Document XRpO71NML1b9X9m7Yp6zgk9eg

TO: Dave Penny - Austin h - /c Interoffice Communication FROM: DATE: SUB J: T. G. Grumbles October 18, 1989 LAB SKIN PAINTING ISSUE VISTA Welcome to the jungle? For starters we have a "blossoming" issue to deal with regarding recent skin painting studies done with LAB. CLER is looking into the issue and discussing potential studies to do in response to recent results. We are discussing joint activities with ECOSOL. Hopefully, Roheim has told you we have a conference call on Friday, October 27, 1:00 - 2:00 p.m, with Monsanto, to discuss the issue and prepare for a meeting with ECOSOL in Houston on November 2. I need your participation in both. I've attached information on the issue. Please call me to discuss. T. G. Grumbles dlj Attachment cc: 0. C. Kerfoot, J. Roheim-Austin VVV 000014254 Monsanto FROM (name-location-phone) James P. Mieure (G4WR) (4--4837) DATE: REF: TO: October 9. 1989 cc: Iversen's First Skin Painting Publication C.B. T.G. R.D. T.M. E.C. Beckmann - G4WA Grumbles - vista Hogue - 04B King - 04C Robinson - A3ND (Houston) I just received a copy of Iversen's first publication of his skin painting studies on Alkylbenzenes. A copy is enclosed. There are a few minor revisions from previous drafts. Note the comment in the second paragraph on p616 that "small mis-dosings of animals is of no significance for the overall results"! The allusion to possible PAH impurities in alkylbenzene (p616) is troubling. J.P. Mieure VVV 00001>255 Shell Internationale Petroleum Maatschappij B.V. Dr J.P. Mieure Monsanto Chemical Company 800 N. Lindbergh Boulevard St. Louis, Missouri 63167 U.S.A. Your ref.: Our ref.: HSE/223/jjv 26th September 1989 *{070) 77 26 37 Dear Jim, Re: Genotoxicity Studies on Linear alkyl benzenes (LAB) Further to our recent telephone conversation, 1 enclose a copy of Iversen's paper on dodecylbenzene (Br. J. Ind. Med. 46, 608-616, 1989). I believe that it would be mutually beneficial to explore the possibility of harmonising publication of the Shell and Monsanto genotoxicity studies. Various options exist (e.g. integration of data and publication of one joint manuscript, back to back publication of the separate papers in the same journal issue), but I think we must first each assess the amount of work and time involved, before making any decision. To assist in this exercise I agree on the suggestion of a reciprocal exchange of draft manuscripts. I will today explore the possibility of our sending you a copy of the Shell manuscript (ref SBP.50,88). This will be handled through our London headquarters (via Bill Fernley) and, although I anticipate no major problems with clearance, this may take a couple of weeks. 1 understand that you will endeavour to organise despatch of a copy of the Monsanto draft, subject to the same potential clearance limitations. As I mentioned, it is our intention to meet with Iversen within the next few months to establish the status of his second manuscript. Ideally this meeting should precede any meeting of toxicologists/technical managers representing ECOSOL/CLER and the LAB consumer companies. 1 understand that such a meeting has not yet been finalised. WV 000014256 May I thank you for the opportunity to talk with your colleague. Dr. Ellen Robinson, and 1 look forward Co meeting you both in the near future. With best wishes. Yours sincerely, Shell Internationale Petroleum Maatschappij B.V. Health, Safety and Environment Division K.R. Huckle cc. G.W. Femley CTMAP/42 Shell Centre London attn.: 1 VVV 000014257 l British Journal of Industrial Medicine 1989:46:608-616 i i r J*- C Studies of the carcinogenesis and tumorigenesis of skin applications of dodecylbenzene on hairless mice O H IVERSEN From the Institute of Pathology, University of Oslo, and Department of Laboratory Animals, The National Hospital (Rikshospitalet l, N-0027 Oslo l, Norway abstract Dodecylbenzene in various concentrations, dissolved in acetone to a final volume of 50 was applied twice a week for 78 weeks to the back skin of hairless mice, with or without pretreatment with 51-2 p% 9,10-dimethyl* 1,2-benzanthracene (DMBA). A negative control group was painted with acetone only and a positive control group was given a single application of 51-2 fig DMBA. A few tumours developed but there was no significant skin tumorigenicity--that is, occurrence of benign or malignant tumours--by acetone alone, or by 16% dodecylbenzene. More tumours developed in the group treated with 80% dodecylbenzene than in the group treated with acetone alone or with 16% dodecylbenzene, but there was no significant difference between treatments with 16% and 80% dodecylbenzene. There was only a suggestive increase in tumorigenesis in the group given 51-2 pg DMBA and