Tobacco Documents Online

S T A T E\1 E N T 0 F R 0' B*E R T C A S A D H 0 C K E T T I aim Research Diirector of Thie Council for Tobacco- Research -- U.S.A., Inc. I have been employed by the council and i ts predecessor, the Tobacco Iniduistry: Research! C'ommittee, since 1975, 'first as Associate Scientific Director, then A:ting Scientific D'irector, and in my present capacity. My Curriculum Vitaei i s attached~. In, my opinion, the propose& amendment to the Federal Cigarette~ Labeling and! Advertising Act will cause justifiable consternation among many e arnest and able investigators o f can- cer, heart d'isealses and emphysema at the experimental and clini- cal ca1 levels. These scientists labor in the hope of clarifying th e etiology an d pathogeni c step s aind stage s in th e: development of these d'isorders with a view to preventing or delaying, -their clinical appeairance. Contrary to the "findings"' in the bill, however, th,e cause or causes of these diseases have not 'been scientifilcally established, and much work remains to be done., Sta~tistical Ass~ociation~ The c laims a g,ainst s moking a re b ased largely oni epide- miologic studies. Positive statistical associaltion, hokeveT,, -- as epidemiologists are: aware -- does not mean causztion. .According,ly, these studie s do no t suppo rt th e'°find'ings" i 1 the . O ~ V. W ~ . ~ ~D'

proposed amendment. Students of logic have 1ong recogni:zed that a n:ega,tive statistical association between an event and a putative cause may be conclusive. Thus when it was suggzsted that some hair dyes maight produce cancers in women, a, study of 5,,000 beauti- cians who had been e xp osed to such d'yes for twenty years s howed that these persons had not experienced any excess of cancer., This negative correlation appears quite conclusive withl respect to dyes that ha:ve long been in use:., Similarly a concern, regard- ing the SV-40 v,irus,, which causes bladder cancer in monkeys and -was found to be present in some polio vaccines, was considered resolv:ed, wh~en no, h igher cancer rate was found in the vaccinated humans thani in the unvaccinated ones. On the other hand,, a, positive statistical association - between an exposure and a disease condition is well known to be only evidence of some sort of a relationship, the nature a of which remains to be discovered. For example„ there is a statis- tical association of severe overweight in women with high~~er uterine- and ovarian cancer rates and in men, of highier colon-rectum and prostatic cancer. In, the same category is the positiv:e correlation between early marriage (or sexual activity). of women and eventual cancer of the cervix., Such positive associations only indicate a need for other kinds of investiga- tion in order to elucidate the:nature of the relationship.

Still more s triking i s the r eported epi3eTio logicali finding that men and womlen who averalge seven hours of sleep per night have 1 ower d!eath r ates f rom c oronary h eart d i~~sease than those getting either more or less sleep. Those getting lesss than f ive h ours h ad v ery h igh d eathir ates ainde those s leeping t en hour s o r mor e pe r n ight h~a d hi'ghe r tha n averag e rates .. I h ave pr eviousl y p o inted o ut thlalt m erely changing the hours in b.ed of the short-term or long-term sleepers to seven per night, w~ould be unlikelytochangethispActurera~dically.,' More likely the amount of' "useful sleep" that an individual cann achieve i s a reflection of a psychophysiological diath,esis that may be the basis of the difference and which may or may not be alterabSe . Studde s shiow~ that many persons are able to achieve "'usefull and~ effective sleep" in a relatively short period.. Others may have difficulty in getting to sleep or sleeping rest- fully on account of psychologicall tensions or emotional prob- .. Lems. Whether such persons can be treated to reduce such, ten- sions is a question many investigators are attempting. to attack.' Surveys have: shown that.many smokers testify to the feeling, that smoking cigarettes helps to arouse them when they are drowsy but to relax them when they are tense. Confirmation cf these contradictory impressions by objective pharmalcologicall method!s h as b een &iff icult, b ut s tudie s o f, e1ectro-encephalo- grams have clailmed correlations between certain types of brain waves and psychological characteristics of the subject. Some. a3ss3s01

