Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
THE QUESTIONS OF BENEFITS AND RISKS Reginald F. Crampton Perhaps I should not have to remind such an audience as this that in a democratic society it is recognized that there should be freedom to undertake risks and to accrue personal benefits that might arise as a result of undertaking those risks. It was that fundamental concept that indicated to at least one philosopher that democracy was not a very good form of government. Socrates postulated that the inevitable re- sult of democracy would be tyranny, which, as you know, can take the form of either personal dictatorship or that exercised by bureaucracy. One might bear this in mind when one is talking about the risk-benefit analysis of any substance, including saccharin. One can divide risks and benefits into two broad categories. There is firstly the qualitative assessment. One can say: "Does saccharin cause bladder cancer in man? Please tick 'yes', 'no'." "Is saccharin good for fat people? Please tick 'yes', 'no'." Much of this sort of assessment of the risk-benefit, of course, is motivated not so much by science as by personal experience and emotive input. However, to be serious about a risk-benefit analysis one should deal essentially with quantitative data and try to find, no matter how difficult it may be, some units in which to express the types of risk and the alleged and defined benefits. So, Mr. Chairman, I would not entirely agree with you that such a meeting as this should be devoted to a nonscientific and non-data- producing procedure. One often hears that inasmuch as a compound has been in use for fifty years and no one has seen any adverse effect, therefore those effects do not, in fact, occur. Now this may well be true of adverse effects that are catastrophic, where the identification of them is apparent even to the most humble citizen. But in order to validate that conclusion I would suggest that such a statement should l27
l28 have, as its basis, an active search for specific effects and a list of data which show that no effects had been observed -- negative data, if you like. On that basis I think the statement might have some justifi- cation; but otherwise, apart from acute and catastrophic episodes, I do not think it has very much. I would like to put before you the following types of benefits: direct personal -- subjective and objective -- economic, technological, international trade, and regulatory. This lists in some rather general terms the sort of benefits that might accrue from the use of any agent or the adoption of any procedure. They are perfectly straightforward and simple. What I propose to do is to go through this list in terms of saccharin and to pose some questions and indicate some possible basis of answer to them. The last two items on the list may be a bit contro- versial, but I will say more about that when we get to them in the discussion. Let us take the personal benefits. We heard yesterday that the sub- jective craving for sweetness has a physiological basis. If satisfy- ing this need for sweetness is to fulfill a physiological requirement, then it is reasonable that it be regarded as a benefit. The question remains as to how to measure the benefit. What sort of units of mea- surement can be derived? The following data in Table l indicate the relative importance of the benefit as assessed by government. Europe, which as you know has nearly destroyed itself over the last 80 or 90 years in various stupid ways, has produced as a result of this some interesting data. The two items I would draw to your attention are the impact of the two world wars on saccharin production in Germany. It would be reasonable to assume that in l922 and l944 the vast major- ity of saccharin produced was for human consumption. The importance of these data is to demonstrate that when the usual source of sweetness (sucrose) is diminished, an alternative (saccharin) is used. This illustrates that a government in time of war, when facilities are strained, is willing to recognize a need and devote time, planned dis- tribution costs, and so on, to the satisfying of this need. So here, in some general way, is a kind of assessment in quantita- tive terms of the benefits that at least one country ascribed to saccharin. The figures for the U.K. follow a similar path. I do not TABLE l Saccharin Production in Germany (kg x l,000) l894 30 l922 300 l934 96 l944 500 l965 27
l29 have the kilogram or tonnage figures, but a curve plotting production against time would be quite similar to the German one. During the war period the U.K. government adopted a similar policy. I suggest to you that sugar, whether sucrose or the other sugars discussed yesterday, is bought by the population for sweetness rather than for its calories. In that respect a comparison of sugar and sac- charin can be made directly in terms of the motivation that underlies the purchase of the two. In Table 2 we see that about 53 percent of consumers of artificial sweetening agents are motivated by medical or paramedical reasons. Another l2 percent come under the heading of "weight watchers," a par- ticularly American term. The remaining 35 percent must be assumed to take artificial sweeteners to augment their sweetness intake and possi- bly are motivated by economic considerations. Such a rough estimate of the types of uses of saccharin does indicate that in some areas of, or related to, medical practice, there is a use for saccharin, and this has to be classified as a beneficial use. We heard yesterday about the role of sugar and its relation to four major categories of medical problems. I would like to make a few com- ments about saccharin in these areas. The first one that comes to mind is diabetes mellitus. As you know the incidence of this is in the order of l.5 to 2 percent of the population, and we won't argue the point about how you specifically define diabetes mellitus. I think there are two questions related to saccharin in diabetes. First, is the control of diabetes more readily accomplished if one has the facility of using a noncaloric sweetener? The answer to this question, I would suggest, would come not from those people who are working on the biochemistry of diabetes and others seeking to unravel its etiological basis, but from those physicians who actually run dia- betic clinics, who see the patient, who talk about the types of food he eats and how often he eats them. This is the best source from which the most realistic assessment can be made as to whether saccharin is of benefit in the day-to-day management of the diabetic. The same would be true of other noncaloric sweeteners. The second question is perhaps more open to criticism, as it relates to the quality of life. One could ask if the quality of life for the diabetic is improved by the use of noncaloric sweeteners. I would sug- gest that diabetic doctors themselves would provide the best answer to this question. Although one may criticize the idea that the enjoyment of food is a benefit, it is a fact that the majority of the population of North America and Europe do, to some extent, live to eat. Another topic discussed yesterday was obesity, and it was said that the place of noncaloric sweeteners had been disappointing in the manage- ment of obesity. Nevertheless, it is true that every time someone re- quires or desires a sweet cup of tea or coffee and uses saccharin or, in some countries still, cyclamate, they are in fact not consuming so many grams of sucrose. Perhaps the relative failure of the noncaloric sweetener in the management of obesity and weight control lies in the standard of management of the patient rather than in some property of
l30 TABLE 2 Artificial Sweeteners U.S.A. Census l967/68 Consumed by 20 x l06 (I) Medical, dietary restriction l0.6 x l0^ (II) Weight watchers 2.4 x l06 (III) Not specified 7 x 1O6 the noncaloric sweetener. It would be possible to devise experiments or surveys that would give some quantitative assessment of the value to different sorts of people of noncaloric sweeteners in terms of being able to control and/or reduce their body weight. The next subject, and one which provoked the least degree of contro- versy yesterday, is dental caries. While I would support the conclu- sion that no one single preventative measure is the answer to dental caries, some control of sugar intake is an important factor. The consumption of sugar, depending on the form in which it is presented, particularly to the child, is an etiological factor. There are some data on this as a result of the sugar shortage that occurred in Europe during the last world war. Particularly in the U.K. and Scandinavia, the incidence of dental caries did drop remarkably. This was, in a way, a self-controlled study in that fluorination of water supplies had not then been adopted. To what extent can one say that the use of saccharin could be of benefit in the prevention of dental caries? It is difficult to answer this question quantitatively in the absence of any controlled trials. Could, for instance, the supply of sweet products for children reduce exposure to sucrose, assuming that the use of saccharin was compatible with safety and consumer acceptance? These are questions that could be answered by programs of technology and long-term consumer research that might be operated jointly by industry and government. The last subject was cardiovascular disease, and what emerged yester- day was the agreed complexity of the interrelationships between obesity and diabetes, coronary thrombosis, and hyperlipemia. So one cannot really talk seriously about this problem in terms of saccharin, in terms of sugar substitution, in any way that is meaning- ful. Yesterday there seemed to be a generally agreed conclusion that no one would be overly concerned if the consumption of sucrose was in the order of 2 g/kg body weight per day or less. But there was some, perhaps almost unspoken, degree of concern that if it exceeded this figure by a large amount some apprehension would become evident about adverse effects. One should ask the sensory physiologist if this craving that every- one agrees we all have, and for which there is a physiological
l3l explanation, does in fact have a threshold. I know of no data indi- cating that individuals having unlimited access to sweet foods demon- strate any self-imposed limit on their sweetness intake. One rather assumes that there must be such a limit. If this is true, for many individuals it is certainly above the level of 2 g/kg body weight/day of sucrose. The benefits that might accrue by the use of saccharin or other agents to meet the need for sweetness in excess of that level have yet to be assessed. The relevant question on economic benefits, the next general heading under types of benefits, might be: Would the availability of noncaloric sweeteners enable the food industry to produce food at lower cost to the consumer? The answer to this question must obviously come from the industry itself. This particular question is relevant in those areas of the world where lack of food is not a pressing problem. It may also have some relevance when the price of sugar and its availability show marked changes, and this has occurred in recent times. The other aspects of the benefits I can quickly dismiss. The tech- nological benefits, again, I think are a question for industry on whether it could produce better things if more basic materials were available to it. The trade and the regulatory benefits I will pass over because the remaining time can be better devoted to some aspects of the risk factors. Although these will be covered more fully by Dr. Coon, discussing the NAS report on saccharin, there are just two observations I would like to make because they are not contained in that report. They are related to two studies that have been carried out in the U.K. by Sir Richard Doll and Dr. Armstrong at Oxford. These were epidemiological surveys, and Dr. Armstrong was kind enough to give me permission to talk about them very briefly. The first study compared the possible impact of cigarette smoking and of saccharin on the incidence of bladder cancer in populations born in l870 and thereafter. Briefly, there has been a 36 percent rise in bladder cancer in the male, and about l2 percent in the female. The investigators' analysis of data took as its basic premise the assumption that this increase was due to an environmental agent, to wit saccharin. Did the data support or deny this assumption? The conclusion reached was that there was no evidence to support this assumption, but that the increase in bladder cancer was related to smoking. This study has been published (Br. J. Prev. Soc. Med., 28:233-240, l974). The second study will be published later this year. This is a com- parison of bladder cancer in diabetics and nondiabetics. The diabetic male consumes about ten times more saccharin than the nondiabetic; the diabetic female, about two times more. I will discuss the results very briefly. First, there was no increased incidence of bladder cancer in diabetics. Second, there was a lower incidence of bladder cancer in diabetics, but this was not statistically significant. Third, diabetics smoked less than nondiabetics, and the nonsignificant lower incidence of bladder cancer in diabetics would fit the hypothesis developed in the first paper that cigarette smoking is far more likely to be an etio- logical basis of bladder cancer than the intake of saccharin. These
l32 conclusions have some qualifications, the most important of which is the assumption that the induction period of bladder cancer if it were caused by saccharin would be less than thirty years. In summary, data do exist on which one could form some risk-benefit analysis. More data could be generated relatively easily that would make the risk-benefit analysis more complete. Perhaps the most impor- tant question that one should ask is whether the community is willing to accept a rational risk-benefit analysis as a basis for decisions. And further, is the community willing to support the effort that is needed to develop such analyses? From a European standpoint it seems that the answer to the latter question in the United States is no. The United States did set up an organization devoted to this end, i.e. the Citizens' Commission on Science, Law and the Food Supply, but appar- ently it has run out of steam only because it has run out of money. Finally, Mr. Chairman, may I express my appreciation and thanks to the Academy for inviting me to address you. DISCUSSION PFAFFMANN: In view of the fact that we have a set series of substantive matters to be presented, I think the comments on this first presen- tation should be directly to the substantive issue, rather than the debatable aspects of the matter. DAVID KIM, Mitsui & Company, New York: Dr. Crampton stated that in relation to obesity, the failure in management of the diet may be attributed to the patient rather than to some property of saccharin itself. I would like to ask to what data are you attributing this explanation. CRAMPTON: The point I was trying to make arose from the comments yesterday that in spite of the use of the noncaloric sweeteners, they did not seem very successful in promoting the weight reduction or weight control in individuals. It was suggested that perhaps it is a psychological effect. You give saccharin to a person, and he feels that by taking it he is adopting a weight-reducing regimen, and so he can go ahead and eat all his pies and pastries when he gets back home. If he does this, I would suggest that this is the result of dietary mismanagement of the patient, because it is un- doubtedly true that for every milligram of saccharin he takes, he is saving the caloric intake equivalent to 300 milligrams of sucrose, or thereabouts. It was just a simple way of saying that the use of saccharin or other noncaloric sweeteners is really dependent on the extent to which they were part of a regimen and not an end in themselves.