thereafter painted with 40% dodecylbenzene twice a week compared with the group given 51-2 pg DM BA once. As regards histologically malignant tumours-- that is, carcinogenicity--the group treated twice a week with 16% dodecylbenzene alone developed two skin malignancies, whereas only one carcinoma was observed in the group treated with 80% dodecylbenzene; both results are non-significant. There was only a suggestive increase in the occurrence of skin malignancies in the group treated with DMBA followed by 40% dodecylbenzene compared with that treated with DMBA alone. As regards other tumours, treatment with 80% dodecylbenzene led to six lymphomas in 56 animals, whereas the acetone control group had two lymphomas in 56 animals, a difference which is only suggestive. DMBA alone and DMBA followed by dodecylbenzene gave five and four lymphomas, respectively. There was no significant difference between controls and dodecylbenzene painted animals for lung adenomas or other tumours. A pronounced epidermal hyperplasia, an increase in melanin pigment ("blue spots"), pigment leakage, and skin ulcerations were seen, mainly after 80% dodecylbenzene and alter DMBA followed by 40% dodecylbenzene. There was an obvious increase in amyloidosis after continual treatment with 80% dodecylbenzene. The results indicate that 80% dodecylbenzene alone is weakly tumorigenic but not carcinogenic in the skin of hairless mice, and it tends slightly to enhance DMBA initiated tumorigenesis and carcinogenesis. Dodecylbenzene may be a weak inducer of malignant lymphomas. It is a fairly strong skin irritant and may increase amyloidosis. Dodecylbenzene is an alkylbenzene manufactured from benzene and alkenes. Alkylbenzenes are used as raw material for sulphonate detergents, as plastisisers for polyvinyl chloride floor covering, and as dielectric fluids. It is an important component of low viscosity cable fluids and is widely used as insulating cable oil.1 Since it is used as an insulating fluid in cables in the oil industry in Norway, and since some workers may Accepted 18 October 1988 occasionally be heavily exposed, it was considered of interest to determine the possible skin carcinogenicity of a particular dodecylbenzene produced by Essochem. A complete toxicological investigation was not intended. The study was originally directed only at determin ing the frequency and type of skin tumours. Amyloidosis, signs of skin irritation, and swelling of lymph nodes, however, often became so obvious that we began to record them. Other tumours and lesions in lungs, liver, spleen, and kidneys were occasionally found and registered. Studies ofthe carcinogenesis and tumorigenesis ofskin applications ofdodecylbenzene on hairless mice 609 Materials and methods Male and female mice of the hrjhr Oslo strain, obtained from Gamk Bomholt Gaard, Aarhus. Den mark. were used. Spontaneous skin tumours have not been observed in these animals (personal experience). They are known to react normally to the classic two stage treatment protocol for production of skin tumours/ The animals were 60-90 days old at the beginning of the experiments and weighed 20-25 g. All the mice were housed in plastic cages, eight in each box, using dust free, sterilised wood shavings as bedding. The room had a constant temperature (25*C), 50% humidity, and a 12/12 h light/darkness rhythm. The mice were fed a standard diet (pellets from Feileskjapet A/S) with free access to water. The cages were cleaned at noon once a week. All the experimental animals, including the negative controls, were kept in the same room. The animals were individually identified by a system of ear and tail markings when they developed a lesion. CHEMICALS The acetone used was of reagent grade. The dodecylbenzene was a 00-05 (predominantly CI2) bran ched chain alkyl benzene, and provided by Exxon Chem, Belgium, identification No MRDE-7. This product is a complex mixture of many different isomers of 00-05 branched alkylbenzene and it is not possible to provide either purity or formulas. No impurities of toxicological significance are expected to be in the product--for instance, benzene will not be present in view of boiling range of 270-305*C. It is a clear liquid and its boiling range was 270-305*C, specific gravity 0-869 (20/20`C), viscosity llcSt at 20*C. and vapour pressure 0-001 KPa at 20'C. The 7,12>dimethylbenz(a)anthracene (DMBA) was from Sigma Chemical Company, St Louis, Mo, USA, and was used as purchased. TUMOUR INDUCTION EXPERIMENTS Topical skin applications were made with a graded pipette on the skin of the middle of the back in a final volume of 50 ji) of either reagent grade acetone, acctone/dodecylbenzene, or DMBA in acetone. Applications were done twice a week for the