, immune surveillance, epithelial unjamming. They've published it all, proven it all. But without understanding that it’s ultimately a clearing issue, they've studied each piece in isolation, missing the breathtakingly simple whole. And this isn't a criticism of researchers - they've done extraordinary work mapping these mechanisms. But when you're deep in the molecular details of your specific pathway, it's hard to step back and see that it's all the same pattern.

And to be clear, there are undoubtedly individual physicians and researchers who see pieces of this pattern. Some may even see all of it. Gatenby clearly understands the evolutionary dynamics. But as we talked about above, this understanding isn't reflected in how medicine teaches, researches, or treats cancer. Medical students still learn that cancer is caused by genetic mutations. Research funding still flows toward finding new ways to kill cancer cells rather than asking why clearing failed.

So even though there are likely many who see pieces or the whole, a radiation oncologist who suspects treatment triggers metastasis still has to follow standard protocols. A researcher who understands clearing failure can't get funding to study it when grant committees expect molecular pathway research. The system's inertia overwhelms individual insight.

And this pattern that we are seeing here, of medicine finding the mechanism downstream, but not even asking about the upstream cause, is not unique to cancer. So now we have another piece of the elephant. Medicine is missing that clearing failures cause cancer, and all of the “mysterious” cancer findings throughout history are just what you would naturally find when regular cells keep regular cellular processes as they go through the natural selection process. So then the next obvious question is: why is the clearing failing? Let’s look at that next.