July 1, 2013

What’s a Group 1 carcinogen?

Stuff has a (mostly reasonable) story on alcohol and cancer, quoting Prof Doug Sellman

“The ethanol in alcohol is a group one carcinogen, like asbestos,” 

Many of the readers of this story won’t know what a “group one carcinogen” is.  Given the context, a reader might well assume that “group one carcinogens” are those that carry the largest risks of cancer, or cause the most serious cancers. In fact, all it means is that an additional hazard of cancer, whether high or low, has been definitely established, because that’s all the IARC review process tries to do. The Preamble to the IARC Monographs that define these carcinogens says

A cancer ‘hazard’ is an agent that is capable of causing cancer under some circumstances, while a cancer ‘risk’ is an estimate of the carcinogenic effects expected from exposure to a cancer hazard. The Monographs are an exercise in evaluating cancer hazards, despite the historical presence of the word ‘risks’ in the title. The distinction between hazard and risk is important, and the Monographs identify cancer hazards even when risks are very low at current exposure levels, because new uses or unforeseen exposures could engender risks that are significantly higher.

Group 1 carcinogens tend to either be very common exposures or to cause very specific types of cancer, because those are the two scenarios that make it easy to establish definitely that there is a risk.  They include asbestos, arsenic, sunlight, birth control pills, plutonium, diesel exhaust, and wood dust.

Some group 1 carcinogens, such as tobacco and hepatitis B, are responsible for large numbers of cancer deaths worldwide. Others, such as plutonium and diethylstilbestrol, are responsible only for small numbers of deaths. Some group 1 carcinogens cause aggressive, untreatable tumours; for others, such as human papillomavirus, disease is largely preventable by screening; still others, such as sunlight, sometimes cause serious disease but mostly cause relatively minor tumours.

The phrase “group one carcinogen” is only relevant in an argument over whether the risk is zero or non-zero. Its use in other contexts suggests that someone doesn’t know what it means, or perhaps hopes that you don’t.

avatar

Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »

Comments

  • avatar
    Stuart Dodd

    TL seems to be implying that health professionals are deliberately misleading ev’ryday folk as to the risk of alcohol consumption and cancer but fails to mention that alcohol kills 200+ Kiwis a year through cancer alone (not to mention the many hundreds more who survive this entirely preventable cancer).
    In fact by not referring to the cancer deaths caused by alcohol the article was more misleading than the phenomenon he was criticising as you would come away from the article thinking alcohol’s contribution to cancer was not significant, when in fact alcohol is in the same league as tobacco and HepB? I note there is no specific grouping or term for carcinogens that kill a lot of people? Saying that a carcinogen is Group 1 is not misleading – it’s simply a shorthand for highlighting the fact that it very definitely is carcinogenic.

    10 years ago

    • avatar
      Thomas Lumley

      No. Stating alcohol is a Group 1 carcinogen is unhelpful because (a) very few people know what the term means, and (b) it doesn’t mean what the usage implies. Some Group 1 carcinogens cause lots of deaths, some cause very few.

      It’s true that alcohol is responsible for lots of deaths. If you want to say that, say it, don’t hide behind jargon that really means something different.

      10 years ago

      • avatar
        Thomas Lumley

        And alcohol’s contribution to cancer is not in the same league as tobacco. Worldwide, about 3.5% of cancer deaths are attributed to alcohol; about 30% are attributed to tobacco. Hepatitis B is more comparable to alcohol.

        10 years ago

  • avatar

    I took azathioprine for exactly 5 weeks and developed leukemia. I was not a transplant patient but had taken every test in the world to get on the list. When the oncologist wanted to admit me to the hospital for bone marrow tests etc, I said no and went home. I did not at all believe that this drug could have caused me to develop leukemia in only 5 weeks. I wish I had been advised as to the possible dangers of this drug, particularly when it had been taken with prednisone. Maybe to you a group 1 carcinogen means little because not that many people take pred and zath except transplant and other disease. However I would have appreciated the information as I could have researched it on the internet. Maybe for you it is not worth giving the warning But if you are that one person that develops leukemia, then it would have been worth millions.

    10 years ago

    • avatar
      Thomas Lumley

      That’s my point, I think. People want to know the potential for risk, not whether it’s Group 1 or Group 2a.

      10 years ago

      • avatar
        Ian Dee

        Yes. I agree with you Thomas. What would be more helpful is words like “is a carcinogen”, “probable carcinogen” etc.
        Group 1 carcinogen isn’t shorthand at all. At last count “is a” 3 characters more.
        Statements like “Alcohol is akin to asbestos” are misleading and don’t tell us anything about the relative risk. Although getting enough data to make meaningful decisions takes a lot of research. In the end that information needs to be easily accessible to everyone so better decisions can be made.

        10 years ago