Tobacco Documents Online

Review of Chapter 8 by D. Hoffmann, K. D. Brunnemann, and N. J. Haley of the draft Compendium of Technical Information on ETS edited by the Environmental Protection Agen.cy. L H Primary Reviewer: J. Donald deBethizy, Ph.D., D.A.B.T. Secondary Reviewers: Riley A. Davis, M.S. and David W. Eaker, Ph.D. Overall, Chapter 8 entitled "Absorption of smoke constituents by nonsmokers" by D. Hoffmann, K. D. Brunnema:ln, and N. J. Haley is well written and informative. The authors provide a concise and critical perspective on the many markers. that have been reported in the literature. However, the chapter can be improved considerably by addressing the following criticisms: . 1. Introduction. This section is somewhat anecdotal and needs to be tied to the literature better. For example, the definition of ETS in the first paragraph should be referenced. In addition, the discussion at the top of p. 96 about the SS smoke yield of ultra low and low yield cigarettes needs to ba referenced and more specific. The paper by Nelson et al. (1989) compares the ETS generated by smokers smoking cigarettes that vary in MS smoke yields. In fact, using this reference the a-.ithors could state that RSP, nicotine, formaldehyde, and acetaldehyde were 20 % lower in ETS generated by smokers smoking ultralow "tar"-yielding cigarettes compared to an equal number of full flavo::- low "tar" cigarettes (1R4F) instead of saying "a somewhat lower yield of SS is expected from the low-yield cigarettes". 2. Table 1. The tobacco-specific nitrosamine numbers reported in Table 1 may be artifactually high. The authors should point out that these data were collected prior to a report by Connors and Caldwell (1989) which reported artifactual nitrosamine formation on the Cambridge filter pads used to collect the SS smoke. These authors describe a method for eliminating artifactual formation of TSNA's by impregnating the pad with ascorbic acid. This paper should be referenced. 3. p. 96. lines 11, 12, 16, 22. These sentences should be rewritten to avoid the use of "polluted". This word conjures images of smokestacks belching black soot <<nd is not necessary in this document. 4. p. 96. lines 11 - 12. The authors should be specific about how much dilution of SS smoke takes place in indoor environments. It is important for readers of this chapter to realize that even though the yield of some chemicals from cigarettes is higher in the SS than in the MS smoke on a per cigarette basis (See Table 1 of the chapter), the dilution of the SS smoke in the environment is considerable. The dilution of these SS conponents in air reduces the exposure of nonsmokers by orders of mag:zitude when compared to L

the exposure of smokers to these same chemicals. For example, the SS yield of nicotine per cigarette is reported in Table 1 to be 2.5 to 21 times higher than the MS yield. However, the exposure to nicotine from ETS is generally 200 to 300 fold lower than smokers (See conclusion in Jarvis et al., 1984, ref. 41 in this chapter and summarized in Table 3 of this chapter). These data indicate that there is considerable- dilution of SS smake components in the environment and that its important to stress this fact.when presenting data shown in Table 1. Otherwise:, the public is done a disservice when reports appear in the lay prass that SS smoke is 10 times more toxic than MS smoke. r U L* I L 5. 'p. 96. line 25. The optimal assessment of ETS exposure would include bot analysis of physiological fluicls and the environment; not just fluids as suggested by the authors. It is important that the biomarker concentrations observed in the passively exposed individual be related to the environmental concentrations. Since the half-life of some of the biomarkers is very long (e.g. cotinine t 1/2 = 15 to 25 hr), there is the danger that ETS exposure could be attributed to the wrong environment if the relationship between environmental concentrations of ETS components and the physiological fluid concentrations are not confirmed. The limitations of saliva and blood for discriminating non-smokers and passive smokers has been discussed by Wall, et al-., 1988. These autors conclude that saliva and serum are nat appropriate matrices for discriminating non-smokers and passive smokers, but could be used to define active smokers. Factors which contribute to this lack of discrimination include: a) variance in nicotine metabolism, b) the time of day for sample collection, c) under-reporting of active smoking, d) adjustment of cigarette consumption for nicotine content, and e) over- and under-reporting of passive cigarette smoke exposure. 6. p. 96. The statement that the presence of nicotine and/or its metabolites in biological fluids is entirely due to the exposure to tobacco smoke may not be true. Sheen (1988) and Castro and Monji (1986) have demonstrated that nicotine is present in solanaceous vegetables such as potato, egg plant, tomato, and green pepper which are commonly consumed by humans. The contribution that nicotine in the diet makes to the baseline nicotine/cotinine in physiblogic fluids remains to be determined. However, with this data in the literature one cannot assume nicotine is "entirely" derived from tobacco. The statement should be modified to include references to these results. 7. p. 97. The statements about HPLC an<<lyses of nicotine and cotinine should be updated to include recent: literature. Cummings et al. (1989) used the Machacek and Jiang (1986) HPLC method to screen the urine of 663 never and ex-smokers for nicotine and cotinine. However, the mean cotinine concentrations were unusually