TOXICOLOGY THE REPORT OF THE NATIONAL ACADEMY OF SCIENCES Julius M. Coon I think it can be said in regard to the toxicology of saccharin that the issues are the uncertainties, and uncertainty is the main issue. My assignment is to talk about the so-called Academy report entitled "The Safety of Saccharin and Sodium Saccharin in the Human Diet." It has been emphasized that the Forum is not supposed to be too scientific; but I have to point out that this report is almost purely a scientific document, and in talking about it, I don't know how I can avoid being somewhat scientific. First, I want to enter a few disclaimers. The first is that the report I am going to talk about is not up to date on the saccharin tox- icological data. The report was completed in July, l974. It made rec- ommendations for further work, some of which was already in progress at that time and is still in progress. I hope we will hear about some of that work a little later this morning. Another disclaimer is that the report itself is not all-inclusive. At the time it was prepared, it did not include all of the toxicologi- cal data that came to the attention of the committee. It summarized primarily the data that made it necessary to prepare the report in the first place. On that basis its main focus is on the question, Is saccharin carcinogenic? and even more specifically, Is saccharin a urinary bladder carcinogen? I should indicate that the charge to the committee from the Food and Drug Administration was to determine when the experimental findings are sufficient to conclude that saccharin is or is not carcinogenic when administered orally to test animals, and to prepare and submit a report to the Food and Drug Administration on the safety of saccharin and saccharin salts as they are used in the human diet. l33
l34 Dr. Crampton made a historical note, and I would like to make one too. Saccharin was discovered in l879, and it was first used in the l880s as an antiseptic and as a food preservative. It was in the mid- l880s that the material was first used by diabetics. It was not until l907 that the canning industry in this country began to develop an in- terest in sweetening their canned food products with saccharin. A little story has made the rounds in that connection. In l907 Dr. Harvey Wiley, who was Chief of the Bureau of Chemistry of the Department of Agriculture at that time, was advising President Theodore Roosevelt on the use of saccharin in canned foods, and in regard to canned corn he said, "Everyone who ate that sweet corn was deceived. He thought he was eating sugar, when in fact he was eating a coal tar prod- uct, totally devoid of food value, and extremely injurious to health." In response, President Roosevelt said, "You tell me that saccharin is injurious to health? My doctor gives it to me every day. Anybody who says saccharin is injurious to health is an idiot." So as long ago as l907 saccharin was, so to speak, fully evaluated for its safety. Teddy Roosevelt, of course, still occupies a place in history as one of the great Presidents of the United States, perhaps because he usually said what he meant. He proceeded to appoint a board of scientific advisors, which several years later, in l9l2, concluded that three-tenths of a gram of saccharin per day was safe and that one gram per day may cause digestive disturbances. The latter amount today is the FDA recommended limit for the daily consumption of saccharin. Saccharin continued to be used widely, with peaks of use during World War I and World War II, throughout the next 60 years, and there has yet been no evidence of injury to public health as a result. The committee report emphasizes that the long-continued and widespread use of a chemical substance without any harmful effects coming to light does not in itself provide proof that it has not produced some subtle, insidious, harmful effect. This concept, of course, applies to saccha- rin. But the report also claims that such apparent absence of harm should be put in the balance and weighed accordingly in the evaluation of safety. There are two synthetic processes for producing saccharin. One is called the Remsen-Fahlberg Process, which starts with toluene as one of the reacting materials. The products from this manufacturing process contain an important impurity, which you will hear more about later. It is usually referred to as OTS, orthotoluene sulfonamide, and the various commercial samples that have been used in most of the toxicolog- ical tests contain this impurity in concentrations ranging from ll8 to 6,l00 parts per million. The other process, the Maumee Process, does not start with toluene, but with phthalic anhydride, and the OTS impu- rity amounts to only l to 3 parts per million in the commercial product. In the Remsen-Fahlberg Process there are other trace impurities, but at the present time the main focus of attention is on the OTS as the major impurity. Table l illustrates certain things that constitute some of the main issues in the problem of the toxicology of saccharin. These data show
l35 TABLE l Saccharin Carcinogenesis Tests -- Rats 2 Yrs (F,.-F Generation Feeding; In Utero Exposure) Lab. (Date Finished Saccharin in Diet (%) No. Rats Started No. Bladder Tumors Incidence (%) No. Rats (Sex) FDA (l973) 0 35 l/25 (M) 4 7.5 35 7/23 (M) 32 0 45 0/24 (F) 0 7.5 45 2/3l CF) 7 5 -- Negative (M 5 F) 0 WARF (l972) 0 20 0/l0 (M) 0 5 20 4/l5 (M) 27 0 20 0/l0 (F) 0 5 20 0/l2 (F) 0 the results, in very summary form, of two chronic toxicity tests in which bladder tumors were being looked for especially. One test done at the Food and Drug Administration (FDA) was finished in l973; the other was done at the Wisconsin Alumni Research Foundation (WARF) and completed in l972. The results of these two tests were statistically positive as deter- mined by the committee. Notice the subtitle, in parentheses, which says: "FQ-FJ Generation Feeding; In Utero Exposure." This should be emphasized because it means that both males and females (Fg generation) were fed saccharin from the time they were weaned, and they were mated to produce the FI generation. The females that became pregnant were fed saccharin throughout pregnancy, throughout lactation, and during the preweaning feeding of the young. Then the weaned animals were fed saccharin throughout their lifetimes at the diet levels indicated in the table. In other words, the animals in these tests were conceived, developed in utero, were weaned, and subsequently lived throughout their lives in a saccharin environment. It will be noted that in the percent incidence column the results are not highly significant, but they are significant enough to warrant concern as to the potential bladder tumor genicity of saccharin. The FDA study was carried out at levels of 7.5 and 5 percent in the diet. Only the male animals fed 7.5 percent saccharin in the diet showed a significantly greater incidence of bladder tumors than did the controls. Lower levels were also fed in each of these tests.
l36 In the WARF study there was a significantly positive result, again only in the males with the 5 percent feeding of saccharin. From the results of these two studies it seems apparent that the male is more sensitive than the female to whatever it is that caused the bladder tumors. In the FDA study, the OTS impurity in the saccharin used varied from approximately 250 to 5,000 parts per million. In the WARF study, the OTS impurity ranged from about 200 to 370 parts per million. The interpretation of these results is difficult for several reasons. First, the weaning weights of the animals in the 7.5 percent saccharin test were depressed up to 20 percent in males, and 29 percent in fe- males, thus raising suspicion that toxic effects may have influenced the final outcome. Secondly, bladder parasites or bladder stones were not ruled out as potential contributing factors in the positive WARF test with 5 percent saccharin. Third, the commercial samples of sac- charin used in these two studies contained 200 to about 5,000 parts per million of the OTS impurity, which itself is suspected to produce blad- der stones, which in turn produce bladder tumors in rats. Table 2 indicates the worldwide nature of the saccharin testing pro- gram, especially in the last four years. Here it is seen that chronic toxicity studies have been done in England, Germany, Japan, Canada, and Holland, as well as the United States. Emphasized again in this table is the subtitle: "Feeding Started at Weaning." Since all of these studies were negative statistically, the difference between the design of these tests and that of the FDA and WARF is considered of importance, although its significance is not fully understood. All these tests, except those of Shubik and Golberg, were done with commercial samples of saccharin containing the OTS impurity in amounts ranging from about 200 to almost 6,000 parts per million. The Litton- Bionetics study used saccharin containing the highest concentration of OTS impurity, namely 3,000 to 6,000 parts per million. The Shubik and Golberg studies were done with samples of the saccharin containing only 2 to 3 parts per million of the OTS. It will be noted here that there are ten studies run at the 5 percent level, seven in rats, and three in mice, all of which were statistical- ly negative as far as bladder tumors were concerned. Most of the tests were done with sodium saccharin, though the free acid was used in the Litton-Bionetics Test. Whether or not this difference has any signifi- cance in the chronic toxicologic evaluation of saccharin is not known. An attempt is made in Table 3 to provide perspective as to the con- sumption of saccharin in the United States. The estimate of 5 million pounds as the total consumption per year is probably a little high be- cause about a half million pounds are used for nonfood purposes. But as a rough estimate the average per capita consumption is something of the order of 30 milligrams a day. It should be emphasized, however, that estimates of averages of this kind are rather meaningless when it comes to safety evaluation because we should be primarily concerned with how much saccharin is consumed by about the l percent segment of the population that consumes the greatest amount.
