Selenium Supplement Cuts Cancer Death Rate in Half: An interview with Dr. Larry Clark



Part II: Clinical Study details and basic selenium facts.

By Richard A. Passwater, Ph.D.

In Part I, Dr. Larry C. Clark, M.P. H., Ph.D., the Director of the Arizona Cancer Center's Epidemiology Program of the University of Arizona (Tucson) chatted with us about the results of his clinical study showing that supplementation with high-selenium yeast cut the cancer death rate in half. His clinical study was reported in the December 23, 1996 issue of the Journal of the American Medical Association. In this second part, we will discuss some of the details of his clinical study along with some background information on the trace mineral selenium and selenium supplements.

Passwater: Dr. Clark, both the clinical results and the statistics (p values) for these results in your clinical study were impressive. Another important observation is that six of seven centers found consistent results for total cancer incidence and mortality. I would imagine that the National Cancer Institute (NCI) would be ecstatic at your results. I assume that NCI is well aware of your results. What is their reaction so far?

Clark: Yes, they are aware of our results. I think right now they are being extremely cautious because of what has happened with the recent beta carotene studies. I think we need to be somewhat cautious because this is one study and we need to replicate it again in the same population and in other populations to see how generalizable the results are.

Passwater: Are you able to get the funding you need now to continue with these other studies?

Clark: That has been difficult to date.

Passwater: That is what I find amazing. Perhaps after the publication of your results will attract some interest.

Clark: I think after the publication in the Journal of the American Medical Association, I think it will be easier.

Passwater: If the scientific community still doesn't respond I am certain that the public will pressure the institutions to fund you.

Clark: I hope so. We need a number of studies to convince everyone. If there were five or six studies in progress, being conducted in different populations and looking at selenium in different ways, that would be very useful. Hopefully with publication of our results a number of new selenium clinical trials will be getting under way.

Passwater: At this time, you have given two preliminary oral reports. Scientists attending two major scientific meetings have heard your preliminary results. Abstracts have been printed in scientific journals. What has been the response so far of scientists in the audience? Are they saying, gee, I want to go out and start a study of my own or are they saying, "that's interesting" and walking away. What's the reactions been from these meetings that you discussed your results.

Clark: I think it runs the gamut from people who are extremely pessimistic of the results and cannot believe them thinking "the results are too good to be true" to people who find it very interesting and think that it needs to be replicated and want to design their own studies at their own institutions to confirm them.

Passwater: Maybe they will. There are already several researchers involved. In your latest study, there were more than 30 researchers at seven clinics, plus your own core team, who participated. I am sure that they will want to continue this line of research.

Clark: Yes, besides those of us from the Arizona Cancer Center of the University of Arizona and Cornell University, others involved in the clinical study include researchers from the Medical College of Georgia, University of South Carolina, Georgia Dermatology and Skin Cancer Clinic, Columbia Skin Clinic, Coliseum Hospital, Quadrangle Medical Specialists, Wilson Dermatology Clinic, University of North Carolina, University of Connecticut School of Medicine (UCHC), Eastern Dermatology (Greenville, NC), Columbia Urological Associates (SC), Dixie Medical Center (ST. George, UT) University of Miami School of Medicine, Veterans Administration Medical Center of the University of Miami, South Suburban Gastroenterology (South Weymouth, MA), East Carolina University, Columbia Gastrointestinal Endoscopy Center (SC), Wilson Gastroenterology (NC) and private physicians.

Passwater: Let's go back to the beginning for a moment. Things just don't happen --there is a reason for everything that happens. I am interested how your thinking was shaped and I think I see a thread that I want to follow for a moment. Your selenium research started at the University of North Carolina. You then moved to Cornell where several researchers have been key pioneers in selenium research. In early 1960, I incorporated the research of Drs. Milton Scott and several of his colleagues who were researching the inter-relationships of selenium and vitamin E into my research on antioxidants. They were elucidating the role of selenium in the kidney and liver health of chickens. The selenium-Factor-3 of Dr. Klaus Schwarz had a synergism with vitamin E in some liver and kidney functions, and I thought that factor-3 might also help explain the antioxidant synergism that I had noticed with vitamin E and selenium in my longevity experiments.

If I remember correctly, your associate, Dr. Gerald Combs, Jr was already at Cornell. He was a colleague of Dr. Scott doing research on the pancreas of selenium-deficient chicks in 1972 or 1973. I knew his dad who was a professor of nutrition at the University of Maryland from 1948 to 1969. The poultry industry where I live in Delmarva owes a lot to both Dr. Combs and Dr. Scott. Dr. Combs, Sr. was awarded the Meritorious Service Award by the Delmarva Poultry Association. The Cornell connection is that Dr. Combs, Sr. earned his Ph.D. from Cornell in 1948 after taking a hiatus from his education to serve our Country. He earned the Bronze Medal in World War II.