{ I L: .r high (9.5 ng/ml) for people who claimed to :not be exposed to ETS. One explanation for these high concentrations of cotinine may be interference of cotinine with caffeine using this HPLC method. Thuan et al. (1989) found that caffeine eluted between cotinine and the phenylimidazole internal standard causing interference with the cotinine determinations. This section should be expanded to include statements that describe the use of HPLC for nicotine and cotinine and point out its potential pit falls (Hariharan, et al., 1988). 8. p. 97. line 16. 3-hydroxycotinine must be derivatized because it is not sufficiently volatile for GC analysis and not readily soluble in organic extraction solvents. The line stating " not readily soluble" should be rewritten. 9. p. 97. The section on RIA techniques should address the concerns about the use of RIA for nicotine and cotinine at concentrations found in nonsmokers exposed to ETS which have been raised by Van Vunakis et al. (1987). The:se investigators have questioned the accuracy of the RIA, which wa.s optimized for use in active smokers, at the low concentrations encountered in the ETS exposed population. They state that cotinine concentrations below 10 ng/ml require the use of large sample volumes and data from a sub-optimal portion of the standard curve (< 5% inhibition by the antibodies). These limitations of the assay should be discussed and referenced. The detection limit for the RIA is frequently stated as being between 350 to 390 pg/mi for nicotine and cotinine, respectively. However the more important Limit of Quantitation has not been defined statistically. By definition there is a 30% chance of inaccuracy at the 95% confidence interval at the value of the Limit of Quantitation in any analytical technique. It is not unreasonable for the Limit of Quantitation to be stated for the application of,the RIA by the investigators reporting the data and the conditions from which the LOQ was derived. According to Van Vunakis, et al. (1987) increased sensitivity can not be derived from increasing the sample volume above 20 ul for urine samples and 0.2 ml for plasma samples. 10. p. 98. line 9. This statement should be expanded to include the reported cross reactivity of the cotinine antibody used in the RIA analyses with 3'-hydroxycotinine (301:; Schepers and Walk, 1988). 11. 1 p. 98. lines 10-11. It is unlikely that there is loss of cotin-ine during extraction prior to GC ana:lysis since all of the commonly used methods employ internal standard that would correct for losses. Thi,s statement should be delets:d or modified to point this out.(Davis, 1986; Curvall, Kazemi-Va:1a, and Enzell, 1982; Jacob, Wilson, and Benowitz, 1981). 12. p. 98. References to reference 33 in the second paragraph should probably be reference 32. 13. p. 98. Last paragraph. There is a broad range of values for