CC sub jects c an l earn t o pr oduce various t ypes o f b raiin waves (electro-encephal'ograms) at will. Others have difficuilty in prodiucing the types reputed'~ to reflect inner repose. Smoking has been claimed' to promote this kind of tranq,uiility but thie evidence so fa!r is widely regarded as "soft". The discovery of' centrally active small peptides (endorphlins, encephailins and' related substances) may eventually provide an answer to, these q,uestions. Lung Cancer Diagnosis In 1912, a book by Adler called attention to the occurrence of primary carcinoma in the lung and apparently createdi the impression that this disease was a new development whos e caus e mus t b e sought. By co ihc idenc e thi s boo k appeare d near the t ime when d evelopment o f the b lended c igarette in the United! States was stimulating a vast expansion of cigarette use in this country, which wals further stimulated by the social cond'itions brought about by World War I. Thus, a basis~ was laid for the claim promulgated a few years later that an '''epidemic" of lung carcinoma had been engendered by cigarette smoking.. Meanwhile, h owever,, the 1 ate Dr. Milton B. Rosenbla tt had'made a very intensive and extensi~ve study of lung, cancer as described in the medical literature, mostl'v European, of the nineteenth century where the art and science of pathology was d'eveloped and practiced r elatively e arly. Th e appl'icati on o r . 03613502 -4

C post-mortem e xaminations f or c aluse o f' death was made q uite routinely ini some central European hospitals long before it became prevalent in other aireas. Carcinomal of the lung was recogniized,, post mortem, in a great many cases where it had been missed altogethier in clincial pre-mortem diagnoses. In fact, it was missed clinically as often als 90-95% of 'the time. In hospitals where post mortem h istological diagnosis w~as practiced reg,ularly, carcinoma of the lung constituted a proportion of total male cancers s imila'r •to or greater than that reported, today. Yet in that era cigairettes were practically unknown. In retrospect, it is not strange that lung cancer was missed in clinical diagnosis. Tuberculosis~ was common as a sourceoftbleedi ngf'rom t hel ung: and &eathf rom, pneumoniag,enerally occurred in the lung carcinomia patient so that cancer was not s uspected. As betterr clinical d!iagnostic miethods came into use,. such as rad'iography, bronchoscopy, exfoliative cytology, exploratory thoracotomy and' others, the gap between clinical diagnoses and post-mortemi discovery of lung, cancer was closed gradually over several decades. The introduction olf antibioticss eventually played a role also since a pneumonia might be r es olved,, e xpos ing a c arc i'noma t o d i scovery. In the United States, this European work was apparently little known. The dogma, was that all lung, cancer miust be metastatic so that the -primary lesion must be found i f the 03613503 -s-

origin, of: the neoplasm was to be knoun. Koreover, post mortem examination was mu!ch less frequently made than in Durope. A1 though h istological e xamination o f' cancer tissue by an experienced paithologist still provides, thie most reliable diag,nosis available, the developnent of clinical methods that are quicker and easier apparently discouraged the application of' painstaking post-mortem s tudy in this country, and hias promoted th e genera 1 us e o f les s reliabl e cliniea l methods . The lung is a frequent target for metastases from other organs. Dr . Shield s Wa rre n reporte d tha t abou t one-thi rd' o f al l cancers arising, in other regions of the blody, eventually mietaistasize to the lung,, often at a early stage. Dr. Rosenblatt thought it very unlikely that such metastatic cancer present in the lung but not originated there was "Icaused"' by tobaccoo smoke.ii'efelti tlog,icaI to: exclud'emetasttat ic c'ancerin thee lung, from any statistical study of correlaition with smoking,. But the dlistinction bletween primary and metastatic cancers in, the lung is not always easy and a distinction is no longer require'd on death certificates or attempted in statistical records. While the distinction appears to be very i~mportant for the s tudy o f caus ative f'actors and etiolo~gy the necessary data are not easily avalilable. Nevertheless in a careful r eview o f c ontemporary d'eath certificates and hospital records in a few individual hospitals, Rosenblatt f'ound considerable clinical over-diagniosis of lung 03613504

trie3 also to show why dogmatic positions are inappropr.iate and restrictive in the present state of knovIedge.

