26 November 2017
Hi Everyone,
I know it’s been a while since I did a blog post in the spirit of my old “cancer for dummies” blog, but I thought it would be fun to tackle a question about cancer prevention using ginseng.
Ginseng is a family of herbs that grow mostly in colder climates, and feature prominently in eastern medicine, believed to aid the healing or amelioration of effects from a myriad of conditions including: cardiac function, sexual function, glucose metabolism, pulmonary disease, cerebrovascular disease, athletic performance, and warding off the development of cancer: https://www.ncbi.nlm.nih.gov/pubmed/21704950
Nutrition research is notoriously difficult, as touched upon in a fantastic post by FiveThirtyEight: http://fivethirtyeight.com/features/you-cant-trust-what-you-read-about-nutrition/
In short, for observational / questionnaire studies is very difficult for a study subject to keep track of intake with a high degree of precision, and the statistics surrounding nutritional studies must be handled with great care. With so many potential confounding factors, it’s often very difficult to isolate a particular variable with a high degree of certainty. Exploratory analyses can also be easily confounded by statistical multiple testing problems; the more time one tests for correlations, the higher the probability that a spurious correlation will come to the surface. From the FiveThirtyEight piece:
Nutritional questionnaire studies evaluating cancer incidence are also notoriously difficult because most of the study population will not get cancer in a particular time period; those following patients over time must follow patients for a *long* time: on the order of magnitude of decades.
Three studies in particular piqued my interest, by the authors Yun and Choi, a Koren-Chinese collaboration studying the effects of ginseng consumption in relation to cancer incidence that spanned over 15 years.
The first study was produced in 1995, using 1,987 case-controls (3,974 patients) comparing ginseng users to non-users. The authors found that ginseng users had a 50% lower risk of having developed cancer (Odds Ratio: 0.50, 95%CI: 0.44 – 0.58). The authors specifically sought to investigate cancer incidence, not a myriad of potential correlations that might cause spurious correlations, so there is less concern for multiple testing error: https://www.ncbi.nlm.nih.gov/pubmed/7655337
The second study, produced in 1998, utilized questionnaires to quiz 4632 people about their frequency of ginseng intake, kind go ginseng, etc. The authors reported that there was about a 60% lower risk of dying from cancer in the ginseng-using group compared to the non-using group on a whole: (RR: 0.40, 95%CI: 0.28 – 0.56) https://www.ncbi.nlm.nih.gov/pubmed/9698120
While these studies are fairly large, in questionnaire studies it’s very difficult to escape the correlation vs. causation problem: do people who consume more ginseng get less cancer because it prevents cancer, or are these people generally more proactive with their health in many areas of their lives? The observational study subjects are from a culture that respects and reveres ginseng as a panacea, and those more proactive about their general health might be more apt to consume more ginseng, while at the same time might also be more apt to exercise more, better manage chronic stress, or perform a multitude of small things that all might add up to a lower risk of cancer over time.
However, the authors did not stop there: in 2010 they produced results from a randomized, double-blind, placebo-controlled study evaluating the consumption of 1g of red ginseng extract powder per week for three years, or placebo, and followed the patients for 8 years. This type of study design is able to detangle the correlation vs. causation conundrum. The authors enrolled 643 chronic atrophic gastritis patients, presumably for a patient population with a higher risk of developing gastrointestinal tumors. Often, trials evaluating cancer prevention will enroll patients at higher risk of developing cancer because studies evaluating the incidence of cancer in unselected populations can take decades. This has the advantage of a shorter potential trial, but might potentially limit the breadth of application of results (i.e. can the results be applied to everyone, not just high risk people?). In the eight year period, 24 patients developed cancer. Unexpectedly, most were not of gastrointestinal origin. But, fewer patients in the ginseng vs. placebo group developed cancer: the observed odds ratio of the declared primary endpoint was 0.54 with 95% confidence intervals of 0.23 to 1.28, p = 0.13. This indicates that there is a 13% chance that the observed differences are due to random chance, and that there is 95% confidence that the true effect of 1g of red ginseng extract consumption per week for three years is between a 77% reduction in risk of developing cancer, to a 28% increase risk in developing cancer, with a mean of 48% reduction of risk. https://www.ncbi.nlm.nih.gov/pubmed/20521975
The prospective study followed patients for only 8 years, and additional follow-up would most definitely provide more power to see if the effect gets stronger with time (it should if it’s real). The study was published in 2010. It is not uncommon to publish a follow-up study from trials like this after more time has passed, and more potential cancer cases have been observed. Unfortunately, a search for this potential follow-up analysis came up short. It is my hope that the authors of the original study continued to follow patients over time and if so, will share these results.
My co-workers and colleagues often joke that I am a bit too much of a “data skeptic” and that perhaps my bar for evaluating research is a bit too high. I am often highly skeptical of nutritional studies (as you could probably tell from my above commentary) but I will say that the result of this literature dive has made me give a fresh look at ginseng. No medical intervention is without risk, but ginseng is generally well tolerated and few people experience side effects. https://www.ncbi.nlm.nih.gov/pubmed/21704950 I personally seldom take supplements, but even this data skeptic is considering purchasing red ginseng extract tablets!
If you have followed me this far I would like to thank you, dear reader!
Ryon