c t Ii L U L the half-life of cotinine in smokers and nonsmokers reported in the literature as implied by the authors. Since this paper will be read by those interested in using cotinine as a marker, the range of values reported in the literature shoL.ld be provided. The authors also imply that the difference in the cotinine half-lives between nicotine gum users and smokers is significant without providing some assurance that the difference is statistically significant. The statement on page 98 that.the longer half-life of cotinine in nonsmokers suggest that the residence time of nicotine and its metabolites are longer in nonsmokers: is overstated. There is very little 'difference, if any, in 'the rate of nicotine elimination between smokers and nonsmokers (Kyerematen et al., 1982). Thus the residence time of nicotine atself is not likely to be significantly longer in nonsmokers. The statement should be limited .to cotinine unless evidence of a longer half-life for nicotine is presented in the paper. The statement that the longer residence time of nicotine metabolites in nonsmokers could "conceivably increase the possibility of endogenous formation of carcinogenic tobacco-specific N-nitrosamines" is unsupported. Cotinine and its metabolites are very unlikely candidates for nitrosation and when cotinine is nitrosated, the resulting product is not a mutagenic/carcinogenic N-nitrosamine (See report from Brunnemann's laboratory at the 1989 TCRC meeting). In addition, there is no evidence for the in vivo formation of tobacco-specific nitrosamines from.nicotine or any of its metabolites. Since this statement is speculation, it should be either modified to include the latest information or eliminated from this type of document. 14. p.99. last paragraph and p. 101 second paragraph. A criticism of this work was described in Van Vunakis et al. (1987). This should be discussed and referenced at both of these points. The discussion on p. 101 about creatinine should reference p. 322 of the Van Vunakis paper. 15. The sections on carbon monoxide and th.iocyanate are accurate and well balanced. 16. The sections on hydroxyproline, N-nitroso-amino acids, aromatic amines, thioethers in urine, genotoxicity of physiological fluids, and adduct formation should all be grouped under a heading Indirect Markers of Uptake similar to how these authors treated this same material in another paper on the same subject (Haley et al., 1989). This would reduce the importance of these assays for assessing ETS exposure in the context of this chapter. Collectively these assays suffer from a lack of specificity for ETS constituents and a lack of sensitivity. Specific comments about these assays are listed below: p. 103. Aromatic amines. Even though the yield of some aromatic amines may be higher fro:n the SS than from the MS of some cigarettes, one shou.1d not imply that the exposure will be greater. SS is highly diluted in air before it is breathed by nonsmokers (See discussion in point #3 above). The amount of these aromatic amines

contributed by ETS compared to the amount being excreted in urine is a very small percentage. This section should be deleted because the contribution from other dietary and environment sources would confound this assessment. r f ; p. 103. Thioethers. These data are too preliminary to include in this document. The study can be referenced, but thioether excretion is notoriously insensitive to exposure to low doses. There are too many other xenobiotics that could react with glutathione in vivo and alter thioether excretion in the urine. p. 103. Items 4 and 5. It is, unlikely that either hydroxyproline or N-nitrosoamino a:cids could ever be used to assay personal dosimetry of ETS exposure. The paragraph on p. 8 discussing the use of hydroxyproline could include the recent finding that the diets of nonsmoking women married to smokers is different than the diet of nonsmoking women not married to smokers. These differences in diet could exp:lain the "surprising" differences in hydroxyproline excretion. These sections should be eliminated from the paper. p. 104. Urine mutagenicity. This should.be eliminated from the paper. The contribution of the diet to urinary mutagens is large enough to preclude the use of this assay for monitoring ETS exposure. p. 104. Adduct formation. The statement that urinary mutagenicity could be utilized to assess uptake of ETS should be removed from paragraph 1 of section C for the reason stated above. 17. p. 106. Site Ogden and Maiolo, 1988 and 1989 papers here along with reference 101. 18. Table 3. This table is a composite of Table 1 and 2 from the cited reference. The plasma concentrat:Lons for nicotine and cotinine in the plasma in the nonsmokers is below the limit of quantification for the assay. This leads to an erroneously high value relative to smoker's plasma. The authors of the original paper state "The average concentration of cotinine, whether measured in plasma, saliva, or urine, lay between one-third to one- half of 1 % of the levels found in the smckers in this study and migYit therefore be regarded as trivial". The statement on p. 99 referencing this table should reflect the conclusion drawn by the original authors. The statement used in this paper on line 13 of p. 99 is not consistent with the conclusion in the original paper. L

