Tobacco Documents Online

..:::..: .. ... ..... -- . : .. . .. ... . .. ._. - _ --- - _ -.-- - - : . ..-...... . : : r. - - - - -- Rko -_ - _ - -- - _ -- - -___. -__- _- -_r~~. - - - - _~--- _~_- - - - -- -- - --_-- . , .. ...__: _. ._._..... .... -- -.. , -- - - ... .. ...~. ....-- ------ -- .. ---~- -- - --- __.... .. _ . =- ~ - - .- . - ----,... . ~~-_- .- ~--- -- -_ _.--- ..~-. - - - ~- =r:: - - - -- ---- =}=- --- --- _ -- -- -__ - - ~ -- :-{: - -~-- - --- - -- -- -- -- --- -+- - _ . . r - -- ,,.... . _. _ .. .._. _.~ ~ -- 1- - -- -- =- ~ = ~ - _ ~~ ~ _- -- - - - - - - _ - - - -- - _ - - ~-,~-- - - -_ - --~- --__~_=---~--~ - - - _ -~-- _ In particular, we can iind that value ql* such that there is 95% probability that ql < ql*. However, it is important to note that the uncertainty distribution so plotted contains only the uncertainty due to the numerical size of the experiment -- the uncertainty that arises because we used a small number of animals, instead of an infinite number. It does not include the uncertainties which must be present because of the shakiness of all our assumptions. 7. The Two Major Extrapolations The assumptions made so far have allowed us to parametrize an animal dose-response relationship, obtaining values for the parameters which are •,presumably reasonably appropriate for high doses. Strictly speaking, this parametrization of the dose-response curve only enables us to estimate the results we would expect to see at high doses in animals - the dose-resposne relationship can only be relied on to interpolate between high doses and perhaps to extrapolate a short distance outside the experimental range of doses. The problem now is to perform two extrapolations - from animals to humans, and from high dose to low dose: ~ Animal ~ Human ( ------------- ~ ~ ~ High Dose ~ Observed ....~.......:. ~ I I I V I ---------------------- ~------------------ ~------------ I I I I I Low Dose ~ ---------- ~----- >Required ( I I I ------------------------------------------------------- WGICALLY there are two distinct routes to follow in this extrapolation, since there are logically two distinct dose-response -11-

curves involved (see below). One can extrapolate from high dose to low dose using the ANIMAL dose-response curve, and then extrapolate to humans (dashed lines), o,r extrapolate to humans at high doses and then use a HUMAN dose-response curve to extrapolate to low doses. We have seen how to estimate the parameters of the (high dose region of) the animal dose-response curve. In practice, the same curve (with the same parameters) is is used to extrapolate to low doses, by building ir.to the mathematical structure of the dose-response curve all our assumptions about low dose behavior. How is this relevant for estimating human risk? Consider a generalized situation in which we wish to estimate the response (R) of humans to some dose (D) of material, when there is a response (r) in some animal at dose (d). Notice that nothing implies that r, R measure the same sort of response - they could be completely different (r could be acute toxicity to the lung of a mouse, R could be skin rashes in hwnans). Similarly, the dose measures d, D may be completely different. In the case immediately at hand, r is the lifetime probability of tumor in animals, and d is a dose as measured in the animal experiment. There are other cases of practical importance however - r might be some measure of response (such as number of revertants per culture dish) in a mutagenesis bioassay, with d the dose applied to each culture dish. Animal Human Response: r R (lifetime probability of tumor, p) Dose measure: d D (as used in experiments) Dose-response curve: r= f(d; a,b,c,...t ) R= F(D; A,B,C,....T ) [p = 1 - exp{-(q0+q1.d +...)} ) What is required is some connection between the parameters a,b,c,... of the animal dose-response relationship and the parameters A,B,C,... of the human dose-response relationship. These parameters presumably include those mentioned in section 6, and I have explicitly included age amongst them. Given such a connection, the extrapolation to humans of the results in the animal studies is perfectly straightforward. The probLem lies in finding the connection. Once such a connection is found (by whatever means) we have the methodology for the two extrapolations required. Notice the difference between what is done in the two distinct pathways of extrapolation ment:ioned above: In the first, the shape of the dose-response curves are examined, and it is decided how they may be (separately) extrapolated to low doses. Then some relationship is postulated between the parameters of the dose-response curves at low doses (it has to be postulated, since nothing can be measured at such low doses). One potential advantage of this approach is that the animal dose-response curve could be measured,, in principle and by heroic experimentation, down to lower response rates than usual (and this has been done in some cases) - allowing greater confidence in this extrapolation to low dose. -12-