l37 TABLE 2 Saccharin Carcinogenesis Tests (Feeding Started at Weaning) Lab. Year Reported Species No. in Group Max. % in Diet Duration Fitzhugh et al. (FDA) l953 Rat 9 MF 5 24 mo. Lessel (England) l959 Rat 20 MF 5 24 mo. Roe et al. (England) l970 Mouse 50 F 5 24 mo. Miyagi (Japan) l973 Rat Mice 54 M 50 MF 5 5 28 mo. 2l mo. Schmal (Germany) l973 Rat 52 MF 0.5 sac.+cyc. 24 mo.+ Munro (Canada) l973 Rat 60 MF 5 28 mo. Litton- Bionetics l973 Rat 26 MF (dup.) 5 24 mo. Van Esch (Holland) l973 Mice 50 MF 0.5 20 mo. Shubik (Omaha) l973 Hamster 30 MF in water l.25 80 wk. Golberg (Albany) l974 Monkey 3 MF 500 mg/kg/day 6 yr. (cont.) Bio-Research Institute l973 Rat Mouse 25 M (dup.) 25 MF (dup.) 5 5 24 mo. 24 mo. FDA l973 Rat 48 MF 5 28 mo. An approach to such an estimate of individual maximum intake rate can be based on the consumption of soft drinks in the U.S., where 70 percent of the saccharin is consumed in soft drinks. Surveys have shown that about l2 million people consume saccharin in soft drinks and that the top l0 percent of those consumers drink an average of 42 ounces of soft drinks a day. This represents about five ordinary bottles of
l38 TABLE 3 Saccharin Consumption in U.S. (Projected Estimate) Total consumption per year = 5,000,000 Ib Ave. consumption per capita per day = 30 mg 70% of saccharin is consumed in soft drinks l2,000,000 people consume saccharin in soft drinks Top l0% drink 42 oz (ave.) = 365 mg/day FDA recommended limit = l000 mg/day = 0.05% of diet Maximum human consumption = 0.02% of diet 5% saccharin in rat diet = 2,500 mg/kg/day 0.02% saccharin in human diet = 6.4 mg/kg/day Rat dose = 390 x human dose 7.5% saccharin in rat diet = 600 x human dose soft drink, in which there are 365 milligrams of saccharin. It is in- teresting that this is very close to the three-tenths of a gram that President Roosevelt's panel concluded was safe in l9l2. The Food and Drug Administration recommended limit of consumption of saccharin is shown here as l,000 milligrams a day, almost three times as much as that estimated as the actual maximum consumption. Since l,000 milligrams constitutes about 0.05 percent of the average human daily diet the maximum estimated human consumption of saccharin then constitutes about 0.02 percent of the diet. Of course it should be kept in mind that peak daily soft drink consumption by any individual is not continuous through life; it tends to be intermittent, with sea- sonal variations. Whereas 5 percent saccharin in the rat diet is equiv- alent to 2,500 milligrams per kilo per day, the 0.02 percent saccharin in the human diet provides about 6.4 milligrams per kilo per day. Thus, the rat dose at the 5-percent level in the diet is 390 times the maxi- mum human consumption. And at the 7.5-percent level in the rat diet the saccharin intake would be about 600 times the human dose. These considerations are summarized in Table 3. Dr. Crampton mentioned some of the epidemiological aspects of the consumption of saccharin by diabetics. It is usually assumed that the maximum consumption of saccharin is in diabetics. This is true from the standpoint of the duration and consistency of the intake, but the
l39 maximum intake in any given short period of time is no larger in dia- betics than it is in the top l0 percent of consumers of saccharin- sweetened soft drinks. Several questions, issues, and uncertainties that were posed in the saccharin report were followed by a list of recommendations. I have already alluded to some of these questions. First, what is the signif- icance of the impurities, or the main impurity, OTS, as far as the production of bladder tumors is concerned? Is it possible that OTS it- self is responsible? Second, what might be the toxicological interac- tion between these impurities and saccharin itself? The problem of the toxicological interactions between two or more substances is encountered frequently. Either of two substances alone, for example, may have no effect but the two together may have a definite toxicologic effect. Third, there is the question of the possible role of the bladder calculi or the parasites or both. Does 5 percent saccharin in the diet, for example, or the impurity therein, produce bladder calculi that then produce tumors? Or does the prior presence of the calculi or parasites more readily promote tumor formation when the saccharin or its impurity is being taken in the diet? Fourth, as far as the species specific histologic characteristics of the rat bladder is concerned, the question arises, is the rat an appro- priate species for bladder cancer studies? I will not attempt to an- swer that question. Other speakers may comment about that later. Finally, the use of such high levels of saccharin in the diet appar- ently produces generalized toxic effects, which are reflected in the FDA study at the 7.5-percent dietary level by a marked weight deficit in the weaned rats. This constitutes a major issue in the interpretation of the results of such studies. The final conclusion of the saccharin report was that the results of the toxicity studies thus far reported have not established conclusive- ly whether saccharin is or is not carcinogenic when administered orally to test animals. Because that question could not be answered, the following recommendations for further research were made: Carcinogenesis studies of pure impurities, especially OTS Carcinogenesis study of pure saccharin Carcinogenesis study of mixtures of known amounts of saccharin and OTS Study of interaction of stones or parasites in bladder and saccharin in diet Study of urine composition as affected by high saccharin and of OTS in diet (Na, pH, etc.) Study of significance of parental and in utero exposure in carcinogenesis studies Continued epidemiologic investigations It is hoped that the results of one or more of these studies might be enough to settle the issue one way or the other. Some of the in- vestigations recommended are already in progress, and I hope we will
l40 hear of progress in some of these studies later today or at least in the near future. DISCUSSION KRAYBILL: Dr. Coon, has any of these laboratories attempted to calcu- late out on a weekly, monthly, or annual basis what the concentra- tion or consumption of OTS or these contaminants are over the span in which these animals were fed? COON: Do you mean in terms of parts per million in the food? KRAYBILL: Well, if you gave the parts per million in the diet, then you would have to go back and figure out what the intake was, and whether anyone had calculated values for the actual intake of OTS either on a monthly, annual basis, or for the two-year period. May- be this will be answered later. I don't know. Maybe I am scooping somebody. COON: I see that Dr. Grice is interested in this particular question, so I will pass it on to him. GRICE: I will talk about that during my presentation. ROSS HALL, McMaster University: I have a question I would like to address to Dr. Coon. One of the concerns that comes out of this study is the interaction between the impurities and saccharin itself. But in terms of the human experience -- and you mention that some people are drinking as much as 42 ounces of soft drinks a day, and soft drinks contain a lot of other chemicals -- there is bound to be interaction between these chemicals and the saccharin and the impu- rities in the saccharin. Is there any way in which soft drinks or other foods in which the saccharin is contained could be studied, instead of just studying saccharin by itself? COON: The essence of the answer is no. In connection with the inter- action problem, if soft drink or foods of man were used as vehicles for saccharin in animal studies the concentration of saccharin in these foods would have to be so greatly increased that the true in- teraction picture would very likely be greatly distorted. Also it is a basic principle that it is impossible to study the toxicologic interactions between just two things when they are added to food. As you just implied, there are too many other dietary components that may enter into the picture in possible interactions with the two substances under study when they are added to the diet. So it is not a simple matter of the interaction between two things, but a complex matter of the interactions among innumerable things.
l4l HALL: In terms of the question we are addressing ourselves to at this meeting, it seems to me that the human experience in terms of diet is very different to that which your rats are exposed to in the lab- oratory, and that if we are really addressing ourselves to the ques- tion of whether saccharin is harmful or potentially harmful in the human diet, then somehow or other we should be able to design exper- iments in which this kind of question can be studied, instead of just studying the saccharin under very precise laboratory conditions in which the diets are very different from that of the human experi- ence. Somehow we have got to bring the kinds of diets that humans are eating into the experimental situation in order to answer these kinds of questions effectively. COON: You mean use a human diet as a diet in experimental animals? HALL: Well, why not? COON: I would anticipate nutritional difficulties in species such as the rat and mouse if they were fed simulated human diets for a long period of time. Some animals, such as the dog and monkey, could be given a diet that is much the same as the human diet. The study of Golberg at Albany (see Table 2) has been going on for about six years now and is still continuing. In that study, monkeys have been getting up to 500 milligrams per kilo per day for six years and there has yet been no indication of harmful effects in those animals. However, I do not know how similar to a human diet is the diet of those monkeys. That is about all I can say to the question. MICHAEL JACOBSON, Center for Science in the Public Interest: I have a brief question for each of the gentlemen up there. Dr. Crampton, you mentioned a study still unpublished, showing essentially no difference in the incidence of bladder cancer between diabetics and controls. Am I right? CRAMPTON: Yes. JACOBSON: Were the controls controlled for sex, age, race, and all dietary aspects, especially if they have protein and carbohydrate in their diet? CRAMPTON: The study has been finished. The results will be published later this year in the Journal of Preventive and Social Medicine, and I did not do the work. I suggest that you read the paper when it is published. I am not familiar with all the details of this study, but I have talked with those who are. I thought that this audience would like to hear of the major findings. I am sorry I cannot answer your specific questions.
l42 JACOBSON: I am just trying to get the significance of this study. Also, did you recall from reading it what the level of sensitivity was? Could it have picked up at 50-percent increase of l/l00 of a percent increase, or roughly how sensitive was the study? CRAMPTON: I don't know. JACOBSON: Thank you. Dr. Coon, you finished up your talk by mentioning that the evi- dence is not yet clear as to whether saccharin causes cancer in ani- mals or does not cause cancer in animals. Am I right? COON: Right. JACOBSON: Assuming that is the case, do you believe that the use of saccharin in foods should still be permitted even though its safety is not proven? COON: Do you want my personal opinion? JACOBSON: I guess two opinions: One, your personal one; another, if you were head of FDA. COON: I cannot imagine what my attitude would be if I were head of FDA. But my personal opinion is that saccharin is safe as it is permitted to be used today. As a member of the committee that prepared the report, I have to insist on scientific grounds that we do not have the evidence to say whether or not saccharin is a carcinogenic agent in test animals. Now, in one case I am being a pure scientist, and in the other case, I am stating, on the basis of my gut feeling, that there is no risk or hazard in the consumption of saccharin as it is used today. JACOBSON: I guess the spirit of the law is that safety should be dem- onstrated before it is permitted in the food supply. The way that the thrust of these experiments seems to be going is that we are going to have to wait until we prove that OTS is the culprit or that there is some relationship between calculi and the saccharin. It is as if we are looking for every possible excuse to keep saccharin in; but I wonder if it is not possibly due to industrial pressure to continue permitting its use in the food supply. COON: I cannot comment on the industrial pressure factor. But I would suggest that when something has been widely used for 80 years with- out evidence of injury to the consumer, then it should be demon- strated to be unsafe before it is deleted from the food supply.
COMMENTARY AND DISCUSSION Harold C. Grice Philippe Shubik Ernst L. Wynder HAROLD C. GRICE As my contribution to this Forum, I believe it would be useful and informative to discuss current and future Canadian research and regu- latory concerns as they relate to the NAS report. First, I will briefly review the regulations. The Canadian Food and Drug Regulations allow the use of saccharin and its ammonium, calcium, and sodium salts as nonnutritive sweetening agents in certain dietetic foods. Such foods are carbohydrate or calorie reduced to meet the re- quirements of our regulations. In addition to these restrictions to the use of saccharin, we have label requirements that read as follows: "A food containing saccharin or its salts shall carry on the label a statement to the effect that it contains a nonnutritive artificial sweetener." We have another regula- tion that reads: "No person shall sell a food containing a nonnutritive sweetening agent unless (a) that food meets the requirements for spe- cial dietary foods, and (b) the label carries a statement implying a special dietary use." With reference to specifications, our regulations require that sac- charin meet the specifications set out in the Food Chemicals Codex. This limits the amount of orthotoluene sulfonamide (OTS) to l00 parts per million. Since l970, saccharin has been the only nonnutritive sweetener permitted as a food additive in Canada. Under our drug regu- lations, however, both saccharin and cyclamate are permitted. Combina- tions of the two may also be sold. A preparation containing saccharin or its salts is required to carry a statement on the label to the effect that it is a chemical substance without nutritive value and should be used in moderation. After careful study of the NAS report, and in light l43
l44 of our current research studies, we have not found reason to further restrict use of this nonnutritive sweetener at this time. Now let us turn to our research studies and try to answer some of the questions that have come up this morning. We have been involved in the analysis for impurities of the saccharin samples used in the stud- ies mentioned in the NAS report, including those conducted in the United States, Canada, England, the Netherlands, and Germany. OTS was the major impurity and was found in amounts varying from ll8 to 6,l00 parts per million. Other impurities were present in considerably less- er quantities, some less than one part per million. The water-soluble impurities from saccharin samples used in four different animal studies -- at the Health Protection Branch, FDA, the Eppley Institute, and WARF -- are graphically presented on Figure l. The large spot in the middle of the chromatogram is of sodium saccharin. The compound at the top with the Rf of about 0.9 is orthotoluene sul- fonamide. Among water-soluble impurities that we isolated from the different saccharins so far are: o-Toluenesulfonamide, p-Toluenesul- fonamide, o-Sulfamoylbenzoic acid, p-Sulfamoylbenzoic acid, 5-chloro- saccharin, tf-Methyl, o-toluenesulfonamide, Ferrous sulfate. At least five more impurities have been isolated, and some of them tentatively identified as sulfones. The results of our investigations to date do not suggest there was a common impurity other than OTS present in a high concentration in the Solv. Front Start FIGURE l Graphic presentation of the water-soluble impurities from saccharin samples used in four different animal studies -- the Health Protection Branch (A), FDA (B), the Eppley Institute (C), and WARF (D). Solvent System: n- butanol: 95 percent ethanol: H2O (40:4:9).