whole experimental period (80 weeks). Under such circum stances, possible small dose variations at each applica tion are obviously negligible and ofno importance for the overall results. Separate pipette tips were used for acetone, dodecylbenzene, and DMBA. A negative control group (group 1) of 56 mice, 28 males/28 females, was exposed to 50 p\ reagent grade acetone twice a week. An experimental group (2) of 56 mice was given 16%, and a third group (3) of 56 mice received 80% dodecylbenzene in acetone twice a week. These groups each contained 56 mice because few tumours or none at all were expected to occur. A positive control group (4) of48 mice. 24 males/24 females, was given a single application of 51-2 pg DMBA in 50 p\ acetone and thereafter no treatment. All the animals exposed to a single application of 51-2 ^g DMBA at this institute during the past three yean (128 mice) represented a fifth, historical control group (5) which was only used in the statistical analysis. A sixth group (6) of48 mice received 51-2 pg DMBA in acetone and after one week was painted with 40% dodecylbenzene in acetone twice a week. OBSERVATION OF SKIN PAPILLOMAS, SKIN CANCERS. OTHER TUMOURS. AND CERTAIN OTHER LESIONS The animals were examined once a week for 78 weeks. A drawing of each animal was made and each tumour or lesion charted on the drawing. Each tumour was registered as a tumour when it measured more than I mm1, and was present for more than two observa tions. As usual, some animals died during the experiment and others were killed when an obvious malignant skin tumour had appeared. The remaining animals were killed after 78-80 weeks. Whenever possible--that is, except when precluded by extensive autolysis--a necropsy was performed and sections were taken from all the skin lesions, from the lungs, the kidneys, the spleen, the liver, and from enlarged lymph nodes. All tumours registered as malignant were thus histologically verified. Cellular atypia and infiltration below the musculus panniculus was used as a criterion of malignancy for skin tumours. The histological investigations were performed by the author. The term skin tumorigenesis covers ail the skin tumours (papillomas, keratoachantomas, carcinomas, and possible sarcomas). The development of malig nant skin tumours (carcinomas and sarcomas) accord ing to histological criteria is called skin carcinogenesis. Spindle cell squamous cell carcinomas are known to occur1 and are difficult to distinguish from dermal sarcomas with routine histology. Hence, these are grouped together. Most, if not all of them, are probably spindle cell carcinomas. STATISTICAL EVALUATION On the basis of the crude incidence of skin tumours, elaborate statistical calculations were made according to the recommendations of the IARC.4 The results are illustrated as tumour rates (the percentage of tumour bearing animals in relation to the number of animals alive at the appearance of the first tumour related to time)--and tumour yields (the cumulative occurrence of all skin tumours related to time) in all groups. To evaluate significant differences in tumour rates, we used the method for non-incidental tumours first VVV 00001*^259 l i i :h a iA u i 610 Table 1 Percentage survival tactual number of surviving mice in parentheses) Iversen Trtaunent 1 Acetone alone 2 16% Dodecylbenzene in acetone 3 80% Dodecylbenzene in acetone 4 51 2 pg DMBA once 6 51-2 ug DMBA once followed by 40*/* dodecylbenzene Week 0 20 25 JO )J 40 45 50 SS 60 65 70 7J SO too 98 98 96 94 92 92 92 90 90 90 90 90 90 (56) (55) 155) (54) (53) (52) (52) (52) (50) (50) (50) (50) (50) (50) too 98 98 98 98 96 95 95 95 95 90 90 90 90 (56) (55) 155) (55) (55) (54) (53) (53) (S3) (53) (50) (50) (50) (50) 100 100 98 98 98 96 94 94 94 94 92 92 90 84 (56) (56) (55) (55) (55) (54) (53) (53) (53) (53) (32) (52) (50) (47) too 100 97 97 97 97 94 94 94 88 85 83 79 73 (48) (48) (47) (47) (47) (47) (47) (45) (4$) (42) (41) (40) (38) (35) too 98 98 96 % 94 94 90 88 83 79 79 79 79 (48) (47) 147) (46) (46) (45) (45) (43) (42) (40) (38) (38) (38) (38) described by Petos and elaborated by a computer based test program by Peto et at.* This program takes into account varying mortality rates between the experimental groups and assesses both the number of tumour bearing animals and the time to the first tumour in each animal. To evaluate the difference in tumour yields we have used the method of Gail ei al based on multiple times to tumour (method }).* This method assesses the number of tumours appearing, the varying mortality rates between (he groups, and the time of appearance of each tumour. We used the cht-squared test to evaluate the final occurrence of lymphomas and skin cancers. Statistical significance has been defined as follows: 010 > p > 0 05 * suggestive. 