In the second, some relation between the parameters of the dose-response curves is obtained at high doses (and this may be done experimentally, in principle, since at high doses the responses are measurable). Then it is decided how the human dose-response curve should be extrapolated to low doses. The advantage here is the possibility of direct comparison between species, albeit at high dose. The difference between these two logically distinct routes of extrapolation might be important in some circumstances. For cancer risk assessment based on animal carcinogenesis bioassays, however, the distinction is glossed over (one might even say, ignored), by the practice of assuming the same mathematical form for the dose-response curve in both humans and animals (or more generally, in all species), and interpreting the parameters in the same way for both compared species. In the general case, however, what is required is some sort of relationship between the parameters of the dose-response curves: Animal Human r= f(d; a,b,c...t) R- F(D; A,B,C...T) We need to be able to derive the parameters A,B,C... from the values a,b,c which can be estimated from experiments, and then use the human dose~-response curve to extrapolate to low doses. The practical approach is to seek parametrizations of the dose-response curve which result in the derivation of A,B,C... being simple given'a,b,c... Consider the case of acute toxicity, for example. It is found that the shape of the dose-response curve for acute toxicity, in which the response is death, is very similar for a large number of toxins and for many different species. There is, in this case, a t:hreshold-type dose-response curve which can be nicely parametrized by two values: the dose at which 50% of the animals tested can be expected to die (under suitable conditions), and the slope of the dose-response curve at this dose. The first parameter is known as the LDSO (the second has no special name). 1'7hy is this parametrization useful? If the LD50s of various mate:rials in one species are plotted against the LD50s of the same materials in another species, one finds approximate proportionality between them (the plot is a straight line). This can be expressed as, for example, LD50(rabbit) is proportial to LD50(mouse). Even more remarkable, it turns out that if the dose is measured in a suitable way, as (amount)/(surface area of animal), then approximately we actually have LD50(rabbit) - LD50(mouse) - LD50(other species) and it is this approximate equality which explains the utility of the LD5C1. The other parameter used in defining the dose-response curve, the slope of the curve at the LD50, is not involved in this relationship. -13-

Had we chosen some other method of parametrization, it is quite possible thO required interspecies relationship between parameters would be much mor'e complicated. 8. Interspecies Comparison - Constant Relative Potency What is sought is a simple relationship between the parameters of dose-response relationships in different species. When it is assumed that the dose-response relationship includes a term linear in dose, there is a simple measure of the strength of a carcinogen - the carcinogenic potency (the slope of the dose-response curve at low dose). The simplest hypothesis is that for different species, the ratio of carcinogenic potencies is constant for different materials, so that if material A is twice as potent a carcinogen as material B in species 1, it will also be twice as potent as material A in species 2. This is the idea of constant relative potency, as applied to carcinogenesis, and it underlies the standard approaches to estimating human risks from animals. There is even some data which supports this idea! There have been several hundred bioassays performed simultaneously on rats and mice, and when the results of these are parametrized using a dose-response relationship which includes a linear term, we can estimate the potency in t:wo species for each material tested. Plotting the potency measured in rats versus the potency measured in mice for each material then gives the figure shown. Notice that each measurement is uncertain to greater or lesser degree, due to the relatively small numbers of animals tested. If the idea of constant relative potency were exactly correct, these points would all lie on a straight line on the figure - or at least, all would lie sufficiently close to such a line that the measurement uncertainty bars on each point would encompass the line. From the figures, one can see that: (1) On average, potency in one species is proportional to potency in the other species. (2) There is a large scatter of the points around the lines of exact prop,ortionality - a scatter bigger than would be expected from the measurement errors alone. A similar comparison can be attempted between the potencies measured in animal experiments, and those observed in humans. These cases have arisen in the past where humans have been exposed to materials before they were known to be carcinogenic. We can make use of other's misfortune to estimate how potent each such material is in humans, and compare with estimates obtained for mice and rats'~in laboratory experiments. In this case, the uncertainties are so large that little can be quantitatively states, although qualitatively the idea of const:ant relative potency does not seem to be disproved. A more recent and much more thorough study of comparisons between humans and animals has b~een carried out for the E.P.A. by Dr. Kenny Crump, and we can expect that to be published soon - I understand that conclusions are qualitatively similar.