Was the fact that Cornell was strongly involved in selenium research the reason why did you left Chapel Hill to go to Cornell in 1982?

Clark: Yes, Cornell was appealing to me because Dr. Jerry Combs, Jr. was there then. As you know, he is an outstanding nutritional biochemist and expert in selenium. He, and our statistician, Dr. Bruce Turnbull, and I have been a good team. We enjoy working together. It has really been the three of us who form the executive committee which runs the clinical studies.

Passwater: Okay, so it is the three of you and as the studies expanded others have participated in the studies.

Clark: Yes. We formed the executive committee and we also had a number of really committed dermatologists especially Dr. Gloria Graham of Wilson, NC, which is where I did my original research and where we started the intervention trials. I have to say she and our other dermatologists really made it all possible.

Passwater: Was your original objective to study selenium and skin cancer incidence?

Clark: Yes.

Passwater: But, apparently, your data does not seem to support your earlier study that selenium was protective in skin cancers.

Clark: Well, to date, the overall information suggests that there is no significant difference between the treatment and placebo group. It will be interesting to monitor this over a longer period of time.

Passwater: Were the subjects enrolled in your clinical study limited to persons who(m) have already had cancer?

Clark: These are people who have already had a history of either two carcinoma of the skin, basal cell or squamous cell carcinoma.

Passwater: Would you expect any difference in results when the general population rather than just those who have already had an incidence of cancer are studied?

Clark: That is a very important question and that is why we need to expand the studies beyond people with a history of skin cancer.

Passwater: Didn't your original research originally examine selenium and skin cancer in the general population?

Clark: The case control study was in a dermatology population. It took a group of cases and compared it to a group of non cases. The rest of the studies have been in people with a history of non-melanoma skin cancer.

Passwater: What gave you the presence of mind to look at all the other cancers. Here you are looking mainly at a very narrow end point. You could have set up your study to look only for various skin cancers, but you decided to look at all these other cancers.

Clark: Well we were certainly aware of the literature that the effect would probably go beyond skin cancer but we originally never dreamed that we would have enough cases to analyze it. Our concern was really a safety concern. We wanted to well-document that there were no adverse health effects from this dose of selenium. One of the real problems in getting funding for selenium research is concern about its potential toxicity. So we bent over backwards to get as complete a health assessments on our patients as possible. So not only did we collect information on illness, but we also collected information on medication to complement that database.

Passwater: Will you be publishing any of the data from your investigation into the safety of selenium?

Clark: That is one of the manuscripts we plan to publish down the line. It is not a real exciting paper because toxicity is not a problem at 200 micrograms per day, which is the level that we found was very protective against cancer.

Passwater: What type of criticisms have the journal peer reviewers expressed? There is little doubt that the results are not do to chance. The p value trends are very reassuring.

Clark: Some of those p values are very small.

Passwater: Yes, p=0.0001 for total carcinoma incidence and p=0.0003 for total cancer and p=0.0003 for prostate cancer incidence for examples. Rarely do we see p values that small in a human clinical study. So it will be hard for reviewers to find much to be concerned about.

Clark: Their concern is that we saw the effect for secondary endpoints and not for primary endpoints. They also wonder how is it possible that we could see an effect on lung cancer, colon cancer, and prostate cancer but not see it on skin cancer? The more extreme view is basically saying it sounds like you have a therapeutic agent rather than a preventative agent. All we can say is that this is the data that we observed in our study. I am not sure that when the study is replicated that other people are going to find as strong a protective effect as we found. I am not saying that we found a 50 percent reduction so therefore the true estimate of risk reduction is a 50 percent reduction in cancer mortality or incidence. What I believe is that we have a significant reduction, the real magnitude will need to be determined in other studies. The magnitude may also vary due to the population studied and the age that they start taking selenium.

Passwater: I am not a betting man but I would certainly wager any money I could on the fact that the other tests will confirm what you have done.

Clark: Yes, I think that when they are done, the majority of them will confirm that. However, there may be populations where selenium will not appear to have quite this much of a protective effect.

Passwater: Do you feel there is a threshold level?.

Clark: There are a couple of things that could happen. I think this is important to discuss because you have to be realistic. For one person to fail to find it and stand up and say I have the truth and everybody else has been wrong is not appropriate. I think if you put it in perspective before they do that you are in better shape. For instance in the Physicians Health Study, the risks of disease in these physicians was so low that it is possible that selenium might not be able to show much of an effect there. If a population that has a relatively high selenium diet, you also may not see a protective effect of a nutritional dose of selenium.