0 05 > p > 0 01 = significant, 0-01 > p > 0 00 = very significant. Pronounced biological variations are known to occur in biological skin painting experiments.'* Hence, statistical significance is not always the same as biological significance and some small, statistically non-significant differences may still be real. Results SKIN TUMOURS The survival in actual numbers and as percentages is presented in table 1, the crude and adjusted (for groups 4 and 6) end results in table and tumour rates and yields in figs 1 and 2. Statistical evaluations are shown in tables 3 and 4. To make the results comparable the end results and the curves for tumour yield, for groups 4 and 6 (fig 2). have been adjusted to show the total number of tumours occurring in a group starting with 56 mice. Hence, the values observed for the two groups com prising only 48 mice have been multiplied by 56/ 48 * 1-1667. The values shown in table 2 and fig 2 thus represent the number of tumours that actually appeared in the groups of mice, and those that would have appeared in a group of 56 mice with the same tumour yield as that actually observed in the 48 mice. The final occurrence of skin tumours in table 2 (illustrated with respect to time in figs I and 2) shows that the two groups initially painted with 51-2/ig Time (weeks) Fig 1 Tumour rate t tumour bearing animals as a percentage of those alive at appearance offirst tumour related to time l during observation periodfor each offive experimental groups. Final number ofmalignant skin tumours is given in parentheses at end ofeach curve. For various treatment schedules see text, materials and methods. VVV 00001*260 ..'.v.-ist/fy; - Studies of the carcinogenesis and tumongenesis ofskin applications ofdodecylbenzene on hairless mice Table 2 Tumorigenicity of dodecylbenzene: tumour analysis andpathologicalfindings 611 Group No So of mice m group Treatment / 56 Acetone 2 56 Dodecylbenzene 16% J 56 8054 4 48 (56) DMBA once 6 48(561 DMBA + 40% Dodecylbenzene Tvpe of tumour Skin tumours: All skin tumours Carcinomas Sarcomas Skin malignancies Lymphoid tumours: B-cell lymphoma Lymphoma NOS Sum ivmphomas: Reticuloses Sum tvmphoid lesions: Lung adenomas: Per cent of ail necropsied Other tumours: Abdominal sarcoma Angiosarcoma oi liver Hepatoma Amvloidosis: Per cent of alt necropsied Skin toxicity: Pronounced hvperpiasia Ulcerations Piemen: disturbances No of tumours (lesions)No of animals with tumours (lesionsl 3,2* 0 0 0 4/4 LI I/I 2/2 13/9 53/23(62/271 I/I 5/5 (6/61 0 0(0) I/I 5/5(6/61 00 2>'2 2'2 2/2 2/2 3'3 4:4 S/S 6-6 4/3 S'6 10 20 0 1/1 (I'll 6.6 3/3 (4'4) 6/6 4/4(5.51 1 1 2/2 (12) 7-7 6/6(7-71 6.6 14 10(16 12) II 27 11 11 0 0 00 1J 0 00 11 0 2 9-9 22 22 12.12(14 14 5 23 39 23 0 + + 0 0 0 4- 0 0 + ++ 0 68/22 (79'26) 5/5(6/61 6/6 (7/7) 11/11(13/131 0(0) 3/3(4/4) 3/3(4/4) 2/2(2/21 5;5(6'6) 6-4(7/51 13 11 (I'l) 11 (ID 0 5 5 <6 6| 11 f + +++ Skin tumours. other tumours and tome lesions found in hairless mice painted with acetone. DM BA. and dodecylbenzene as indicated. When numbers are given as fractions the numerator indicates number of tumours or lesions observed in groups of 48 or 56 mice (see te*i). and the denominator indicates number of affected animals. The figures in brackets for groups 4 and 6 are adjustments to 56 mice, for comparison. Croup 5 is a historical group used only for statistics and hence not depicted in the table. `Three very small, entirely benign papillomas. NOS " No other specification. For more detailed explanation, see text. DMBA had the highest tumour rates and yields, whereas the three groups with dodecylbenzene in acetone or acetone alone had a low occurrence of tumours. The four upper rows of results in table 1 show the numbers of the various lumours per number of animals with tumours. Table 3 shows the statistical assessments of skin tumorigenicity. in this table we have not only taken into consideration the specific, positive control group used in this experiment (group 4) but also compared it with another large group comprising 128 mice which had been painted once with 51-2 jig DMBA at this institute during the past three years (group 5). There were no statistical differences in tumour rates or yields between these two latter experiments (groups 4 and 5, respectively). The t value for positive trend for tumour bearing animals in this comparison was 1-92. with a one tailed p value of0-32. and the chi squared statistics for heterogeneity gave a p = 0-65. For tumour yield the chi squared was 2-64 (0-20 > p > 010). Hence, the larger group may also be used for statistical comparison with the experimental results ofthis study. There was no significant difference between group 1, painted with acetone