9. Interspecies comparisons - practical and theoretical The measure of carcinogenic potency introduced above was roughly def.ined as the ratio of (excess tumor probability)/(dose), at low enough dose. For the E.P.A. model usually used in risk assessments: p = 1 - exp{ - ( q0 + ql.d + q2.d2 + ... + qk-1•dk-1 ) } the corresponding measure is ql. When this dose-response relationship is used with real data, it is usual to use an "upper 95% confidence limit" estimate q1* of ql as the measure of potency, since such an estimate is always non-zero (while, for example, the maximum likelihood estimate is oft:e.n zero). The "upper 95% confidence limit" is with respect to the numerical uncertainties of the experiment only, and so this estimate of potency is in no sense an upper limit with respect to all the other uncertainties involved. To compare humans with animals, the approach taken is to use a similar dose-response relationship in both cases: Animal Human p = 1 - exp( - ( qO + ql.d +...)} p = 1 - exp( - (QO + Q1.D + ..)} and then the constant relative potency hypothesis suggests that Ql is proportional to q1, or to our estimate ql* of it: Q1 = const. q1* where the constant depends only on which animals species is used. We expect the constant to be different for different animal species - it will presumably depend on how we measure dose, on the relative lifespans of animal and human, on relative metabolic rates, and a whole host of other factors. With enough experiments, we could measure the constant in this relationship - at least in comparing animal with animal, rather than human with animal - and (in theory) empirically determine how it vari'_ea with these factors. The graphs above suggest that the constant is not completely constant, but that there is some sort of random uncertainty built in (or at least, an uncertainty that we can treat as random), amounting to an average factor of about 5. If we are very lucky, it may be possible to find some way of measuring dose so that the constant in the above relationship is numerically equal to 1, so that the potency is equal in different species (up to the uncertainties) - just as it was possible to find such a measure in the case of the LD50• In practice, the E.P.A. assumes that the constant is exactly unity if the dose is measured as a (daily average amount)/(surface area of animal), by analogy with the LD50 case. (The'graphs shown above actually suggest that it would be better to assume an average factor of unity, with an uncertainty factor of about 5, when the dose is measured as a (daily average amout)/(bodyweight of animal)).

10. An example - 1,2 Dibromoethane As an example of the procedures usually adopted, let us look at the case of 1,2-Dibromoethane. What follows is by now means complete, but it indi'_cates the sort of analysis which has to be performed. This example is confined to analysing just one result out of many, in a single bioassay (of about 5). In practice, it is essential to look at all the results. The bioassay I have chosen was an inhalation bioassay in the National Toxicology Program series. A summary of the study design is: Initial number of Concentration Time on study animals ppm exposed observed (6 hrs/d, 5 d/wk) (weeks) ------------------------------------------------------------------------ Male rats control 50 0 0 104-106 :Low-dose 50 10 103 1 high-dose 50 40 88 0-1 Fema l.e rats control 50 0 0 104-106 low-dose 50 10 103 1 high-dose 50 40 91 0-1 And similarly for mice We wi11 look only at the results in female rats. First, their survival was not as good as might be desired in such an experiment, but the early mortality was probably due to the cancers appearing in the study, so it is acceptable - we can use (at least initially) the simplest analysis based on "end-of-life" data, without having to worry too much about the age dependence (this should always be backed up by further analysis, of course). TIME ON STUDY IWEEKSI a -'- ~. _ tt FEMALE RATS 'p UqTN[AT[OCOHT/Wl - O tOw00ti HWOON ~ N w ao n TIME ON STUDY IWEEKS) Firyro 2. SurviYaf Curva for Rstt Exposad to Air Cont.ining 1, 2-Dibramoath.m -16-