l45 WARF and FDA saccharin samples that we have analyzed. The NAS report mentions that orthotoluene sulfonamide was found to be a major impurity in all of the saccharins prepared by the Remsen-Fahlberg procedure. It is interesting to note that in the investigations conducted by Litton- Bionetics, the German Cancer Institute, and the Netherlands National Institute for Public Health, the saccharin samples contained relatively high amounts of OTS: 6,l00, 3,075, and 5,050 parts per million, re- spectively. The Netherlands group used saccharin in the free acid form, while sodium saccharin was used in studies in the United States and Germany. Although the relative amounts of OTS were higher in these three studies, no tumors attributed to saccharin treatment were found. The amount of saccharin and OTS received by animals in the various studies, are presented in Table l. Here are the milligrams of OTS consumed per rat per day in the Health Protection Branch study, the WARF study, FDA, and German National Cancer Institute. There were three samples submitted to us for analy- sis from WARF and five from FDA. These figures represent the high and low values of OTS in those samples. We do not know when or for how long the various samples were fed. Apparently the high sample from FDA, that is, the 8.79 mg of OTS/rat/day, was used for the latter few months of their study. That is to say, the low OTS samples were used TABLE l The Amount of Saccharin and OTS in Different Animal Studies. WARF and FDA Studies Used Four and Five Different Lots of Saccharin, Respectively. The Amounts of OTS Shown Here are the Lowest and the Highest Amounts of OTS Submitted to us for Analysis out of Three from WARF and Five from FDA. Grams of Maximum Saccharin Saccharin Consumed in Diet Rat/Day (%) (25 g food) OTS Content in Saccharin OTS (ppm) w OTS in Diet (ppm) Consumed Rat/Day (mg) HPB 5 l.25 ll8 0.0l2 6 0.l5 WARF (4) 5 l.25 2l3 0.02l l0.6 0.26 5 l.25 336 0.033 l6.5 0.4l FDA (5) 7.5 l.87 245 0.024 l8 0.45 7.5 l.87 4660 0.46 345 8.79 German Cancer Inst. 0.5 0.l 3075 0.30 l5.3 0.37
l46 during the critical periods, if you will, of pregnancy and lactation, and the early formative years of the rat's life. I will now discuss a survey we conducted of saccharin impurities in marketed saccharin products. Last year we analyzed for the OTS con- tent in a number of saccharin-containing table sweeteners manufactured or distributed by l5 different companies. These products obtained on the open market in Canada were in the form of tablets (T), liquids (L), crystals (C), and blends (B), as shown in the Table 2. Two values for OTS content indicate that the same brand products, having two different lot numbers, were analyzed. In some instances, the OTS content between the two lots was almost identical, while at other times there was a considerable difference. I would like now briefly to consider the question of placental trans- fer of OTS and saccharin, their uptake from the neonate from dam's milk, and the residence time in the neonatal bladder. The importance of placental transmission of saccharin, the slow rate of its fetal clear- ance, and the possible accumulation of saccharin in some tissues is briefly mentioned in the NAS report, pages 22, 24, 25, 50, and 5l. On page 22, it is noted that the accumulation of saccharin in the bladder of adult rats is l9 times higher after multiple dosing than after a single dose. On page 24, it is noted that the slow rate of fetal clear- ance of saccharin, coupled with repetitive maternal ingestion of saccha- rin during pregnancy might lead to accumulation in the fetus. What about OTS? Does it cross the placenta? It is our understand- ing, from unpublished work in Great Britain, that OTS does cross the placenta. If this is so, does OTS accumulate in the fetal bladder the same way that saccharin does? What about the milk? Saccharin is present in dam's milk, and we have recently determined that OTS is also present. What is the resi- dence time in the neonatal bladder of saccharin and OTS derived from dam's milk? It is apparent that there are a number of unanswered questions that relate to the pharmacodynamics of saccharin and OTS in the neonate and the fetus. I would like now to briefly outline the animal studies with saccha- rin and OTS that recently have been completed or are currently underway in our laboratories. On page 52-A of the NAS report, the inhibition of carbonic anhydrase is mentioned, and Dr. Coon made reference to this this morning. The hypothesis for the development of bladder cancer from OTS is that OTS inhibits carbonic anhydrase in the kidney, increas- ing the excretion of bicarbonate. This action could produce an alka- line urine, favoring the production of stones in the kidneys and the bladder. Irritation from stones over a long period of time could pro- duce hyperplasia and finally tumors. Prior to our undertaking the cancer study, preliminary dose range- finding studies were initiated in which OTS was administered by gavage to pregnant females from day l of pregnancy to day 2l after parturition. It is important to note that dosage was by gavage, since this may well have aggravated the effects observed. After weaning, OTS was incorpo- rated into the diet of the pups, and various levels were fed up to 250
l47 TABLE 2 OTS Found in Saccharin-Containing Table Sweeteners Saccharin0 Content Sample Type& £%) In Whole Product In Saccharin ' o Source of Sample l T,Na l00 89 89 2 3 T,Na T,Na l00 l00 57 67 57 \ 67 f Same manufacturer, same distributor 4 T,NG 24 377 l569 5 T,E,NG 30 ll44 38ll 6 T,E,NG 23 40 589 l055 256l ) 2638 / Same distributor 7 T,E,NG 8 T.E.NG 37 l87 505 9 l0 T,E,NG T,E,NG l8 l9 2l l43 ll7) 750 J Same manufacturer, different distributor ll T,NG 67 583 870 l2 l3 T,E,Na T,E,Na 38 36 l08l-f 7ll) 3003) Same company, different plants l4 L.NG 2.3 544^ 544 l5 T,E,Na 2l l87 890 l6 l7 L,Na C.Na 25 98^ 89 98 ) 89 1 Same distributor l00 l8 l9 B,NG B,NG 3.8 4.0 33 3.5 855) 88) Same manufacturer, different plants (countries) 20 B,Ca 3.2 6.8 2l2 SOURCE: Stavric, B., and Klassen, R., O-Toluenesulfonamide in saccharin preparation, JAOAC, May l975 (in press). Samples identified by manufacturer and/or distributor and a lot number. Samples l6, l7, and l9 had no lot number. *T, tablet; L, liquid; C, crystals; B, blends; E, effervescent; Na, sodium salts; Ca, calcium salts; NG, form of salts not given. ^Saccharin content as labeled. All results represent average of duplicate determination and 2 injec- tions into chromatograph, except for samples 2 and l5 where a single determination was performed. eo-TS in saccharin = o-TS in whole product x l00/saccharin content as labeled. /"Approximate value," explanation in text. liquid samples: in the evaporated material.
l48 milligrams per kilogram per day. The pups were killed at varying time periods up to l05 days of age, and microscopic examination of the urine for stones was done. Histopathologic examinations of the bladders and the kidneys were carried out. A higher incidence of bladder stones in all groups com- pared to the control was observed. There were no differences between males and females. Hyperplasia of the bladder epithelium was observed in animals on the l00 milligram per kilogram per day, and 250 milligram per kilogram per day meals, while no hyperplasia was seen in the females. Urinary stones measuring l5 to 30 microns in diameter were found on a millipore filter after filtering the urine. If bladder stones can stimulate hyperplastic change in the rat urinary bladder, and if this hyperplastic change can ultimately lead to cancer, then the presence or absence of stones in the rat bladder carcinogenesis studies should be determined. We simply do not know what the role of stones is. We do not know the residence time of these stones in the bladder and what effect they could have either as a contributing or a direct cause to bladder carcinogenesis. Because of the results of the study I have just described, we have a similar study underway using lower doses of OTS administered in the diet both to the dams and to the young throughout the entire experiment. We do not have the results of these studies. A chronic study involving both the FQ and Fj generation has been in progress since February l, l974, to evaluate the toxicity and carcino- genicity of OTS and OTS-free saccharin. The protocol for the FQ, Fj generations is as follows: there are 6 groups - - 50 males and 50 females per group. These are control, 2.5 mg/kg/day OTS, 25 mg/kg/day OTS, 250 mg/kg/day OTS, 5 percent saccharin, 250 mg/kg/day of OTS + NH^Cl l percent in drinking water. Since OTS is a weak carbonic anhy- drase inhibitor, ammonium chloride was added to the drinking water in the one group to maintain a slightly acidic urinary pH. The saccharin used for this study contained 0.4 parts per million OTS. The FQ genera- tion has been on test for 58 weeks, and the Fj generation on test for 38 weeks. To date, mortality in the FQ generation is 4 percent and in the Fj generation is less than l percent. No gross tumors of the blad- der or kidneys have been observed in moribund animals that have been sacrificed. Table 3 summarizes our research work in relation to the recommenda- tions for research that Dr. Coon mentioned. The first two recommenda- tions are being covered in part in the chronic study that I have just described. As for the comparative studies of the role of stones and parasites in the bladder in the induction of bladder tumors in labora- tory animals, the design of the chronic study involves examining for their presence in fresh, voided, samples of urine of animals treated with OTS and saccharin. In studies of the change in urine composition at high levels of sac- charin intake and the relationship of such changes to the induction of bladder stones or calculi, we are measuring urinary pH from OTS and
l49 TABLE 3 Summary of the Work at the Health Protection Branch in Relation to the Recommendations from the NAS Report NAS Recommendations Health Protection Branch Research Investigation of the question of transplacental carcinogenesis of saccharin and its impurities Investigation of the toxicologic significance of impurities in commercial saccharin preparations Comparative studies of the roles of stones and parasites in the bladder in the induction of bladder tumors in laboratory animals Study of the changes in the urine composition at high levels of saccharin intake and of the relation of such changes to the induction of bladder stones or calculi Chronic toxicity and carcinogenicity study (FO-FJ generations) with OTS and saccharin has been in progress since Feb. l974 As above Examination of fresh voided samples or urine of OTS and saccharin treated animals for the presence of stones and parasites using the millipore filter technique Measurement of urinary pH in OTS and saccharin treated rats which is correlated to the incidence of bladder calculi saccharin-treated animals, and this will be correlated with the inci- dence of bladder calculi. Hopefully, our studies will therefore answer some of the questions posed in the NAS recommendation. PHILIPPE SHUBIK I am going to direct my rather brief remarks strictly to a commentary on the report of the National Academy of Sciences and try to stay within that frame as advertised in the program. I had assumed that everybody in this audience would have read the report -- not only have read it, but read it in detail, have checked it, checked the references, and so forth. But I find to my astonishment that this is not really the case. To comment on a report that I think has not been thoroughly digested is quite difficult because of having to repeat much that is in it over and over again. The toxicological investigation of saccharin may seem to some people to have been grossly exaggerated as an exercise, a misplacement of emphasis, perhaps a misuse of toxicological resources. And indeed, the
l50 recommendations of the committee of the National Academy that you have heard reported today may appear to be compounding the situation by recommending that even more studies be performed in addition to the vast number already undertaken. This is a matter that was discussed in the working sessions before the Forum and one to which I previously have given some thought. It merits discussion. I would like to say at the outset that I for one commend the National Academy's committee on their report and hope that their recommendations will be acted upon and implemented to the fullest extent. I believe that answers to the questions that have been posed not only will re- solve the specific problem before us, but also will be of considerable importance in adding to our basic knowledge of toxicology and will as- sist us in the future interpretation of problems of this sort. I think that this is the primary justification for many of the additional stud- ies that are recommended. As a practical matter, I happen to believe that we should not be con- cerned about the potential hazards that may exist from saccharin. That is my personal view. It is a view I would be willing to state at any time and for the reasons that I now would like to discuss. First, I think that the epidemiological studies, particularly those by Armstrong and Doll and subsequent studies by Kessler, are extremely impressive. In spite of the questions this morning about the details of those studies, I myself have considerable faith in the ability of Richard Doll and Bruce Armstrong to undertake studies in which all the various facets are taken into consideration. I think that when these studies are read in detail, most people will have no difficulty in be- lieving in them. The fact that an epidemiological study has been done on a food addi- tive in relationship to its potential chronic effects is a first. It has demonstrated that imagination on the part of scientists will often overcome problems that have been said over and over again to be hope- less. I had been told many times that it was extremely unlikely we would ever obtain any meaningful epidemiological intelligence on the chronic effects of food additives. Doll, his coworkers, and others have demonstrated that this is not so, because in this instance we are, I suppose, in a way lucky enough to have a special population of diabetics. But I think that by no means should this sort of thing be restricted to this one additive. When one starts to think about various other additives, occupational groups, special groups on various diets of one sort or another, there is a huge field to be opened up in which a great deal more epidemiology will be brought to bear on problems dealing with food additives. This is one of the good spin-offs of this study. The second reason that I believe that saccharin is most unlikely to be carcinogenic is that it is unequivocally not metabolized. The data, now confirmed over and over again by some of the best toxicological bio- chemists in the world, have demonstrated without any question that sac- charin is excreted unchanged. I do not know of a carcinogen that is excreted unchanged, so it seems to me that it is most unlikely we have