alone, and group 2, painted with 16% dodecylbenzene twice a week, but there was a significant difference between group I (acetone) and group 3. painted with 80% dodecylbenzene. the latter resulting in more skin tumours. There was no sig nificant difference between groups 2 and 3 painted with 16 and 80% dodecylbenzene. respectively. There was naturally a very significant difference between the acetone painted negative (group 1) and the positive (groups 4 and S) control groups painted once with 51-2 jig DMBA. but there was only a suggestive or nonsignificant difference between group 4, painted with 51-2 jig DMBA alone, and group 6, painted with DM BA followed by twice weekly applications of40% dodecylbenzene. There was. however, a suggestive (for tumour rates) or significant (for tumour yields) difference when group 6. painted with DMBA + 40% dodecylbenzene, was compared with the large group 5 of historical controls. Hence, there is a slight indica tion that painting with 40% dodecylbenzene twice a week for 78 weeks may enhance DMBA induced tumongenesis. As regards skin carcinogenesis--that is, the induc tion of malignant tumours ofthe skin (table 4)--there was no significant difference between group 1, painted WV 00001*261 612 Tversett ? F' m x y r im f' nx. ` 1 - Fig 2 Tumour yield (total number oftumours appearing adjusted to groups of56 mice related to time i during observation periodforfive experimental groups. Forfurther explanation, see text tofig l. with acetone alone, and group 2. painted twice a week with 16% dodecylbenzene. Only one skin malignancy developed in group 3, painted twice a week with 80% dodecylbenzene, which does not differ significantly from either group 1 or group 2. There was a significant difference between the acetone painted group l on the one hand, and group 4, painted with DM BA alone, and between group 1 and group 6, painted with DMBA + 40% dodecylbenzene. The difference be tween the groups painted only with DMBA (group 4) and the group given DMBA followed by 40% dodecylbenzene (group 6) was only suggestive. Hence, the results show that continual treatment of hr/hr mice twice a week with 80% dodecylbenzene is a weak skin tumorigen. but not a carcinogen, for hairless mouse skin, whereas 16% dodecylbenzene alone given twice a week has no significant tumorigenic or carcinogenic potency. DMBA alone given once is confirmed as a tumorigen and a carcino gen. When 40% dodecylbenzene is given twice a week Table 3 Sraruricdf analysis ofskin tumours Tumour rates rumour yields 1 vJ 1 v4 t vS 6 94 6wS 1 *3 / *4 1 vS 6 v4 6 v5 ObsiExp ObsiExp ObsiExp ObsiExp ObsiExp Best odds Best adds Best odds Best odds Best odds 1 Acetone alone 0-26 2 16% Dodecylbenzene in acetone 3 80% Dodecylbenzene in acetone 194 4 SI-2 jig DMBA once (ibis study) 5 SI-2 jig DMBA once (128 mice) 6 SI-2 jig DMBA once followed by 40% Dodecylbenzene 2-9t One tailed p value for positive trend 00130 Chi-squared for heterogeneity 496 Degree of freedom 1 p Value for heterogeneity 00260 Best p value for tumour yield Condusion s 0-08 0-06 239 11-64 <0-0001 29-26 1 <0-0001 1 49 15-70 <0-0001 23-97 1 <0-0001 0-89 112 2-43 0 2274 056 0-4547 vs vs ns 0-89 1 37 6-18 0-0406 3-04 1 0-0811 sugg 100 8-69 (week 57) 0-03 1-00 004 too 100 1-00 1 56 1-55 (week 38) (week 39) 5 45 1 0-025 > p>(H)IO s 28-82 1 <0-0005 vs 25-22 1 <0-0005 VS 3-34 1 OI0> p005 sugg 4-42 1 OOS> p> 0-025 $ i -- Significant vs * very significant ns - not significant sugg -- suggestive. Fordetailed explanation of statistics, see refs 12,13.14. All other comparisons were not significant. VVV 000014262 Studies of the carcinogenesis and tumorigenesis ofskin applications ofdodecylbenzene on hairless mice 613 Table 4 Chi-squared assessment of lymphomas and malignant skin tumours (one tailedp/ Groups tested Lsmphomas Malignant skin tumours 1 r2 1 v3 1 -4 1 16 4 vf, f ~ 2-04. 1 DF.0-20 > p > 0 10 not significant r * '88. 1 DF.O-IO > p > 005 suggestive r - 1-43. f - 6T8. 1 DF. 0-15 > p > 0-10 1 DF.0 010 > p > 0 005 non-significant very significant r - -08. T1 - 13-33, 1 DF.0IS > p > 010 1 DF.p > 0-005 non-significam very significant r " 213. 