REFERENCES 1. Castro, A. and N. Monji. Dietary Nicotine and Its Significance in Studies on Tobacco Smoking. Biochemic3l Archives 2, 91-97, 1986. 2. Connor, J. M., and W. S. Caldwell. Artifact Formation during Smoke Trapping. An Improved Method for the Determination of N-nitrosamines in Cigarette Smoke. Presente:d at the 43rd Tobacco Chemists' Research Conference, Richmond, VA, October, 1989. 3. Curvall, M., E. Kazemi-Vala, and C. R. Enzell. Simultaneous Determination of Nicotine and Cotinine in ]?lasma using Capillary Gas Column Chromatography with Nitrogen-Sensitive Detection. J Chromatog 232, 283-293, 1982. 4. Curvall, M. and C. R. Enzell. Monitoring Absorption by Means of Determination of Nicotine and Cotinine. Archiv. Toxicol. Suppl., 88-102, 1986. 5. Davis, R. A. The Determination of Nicotine and Cotinine in Plasma. J Chromatog. Sci. 24, 134-141, 1986. 6. Haley, N. J., D. W. Sepkovic, K. E. Brunnemann, and D. Hoffmann. Biomarkers for Assessing Environmental Tobacco Smoke Uptake. Presented at the Air Pollution Control Association Specialty Conference on Combustion Processes and the Quality of the Indoor Environment, Niagra Falls, NY, September, 1988. 7. Hariharan, M., T. VanNoord, and J. F. Greden. A High Performance Liquid-Chromatographic Methcd for the Routine Simultaneous Determination of Nicotine and Cotinine in Plasma. Clin. Chem. 34, 724-729, 1988. f 8. Jacob III, P., M. Wilson, and N. L. Benowitz. Improved Gas Chromatographic Method for the Determination of Nicotine and Cotinine in Biologic Fluids. J Chromatog. 222, 61-70, 1981. 9. Kyerematen, G. A., M. D. Damiano, B. U. Dvorchik, and E. S. Vesell. Smoking-Induced Changes in Wicotine Disposition: Application of a New HPLC Assay for Nicotine and Its Metabolites. Clin. Pharmacol. Ther. 32, 769-780, 1982. ~. ~. 10. Machacek, D. and N. Jiang. Quantifi.cation of Cotinine in Go Plasma and Saliva by Liquid Chromatography. Clin. Chem. 32, Go s 979-982, 1986. ' ~ c1i W co

11. Nelson, P. R., D. L. Heavner, and B. B. Collie. Characterization of the Environmental Tobacco Smoke Generated by Different Types of Cigarettes. In Present and Future of Indoor Air Quality, Proceedings of the Brussels Confe::ence, 14-16 February, 1989, edited by C. J. Bieva, Y. Courtois, and M. Govaerts. Excerpta Medica, Amsterdam. Pps. 277-282. 12. Ogden, M. W. and K.C. Maiolo. Gas Chromatographic Determination of Solanesol in Environmental'Tobacco Smoke(ETS). J High Res. Chromatogr. Chromatogr. Comm. 11, 341-343, 1988. 13. Ogden, M. W. and K. C. Maiolo. Collection and Determination of Solanesol as a Tracer of Environmental Tobacco Smoke in Indoor Air. Environ. Sci. Technol. 23, 1148-1154, 1989. 14. Schepers, G. and R. A. Walk. Cotinine Determination by Immunoassays May Be Influenced by Other Nicotine Metabolites. Arch. Toxicol. 62, 395-397, 1988. 15. Sheen, S. J. Detection of Nicotine in Foods and Plant Materials. J Food Sci. 53, 1572-1573, 1988. 16. Thuan, N. T. L., M. L. Migueres, D. Roche, et al. Elimination of Caffeine interference in HPLC Determinat:.on of Urinary Nicotine and Cotinine. C1in..Chem. 35, 1456-1459, 1989. 17. Van Vunakis, H., H. B. Gjika, and J. J. Langone. Radioimmunoassay for Nicotine and Cotinine (Method 16). Environmental Carcinogens Methods of Analysis and Exposure Measurement, Volume 9-Passive Smoking. IARC Scientific Publications No. 81. eds. O'Neill, I. K., Brunnemann, K. D., Bodet, B. and Hoffmann, D. Oxford University Press, New York. Pps. 317-330. 1987. 18. Wall, M. A., J. Johnson, P. Jacob, and N. L. Benowitz. Cotinine in the Serum, Saliva, and Urine of Nonsmokers, Passive Smokers, and Active Smokers. Amer. J Puba. Health 78, 699-701, 1988.