l5l here an example of something that is entirely different from substances that we call carcinogenic. Third, there are clearly a large series of negative studies on record, and we are faced with having to explain two studies that are positive. In the instance of both these studies, first of all, extraordinarily high levels of material -- particularly when considered in terms of the embryonic rats exposed -- were given to the animals subsequent to birth and to the newborn. The type of experiment used -- feeding the mothers, feeding the newborn, going straight on through -- to my mind is, in any case, a generally inappropriate experiment. I believe that without doubt many of our studies in toxicology are inadequate in that we do not have good data on transplacental exposure. Exposure of the newborn under appropriate circumstances is something we must have in order to make a complete decision. But I don't think these things should be done in a hammer and tongs way in the same experiment. It is completely impossible to sort out whether the effect was a transplacental one, whether it was an effect on the newborn, or whether it was an effect on the adult. These three stages of the experiment should be separated so that we can, in fact, really see what we have done. It is an experi^ ment that is extraordinarily difficult to analyze from that standpoint. Insofar as the levels are concerned, these remind one of prior stud- ies in which very high levels of material were fed. In the case of a food additive, MYRJ 45, polyoxyethylene stearate, tested many years ago, actually 20 percent of the material was fed in the diet. It had an impurity in it, ethylene glycol, which produced bladder calculi and subsequently bladder tumors. The test was held to be inappropriate to perform for the material itself at this high level. The probability that OTS is the offending substance in saccharin is made even more probable by the fact that related sulfonamides have been shown to produce bladder calculi and, subsequently, tumors. In an ex- periment performed relatively recently and published by a colleague of mine, Dr. Clayson, another sulfonamide was shown to have this property, which could be inhibited by changing the acidity of the urine with ammonium chloride. That is a situation in which a mechanism can be clearly demonstrated as to why this substance does what it does. This is a l-2-3 business in which there is no mystery. I believe that it is more than likely to suppose that the same sort of situation will hold true with saccharin. I should like to add one additional caveat to the National Academy report. It is my belief, in view of the evidence accumulated on the potential importance of OTS, that we have been extraordinarily remiss to have had on the market a substance with the level of impurities present in saccharin. When we undertook a test several years ago, which was planned at the National Cancer Institute with a group of other people, none of us was familiar with the fact that saccharin had any impurities in it. Indeed, we did not know that there were two pathways of synthesis, and there were no proper standards set up for saccharin at that time. Had that
l52 been the case, the test would have been designed properly at the beginning. As it stands now, it would be my view that one thing that should be added to the recommendations is that the levels of OTS in saccharin should be as low as possible and that the saccharin with a very low level of OTS should be the only kind that finds its way to the market- place. A survey, such as that presented by Dr. Grice, should under no circumstances ever again demonstrate that we have these enormous levels of this impurity. Lastly, the question arises as to the meaning of this sort of car- cinogen in its own right as a carcinogen, the question of carcinogenic- ity and different sorts of carcinogens. Some years ago we were faced with a situation in which a number of food additives were tested by subcutaneous injection. A variety of substances were produced, includ- ing sarcomas. It was deemed by a wide variety of people that experi- ments by subcutaneous injection are not appropriate for food additives, and I think that was an absolutely correct decision. At a recent meeting of a scientific group at the World Health Orga- nization, it was pointed out that there are a variety of different sorts of carcinogens. I think that nobody looking at these results could possibly equate the sort of thing one has with saccharin and OTS with, for example, aflatoxin. Aflatoxin is carcinogenic in microgram quanti- ties. We are talking about grams of this material. Aflatoxin is a compound that reacts with, for example, nucleic acid. It is a carcino- gen that is immensely powerful. Nitrosamines and various other car- cinogens are compounds of this sort, which are extremely potent and act in very small quantities. Then we come to the other end of the spectrum. We come to a variety of compounds that are perhaps secondary agents. They do not, in fact, produce tumors, as far as one can tell, by acting directly in the same manner. This report of the World Health Organization points out that the time has come for us to subdivide carcinogens into different sorts. In the instance of a compound that produces tumors via bladder calculus forma- tion, it clearly is a compound in which you can quite logically deter- mine a level at which activity will occur. You know how to do this. The Delaney philosophy is based on the compounds in which this is not possible and in some instances, I think, is entirely justified. But I think the time has come for us to look at carcinogens as individual compounds, to look at the way they work, and to not just heap every- thing into one great box. The problem we face, indeed, is the word carcinogen, which in many instances is used so loosely as to have lost its real meaning. ERNST L. WYNDER As I sat here and listened to the discussion, and as I read the very excellent report by Dr. Coon and his colleagues, I recalled something
l53 that we used to say in medical school: "Common diseases occur com- monly while uncommon diseases are being discussed in grand rounds." It seems to me that we are discussing something, at least from the point of view of epidemiology, for which it has been well shown that there is still no evidence that saccharins or other sweeteners relate to bladder cancer in man. If we had the same conference on tobacco and bladder cancer -- and it has been well established that one-third of all bladder cancers in man relate to cigarette smoking - - I dare say that either we would not have a conference here or the conference hall would be empty, because we, as citizens, rarely try to blame ourselves for something that goes wrong, but rather like to blame George. In spending our limited research funds and our limited research per- sonnel, it is important that we concentrate our efforts on those car- cinogens in our society that have the largest impact on cancer in man. It has been suggested that it is the dosage that makes a poison. It seems to me that to apply a large amount of material to animals, and draw from those effects conclusions applicable to man is not a good scientific practice. At the American Health Foundation we use an inter- disciplinary approach for cancer research. Chemists, biologists, and epidemiologists sit together to look at a scientific problem, and among these, the epidemiologists reign supreme. Irrespective of what you show me in a given animal that receives a high dose of a carcinogen, if there is no epidemiological evidence, particularly if the evidence has been established over a period of years, the animal data is inconsequential. Now, if we look at the chemistry of saccharin, it does not appear to be carcinogenic. If we look at the high-dose experiments, saccharin may well produce bladder tumors, as has been shown in rats but not necessarily in other animals. This may result from the impurities we have heard about. But you and I have to ask ourselves: "To what extent is man a rat? Are we more like a hamster? Are we more like a primate? Or is a rat, perhaps because of the presence of parasites, particularly likely to form more parasites, stones, and subsequently tumors?" But the key, as I said, is epidemiology. It so happens that saccha- rin has been in use for nearly a century, so it did not come about just l0 or l5 years ago, like the pill, where we have a different problem. Kessler, in reviewing the epidemiology of diabetes and bladder cancer, showed no correlation. I recently discussed this subject with my friend, Sir Richard Doll, in Oxford, and I am reading to you the summary of the paper that he produced with Dr. Bruce Armstrong and that will appear later in the British Journal of Preventive Medicine. He concluded, in a summary: "The frequency with which diabetes mellitus was mentioned on the death certificates of l8,733 patients dying from bladder cancer has been compared with that of l9,709 patients dying from other cancers." I am emphasizing that we are not dealing with small numbers.
l54 "The estimated relative risk of bladder cancers in diabetics was 0.98, with a 95 percent confidence limit. There was no increase in the risk of bladder cancer in patients with diabetes of long duration." He studied the incidence of bladder cancer in individuals who had been diabetics for more than 20 years. Diabetics were shown, by ques- tionnaire, to consume substantially more saccharin than nondiabetics, and the duration of regular saccharin used by diabetics was highly cor- related with the duration of diabetes. There was, therefore, no evidence from this study that consumption of above average amounts of saccharin had led to bladder cancer in men or in women. It seems to me we must look at the available human data, and it is unlikely that we will have a study of this magnitude again for some time to come. In our own studies we have been taking saccharin data now for some time. In the last l40 cases, compared to 280 controls, the amount of saccharin taken and the number of saccharin takers was identical be- tween study and control group. We have had only two instances where in our epidemiological studies we have shown a correlation of diabetes to a given cancer - - kidney cancer in women and pancreas cancer in women, but not in man. We concluded in these instances that it had to do with the fat metabolism that relates to kidney and pancreatic cancer, rather than with the diabetes itself. So epidemiologically, ladies and gentlemen, we have at this time no evidence that saccharin relates to bladder cancer in man. I will not talk about the cost-benefit problem that was well dis- cussed by Dr. Crampton. Clearly we have to consider in our final evaluation the benefit as well as the cost to society of a sweetener as opposed to sugar. We also have to consider whether the l7 to 20 per- cent of excess calories we get from sugar may have an effect on obesity. It may have an effect on all those diseases that relate to excess weight. Now what about future work? Future work, I think, could well, in a very limited way as we heard from our colleague from Canada, concern with the impurities as it relates to the material in saccharin, and I think additional studies on sweeteners, at least in my institute, would not be worthwhile to undertake. In the case of epidemiology, I learned that perhaps we should get more data on soft drinks. These have been added to our questionnaire, and will, in a relatively short period of time, because we have a large team of interviewers throughout the country, we can well answer the question whether individuals that drink more soft drink than others have a higher incidence of bladder cancer. Obviously we recognize that these data have to be standardized against smoking history, against occupations, and whether our data will show a correlation of fat cho- lesterol intake to bladder cancer as well. Our recommendations, therefore, would be in line with reducing the impurities of saccharin as much as we can, and I certainly agree with the Coon report, namely, that there could be a limit on the amount of saccharin that people should take. Beyond this I would make no
l55 recommendation, and I would like to see that we, as a society, limit the amount of research that we do. In conclusion, I think that in life we strive for the ideal, but we have to accept the possible. In scien- tific research, it seems to me, we ought to concentrate on those areas where epidemiological evidence suggests that certain factors have the greatest impact on human life. Perhaps what this society needs are more people like President Theordore Roosevelt, who instinctively may have made the right decision, because in science, as in politics, sooner or later we must make a decision on the majority opinion; future data will prove whether we have been right or wrong. If I had to make a decision on saccharin and cancer in man on the basis of epidemiology, I would say that we have the data on hand and that no more good will come out of additional data. I would recommend that we conclude the chapter and concentrate our efforts, as well as those of the Academy, on the factors in our life that have the greatest impact -- cancer, heart disease, and other major diseases. DISCUSSION MILTON WESSEL: My concern and interest are in the communication process between science, law, and the public. My question and comment are directed specifically to Dr. Coon and Dr. Shubik, but also to the Academy as a whole. It is my opin- ion that a disservice has been done in the choice of words that mean one thing to those who write them or who speak them to each other, but may mean very different things to the rest of us. I hope I can communicate this concern. It is characterized by Shubik's comment that "the level of OTS in saccharin should be as low as possible." I ask him if he can tell me what that means. He might be able to define that, but he has not yet done so. It could cost $20 million to reduce it down to the one molecule level! The Academy report states that to resolve the question of whether saccharin is or is not carcinogenic, "The following studies must be carried out." It does not state, as Dr. Shubik has indicated, that all these studies are necessary to advance the state of our general toxicological learning. The report is a response to a request for guidance from a regu- latory agency. Mr. Ronk himself says that he is not scientifically expert in this area. The report presents conclusions that my wife and children might interpret as indicating that saccharin may well be a carcinogen; it has not yet been conclusively established. Now, the fact is, as both Dr. Coon and Dr. Shubik will tell me, that it is impossible to establish anything conclusively; that when these tests are finished, it will not be conclusive. Perhaps we will have increased our confidence level from 99.3 percent to 99.6 percent.
