[♪ INTRO].

Sometimes, it seems like everything around me causes cancer, or at least somebody’s saying that. There’s my tasty steak, my morning latte, half the chemicals floating around in the air.

Not to mention the sun. Sunlight. I love you!

You’re great; you’re the reason life exists. But I’m worried about this mole. And it doesn’t help that the agencies that do risk assessments for carcinogens, or the things that cause cancer, aren’t always very clear about what they mean.

Sure, we know that cigarettes are bad, but bacon is in the same category as cigarettes. Really? The confusion comes from the fact that most of the labels on substances really are only talking about the quality of the evidence that it does or doesn't cause cancer, not how likely it is to sprout a tumor on you.

And some of the determinations you hear about are more legal than scientific. I’m looking at you, California. The first step to understanding what these labels really mean, and how to interpret them, is to establish what cancer actually is.

Cells are considered cancerous when they grow uncontrollably, which can cause tumors. And in many cases, rogue cells will break off to form new tumors in other parts of the body. Figuring out the environmental exposures and lifestyle factors that increase your risk for cancer can be complicated, especially because each person might have different genetic predispositions to various types of the disease.

But with careful studies, scientists have been able to identify dozens of things that are a good idea to avoid. In many cases, clues to what might be harmful come from noticing patterns. One of the first times doctors did this was back in 1775, when a London surgeon named.

Percivall Pott noticed an alarming number of former chimney sweeps developing skin cancer on their scrotums. He proposed that chimney soot was getting stuck on their skin and eventually leading to tumors. If you wanted another reason to not go into that particular field.

Pott’s approach was essentially an early epidemiological study: looking at disease rates in different populations to suss out potential causes. Of course, the main problem with these types of studies is that they rely on correlations. They can’t prove that chimney soot is the culprit, rather than something else the chimney sweeps might be prone to, say, elaborate song and dance numbers, but they can be suggestive.

And they can still play a huge role in public health, even if they’re not definitive. For instance, just three years after Pott published his observations, some clever Danes decided to recommend daily baths for chimney sweeps. Even 100 years later, continental Europe had much lower rates of scrotal cancer than the.

UK, which did not adopt those bath recommendations. Today, epidemiological studies are just one type of evidence scientists use when evaluating potential carcinogens. Another approach is to do testing in animals.

In the early 20th century, Japanese scientists tried putting various substances on rabbit ears, and seeing whether those could form tumors. In many cases they did. We now often do similar experiments in rats and mice, sometimes feeding them the chemicals or using other exposure routes.

This method is powerful, because it directly implicates whatever you put on the ear, or feed to the animal, and it’s much faster than waiting for cancers to spontaneously develop in the population. Scientists can also drill down and test different elements of substances to identify specific molecules that might be a problem. As we’ve learned a lot more about cancer in the last few decades, we’ve come up with even faster methods.

Biologists now consider cancer a genetic disease. That doesn’t mean that it’s necessarily tied to the genes you inherited from your parents, although it sometimes is. But no matter what causes it, cancer is considered genetic in the sense that something has to happen to a cell’s DNA to allow for that uncontrolled growth.

In many cases, those changes come from carcinogens mutating DNA, they’re mutagenic. So, scientists can do tests to see whether a compound might be a carcinogen, based on what it does in cells. One of the simplest tests is called the Ames test, which is done in bacteria.

You start with a strain of Salmonella that can’t actually grow on its own; it’s a mutant. You apply your test substance, mixed with some rat liver enzymes to mimic how the human body would metabolize the chemical. Then, you try to grow the bacteria in a petri dish.

If you don’t get much, that’s great, it means your test substance isn’t very good at mutating DNA. If you see a lot of colonies, that’s a sign that whatever you added was pretty good at undoing the original mutation, and allowing the bacteria to grow again. You might want to stay away from that.

In the 1970s, the Ames test was a huge advance, because it was super fast. It’s much faster than even the animal tests, and it also gives a rough estimate of how mutagenic a compound is. There are, however, flaws with each of these methods, which we’ll get into.

But the basic idea of carcinogen evaluation is to synthesize all of the available information on a substance, and make a conclusion about whether or not it can cause cancer. The most famous and influential group that does this is the. International Agency for Research on Cancer, or IARC.

It’s a part of the World Health Organization. Over the past half century, they’ve evaluated 1004 potential carcinogens, putting them in one of 5 categories. If the scientific evidence is very strong and consistent that something can cause cancer, it goes in group 1.

If there is less evidence, but it’s still fairly strong, an item might get classified in group 2A as ‘probably’ carcinogenic, and so on down the list. Group 2B is ‘possibly’ carcinogenic, while 4 is ‘probably not.’ Group 3 means there isn’t enough evidence either way. This actually includes half the items.

In the US, there’s also the National Toxicology Program, or NTP, which puts out a list of things either “known to be human carcinogens” or “reasonably anticipated to be human carcinogens.” Various other systems exist, but they tend to look a lot like these two, where there’s some attempt at putting a hierarchy on the strength of the evidence, not the degree of risk. Typically, human studies carry more weight than animal studies, so if something is suspicious only in animals, it’s ranked lower. The main problem with this approach is that it’s easily misinterpreted.