Tumors were found in many tissues. A summary of those tissues where more: than 5% of the animals in any group were found with tumors is (for female rats): Control Low High Subcutaneous tissue: fibroma 0/50 0/50 3/50 Subcutaneous tissue: fibroma or fibrosarcoma 0/50 0/50 4/50 Nasal Cavity: Carcinoma, NOS 0/50 0/50 25/50 Nasal Cavity: Sqamous cell carcinoma 1/50 1/50 5/50 Nasal Caavity: Adenoma, NOS 0/50 11/50 3/50 Nasal Cavity: Adenocarcinoma, NOS 0/50 20/50 29/50 Nasal Cavity: Adenomatous Polyp, NOS 0/50 5/50 5/50 Nasal Cavity: Papillary Adenoma 0/50 3/50 0/50 Nasal Cavity: Adenoma,NOS; Carcinoma, NOS; Adenocarcinoma,NOS; Papillary Adenoma; Adenomatou s polyp,NOS; and Sqamous cell Carcinoma 1/50 34/50 43/50 Lung: Alveolar/Bronchiolar Carcinoma 0/50 0/48 4/47 Lung: Alveolar/Bronchiolar Carcinoma or Adenoma 0/50 0/48 5/47 Hematopoietic System: All leukemias 6/50 7/50 1/50 Hematopoietic System: Monocytic leukemia 6/50 5/50 1/50 Circulatory System: Hemangiosarcoma 0/50 0/50 5/50 Circulatory System: Hemangiosarcoma or Hemangiosarcoma, invasive 0/50 0/50 5/50 Liver: Neoplastic nodule 2/50 0/49 3/48 Liver: Hepatocellular carcinoma 0/50 1/49 3/48 Liver: Neoplastic nodule or Hepatocellular carcinoma 2/50 1/49 5/48 Pituitary: Adenoma, NOS 1/50 18/49 4/45 Pituitary: Chromophobe adenoma 20/50 0/49 0/45 Adrenal: Pheochromocytoma 3/50 1/49 0/47 Thyroid: C-cell Carcinoma 1/49 3/48 1/45 Mammary Gland: Adenocarcinoma,NOS 1/50 0/50 4/50 Mamnla.ry Gland: Fibroadenoma 4/50 29/50 24/50 Notice especially the various groupings which are employed - this is a matter of judgement. It is clear that the major effect is in the nasal cavity, but observe also the effect on fibroadenomas in the mammary gland., and the negative trend seen in the pituitary. Such negative trends are ignored. Using the combined'results in the nasal cavity, we fit the E.P.A. multistage model and find best estimates of: q0 = 2.699 x 10-2; ql = 6.876 x 10-2; q2 = 0; and obtain an upper confidence limit for ql of ql* = 8.6 x 10-2 in all cases using as doses the values 0, 10 and 40 from the experimental design. In fact, the earlier figure of a distribution of values for ql is taken from this example - you can read the probability of ql being less than any given value from that figure. -17-

Now what do we do with this estimate? That depends on the app:Lication, but we will assume that we wish to make a"tTNIT RISK" est::mate for humans from it - that is, estimate the lifetime risk to a hwnan exposed to 1 microgram/m3 of dibromoethane for life. There are several extrapolations required. First, the animals were dosed for a lifetime, but not continuously. Correcting for continuous exposure introduces a factor of 7/5 x 24/6 (for days/week and hours/day) - but notice the subtle assumptions being made here, that it is average exposure that matters (and not peak exposure, for example). Now we estimate that a female rat will suffer and increased lifetime risk.of about 0.48 per ppm in the air (we assume that we are talking about such low doses that the excess risk is small). 1 ppm for 1,2-dibromoethane corresponds to about 7.6 mg/m3 (one would estimate a lit:tle higher from the perfect gas laws), or 7600 microgram/m3, so that the increased lifetime risk to a female rat exposed continuously to 1 migrogram/m3 is about 6.3 x 10-5 What about humans? We saw before that the assumption made was that humans are just as sensitive as animals - i.e. they suffer equal lifetime risks - if exposed at doses which are equal on an (am.ount)/(surface area) basis. Now it turns out that, approximately, equal concentrations in air lead to exposures which are equivalent on this basis, provided the species under consideration absorb about the same amount from the air they breathe. Thus the extrapolation to humans is simple in this case- one simply takes the same value for humans - a "UNIT RISK" of about 6.3 x 10-5 (i.e. that is the lifetime risk from cont:inuous exposure to 1 migrogram/m3 of dibromoethane in the air). It may be desired to estimate from this the effect on humans of ingestion of dibromoethane. In this case there are actually other bioa:>says in which dibromoethane was fed to animals under various condb:tions, but suppose that we have to make some estimate from the inhalation data. The "standard" human inhales, on average, about 20 m3 of air per day, and so inhales about 20 microgram/day of contaminant from air contaminated with 1 microgram/m3. If we assume that 100% of this contaminant is absorbed, the human's daily dose is 20 micrograms/day, or about 20/70 microgram/kg-day (as a fraction of bod,,N,eight), or 2.9 x 10-4 mg/kg-day in the conventional units used. This results in a risk of about 6.3 x 10-5, as detailed above, so that the potency is just the ratio of these -- 0.22 (mg/kg-day)-l. These short outline calculations have made several assumptions which require examination in any particular case. We have not looked at all the bioassay results, so one cannot expect that the numbers obtained here will correspond with what anybody e1se,.who has done a more thorough job, will obtain -- they are placed here in order to show in outline what is done. In practice, one has to decide that the tumor site and type combinations are appropriate for combination in the animal species. That these tumors are relevant end points for estimating the probable effects on humans. That the route of administration, and method of adninistration are reasonable to produce results that may be extra:polated to humans. And a myriad of other details which have only been Lightly touched upon, or completely omitted, in this sketch. s -18-