Passwater: The amount of harm that selenium must protect against is equally as important. Let's look at skin cancer. If you take a northern climate you might be able to see that selenium in the diet will overcome the smaller amounts of sun exposure that the northerners get but perhaps in the south they are over a threshold and it would take much larger amounts of selenium if it could protect at all.

Clark: Or a longer period of time of supplementation.

Passwater: That's what you are going to continue to look at, I hope.

Clark: Right. That's the other thing that makes this research so difficult to do is that in a highly exposed individual you may have to give it to them for an extremely time before you see an effect. Our data suggests that there might be a limit to how much protection you can get but that is a little bit speculative. But we also know that selenium interacts with other nutrients such as vitamin E so it may be that you will see a better protective effect in populations that have low levels of vitamin E and if you study a population that has adequate levels of C and E and carotenoids and other fruits and vegetables, that you would not see as much of an effect.

Passwater: That is what is so amazing about your research. I have always been a strong proponent of the antioxidant vitamins E and C being partners and you are going to see your biggest changes in those that are both low in vitamin E and selenium and your most protection in those that are highest in both of them. But you were able to show with selenium alone in a classical scientific study of one variable show this tremendous difference. That has really caught my attention.

Clark: We are also going to publish a paper on subgroup analyses. One of the subgroups we are going to look at is vitamin E.

Passwater: Have you analyzed the data in that yet?.

Clark: We have. There is a suggestion of a stronger protective effect of vitamin E on mortality. We are currently analyzing vitamin E's interrelationship with selenium treatment.

Passwater: It sounds as if this is going back to what Dr. Jukka Salonen and his colleagues found in their 1985 report on the North Karelia Project. [British Medical Journal 290:417-20 (Feb. 5, 1985)] In studying over 12,000 Finns for an average of about five years, they found that the relative risk of cancer mortality for the third of people with blood selenium levels below 47 micrograms per liter of blood was 5.8 to 1. But of more importance is the finding that, for people with low selenium levels who also had vitamin E levels in the lowest of values, risk of death from cancer, compared to persons with both selenium and vitamin E levels in the upper two-thirds of blood values, was 11.4 to 1.

Do you plan to look at any other antioxidants?

Clark: As a matter of fact, in our original case control study we looked at carotenoids and retinol in addition to selenium. What we found was that if people were above the median levels for both carotenoids and retinol, then selenium levels didn't really make too much of a difference.

Passwater: You were pretty well ahead of the curve working with carotenoids back then. It wasn't until 1981 that people even began discussing a possible protective role of beta carotene against cancer.

Clark: Dr. Richard Peto's paper "Can dietary beta-carotene materially reduce human cancer rates?"came out just as we were finishing our first study. [Peto, R., Doll, R., Buckley, J.D. and Sporn, M.B. Nature 290(5803):201-8 (Mar 19, 1981)]

Passwater: And you were studying carotenoids and cancer prevention before that paper came out.

Clark: Right.

Passwater: In your clinical study, was "placebo/treatment" code broken earlier than you intended because you uncovered the dramatic difference in the cancer rates between the two groups?

Clark: We unblinded the study and broke the codes because there were major differences between the two treatment groups and the safety committee wanted to know which was which. And, then after they did that, we continued to follow the data for another year or two just to make sure that the trend would continue and to allow us to completely as possible document all the cases of cancer to make 100 percent sure that what we were observing was real.

Passwater: Was this for ethical reasons? Did you feel that you could not withhold selenium from the placebo group?

Clark: Yes! We also felt we would get the most information from the people in the study if we put them all on study and these people had contributed ten years of their life to this project. I sort of felt like we should give them something back for that and give them access to a good high quality selenium product.

At this point what we have done is we have informed patients of the study results and told them what treatment group they were in. We have offered all of them our high-selenium yeast supplements for the remainder of the study period through 1998, rather than letting them go to the drug store and buy a variety of products and at different dosages. If that happened, it would be very difficult to control the study.

Passwater: How about the amount chose 200 micrograms. I assume you needed something to be high enough to be effective and yet within dietary guidelines of the RDA publications. Is that correct?

Clark: Yes, the FDA at the time we started our study had what they called the upper range of the safe and effective dose which was 200 micrograms daily.

Passwater: Speaking of the FDA, it was not easy to get official recognition for selenium as an essential trace mineral. Even after the 1972 announcement at the 56th Annual FASEB Meeting by Drs. John Rotruck, Bill Hoekstra and Howard Ganther et al., that selenium was an integral part of the glutathione peroxidase molecule, and full publication in Science [179:588-90] in 1973, selenium was still ignored. I, and others including Drs. Walter Mertz and Orville Levander, aggressively lobbied the Committee on Dietary Allowances of the Food and Nutrition Board to include selenium in the Ninth Revised Edition of the Recommended Dietary Allowances (1980).