C a material to which human smo:cers are exposed. Many chemical stud'ies have shown that there are quallitative,, as well as quantitative differences bie.tween 1alboratory condensate aind fresh smoke to which humans are exposed. Nevertheless, researchers blegan painting such tobacco- smoke cond!ensate on the backs of mice,, generally dissolved in some solvent. After persistent trealtment over a long period with, enormous doises, neoplasms did appear on the skins of thiese mice. Th is method o f' te sting was s eized upon, by numerous o1ther investigaltors with many variations of technique,, dilfferent species and strains of' an, imals and varying results. S'ome of these experiments were not reproducible. Chemists fractionated the tars and by the proliferating techniques of' chromatography iidentifiedliterally thousands of chemical components of cigarette smoke condensate with a high degree of reliability. Probably no other complex mixture in the human environment has e ver b een s o t horoughly a nalyzed. I was critical of this development from the start. "'typicai"' human smoker draws a puff of air through his cigarette. This ge erates al cloud of smoke that enters his mouth within a fraction of a second following f'ormaition„ remains in h is o ral c avity f or ainother f r action o f a s econd and then i s. drawn into the lunig dilvted with five to ten volumes of air. During this short interval, rapid physical and chenical changes are taking, place. The tiny liquid droplets tha t 03613506

C constitute the b1ue cloud are growing in size by coalescence, which influences deposition in the lung. Numerous compounds in the aerosol aire polymerizing, interacting, combining, breaking down,, and otherwise changing,. If such a mixture enters a cold trap, the most readily condensed substances wil be deposited as- a thick liquid but other phases of the smoke escape into the air. Th is s i tuation c omplicates thie d esign o f d~evices f or - exposing animals to inhalation of smoke comparable in physicall and chenical properties to that inhaled by human smokers. The best we can do is to design devices that will produce the smoke mechanically under conditio s approximating those attained' by humans an d ge t i t t o th e animal " s lung a t a comparabl e ag,e and in monitoired dosages. These considerations led me to characterize the skin- paintin g o f mic e wi th cigarett e smok e condensat e a s applying thie "wrong material in the wrong, f'orm, in the wrong dosage,, to thie wrong tissue o f thie wrong animal." I stiT 1 hol d thie same view. To the best of my knowledge, to date no one hias pro- duced the type of lung cancer that is associated wit'a, human smoking by exposing experimental animals to inhalation of fresh, whole smoke. The considerrations, experiences, trials, tests and findings of twenty-eight years that have been described in extrem e brevi ty wi ll , I hope , mak e i t clea r why Thie Counci l i s now emphasizing studjr of the "'constitutional!l'! diseases as such on a basic level with the hielp olf new researchi tools. I have 103613507'

carcinoma in patients who were known to be cigarette smokers. Simila r findings have been made by Feinste in. Ani,nal Ext)eriments When the f irst 1 arge-s cale epidemiological s tudies of the association between cigarette sanaking and lung carcinoma were described publicly in the mid-f ifties, a, number of investigators undertook to expose many different speciI es of animals, i ncluding m ice,, r aits, hamsters a nd h ens, t o c igarette smoke: inhalation. N'.one of these early inhalation studies produced lung carcinomas. The 1958 Annual Report of the British Empire Cancer Campaign referred to these experiments as adding up to a "striking negative r esult." Thie first wave of smoke inhallation studies .was so unproductive of results that o1ther methods were sought. It was recalled that the black tar ac um ulate d as a by-product from destructive distillation of coal h ad b een shown to produce skin cancers when painted on mice andi rabbits. Accordingly, researchers b ega~n applying to ba=o smoke condensate on the: backs of mice. Condensate was obtained! by puffing cigarettes mechanically and p assing the smoke into a cold trap where it was condensed into a dairk-colored viiscous liquid. Though this is not"'tar'''asd'!efined in the~d4ictionalry and does not bear any close resemblance to coal tar, the terml persists. It must be pointed out that, contrary to popular bielief , tobacco smoke: condensat e-- or „ta,r"' -- is not actually: 03613505 7