to 10 3 2 ~-t LisM~C u d ~ ~ ~ 10 1 -3 10 . 104 tci~~ 162 io1 1 101 402 103 POTENCY IN RAT (mq'Kg d) ®enn0~n~ + / 2 f 1 ~ Lr~ ` _.L_ > > ' 10 '101 Y ~ -3 10 -2 1 -5 -4 -3 -2 -t 0 Log„(o) B6C3F1 / / (p= 0 025) / I + . , . + J_ + ~ T r _1_ f f t f ~ 10-3 10 Z 10 1 1 101 10 2 10 3 POTENCY IN MOUSE (mg" kg d) 5 -4 -3 -2 -1 0 1 2 3 og„(g): B6C3F1 (p= 0.025) Iq

6 TABLE E3. ANALYSIS OF PRIMARY TUMORS IN MALE MICE (Continued) Vehicle Control 500 mg/kg 1,000 mg/kg C.irculatory System: Hemangiosarcoma Overall Rates (a) 4;'50 (8%) 3/49 (6%) 1/50 (2%) Adjusted Rates (b) 10.1% 8.8% 2.6%, Terminal Rates (c) 3/38 (8c/'D 2/33 (K) 1/39 (3%) Life Table Tests (d) P=0.130\ P=0.559\ P=0.169N Incidental Tumor Tests (d) P=0.097\ P=0.408\ P=0.176N Cochran-Armitage Trend Test (d) P=0.134\ Fisher Exact Tests P=0.512\ P=0.181 \ Circulatory System: Hemangioma or Hemangiosarcoma Overall Rates (a) 4,50 (8%) 4,'49 (8%) 1j50 (2Si) Adjusted Rates (b) 10.1~'i 11.8~~ 2.6(/i Tesminal Rates (c) 3:38 (8~~) 3;33 (9ci:) I; 39 (3Si) Life Table Tests (d) - P=0.142N P=0.579 P=0.169N Ifncidental Tumor Tests (d) P=O.I ION P=0.573\ P=0.176\ Cochran-Armitage Trend Test (d) P=0.147ti 1i>her Exact Tests P=0.631 P=0.181\ if.iver: Adenoma Overall Rates (a) 0),50 (0%) 5 49 (10(/i) 13,50 (26~(-) Adjusted Rates (b) 0.0~i 13.0-&Ii 33.3~i Terminal Rates (c) 0: 38 (0-,i) 3; 33 (9cii) 13; 39 (339i) LFe Table Tests (d) P<0.001 P=0.030 P<0.001 lnr.idental Tumor Tests (d) P<0.001 P=0.023 P<0.00I Ccehran-Armitage Trend Test (d) P<0.001 Fis,her Exact Tests P=0.027 P<0.001 Liver: Carcinoma Overall Rates (a) 10; 50 (20c~) 14, 49 (29ii) 12 50 (24(/i) Adjusted Rates (h) 24.3r/i 35.9(/'i 25.8 r Terminal Rates (c) 7, 38 (18Si) 9 33 (27/j) 5, 39 (13S'c) L.ife Table Tests (c!) P=0.427 P=0.183 P=0.463 Incidental Tumor Tests (d) P=0.536 P=0.379 P=0.548K Cochran-Armitage Trend Test (d) P=0.363 Fisher Exact Tests P=0.224 P=0.405 Li.er: Adenoma or Carcinoma Overall Rates (a) 10 50 (20 (,,j) 18 49 (37ii) 23 50 (Wj) Adjusted Rates (h) 24.3ri 45.1~i 49.85"r Terminal Rates (c) 7 38 (18(/i) 12 33 (36Si) 16 39 (41(/i) Lifi: Table Tests (d) P=0.013 P=0.042 P=0.014 Incidental Tumor Tests (cO P=0.009 P=0.098 P=0.019 Cochran-Armitage Trend Test (d) P=0.004 Fisher Exact Tests P=0.052 P=0.005 Forestomach: Squamous Cell Papilloma Ovcrall Rates (a) 3 49 (6~i) 3 48 (6c;i) 9 49 (18~i) Adjusted Rates (h) 7.9~(' 9.1c:i . 23.1~i Terminal Rates (c) 3 38 (8r,(') 3 33 (9rD .9 39 (23c;(') Life Table Tests (d) P=0.038 P=0.597 P=0.065 Incidental Tumor Tests (d) P=0.038 P=0.597 P=0.065 Cochran-Armitage Trend Test (d) P=0.034 Fasher Exact Tests P=0.651 P=0.060 20 Benzyl Acetate