l56 We can do $20 million more work with precious resources and get it to 99.7 percent, and keep on and on and on. But that is not what this report was directed to in the first place, and it is not the way I, the public, the regulators, or the Congress will read it. The fact is that the Food and Drug Administration has been put in a very difficult position. It has a report that states that it has not yet been conclusively established that saccharin is not a car- cinogen. The FDA will be faced with questions such as Mr. Jacobsen has asked. Dr. Coon, when he is on the witness stand at the time the decision of the FDA, whatever it is, is appealed, will be asked, "Just what is it that requires conclusive establishment?" I do not think there is recognition in the communication of this kind of in- formation that the questions you are dealing with are not scientific questions. They have a scientific component, but they have a very large public component, and they have to be dealt with by all of us. When someone says here, "What does it take to stop the sale of something because it may be a carcinogen?" I reply, "What does it take to stop me from being able to use that?" We are dealing with two sides of an equation, the most important part of which in a free and democratic society is the freedom of the individual, the freedom of the public to make decisions. I submit that this report -- al- though I recognize that its toxicological expertise is not being questioned in any respect -- essentially is a report to be evaluated by a public that is not qualified, and that may interpret it as one calling for immediate restrictive action. Yet I know that Dr. Handler, Dr. Coon, and probably every qualified scientist in this room believe the exact contrary. COON: I think the talk that we have just heard makes a lot of sense. As far as the Academy committee that was charged to do a certain job is concerned, the committee took that charge from the Food and Drug Administration and tried to accomplish its mission. The committee did what it could with the available, scientific information, and it gave the FDA the best answer possible on the basis of that data. We were not asked to prepare a communication to the public, or to ex- pound or express our sentiments on social, philosophical, and economic issues. We were asked a scientific question; we gave a scientific answer as far as we could with the data at hand, and then made recom- mendations as to what further information was needed in order to come up with the definitive answer that was originally requested. I would agree with Dr. Shubik that the OTS impurity be reduced to the lowest possible level. When asked what we mean by lowest possi- ble level, we could go further and say, "Well, if we can produce saccharin without any OTS in it, that is fine." It would be on the basis, I think, of good manufacturing practice, that practice by which the lowest possible level of OTS can be achieved and still produce a commercially feasible product. I notice that Canada has limited OTS to l00 parts per million in saccharin, and so has the United States through the Food Chemicals
l57 Codex. However, the Maumee Process for manufacturing saccharin pro- duces a level of only l to 3 parts per million. So I would pose the question, why not use the Maumee Process as good manufacturing prac- tice and produce saccharin by that method? SHUBIK: I think Mr. Wessel's comments are well taken. They are inter- esting, and they make people think about a lot of things they prob- ably did not think about before they came here that relate to changing patterns of approaches in society. I am somewhat surprised that Mr. Wessel, having made the allega- tion that this report is not clear, is cheered by the audience here when most of you have not read it. I would have thought that before you really agreed with him that it is not understandable to you, the least you could have done would have been to sit down and read this thing through from beginning to end. An enormous amount of work has gone into it, and it provides a lot of background. It tells you the story, and it does go into a lot of details that clearly Dr. Coon and I had no time to address ourselves to. Mr. Wessel suggests that this report is addressed to the public. Perhaps it is. If it is, it is the wrong kind of report, and I agree with him entirely. As far as I understood it, this was a report requested by the Food and Drug Administration from the National Academy of Sciences. The committee addressed themselves specifically to a question, and to my mind they answered the question in a proper and scientific manner, as has been the case in the past. But per- haps things have changed, and perhaps these reports should be written in an entirely different way. In addition to that, I would like to suggest that in order for the public to be able to understand this, we should surely intro- duce courses in toxicology at our high schools. SVEDA: I have three comments to make: one to Dr. Wynder, one to Dr. Fredrickson, and one to Dr. Coon. The one to Dr. Wynder is very succinct. Sir, you are a breath of fresh air. We need you. The first thing this morning, Dr. Fredrickson, you made a comment about Dr. Yudkin. I am in correspondence with Dr. Yudkin I don't know him personally, and I don't want to defend him or not defend him. Yesterday, his work was talked about quite often and actually denigrated, in my opinion. Also questioned was the work of Dr. Cohen in Israel, with whom I am in correspondence. I deplore the fact -- and I want to go on record, I want this very carefully on record -- that Dr. Yudkin's work was attacked so often or dis- cussed without his being here to defend himself. Now to Dr. Coon I have a comment, which is scientific in nature, I think. Before the break, we were talking about dosage levels in animals and in human beings and that rats are fed saccharin and cyclamate to maybe 5 percent of the diet. I made this same comment on cyclamates to somebody else several years ago. At that time,
l58 when rats were being given 5-percent cyclamates and some of them were given grams of cyclamates, I said: "I don't object to this. But then you ought to give sugar not on a weight-for-weight basis, but on a replacement basis, which is 40 times as much. This would mean then that you would probably feed the rat more than the total rat weighs." In the same way, if you are going to give 5-percent saccharin to an animal, then 350 times this makes seven times, or whatever it is, the weight of the rat itself I am sure. Dr. Wynder's comment, I think, is most appropriate. What do these things mean? Is my point clear? I think Mr. Wessel's point is very well taken that we do not make these things very clear to the public. If we have some time this afternoon, I would like to present some demonstrations that are for the public and will make some of the issues here much clearer. PFAFFMANN: Any comment? COON: It would obviously be impossible, as Dr. Sveda implies, to com- pare the toxicity of 5-percent saccharin or cyclamate with that of a dietary sugar concentration of equivalent sweetness. That would require a sugar concentration of many times more than l00 percent of the diet. At dietary levels of sugar that would be tolerated by the test animals the sweetness equivalent amount of cyclamate or saccharin would be too low for a rigorous toxicity test. Also, we have no idea what the sweetness equivalence is in animals, and it might be a little difficult to find out. We know that in man, but we don't know it in animals. Furthermore, sweetness equivalence has no relation to toxicologic equivalence. SVEDA: May I ask one very simple question? Why in the world does any- body ever eat saccharin or cyclamate but to replace an equivalent amount of sugar? And this, I think, bears on Dr. Wynder's point. Some of these experiments are ridiculous. MICHAEL KASHA: I want to first make a statement to Michael Sveda. It is clear from the conduct of the meeting that the public is on an equal footing with the scientists on the platform, and everyone has had a chance to say what they wish here. I also would like to indi- cate that Dr. Yudkin was invited to the Forum and was unable to attend. His absence was not planned. It seems to me that the conflict between the request that we heard from the attorney Wessel and the scientist Shubik is based on perhaps what an Academy report tries to do. I think, as I under- stand this Academy report, that it tried to evaluate the validity of a scientific test; and although it clearly showed that the scien- tific test had some validity, it called for extensions of the testing program to establish further what the valid basis is. But I think the public demands more, and I think scientists are not in a posture or in the habit of saying, "I therefore can tell
l59 you, the public, that the tests are final and now you can do any- thing you like with this material (e.g., saccharin)." Perhaps the trouble is that our setup is wrong. It is maybe unfair of the FDA to ask a committee of scientists to look at some experiments and ask, "Are they the best, the finest that can be done?" For example, what bothers me is, I see 5 million pounds of a chemical being used by l2 million people of tremendous genetic heterogeneity, and yet the laboratory experiments are done on small colonies of 50 animals of selected genetic homogeneity. So these experiments, as I see them, are very restricted. The scientists will be very cautious about saying that they apply to the whole population. Perhaps we laboratory scientists should not be the only ones advising the FDA. Maybe the FDA should be approach- ing the medical societies saying, "You are the people who make the ethical-medical decisions. You are the people who address the pub- lic. This is what is known from laboratory tests. Then you decide from scientists' work in the laboratory what human applications can be made and state which are the policies that should be formulated." We are not in the habit of making social decisions as scientists, and yet that is what is demanded of us; I see that as a conflict between Shubik and Wessel. WESSEL: I think we are at the heart of a terribly important point. There is no conflict between Dr. Shubik and me. I admire him greatly. What he said a few moments ago is really the cornerstone of what we are saying-. Times have changed -- and changed radically. The age of consumer- ism is here. Yet we are all consumers. No one has the right to say, "I am the consumer advocate." We all are part of this process. We are all environmentalists. These issues that today are being de- bated here in the National Academy of Sciences tomorrow will be de- bated on the front or the editorial pages of the New York Times or the Wall Street Journal, and the next day in the courts and the administrative agencies. The people read what you say. You cannot isolate yourself from the community. The difficulty is thus in the communication process. I know what you intend and you know what you intend, because here we both are part of one aspect of the process. But then I go into court before juries who are even more laymen than the informed ones who attend this kind of a meeting - - by several orders of magnitude. Words of the kind used here in their scientific context are thrown up at me, and there is no way of furnishing the depth of under- standing that you have. I am trying to suggest that what Dr. Shubik was stating may well be one of the focuses of Academy concern: as it begins to get more and more into the scientific component of a public equation, the Academy should be certain that the scientific component be stated in such a fashion that when it gets into the public forum, it will be adequately understood -- as something recom- mended for the purpose, for example, of acquiring more information
l60 or achieving a greater degree of confidence and not designed for the purpose of reaching a desired safety level. ALFRED HARPER: I would just like to add a comment to this. I feel that we are in great danger of falling into that old American problem of wanting a simple solution to a complex problem, and I think the problem with food additives is not all that different from the prob- lem with nutritional requirements. The people who are doing the nutritional labeling, who are con- suming food, want a figure for how much they need, and how much is present in what they consume. Frequently it is not possible to do this solely on the basis of scientific information, and I think the same thing applies with this problem of food additives, carcinogene- sis, and toxic reactions. Perhaps it would be a good idea if scien- tists themselves recognized more frequently that an element of judgment enters into a decision even when the scientific evidence is very substantial. It seems to me that this is the crux of the problem. There is no possibility of having the public understand all of the information, as Mr. Wessel says. It takes extensive training to understand and interpret a scientific report. But it should become clear after the report has been produced, and it has to serve as a basis for a decision, that somebody has taken the best scientific evidence avail- able, used hopefully the best judgment that is available, and made a decision, because we cannot go along indefinitely without making a firm decision. BERNARD L. OSER: I was warned that I was to be a discussant, but I have a feeling now that anything I might say would be anticlimactic. There is .no doubt that despite all its great achievements in the past 30 to 50 years, science is falling into disrepute, and partly be- cause of public disagreements among scientists. This is not very comforting to consumers. I am going to take the cue from you, Dr. Pfaffman, and emphasize the uncertainties referred to in the title of this program. We should start out by saying that there is no such thing as "no risk" or "zero defects," as the term has been used here. In the same sense, there is no such thing as absolute safety. The regulations define safety as reasonable certainty that no harm would result under the conditions of use. With apologies to those of my friends on the platform and in the audience who are toxicologists, I want to talk a little bit about the uncertainties of toxicology. Toxicology, as Dr. Wynder pointed out, is a multidisciplinary science; but more than that, it is not an exact science. Toxicological methodology has advanced consid- erably in the past few decades. We are paying a great deal more attention to sophisticated techniques, more refined procedures, not only in respect to the number and choice of species, the size of
l6l experimental groups, the duration of experiments, but to the number and types of observations that we make. In the case of histopathology, years ago bladders were not even examined. The early reports on cyclamate and saccharin described no examinations of bladders. We recognize now that there are not only species differences but there are strain differences in animals, and the problem of individual variability. There is the question of the effect of prenatal and preweaning nutrition on test animals. We now emphasize more the effects on reproduction, lactation, terato- genesis, mutagenesis, and so on. All this has taken place within a relatively short span of years. I think it is unreasonable to fault scientists, either in industry or in government, for not taking action on the basis of evidence that has only recently been introduced, and in many cases not even con- firmed. There is quite a difference between experimental findings and facts. There is the subjective element referred to a few moments ago, involving the scientists' interpretation of their own and others' findings. Reference was made in the report of the National Academy of Sciences to additional work that needs to be done. Research goes on and on, and additional work does indeed need to be done to estab- lish the validity of in utero or transplacental dosage, on the ques- tion of renal calcification, bladder stones, and so on, and on factors causing secondary rather than primary carcinogenesis. These are areas of research that really need to be explored. But meantime we must make decisions; and so we make them on the basis of the best available evidence that we have at this time, recognizing that they are not going to be all white or black or necessarily irrevocable. Dr. Crampton referred to the benefit-risk problem. We have to consider not only the nature of the risk case by case, but the degree of risk. By the same token, we have to consider the nature and the degree of benefit. Who benefits? Is it the individual, or is it society? And to the types of benefits Dr. Crampton men- tioned, economic, technological, so on, I would add hedonic benefits. I think that the pleasurable aspects of food are most important and often ignored in relating risk to benefit. In that connection we ought to consider consumer wants, not just consumer needs. Who are we toxicologists to decide consumer preferences? Consumers should have the choice of satisfying their wants within the range of safety. What this boils down to is that these are ultimately societal judgments, in which scientists, toxicologists, should play a role, but not the sole role, and others from various segments of society who are adequately informed should participate in these decisions. As Dr. Wynder pointed out, where long experience indicates the lack of harm in the use of a substance or of a food or of any other environmental component where safety is indicated, one should be very cautious not to take precipitate action on the basis of unverified conclusions. Unfortunately, this situation has occurred, and there are those who encourage such premature action.