1 DF.O 10 > p > 0-05 suggestive All other differences were non-sigmficani. for 18 months after a single dose of 51-2 pg DMBA. it shows only a small, suggestive, but not significant, enhancing activity. LYMPHOID TUMOURS The occurrence of lymphoid tumours is shown in table 2 and the statistical evaluation in table 4. When there was a diffusely proliferating population of lym phocytes without cellular atypia in lymph nodes, and hence not an obvious malignant lymphoma, the lesion was classified as a reticulosis. We followed Wogan's advice on criteria for the diagnosis of malignant lymphoma or leukaemia in mice.' The collection of plasma cells in the spleen or lymph nodes was registered only as reactive changes and not as plas macytomas or mature B cell lymphomas. In many of the obvious cases of lymphoma we observed massive infiltrations in the lungs, liver, kidneys, and once even in the heart muscle (fig 3), showing lymphoma involvement of many organs. It may be seen from tables 2 and 4 that treatment with 80% dodecylbenzene resulted in six lymphomas in a group of 56 mice, whereas the acetone group showed only two lym phomas in 56 animals. This difference, however, is only suggestive with the chi squared test. DM BA alone and DM BA followed by dodecylbenzene gave five and four lymphomas, respectively. If reticuloses were also countixl there seemed to be no significant difference in the incidence of lymphoproliferative lesions in any of the treatment schedules. Hence, dodecylbenzene alone may be a weak inducer of malignant lymphomas and DMBA gave no extra effect. LUNG ADENOMAS There was no significant increase in lung adenomas in dodecylbenzene painted animals versus the acetone treated controls. A single application of DM BA alone significantly increased the number of lung adenomas compared with the acetone control group (table 4). A statistically significant further increase, however, was not seen in the treatment group receiving DMBA followed by 40% dodecylbenzene (table 2). Hence, dodecylbenzene did not induce lung adenomas. OTHER TUMOURS A few abdominal sarcomas, one angiosarcoma of the liver, and a hepatoma occurred as seen in table 2. There was no statistical indication that these were related to dodecylbenzene or DMBA. TOXIC MANIFESTATIONS As regards the degenerative lesion amyloidosis, there was an obvious increase in amyloidosis after treatment twice a week with 80% dodecylbenzene. This was observed in the spleen, in the kidneys, and in the liver. The explanation for the increase in amyloidosis is not clear, but if it is taken as a sign of general toxicity, painting with 80% dodecylbenzene twice a week seems to be fairly toxic for the animals. Why this co-variation disappeared when the mice were also treated with DMBA is unknown. In the skin we recorded ulcerations, pronounced hyperplasia, excess, unevenly distributed pigmenta tion (pigment leakage and some "'blue spots'), and various degrees of chronic dermatitis with many mast cells. The most pronounced hyperplasia and even ulcerations were seen after treatment with 80% dodecylbenzene, and after DMBA followed by 40% dodecylbenzene. Hence, the results show that dodecyl benzene displays strong skin toxicity for hairless mice (fig 4) in addition to its weak tumorigenicity. Discission A review of the physicochemical and toxicological aspects of dodecylbenzenes has been published by Rdnneberg.1 His review (containing detailed referen ces) documents that dodecylbenzene may be absorbed through the skin, the lungs, and the gastrointestinal tract. Significant amounts of dodecylbenzene are still not metabolised after 24 hours. Absorption, distribution, biotransformation, and elimination have not been studied quantitatively. Dodecylbenzene has low acute toxicity by the oral route. LD* for rodents has been reported to be 2000-3000 mg per 100 g weight.10" Inhalation toxicity appears to be higher.' Dodecylbenzene is a skin irritant, and even a single application may lead to an inflammatory reaction and transient hyperplasia and hyperkeratosis in mouse skin. Repealed surface applications may induce severe skis irritation and even ukxration in rodents,11; Other organs are also affected, such as the respiratory tract, lungs, kidneys, blood and haemopoietic tissue, and the i Iversen \ Fig 3 Photomicrograph ofmassive infiltration ofheart muscle by malignant lymphoma ceils in a mouse belonging to group 6. (initial magnification 320 * .) central nervous system.1 Chronic dodecylbenzcne exposure has been implicated as a cause of thymus atrophy, possibly leading to an impaired immuno logical defence.11 There is no evidence, however, of allergic reactions of skin or respiratory tract.