Tobacco Documents Online

. ._._ ,...,,~.s.v..u.o.ws.s.ia. `.,s...i . 46 JOHN 11. FACCI-NI Doet.w .n M*d.cinA Fouow of Roy.l Col" oF Pahobqisb Eap.rt PN.rm.ooloquo-Toaicoiopu. w.o. 19.05.80 .e oe.Q7.e5) Pror"Nw of Pathology. Unrv.rNty of Su.r.y F'r-i~~-:r Otficer for E,T.S ferhni,-al Information Staff ilffice of Health and Environment (RD-6891 1.15 Env ironrnent Frotect ion Agency 4o[ , M Street SW Wa=.hir,qtori DC 20460 27.09.40. H P H [j ?.~r Sir, I have read with interest the preliminary draft document (EPA/600/6-90i006A) entitled "Health Effect=. of F'assive Smoking: assessment of lung cancer in adult~5 and re_•piratory disorders in chi idren" and I wish to submit th.e comment_. below which are related to passive smoking and th~~_ a-=._.es-.ment of l ung cancer. I am a histopathologist with a long experience of surgical pathology in a London teaching ho5pit.91 and the problemss of establishing a definitiue diagnosis fron-, biopsy material. I now have a special interest in experin,entAl carcinogenesis, working as part-time Professor at the F:oben-s [n-.titute of Industrial and Environmental Health and Safety at the University of Surrey. I am struck by the apparent lack of concern at the absence of an adequate histopathological confirmation of the diagnosis of primary lung cancer in many of the cases cited in the report. The Environmental Protection Agency would surely not accept a rodent carcinogen bioassay on an industrial chemical as valid unless all the animal-, had been examined histologically to determine the tumour type and tissue or origin. H F n In clinical practice, the confirmation of a primary carcinoma of the lung - when the tijmoiir in question is an adenocarcinoma on biopsy or cytology - cari only be made with confidence if the lung is subsequently re-5ected or a postmortem examination is performed. Biopsy or cytological diagnosis at best can only be used as a diagnosis to treat a patient with malignant neoplasia in the lung but it cannot differentiate with certainty between a primary arid a metastatic carcinoma. There are many examples in the literature where subsequent more rigorous histopathological examination of tumours diagnosed as primary pulmonary neoplasia haue proved to be metastases (Garfinkel et al., 1985. Auerbach et al., 1984, Trinidad et al., 1963). Most of the cases in these studies are in women and more than 50% were diagnosed to have adenocarcinoma. Furthermore, the commonest malignant tumours.

JOHN M. FACCIl~TI Doct.ur M M6d*otrM f.llow of Roy.l Co/" of P.tholoq/.tb Exp.rt Ph.nn.aobqu.-Toxloowqw w.o. 19.05.30 at 06.03.65) Profs..w of Pathology. UMwraty of Saney in non-smoking women are of the breast, G.I. tract, uterus and ouary: all of these are adenocarcinomaa and have the potential to metastasise to the lungs. This means that there is a defninite risk that an important percentage of the cases in some'of the'=.tudies lacking rigorou,5 hi,~.topathological control cited in thi=s report are in fact metastatic and not primary carcinomas of the lung. I would argue, therefore, that it i=s questionable to combine the results of all these studies to make a meta-analysis. A metastatic adenocarcinoma of the breast, for example, is a different disease process with a different aetiology from that of primary carcinoma of the lunij. I do not consider that one can compare, for instance, the results of the Trichopolous study in which only 18% of the pat ients. had h i t.topatho 1 og i ca l conf i rmat i on of the d i agnos i a, 24.6% cytological confirmation and the remainder diagnosed on clinical grounds alone, with a well-controlled study --~uch a•=, that of Garfinkel et al.. The latter included an experienced histopathologist as one of the authors who reviewed all of the histopathological slides of the cases concerned. He, incidentally, found the initial diagnosis of primary lung cancer to be incorrect in 12.7% of the cases, despite the fact that all these patients had a histopathological biopsy diagnosis - more conclusive than cytology alone. They were consequently riot assessed. In this respect, it is of interest that in the discussion cited in the report (Section 3.7) on the conflicting findings between the cohort studies of Garfinkel and Hiryama, mention is not made of an important difference betueen the studies, namely that Garfinkel took the trouble to verify the initial diagnosis for all patients (34/203 non- smoking women in the ACS study with a diagnosis of lung cancer on the death certificate were reported to have cancers of other sites in the final medical report) whereas only 10% of Hiryama's 200 cases were confirmed histopathologically. r,i

• JOHN 11. FACCINI ~ Ooct.ur .n Milid.oins Fsllow of Roy.l Coll.qs of PstlMaloq/sts ExPsrt Ph.rmacoloqurToxlcoropu. (J.O. 19.05.90 st 06.03.85) prohs.ur of Pathology. Univ.rsfty of Surr.y t I aril also of the opinion that; i nsuff i c ient account has 5o far been taken of the l arge c l us,ter of ca=.ets of adenocarcinoma of the lung among Chinese uomen arld the need to irlueltigate them further, especial ly since the local physician=. studying the-se cases consider that enuironmental factors such as. diet and cookirlg methods are important in the aetiology with risk ratios equiualent to or greater than those reported on epidemiological studie-1 for ETS. I enclose a short review I have pub 1 i shed on the sub ject uh i ch I_•ubrili t for your con:. i derat i ori. Your=. sincerely REFERENCES 1. Auerbach, 0., Garfinkel, L., Parkes, V.R. et al. (1984). Histologic type of lung cancer and asbestos exposure. Cancer, 54, 3017-3021. 2. Garfinkel, L., Auerbach, 0. and Joubert, L. (1985). Involuntary smoking and lung cancer. A case control study. JNCI. 75, 463-469. 3. Trinidad, S., Lisa, J.R. and Rosenblatt, M.B. (1963). Bronchogenic carcinoma simulated by metastatic tumors. Cancer, 16, 1521-1529. m ~ ~ C11 C~1 O W N f