l62 In conclusion, and in anticipation of a question that is sure to be asked, let me say that for reasons so well articulated by Dr. Shubik, and underscored by Dr. Coon, I would go along with the view, based not on epidemiological experience (of which I have no knowledge) but on animal experiments, that both saccharin and cycla- mates are safe under the conditions that they have been and are proposed to be used. It has taken enormous doses of these agents to induce the secondary effects of tumors in the bladders of experi- mental animals. It should be recalled that 5 percent in the diet of the weanling rat is equivalent to 6 grams per kilogram of body weight. It has taken these tremendous doses to induce these tumors, not in all animals, but only in a small proportion of them. In some cases it was not even agreed among pathologists whether or not they were really malignant. In most cases, the neoplasms were not re- vealed except by microscopic examination. They were not gross tumors, as many seem to think, and at less than the maximum doses, they did not occur. So I think that the preponderance of evidence today is that the risk, if any, under the conditions of use of these nonnutritive sweeteners is minimal, and they should be regarded as safe. RITA CAMPBELL, Hoover Institution and Stanford University School of Medicine: I am an economist. I came here because I was first attracted to this area as a member of the National Advisory Drug Committee, and I have been playing around with cost-benefit analysis in drugs and food additives. Dr. Crampton is the only speaker who even spoke to what I would consider a kind of economic approach to the matter. I realize this is a group of health people, primarily physicians and biologists. But it did astonish me when I looked at the program and the list of participants that economists are not represented, even though they do have a theory of decision-making in face of uncertainty - - and uncertainty and lack of data are what I learn about in the scientific world. I was very naive. I thought the data was much hotter than economists were used to, but it is not. I think that if you are going to address these types of policy issues, that if even you don't have biostatisticians present, which also amazes me, that you should have once in a while an economist involved in the theoretical discussion.
HEALTH Considerations of the Institute of Medicine Committee on Saccharin Bryan Williams Kenneth L. Melmon BRYAN WILLIAMS I would like to abbreviate my remarks as much as possible. The Institute of Medicine Committee on Saccharin was asked to consider the need for the introduction of saccharin as a drug should it be removed as a food additive. Obviously the turn of events has made our report -- to use the popular Washington word -- "moot." While we were primarily concerned with the possible use of saccharin as a drug, we thought that several of the issues and the uncertainties that we encountered might be of general interest to you. Let me just list them rather than dis- cuss them. The first uncertainty that we encountered was the necessity to recon- stitute the pharmacopoeia. Saccharin is almost our sole remaining agent to make some drugs palatable at the present time. Without it, we would have to reconstitute many medicines. I think now that most of the physicians in this audience would agree that it is possible to manage diabetics and the obese patient without saccharin. There is no absolute requirement for this agent in their medical management. However, as an internist who has been concerned with the day-to-day management of patients, and in an effort to help them manage their diseases, I will readily confess that the presence of saccharin or another artificial sweetening agent has made the patient's life a good deal more tolerable. It is of secondary importance that the use of saccharin has made my job a good deal easier in helping them to manage their diseases. I think particularly of the juvenile dia- betic who is under such restriction anyway; the availability of an artificial sweetener has made the management of that patient a good deal easier. l63
l64 I would like to omit the rest of my remarks in order that we could consider some very specific issues. I would like to introduce Dr. Melmon and Dr. Navia and Dr. Sebrell in that order. KENNETH L. MELMON As a clinical pharmacologist, I was asked to serve on this committee to consider primarily the available information on saccharin related to its usefulness and acceptability as a prescription drug. This infor- mation was to have been considered if saccharin had been withdrawn as a food additive. In thinking about this, I wonder why the task should have depended on the food additive issue. Why consider this drug as a prescription item only if it were not appropriate as a food additive? Despite what appears to me to be a paradox in reasoning in making our assignment, we approached our task. Just as the Subcommittee on Nonnutritive Sweeteners of the Food and Nutrition Board's Committee on Food Protec- tion could not have made a judgment based solely on the ability of the substance to produce tumors in rat urinary bladders, I do not think that our committee would have been able to judge whether saccharin and its salts products are useful as drugs or drug products on the basis of available studies. Members of this audience may already know that prescription drugs are not evaluated on the basis of the Delaney clause, and that they clearly are evaluated on the basis of entirely different laws that we have to work within. Although the chemicals under consideration were available and on the market about 80 years ago (and therefore are not patentable items and not subject to the Harris-Kefauver amendments) I believe they should still be subject to the same consideration that manufacturers would have to give to a new drug being introduced for prescription use (a legend drug). The decision to introduce any chemical as a new legend drug must be based upon the current requirements being met by any new drug. These requirements are detailed in the l962 Kefauver amendments to the Food and Drug Act. In essence, the amendments state that all drugs or drug products introduced onto the market after l962 must pass predetermined tests to establish both the efficacy and safety of the chemical before it can be granted a New Drug Application (NDA). This is in contrast to the minimal requirements of safety alone for the drugs that were introduced before l962. Our committee knew that manufacturers probably would be unwilling to subject a nonpatentable item to the rigid and expensive test criteria of the Kefauver amend- ments, but we could find no logical or rational justification for the grandfather clause, which is applied to drugs introduced before l938. I believe we would have agreed with the recent report of the Office of Technological Assessment that the grandfather clause has been respon- sible for some of the problems that have occurred when drugs approved before l938 were used in therapy, and that the clause should no longer
l65 apply. We, therefore, would likely have recommended that saccharin meet the Investigational New Drug and Compendial standards before being introduced as a prescription item. We reviewed the report by the committee of the National Academy of Sciences on the safety of saccharin. This report thoroughly and accu- rately reviews the available data on animal and human pharmacokinetics and the toxicity in areas of reproduction, teratogenesis, mutagenesis, and carcinogenesis. It concludes that the chemical itself and even common by-products that are made from saccharin are safe enough to be continued as food additives. However, after reviewing what data is available to our committee, and without attempting to detail the spe- cific requirements for the safety of a chemical under an NDA, we point out that much additional data on animal and human pharmacology and toxicology is required to meet the present-day standards required by the FDA for the passage of an NDA. The data on the efficacy of saccharin or its salts for the treatment of patients with obesity, dental caries, coronary artery disease, or even diabetes has not so far produced a clear picture to us of the usefulness of the drug. We realize that tests of efficacy have not been required for saccharin in the past, and therefore acceptable data may yet appear that will prove efficacy in one or more of the above diseases. We did not have time to dwell on the toxicity that might have occurred when doses sufficient to influence these diseases were administered. Without well-designed studies, which might reveal effi- cacy, and without simultaneous study of toxicity, which would be con- ducted during administration of the drug in repeated doses over long periods in normal and diseased humans, no a priori decisions on the suitability of saccharin or its contaminants, as prescribed, could be made by us. We did not have time to consider whether the alternative sweetening agents have efficacy or toxicity different from saccharin, or whether the addition of saccharin to the prescription drugs or diets effectively and positively influences patient compliance. There simply are no studies that relate to these points. We concluded that prescription drugs containing saccharin or its products should be studied before the drugs are released, and that additional epidemiological information should be sought in Phase IV use of the drug, if this phase ever develops. Brief comment can be made concerning the suitability of saccharin and its products as over-the-counter drugs. The FDA is now reestablishing the criteria that such drugs must meet before they are made available to the public. Until the final FDA recommendations are made, we feel that if a drug is not approved as a legend agent, it should not be made available to the public in "lower dosage forms." The public is not likely to be fooled by low dose preparations and will learn to use the drug in necessary doses, that is, multiple tablets or multiple doses, in order to gain the desired effect. -This would, of course, be particu- larly true of sweetening agents. Thus, prescription amounts of the
l66 drug would be consumed regardless of the dosage form of the over-the- counter agents. These opinions clearly have been my own. The committee never came out with a solid recommendation; such a recommendation was never re- quired of it. However, I do think that it considered this issue well enough that it would have formed a consensus along the lines that I have discussed.