* The evidence on putative genotoxicity is inadequate.14 The possible genotoxicity of three linear alkylbenzenes have been tested with the Ames test, in vivo bone marrow chromosome studies, and the Chinese ham ster ovary cell test. All test results were completely negative (personal communication J S Harding, Mon santo Europe, Brussels). When applied to the skin some is absorbed through this organ and some perorally since the animals frequently lick themselves and one another. Some skin tumours have been reported to develop in mouse skin painting assays, but none of the earlier studies has included negative solvent treated con trols.1 1116 It has been discussed whether the tumorigen- icity is due to contamination with polycyclic aromatic hydrocarbons.1214 There are also a few indications about possible increase in the incidence of lymphomas in alkylbenzene treated animals.1' In studies of tumour production in mice after skin painting there are always large biological variations in the results between similar experiments performed with the same strain of mice at different times, and sometimes even between the various cages in the same experiment, as mentioned above.71 Some of the differences presented here are formally significant and some are only suggestive. A cautionary note has to be struck since there is still a possibility that significant differences may be due to "an unlucky hit" based on large, but random biological variations. It is always difficult to assess the significance of results from animal experiments for human tumorigenesis and carcinogenesis. Generally, man is more resistant than mouse to chemical carcinogens i Studies of the carcinogenesis and tumorigenesis ofskin applications ofdodecylbenzene on hairless mice 615 Fig 4 Photograph ofa mouse belonging to group 6 with moderate pigment changes, scarring, and many tumours. and the latency time is much longer. The present paper shows that dodecylbenzene in the doses used may be weakly tumorigenic, but not carcinogenic, for hairless mice skin, but it only weakly enhances DMBA initiated tumorigenesis and carcinogenesis. Dodecylbenzene is without doubt a skin irritant and seems to increase amyloidosis. It is debatable whether the hyperproliferation and the eczematous inflammatory changes induced by dodecylbenzene may in themselves enhance tumorigenesis, or act synergistically with other carcin ogens in the environment. For a long time it was thought that a tissue in rapid proliferation is par ticularly sensitive to carcinogens, but the situation is obviously more complicated than that. When DNA synthesis and mitosis are inhibited, the tumour yield may increase. "" Possibly this refers only to the transformational stage that causes DNA injury. There is a possibility that critical alterations primarily induced in DNA may be secondarily fixed by a wave of rapid proliferation, and hence one cannot completely exdude that a strong essentially non-carcinogenic hyperpiasiogen may act synergistically with small doses of a carcinogen. Cell injury and cell death, however, may also reduce tumour production by killing transformed cells that could have started a tumour. Tumorigenesis is always a balance between malignant transformation and cellular toxicity. Acquired cellular resistance also occurs.* We found three completely benign small papillomas in two animals in the acetone treated group. This is probably a random observation, since it has not been observed before in our laboratory. The negative control animals, however, were kept in the same room as the DMBA and dodecylbenzene ones. A small skin contamination with the oil in the form of a faint oil mist cannot be completely excluded. Such small papillomas, however, may also be induced by a virus. Some authors have suggested that tumorigenidty due to dodecylbenzene may be due to contamination with polycyclic aromatic hydrocarbons (PAH).1215'* We did not test the possible PAH content in the dodecylbenzene we used. One author has previously reported an increase in the incidence of lymphomas in alkylbeiuene treated animals.17 The small increase in lymphomas seen in the present study may be an effect of dodqrylbenzene. It is impossible, however, on the basis of the available information to know whether this is a direct effect of 616 dodecylbenzene or whether it is due to the activation of a latent, slow virus in this strain of mice. No virological studies of these mice have been performed. They are bred under specific pathogen free conditions but are known to develop reticuloses in old age.21 The few malignant tumours found in internal organs are probably a random occurrence, without any relation to the treatment with dodecylbenzene or DMBA. With twice weekly paintings with SO pi ofa carefully prepared concentration of dodecylbenzene. any poss ible small mis-dosing of animals is of no significance for the overall results. In another study a dodecylbenzene called alkyibenzene C was tested in our laboratory (O H Iversen, in preparation). The two results are not easy to compare, since the doses in the two studies were not similar. Nevertheless, painting with 16% dodecylbenzene in the present study provoked four skin tumours in four tumour bearing animals in a group starting with 56 animals. If we adjust the results of the study to a group size of 56 animals there would have occurred four tumours m four animals after painting with a 20% solution. The