Estrogen Metabolite Ratios: Time for Us to Let Go
by Jacob Schor, ND, FABNO (January 2013)
Response to Dr. Jacob Schor's Article 'Estrogen Metabolite Ratios: Time for Us to Let Go'
by Thomas L. Klug, PhD (April 2013)
In an article published in the Townsend Letter in January 2013, I wrote that the theory of estrogen metabolite ratios; that is, the idea that the ratio of 2-hydroxyestrogen and 16-alpha-hydroxyestrone was predictive of estrogen dependent cancers, had pretty well been disproved. I argued that we should no longer test or use these estrogen metabolites to justify clinical decision-making.
While I anticipated that my article might arouse some backlash, I did not expect to be attacked by so august a person as Jonathan Wright, MD. Dr. Wright and Thomas Klug, PhD, have collaborated to create a rebuttal to my article that was published in the April issue of this magazine. Neither contributor brought forth arguments sound enough that I feel any need to amend my earlier statements.
In my article I reviewed seven recent studies published that did not find a significant association between these estrogen metabolites and breast cancer risk.1-7
I mentioned the 2011 study by Zeleniuch-Jacquotte on endometrial cancer risk, which also failed to support the EMR theory.8 Additionally I pointed out a similar lack of support in the 2012 meta-analyses by Obe.9 As none of these large recent studies support the theory, I suggested that it was time for us to abandon it.
This struck me as rather straightforward.
Given the reputations of Wright and Klug, I am obligated to look closely at their responses and question whether I have been too hasty in my reading of the literature.
Admittedly, several hundred years ago, anecdotal reports were the mainstay of the scientific literature, and Dr. Wright's use of stories to put forth his points would have been acceptable evidence. Over the last few hundred years, medical science has shifted to collecting data on multiple patients and using mathematical analysis to extract information. Conclusions reached from anecdotal observations are now viewed with caution; our brains will often create an association that is not truly present:
… we have evolved brains that pay attention to anecdotes because false positives (believing there is a connection between A and B when there is not) are usually harmless, whereas false negatives (believing there is no connection between A and B when there is) may take you out of the gene pool. Our brains are belief engines that employ association learning to seek and find patterns. Superstition and belief in magic are millions of years old, whereas science, with its methods of controlling for intervening variables to circumvent false positives, is only a few hundred years old.10
While we may enjoy Dr. Wright's clinical anecdotes, his description of a former employee with a striking family history of breast cancer and whose 2/16 ratios were suspiciously low, should not be confused with evidence. We have an ethical responsibility to inform our decision-making on more than second-hand stories. The stories, which Dr. Wright uses to argue his point, miss the target entirely.
That is because Dr. Wright's response centers on the clinical effectiveness of diindolymethane (DIM) or indole-3-carbinol (I3C). In doing so he has shifted the discussion away from whether the theory of estrogen metabolite ratios (EMR) is true to whether or not these two chemicals, DIM or I3C, have clinical benefit. He wants you to think that if DIM works, the theory works. This is not relevant or logical, yet one has to give Dr. Wright credit; it certainly is clever.
Although at one time the EMR theory was used to explain why these supplements were clinically effective, these days, there are other better explanations.
A January 2013 paper described using intranasal DIM to treat lung cancer in mice and reported profound benefit; the DIM significantly reduced the number and size of tumors. The authors do not attribute the benefits to any shift in estrogen metabolite ratios but rather to suppression of the phosphatidylinositol 3-kinase (PI3K)/Akt signaling pathway.11
An October 2012 in vitro study that compared the effect of cabbage juice, I3C, and DIM on breast cancer cells attributes the cancer inhibitory effects to aromatase inhibition, not shifting EMRs.