SWEETENERS AND DENTAL HEALTH Juan M. Navia I would like to address my remarks today to the subject of sugar, saccharin, and dental caries. We have discussed many issues in which different aspects and dif- ferent views have been presented. There is one that is very clear and very well documented: the relationship between sugar and dental caries. No one should come out of this auditorium without the clear under- standing that dental caries is stimulated and enhanced by the abuse and improper use of fermentable carbohydrate. I would like to stress the point of abuse and improper use. The moderate consumption of sugar may not really present a hazard to the individual, but for those who in- sist on abusing the use of different kinds of sugars we have to con- sider alternatives and ways to help them. Saccharin may have a role in this situation. Now, before I say something about saccharin, let me say something about dental caries, because we have not had much opportunity to dis- cuss this oral disease. Repeatedly we have agreed that it is a health problem related to sugar intake, but we have gone over it very quickly. Several characteristics of dental caries should be understood. First of all, it attacks mainly children. Fifteen-year-olds frequently have fifteen or more teeth that are decayed, missing, or filled, and that is very sad. It not only attacks children, however. Older people have gingival recession, exposure of the cementum and dentine; they also are subject to dental caries. It affects a large number of people. More than 90 percent of our population is subject to this disease. It is painful, it is disfigur- ing, and it is costly. There is no way that we can cope with this health problem with the resources and numbers of professional people l67
l68 available today. So the only way to approach the problem is by pre- venting dental caries rather than by treating the disease. It is a disease that is stimulated by the excessive intake of fer- mentable sugars, but it is not a nutritional deficiency. It is not a deficiency of fluoride. Fluoride is an important agent in increasing the resistance of enamel surface to dissolution of its mineral compo- nents, but it is not a lack of fluoride that produces dental decay. We should think, therefore, not only in terms of increasing the resistance of the tooth, but also in terms of reducing the caries challenge to which these enamel surfaces are exposed. Another important factor is that the disease has a multifactoral etiology. In other words, it is determined by the interaction between three factors, of which diet, particularly fermentable ingredients in the diet, is one; secondly, the host factors; and thirdly, the micro- flora, which is the specific agent that attacks the tooth. Of all of these, diet is most important. First of all, it affects the tooth before eruption. Sound nutrition is essential for the for- mation of a tooth expected to withstand the challenges and stresses of a perhaps not so balanced diet consumed after the teeth have erupted. You must realize that all other tissues have an opportunity for repair at different stages during development, but the enamel surface that is formed before eruption under poor nutritional conditions cannot be re- paired. Therefore, adequate nutrition during tooth formation becomes extremely important for dental health. After eruption, diet can stimulate the caries process through the frequent exposure of sugars in foods, the form and the concentration of sugars being important. Concentration does make a difference. There might be variations that we like to play with scientifically, but nevertheless what does make a difference is the form, the frequency, and the concentration of sugars in foods. All of these are going to affect the expression of the disease. The diet also is going to deter- mine the implantation on the tooth surface of the organisms that are going to be responsible for caries. So what can we do, and how can we use some of these different sweeteners? First of all, a new life-style is necessary. People do not like to think of discipline, but it is necessary. If you have a life-style in which you don't eat continuously throughout the day, then I think that there is no special need for saccharin or for any substitute for sugar. In those cases, the chances for preserving a strong dentition are good. However, if people insist on the kind of a life-style in which they sip all kinds of sweet beverages and eat sugar snacks all the time, then for those people who insist on this kind of behavior, an option could be available to use the synthetic sweeteners that have been shown not to be caries-promoting. So to summarize my views, sugars have been found to be the ingredi- ent that is responsible for the caries-promoting properties of food snacks, and nonnutritive sweeteners have not been found to be caries- promoting. Therefore, if you were to substitute completely all the
l69 fermentable sugar in foods for saccharin or for any other kind of synthetic sweeteners, you would definitely improve dental health. However, we have no data on what is the impact on partial substitu- tion of the sugars with saccharin. We need studies to elucidate this particular point. I think that the availability of synthetic sweeteners is useful for a dentist who, for example, wants to manage a patient with rampant caries, where a strict control of his fermentable sugar intake is re- quired, and for them, therefore, this is an important point. Nonnutritive sweeteners can be helpful in the prevention of dental caries. But in my estimation they do not really constitute a unique, essential approach, as there are other approaches and other etiologi- cal factors in caries that are also important for the disease. There- fore, the provision of these synthetic sweeteners as over-the-counter products or as food additives is unlikely to have a major influence in terms of dental caries, although I can see that they might be useful in specific circumstances.
NONNUTRITIVE SWEETENERS AND OBESITY W. Henry Sebrell The point was made yesterday that the use of saccharin or nonnutritive sweeteners by people who are trying to reduce has little or no effect. A person can put saccharin in his coffee and then eat a larger piece of apple pie or otherwise increase his caloric intake. The difficulty is that the individual has not been educated to modify his behavior toward food. So from this point of view, saccharin or other nonnutritive sweeteners have little effectiveness. However, this is a shortsighted view and does not take into account the real importance and essentiality of nonnutritive sweeteners for many people. The importance of a sweet tase was emphasized yesterday, and this is the basis for much of the problem. We are biochemical in- dividualists, and the desire for sweetness varies greatly. In some people, it is very strong. This is well known in the diabetic, but it also is very prevalent among the obese, many of whom are prediabetics. Large numbers of obese people have an intense desire for a sweet taste. I have never seen an obese individual who did not have an internal emotional conflict between the desire to lose weight and the desire to eat. This is the basis of the obese person's problem. In order to combat this, the motivation to reduce must be strengthened, and everything possible must be done to weaken this desire to eat. Quite frequently, it is not only a desire to eat, but it is a desire to eat sweet things. Now the basic problem is how to create a behavior modification to- ward food that is going to last for a lifetime. Another problem is how to get the daily allowances for all the nutrients recommended by the Food and Nutrition Board of the NRC into a limited number of calories and in a food pattern that gives the individual pleasure and satisfac- tion in eating. This is a very important consideration in altering lifetime food habits. l70
l7l In order to get all of the nutrients into a food supply limited in calories, you cannot use sugar. It does not carry the necessary nu- trients. Yet for the food supply to be palatable, it must have some sweetness. The only practical way to fill this need is with a nonnutri- tive sweetener. The quantity should be adjusted to whatever satisfies that particular individual's craving for a sweet taste, as opposed to his desire and motivation to eat. It can be argued that this is a drug use for diabetics. Dr. Williams made the point about its great importance for a juvenile diabetic. But it is also of great importance to the obese individual who wants to lose weight. But how about the use of nonnutritive sweeteners by the general pub- lic? As a former public health person, I like to look forward to a time when we can prevent obesity -- a time when we can create a life- style in which we eat properly, have joy and pleasure with our eating, and still not become obese because of taking too much of the wrong kind of food. So I am very strongly of the opinion that a nonnutritive sweetener, as safe as possible, is of great importance and must be part of an educational campaign on how to use artificial sweeteners properly in behavior modification toward our food supply. DISCUSSION MARSHA COHEN: There is a very serious contradiction here between today and yesterday, and I would like to clear it up. Yesterday, in a question to Dr. Stare, I asked if it were possible for someone who has a small caloric need to follow his view that you can ingest 25 percent of your calories in sugar and still get all the nutrients one needs. He agreed that this could be all refined sugar, not carrying any vitamins or minerals, and that you still would be in no nutritional trouble. Dr. Sebrell has just said that you cannot do that. I would like to ask him, therefore, if he disagrees with what Dr. Stare said yesterday about sugar in the diet. SEBRELL: Well, I have had a lot of experience now in making diets, and you cannot get the NRC's recommended nutrient allowances into l,200 or l,400 calories of food if your total intake includes refined sugar, simply because of the point I made that refined sugar adds to the calories without adding any nutrients. I am not sure that you aren't misinterpreting what Dr. Stare said. I said "refined sugar." Sugars or carbohydrate in the diet is something else. The ordinary structure of a restricted calorie diet can have 40 to 50 percent of the calories from carbohydrates or sugars, but these are natural sugars occurring in foods. These are the starches, the sugars in orange juice, in fruits of all kinds, and in milk. If we reduce the fat intake from 45 percent of our calories
l72 to 30 or 35 percent, in accordance with the Heart Association's recommenation, the diet would consist of 40 to 50 percent from carbohydrates, 30 to 35 percent from fats, and the remainder from protein. COHEN: Thank you, Dr. Sebrell. I think you do disagree with Dr. Stare. I did stipulate in our discussion yesterday that according to his view I could have sugars. When I asked, "Well, what if I choose to have all refined sugars?" I believe he said that was still okay and that I still could get all my nutrients. You say I cannot, and I thought that was the point I was trying to bring out yesterday. MIA TALERMAN: I think that people are tending to confuse calories, refined sugar, and sugars. If you are going to invest in a certain amount of calories per day on a certain diet to keep a certain weight, why in the world would anybody want to take calories from a refined sugar source, which gives them no nutrients at all? I fail to see it. Dr. Sebrell mentioned that there are sugars that we get naturally in foods, such as fruits and vegetables. We are all quite aware of this fact. But in those calories we are getting something for our investment. We are getting nutritious value, and I think this is what should be confirmed. Dr. Stare's comment yesterday concerning 25 percent of sugars did include refined sugars as well as sugars found in normal food. However, I think that this puts too much value on refined sugars, because you are, after all, not getting anything for your money and for your investment. SEBRELL: I am afraid you exaggerated a little bit. You are getting something for your money and your investment. We have not said anything about exercise in this Forum. It is a difficult and con- troversial subject, especially for the obese, and it is a question of intake and expenditure of energy. If our use of muscular energy is increased, we can increase the number of calories in our diet. Again let me emphasize the importance of this sweet taste. The desire for sweetness is a very deep and strong thing, and there is nothing like sugar to satisfy it. However, your energy expenditure must be increased in order to have it without becoming obese. I don't want to say we should do without sugar. I do want to stress the need for nonnutritive sweeteners. TALERMAN: Two other points. I am with Georgetown University, working in research at this level, and I have given some time and some thought to this. You did mention exercise. Although exercise is important for the condition of the body, it is on another level and not to be confused with nutrition. In regard to your statement regarding the desire for sweets, I agree. We are not debating that. But we have to establish two
l73 things: we cannot do everything we like, and we have to have a responsibility for our actions. Now what helps to give us a respon- sibility for our actions? Some factual knowledge. And this is what I think is one of the most important things we should estab- lish here -- that people should have some basic, fundamental factual knowledge so that they, themselves, can reach a reasonable conclu- sion on the evidence. Otherwise people have no knowledge, and there is a capitalization on the lack of knowledge. KASHA: As I get the message from this morning, saccharin if l00 percent pure is a completely safe sweetener, with no possible consequences except for its sensory perception and its excretion. However, there is a deleterious acid present, e.g., OTS, which is quite different chemically; it is a sulfonic acid. I wonder if there is anyone pres- ent who is connected with the saccharin manufacturing industry and who could tell us of the effective removal of deleterious by-products in saccharin manufacture. SVEDA: I think I have an answer to that. I am told that as of about a year ago, Monsanto, who went into business making saccharin through the sulfonation of toluene, is out of that business. I also am told by the man who devised the Maumee Process, who is now Vice President for Sherwin-Williams, that this is the only way that saccharin is now made in this country. So far as I know, this is a fact. Is there anybody who would dispute that or knows any more than that? E. D. COMPTON, Sherwin-Williams Company: I work for the man to whom Dr. Sveda referred. He was a founder of the Maumee Chemical Company. Sherwin-Williams manufactures saccharin by the Maumee Process. As far as we know, we are the only manufacturer in the United States. B. STAVRIC, Research Scientist, Health and Welfare, Canada: I am part of the team that was working on saccharin impurities. I can inform this Forum that we developed a method for the purification of saccha- rin not only from OTS, but also from most other impurities in saccha- rin produced by the Remsen-Fahlberg Process. A patent was applied for under the Public Servants Invention Act. However, because of possible conflict of interest (the same government agency is provid- ing a regulatory requirement for the purity of saccharin), the procedure was not patented. It was recommended for publication, and we intend to do this. But, regardless of that, from some contacts I have with my col- leagues from Japan, Dr. Miyaji and others (saccharin in Canada is imported mainly from Japan), I understand that the Japanese industry is making saccharin available on the market with very low OTS levels. How much the levels are, I don't know; but I believe they could be about 30 or 40 ppm. Dr. Grice has presented data about OTS content in saccharin tablets, liquids, and blends. These samples were ob- tained in the Ottawa region in February l974. I believe that if we
l74 were to repeat the same survey today, we would see much lower con- tent of OTS in a variety of saccharin preparations. May I just add something to illustrate the development of the impurities in saccharin. It happened that we had analyzed three saccharin samples produced by the same company during a l5-year period. When we used the same analytical procedure to check for impurities in saccharin produced about l5 years ago, it was unbe- lievable the number and amount of impurities we found. At that time, nobody was interested in checking the impurities by paper chroma- tography or gas liquid chromatography. Saccharin from the same company produced about 4 or 5 years ago (the same saccharin used in one of those animal tests) had a good number of saccharin impurities, including OTS. However, impurities were less quantitatively and qualitatively than initially. The same company produced saccharin in l974, and we have analyzed a sample of it. This lot of saccharin was from regular production. We found practically no impurities in it. So apparently the indus- try is doing its share in correcting this problem.
OTHER SWEETENERS
