Phytochemicals for Cancer Prevention

by Introspection 3 Replies latest social physical

  • Introspection
    Introspection

    Here are some quote from the February issue of Life Extension magazine, there is a lot of good info in this one. I don't think this issue is available online yet, but if you want to check later their website is http://www.lef.org. I plan to add more to this thread later.

    Studies show that people who eat a lot of red, orange, green and yellow vegetables have a significantly decreased risk of various cancers. The protective effect is due to carotenoids. Most people are familiar with the carotenoid beta-carotene, found in carrots. There are, however, hundreds of other carotenoids--some not even discovered yet. There is lutein in spinach, zeaxanthin in corn and lycopene in tomatoes.

    Lycopene is the most abundant carotenoid in humans. The prostate gland alone contains 14 to 18 different metabolites of lycopene in people who eat tomatoes or other vegetables that contain it. . . The two largest studies [of lycopene and cancer risk] involve 14,000 Seventh-Day Adventists (lacto-oco vegetarians) and 47,894 American physicians. In the physician study, men with the higher level of lycopene in their blood had a 20% reduction in risk. In the Adventist study, eating tomatoes more than five times a week reduced risk of prostate cancer by 40%. Lycopene is good at protecting lymphocytes from DNA damage. In an Italian study, 7mg/day of lycopene reduced DNA damage 50% in the first week.

    Carotenoids work synergistically. Taking several together is better than taking one alone. In the now infamous study where smokers took beta-carotene supplements and nothing else, risk of lung cancer actualyl rose. But a 30% reduction was found in a study of 100,000 people who ate a variety of carotenoids on a consistent basis rather than just one. A 60% reduction was found in the same study for non-smokers. It appears that alpha-carotene, not beta-carotene, is the best carotenoid against lung cancer.

    -----

    [Folic acid] has been involved in so many important cancer studies that it stands in a class of its own. Folic acid (the vitamin version of folate) is a B vitamin typically found in certain green vegetables and legumes. Meat contains very little of it. A serving of steak, for example, contains 3% of the RDA, while a serving of broccoli contains 50%.

    Folate has powerful cancer preventive effects through its role in maintaining methylation. Methylation has two powerful roles in preventing cancer. First, it is crucial for the repair of mutations. Second it is crucial for the activation and deactivation of genes involved in cancer. . . Abnormal methylation is present in all cancers, no matter the type. The critical importance of folate, then, becomes apparent.

    Lung and colon cancer are the first cancers to be linked to folate deficiency. Breast, prostate and pancreatic cancer involve the deficiency as well. Alcoholism, folate deficiency and breast cancer go together. The same is true for colon cancer--alcoholism exacerbates folate deficiency.

  • rem
    rem
    Studies show that people who eat a lot of red, orange, green and yellow vegetables have a significantly decreased risk of various cancers. The protective effect is due to carotenoids. Most people are familiar with the carotenoid beta-carotene, found in carrots. There are, however, hundreds of other carotenoids--some not even discovered yet. There is lutein in spinach, zeaxanthin in corn and lycopene in tomatoes.

    I'd be curious to see what these studies really are and if they are published. Hopefully they are not just epidemiological studies. Epidemiological studies are not sufficient to assume causation. And it's quite a stretch to go from epidemiological studies which can't even identify causation to forming a theory about the protective affect of nutrients that happen to have a corrolation.

    Corrolation does not equal causation. There could be other reasons why people who eat lots of red, orange, green, and yellow vegetables have significantly decreased risk of cancers. Maybe they eat more healthy because they are more wealthy, which means they also have better health care? Maybe those people tend to excercise more or in a different way?

    It would be great if we could rely on such studies, but unfortunately there needs to be a lot more work done before such bold statements are made. (This, of course, assumes that the studies in question really are epidemiological in nature and are not more rigorous control studies)

    rem

  • Introspection
    Introspection

    Hi Rem,

    LEF does provide all references, and from what I've seen they cite reputable journals and have all the their past issues online, if you're interested you can take a look at their website.

    Of course, my purpose for posting this information is not to point to their words as fact, but just for the sake of information itself. My personal approach to health care is to include all aspects: nutrition and exercise, self care along with good professional care, including both objective and subjective approaches. It doesn't make any sense to totally disregard how you feel when it comes to your health, even if the signals might be misinterpreted sometimes. For that matter, I don't believe in relying on any single factor, after all good health is dependent on many different factors. Although I've found nutrition to be helpful in this regard, it would be foolish to think any single thing is the "cause of good health." Each of us needs to determine what specific combination of things work for us. So thank you for pointing that out, I would certainly encourage anyone reading this information to look at the overall picture and what it might imply in terms of good nutrition rather than drawing specific conclusions. Needless to say, if you already have cancer and are contemplating a course of treatment, you're going to want information that is much more specific.

  • rem
    rem

    Intro,

    Your point of view sounds like a balanced approach. I was not so much questioning the quality of the studies, but rather the interpretations of them. It is very easy to jump to conclusions with observational studies such as nonintervention or epidemiological studies, but such conclusions are unwarranted. Observational studies are only preliminary and can only point out interesting correlations that require more study to show causation. Most times epidemiological studies reveal correlations that later, with further study, are found to be negative or not causative. Causes and claims of efficacy can only be identified through robust, randomized clinical trials. It would be interesting to find out what type of studies are being referred to in the article.

    I'm just afraid that the journal may be running a bit wild with the facts as journalists are wont to do with scientific studies. Often times journalists will make claims that the scientists conducting the study never made or even cautioned against. It's extremely common - all one has to do is read in the newspaper about the latest study about cancer, nutrition, illness, etc.

    The specific claims made in the article just jumped out as a red flag to me. I would just be cautious of such strong claims without looking into the studies further to see what they really say. As you can tell, I don't trust journalists much when it comes to scientific claims as they are known to exaggerate.

    rem

Share this

Google+
Pinterest
Reddit