Another paper published in October 2012 reports that DIM activates AMP-activated protein kinase (AMPK), which suppresses the mammalian target of rapamycin (mTOR), and that this leads to induction of apoptosis.12 AMPK activation is the same mechanism of action used to explain metformin's anticancer action. An earlier April 2012 paper had already suggested that DIM was an AMPK activator and that this accounted for DIM's action against ovarian cancer.13
AMPK may not be the only way that DIM acts against cancer. Other papers suggest that it acts through histone deacetylase (HDAC) inhibition, at least in prostate cancer and bladder cancer.14,15
Dr. Wright specifically mentioned Bell's 1999 study to support his opinion. In Bell's study, about half of the 17 women with cervical intraepithelial neoplasia (CIN) who took I3C responded well. While at the time, the resulting dose-dependent shift in the 2/16 alpha-hydroxyestrone ratios got Dr. Wright's attention and still serves as his explanation for the benefits seen, he neglected to mention that the majority of the women in this study, 15/17, were infected with human papillomavirus (HPV).16
We now know that human papilloma viral infection causes cervical neoplasia (CIN) and that DIM prevents or inhibits the progression from cervical dysplasia to cancer. In earlier studies this was partially explained using the EMR theory.17 But in the current studies little mention is made about estrogen metabolite ratios. Instead the explanation is that DIM increases serum interferon-g levels and so fights the infection.18,19 Stimulation of interferon-gamma is now also considered a key mechanism in DIM action against breast cancer.20
The North Shore study by Chen et al. that Wright mentions certainly tells us that DIM kills cervical cancer cells, but the authors of the study make no mention of estrogen metabolites.21
That DIM appears to be beneficial in treating some types of cancer is not relevant to whether or not the EMR theory is true. There are clearly better explanations as to how DIM inhibits cancer growth.
Let's not confuse the issues. Shifting 2/16 ratios by consuming cruciferous vegetables or taking DIM may simply be a measure of consumption of anticancer phytonutrients and an indirect measure of the degree that the other mechanisms of action triggered by the DIM will come into play.
Thomas Klug's "rebuttal" offers even fewer insights into why we should preserve this EMR theory than does Dr. Wright's.
Don't be distracted by Klug's description of the intricacies involved in his development of the EIA tests that he manufactures to measure estrogen metabolite ratios. Don't be distracted by all the big words. Where we need to focus our attention is on the studies that Klug references in support of the EMR theory.
He cites two clinical trials, Kabat et al. from 1997 and Ho et al. from 1998. Ho's roup obtained samples from 65 breast cancer patients and 36 controls. Analysis suggested that women with a high EMR had a 0.1 relative risk of breast cancer, or in other words were 90% less likely to get breast cancer than women with a low EMR.22
Kabat's 1997 study analyzed EMR data from 42 women with breast cancer comparing them ith 64 controls. Again, a significant inverse association was found between EMR and breast cancer. This was exciting at the time and led the authors to write, "Larger studies are needed to confirm these results and to assess the relationship of the EMR and of the individual metabolites with breast cancer. …"23
Well, this is the problem. The larger studies have now been done and they have not confirmed these early results. It was these larger studies that I cite in my article. Let's take a quick look at the numbers:
We should also relook at the meta-analyses by Obe et al. Their combined studies yielded data from 682 premenopausal and 1189 postmenopausal cases of breast cancer.
It is not unheard of for early clinical studies to provide results that are not confirmed by later work. This is how science often works. If initial tiny pilot studies reliably yielded accurate results, why do scientists always ask for larger follow-up studies? In fact, it is well accepted that the impacts demonstrated in early studies often decline or disappear in subsequent studies; that phenomenon has been named the decline effect.30
While Jonathan Wright and Thomas Klug wish us to believe that these newer, larger studies all suffer from errors in methodology, there is little reason to believe that the authors conspired to falsify their data.
The scientific method asks that we trust more recent and larger studies over older smaller studies. Wright and Klug are asking us to do the opposite and ignore the existence of these new studies and cling to the results of the now outdated reports. After all, that is what they seem to be doing.
Science rarely follows a straight line, and we must recognize that knowledge and understanding change over time. Perhaps at some future date, some refinement in testing, data analysis, or understanding may revive this EMR theory and bring it back to life. Some of us might even hope that this happens; it was such a tidy theory. But right now, at this point in time, the theory is not supported by current science and it's time that we stop "believing" in it.
Jacob Schor, ND, FABNO