The agencies are evaluating the quality of the evidence, not how carcinogenic something is. This is how you end up with bacon being in IARC’s top category, along with tobacco smoke and things like chimney soot, asbestos, and plutonium, things you really want to avoid. Bacon and other processed meats are way less carcinogenic than cigarettes.

But in terms of whether each item is at all cancer-causing,. IARC concluded that both have very firm backing. So you should also know that IARC tends to be pretty conservative in their decisions.

Out of more than 1,000 items, only 120 made it to group 1, where they’re basically sure that it’s a carcinogen. Even more amazingly, there’s just one thing in its least concerning “probably not” category. That honor goes to caprolactam, a chemical used in making synthetic fibers, like nylon.

The only thing that probably doesn’t cause cancer. But before you think, ‘great, the people making my toothbrush are safe’! Well, not quite.

Short term exposure to the chemical can burn your eyes and skin. IARC is just assessing cancer risk, not overall safety. So they’re saying that it probably won't give you a tumor.

In broad strokes, these classifications are helpful. But what most people want to know is: what stuff do I need to avoid, for what reason, and how bad are those things? To know that, you need to dig deeper than the category rating.

You could read the full IARC reports, but if you’re not into 500-plus page monographs, you could easily get good information from non-profit cancer societies and government agencies. So, for instance, the cancer drug tamoxifen is listed in IARC’s group 1 because it increases the risk of uterine cancer. But if you already have breast cancer, tamoxifen can be lifesaving.

Like if you’re increasing your chances of getting one cancer by definitely getting rid of the cancer you already have, that’s a good deal. The hormone estrogen also makes the cut, but that doesn’t mean that everyone should try to rid their bodies of estrogen, cuz your body makes estrogen and you need it. Estrogen can actually protect against certain cancers, even if it raises the risk for others.

So don’t assume that its classification means it’s always bad in every single circumstance. While the IARC and NTP systems are science-based, they still involve interpretation of data, which can be hard to do, especially if it’s contradictory. In some cases, agencies get it wrong and have to issue changes.

Take the case of saccharin, an artificial sweetener. In the 1970s, scientists found that male rats given high doses of the sugar substitute developed bladder cancer, leading to IARC initially classifying it as a ‘possible’ human carcinogen. But the effect was only seen in male rats.

There were no problems in mice or monkeys, and no epidemiological evidence that it did anything to people, who had been enjoying it for quite a while in their food. Eventually, scientists realized that saccharin was only a problem for male rats because they happen to make lots of a certain type of protein that can form crystals when saccharin is around. Those crystals then irritate the bladder, something that simply does not happen in humans.

In 1998, IARC lowered its rating. This highlights the problem with putting too much stock in animal studies: they’re not human. Plus, they’re usually getting massive doses that are way beyond what we would normally come into contact with.

So, while these tests are useful for screening, there could be a lot of reasons why substances wouldn’t affect us the same. The super speedy Ames test has a similar problem. It’s very good at picking out substances that mutate DNA, but it misses things that contribute to cancer in other ways.

For instance, take alcohol. It’s a well-established carcinogen, but it doesn’t pop up as a positive on the Ames test because it promotes cancer by killing cells, causing the body to replace those cells more quickly. That mechanism can only be detected in animals or humans.

So, depending on the test, some compounds won’t get flagged as dangerous when they are, and some will get flagged as dangerous when they aren’t. To get the best picture possible of what’s actually happening, you need to consider a wide range of evidence. And as you can imagine, in weighing all of these different sorts of experiments, it’s only natural that scientists will sometimes disagree with one another.

Decisions on substances are usually roughly the same between organizations, but there are discrepancies. Then, of course, you have the decisions that aren’t even all that science-based. You know all those warning labels on everything from headphones to mugs that say they contain a chemical ‘known to the state of California to cause cancer, birth defects, or other reproductive harm’?

They come from a 1986 law, known as proposition 65, that was meant to help California protect its drinking water and its people from health hazards. But anyone can file a lawsuit to get a warning label slapped on products that contain even a negligible amount of a chemical on the state’s list. Including, now, coffee, since it contains a small amount of acrylamide.

The IARC labels it as a ‘probable carcinogen,’ but acrylamide is found naturally in lots of cooked foods. And the label is based on the results of animal studies; there’s no clear evidence that it affects cancer risk in humans. Never mind that coffee as a whole is one of the best studied beverages on the planet, and it’s thought to be, if anything, protective against cancer.

So, if you're worried about product labels, use them as guideposts. And when you see headlines claiming that some new thing has been found to be as dangerous as cigarettes, it’s almost definitely not; look into the context. In some cases, like the state of California's warnings, there are clear reasons why you don't need to heed them, which actually makes the whole idea of the warning weaker.

For the others, it's a matter of looking up the amount of risk associated with each one, or asking your doctor for advice. For the most part, you already know the things that you really need to stay away from, like tobacco smoke, excessive drinking, and too much sun exposure. And sometimes taking the new findings into account may just mean eating, like, fewer hot dogs every year.

Thanks for watching this episode of SciShow! If you're interested in learning more about how cancer works, you can check out our video on why we have not yet cured it. [♪ OUTRO].