My 1975 book, Supernutrition: Megavitamin Revolution reported on my research with selenium in cancer prevention and longevity. Selenium supplements became available and the FDA soon brought cases against suppliers charging it to be a food additive. My first assignment at the Solgar Nutritional Research Center was to assist the NNFA and their attorneys, Bass, Ullman & Lustigman in interrogatories with the FDA and attempt to have selenium placed under 21USC350. On February 14, 1978, the FDA Director of Regulatory Guidance Caesar A. Roy wrote "based upon the FDA's present understanding of the current relevant scientific information, the Agency will not recommend legal action based on food additive charges against products containing safe and suitable forms of selenium."

Passwater: What would be the typical dietary range of participants?

Clark: They averaged about 90 micrograms a day.

Passwater: Do you have any thoughts that maybe more would be even more protective?

Clark: At this point in time our data shows I think pretty clearly that 200 micrograms is effective and safe. We don't have any data from our studies that suggests that a higher dose is better.

Passwater: Now, the big question. How does selenium reduce cancer incidence and mortality? It has been assumed that since cancer develops over decades, it would take decades of selenium supplementation to witness such a marked reduction in clinical cancer incidence or mortality. You saw the cancer incidence cut nearly 40 percent and cancer mortality cut in half in just a few years. What do you think is happening?

Clark: Well, there are eight or ten mechanisms by which selenium may be working. Each of those probably have a different dose response curve and a different time to effect. But I think besides the antioxidants I think one of the most likely possibilities is through the mechanism of induction of apoptosis --and there we have done some tissue culture work with a colleague here, Mark Nelson and his graduate student Claire Redman? In tissue culture they have been able to show for both the lung and colon cancer cell line that selenium enhances apoptosis. This is really pretty remarkable. We are going to be looking at that in cancer biopsy samples that we have from our trials and see if we can get some evidence for that but to me that is the most likely explanation at this point but as you know these clinical trials are not designed to investigate mechanisms. They provide the empirical answer.

Passwater: Do you recommend people to take selenium supplements? Are you taking selenium supplements yourself?

Clark: To make a public health recommendation, you really need two trials to show a protective effect and this is the first trial to show that. We can't make a public health recommendation but we can say that a prudent person who needs to make a decision now on taking selenium on reading this study and the large animal literature and speculate would probably want to take the supplement now rather than wait another ten years.

Passwater: That has always annoyed me. The epidemiologists God love them they have brought us a long way often seem to be completely unaware of the animal studies and when they are made to look at the animal studies they say that is in a rat or mouse that is not in a human. They also seem to be unaware of any data reported before 1970 which is as far back as most computer data bases go.

Clark: We did a review. There is now over a hundred experimental animal studies in 25 plus model systems. Two thirds of the studies show a significant reduction in tumor incidence. In half of those studies, or a third of the studies, showed greater than a fifty percent reduction in tumor incidence. That is remarkable consistency.

Passwater: That's not bad and especially since we don't know exactly the mechanism or what forms of selenium are best in the body. A lot of research to be done yet. We may be able to fine tune it to a better more efficient form of selenium to be ingesting.

Clark: I think that is true for higher risk patients. I think there is going to be tremendous public health benefits just from doing the basic 200 micrograms a day. It appears to be safe and effective, however, time will tell the accuracy of our results.

Passwater: As we have pointed out, food tables are not meaningful because of soil selenium variations in which food is grown. You can eat a well-designed diet rich in organic foods and still be selenium deficient because the soil was deficient. Organic does not reflect on selenium content. The selenium has to come from somewhere in order to get into the soil so that it can get into the plant. Animals are low in selenium unless they eat foods rich in selenium. Fertilizers don't have selenium unless its ingredients are selenium-rich or it has been especially added.

Clark: I think that is one aspect that makes the epidemiology of selenium and cancer so difficult is that we can't use dietary questionnaires to measure people's selenium levels except in very rare instances so we are forced to make biologic measure and we can't be sure we are measuring the right metabolic pool.

Passwater: Thank you so very much. Both for doing the research and for sharing the results with us. I have waited for this confirmation for 25 years, similar to my having to wait 17 years for confirmation of my 1976 study showing that vitamin E was protective against heart disease. How sweet it is!

All rights, including electronic and print media to this article are copyrighted by Richard A. Passwater, Ph.D. and Whole Foods Magazine (WFC Inc.).