results for lymphomas, lung adenomas, and the signs of skin toxicity were also similar. When the two stage protocol was used, however, the present study indicates a weak dodecylbenzene induced enhancement of DMBA induced (umorigenests (a slight additive effect), whereas the other study resulted in a slight reduction in DMBA induced tumorigenesis. Probably the two dodecylbenzenes were somewhat different, possibly due to varying contamination with PAH, as mentioned above. The dodecylbenzene alkylbenzene C may have been more toxic to the skin and hence killed more DM BA transformed cells. Two students Marit Berg and Inge Engeland carried out the skin paintings, the observations and recordings of tumours and lesions, and the necropsies on the animals with great skill and application. Nigel J Sarginson from Exxon Chemical International Inc helped in planning and control. I thank the technical staff at the institute for making ail the histological sections and the secretariat for typing and retyping the manuscript. The study was supported by Exxon Chemical International Inc, Belgium, without any restrictions on the research programme or the publica tion of the results. References I Romeberg A. Health effect* of insolation fluids. Report II. Toxicological properties of alkylbensenes. Monograph. Oslo: fversen Standard Tddbn og KabeUabrik A/S. 1985. 2 Ivenen OH. Hairtoa mouie skin ia two-stife chemical carano- genesm Virchows Arch (Ceil Pathol) 198238:263-72. 3 Turusov VS. ed. Pathology of tumours m laboratory oaimnit. Vol II: Twnourt of the mouse. Lyon: International Agency for Research on Canes. 1979. 4 Peto R, Pike MC. Day NE, tt al. Auks: Guidelines for simple, sensitive significance tests for carcinogenic effects in long-term animal experiments. In: Montesaoo R. Bemet H. Tomatu L. eds. Long-term atdshort-term screening assaysfor carcinogens: a criticalappraisal. Lyon: International Agency for Research on Cancer, monogtapha, Suppl 2, 1980:311-426. 5 Peto R. Guidelines on the analyst oftumour rates and death rate* in experimental animats. Sr J Cancer 1974:29:101-5. 6 Gail MH. Santner TJ. Brown CC. Ad analysis of comparative carcinogenesis experiments based on multiple times to tumor. Biometrics 19803*255-66. 7 FsreG. The influence of number ofmice in a box on experimental skin tumour production. Br J Cancer 1965:19:871-7. 8 Ivenen OH. Ivenen U. Is there a diurnal variation in the susceptibility of mouse skin to the tumorigenic tcuon of methytcholanthrene? Acta Pathol Microbiol Seand 5er A 1976:84:406-14. 9 Wogan GN. Tumors of the mouse hematopoietic system: their diagnosis and interpretation in safety evaluation tests. Report of a study group. CRC Critical Reviews in Toxicology 1984:13:161-81. 10 Hopper SH. Hulpieu HR. Cole VV. Some toxicological properties of surface-active agents. Journal of the American Pharmaceutical Association 1949:38:428-32. 11 Draehev SM. Bylinkina AA. Garshenin VF. The character and rigtuficance of surface-active agents in water. Shorn Vysoke Skoiv Chem Tech Prase Tech Vodv 1965:1:161-6. 12 Horton AW, Denman DT.Trossct RP.Carcinogenesisoftheskin. II. The accelerating properties of aliphatic and related hydrocarbons. Cancer Res 1957;17:758-66. 13 Honor AW, Van Drcal PA. Bingham EL. Physicochemical mechanisms of acceleration of skin carcinogenesis. In: Montagna W, Dobson RL, eds. Advances in biology of skin. Carcinogenesis 1966;7:165-82. 14 Lankas GR, Bacter CS. Christian RT. Effect of alkane tumor- promoting agents on chemically induced mutagenesa in cultured V 79 Chinese hamster cells. J Toxicol Environ Health 1978;4:37-41. 15 Horton AW,BinghamEL.BurtonMJG.TyeR.Carrinogencmof the skin. III. The contribution ofelemental sulfurand oforganic sulfur compounds. Cancer Res 1965;28:1759-63. 16 Saftotii U, Shubik P. Opdyke DL. Carcinogenesis tests on alkyibenzenes and slkylbenzene sulfonates. Toxicol Appl Pharmacol 1962;4:763-9. 17 Tanaka T. Kano-Tanaka K.. The effect of cocarcinogen on transplacental carcinogenesis. In: Seven L. ed. Tumours in early life in mi md mortals. Proceedings of the 6tk Quadrennial Iniernatumal Conference on Cancer. Perugia. Perugia: Perugia University Medical School, 1978:963-71. 18 Ivenen OH. Hydroxyurea enhances methylnitrosoura skin tumorigenesis when given shortly before, but not after, the carcinogen. Carcinogenesis 1982;3:891-4 19 fversen OH. Enhaacnent ofmethylnitrosouiea skin carcinogenesis by inhibiting cell proliferation with hydroxyurea or skin extracts. Carcinogenesis 19823:881-9. 20 Iversen OH. Astnip EG. The paradigm of two-stage caretoo- getMsia: a critical attitude. Cancer Invest 19843:51-60. 21 Laeruzn OD. Reticulum cell neoplasms in normal and benzene treated mice. Acta Pathol Microbiol Seand Ser A 1973*1:57-60. VVV 00001426O