(1) Seminar on "Oncogenes and Experimental Carcinogenesis"
This seminar was held on March 12-14, 1986 at the Sheraton Molokai Hotel, Molokai, Hawaii. The organizers were Dr. Masaaki Terada, National Cancer Center Research Institute, Tokyo, Japan; and Dr. Stuart A. Aaronson, National Cancer Institute, Bethesda, Maryland, U.S.A. There were six participants each from Japan and the United States. The purpose of the seminar was to discuss and exchange information on the molecular mechanisms underlying the development of cancer, with the primary goal being to exchange information on the role of oncogenes in carcinogenesis and to prepare future cooperative efforts between scientists of the United States and Japan, including exchanges of patient materials and reagents, most notably cloned oncogene probes.
In opening remarks by Drs. Terada and Aaronson, each stressed the rapid strides being made in developing important insights into the malignant process, as well as the importance of collaborative investigations between scientists of the two countries in most effectively exploiting these new leads.
March 12, 1986- Morning Session (Drs. Terada and Aaronson, Chairmen)
Dr. Angel Pellicer, (New York University Medical Center, New York, New York) discussed in vivo activation of mouse oncogenes by carcinogenic agents. Dr. Pellicer has been studying murine thymic lymphomas induced in three strains of mice by two different agents, the chemical carcinogen, NMU (nitrosomethyulurea), and X-rays. Tumors were induced in virtually all of the test animals in two strains, RF/J and C57B1/6, by either agent, whereas in the 129/J, only the NMU was effective. Analysis of the thymic lymphomas have been completed in two strains. Forty percent of the tumors, regardless of the inducing agent, induced focus formation in NIH 3T3 cells. To date, two oncogenes have been isolated, K- and N-ras. Analysis of the C57B1/6 tumors has not been completed.
In the initial mutations identified after cloning and sequencing representative K- and N-ras, 1 9 out of the 2 1 K-ras activated oncogenes were found to have the same mutation (G—» A transition in the second base of codon 12 changing glycine to aspartic acid) found in the gene originally cloned. Conversely, only the originally cloned N-ras gene had the G—» A transversion at codon 61; another had a G—» A mutation in the second base of codon 12. Four other tumors contained different mutations not as yet identified. A11 of the NMU-induced lymphomas and 12 of the 1 5 Iymphomas induced by!!
!-rays produced tumors in nude mice. Dr. Minako Nagao (National Cancer Center Research Institute, Tokyo) reported on activated c-raf in the transformants induced by a rat hepato-cellular carcinoma and a human colon cancer. A c-raf sequence was isolated from a transformant induced in NIH 3T3 cells by a 2-amin0-3-methylimidazo[4,5-f] quinoline (IQ)-induced rat hepatocellular carcinoma. Two cosmid clones with transforming activity were isolated from a secondary transformant. Comparison of restriction enzyme maps between normal rat DNA cloned into!!
!phage and the cosmid clones of activated c-raf revealed that the latter involved recombination around exons 7-9. Northern blot analyses of transformed NIH 3T3 cells revealed that the region recombined to the 5' portion of c-raf, expressed as poly(A)+RNA, was the same size as the poly(A)+ RNA of c-raf, indicating that the recombination was a necessary event for activation of c-raf. Since this recombination was not detected in the DNA of the original tumor, activation of c-raf likely occurred during the transfection process. Activated c-raf was also found in a transformant induced by DNA of a metastatic human sigmoid colon cancer. It was determined that activation of the human-derived c-raf also occurred during transfection.
Studies of oncogenesis in transgenic mice were described by Dr. Douglas Hanahan (Cold Spring Harbor Laboratory, Cold Spring Harbor, New York.) Oncogenes were transferred into the germ line of mice in two different configurations: targeted oncogene expression using hybrid genes composed of tissue-specific cellular promoters linked to oncogene protein coding information; and "natural" oncogenes in which the normal transcriptional control sequences are left in their usual configuration with respect to the oncogenes. The presence and expression of transferred oncogenes was insufficient to elicit tumor formation with either approach, indicating "complementary" events were required for oncogenesis. Transgenic mice are predisposed to cancer which eliminates one or two of the steps in the multistage process of oncogenesis, allowing other events to be examined in detail. These mice are also useful in the study of oncogene activities through genetic complementation experiments in which oncogenes can be studied individually in separate lines of mice and then combined through mating to examine the effects in the F1 hybrid progeny. These studies will extend the oncogene cooperation assay into cell types other than fibroblasts in the context of the multicelled organism in which tumors actually arise.
March 12, 1986- Afternoon Session (Drs. Shibuya and Papas, Chairmen)
Dr. Kenji Shimizu (Faculty of Science, Kyushu University, Fukuoka) described the structure of the activated c-raf-1 gene from human stomach cancer. Of 57 kbp of the sequence isolated, a region of 39 kbp was found to be the minimum functional unit for the transforming activity; a cosmid clone harboring this region induced foci upon transfection. Tests for homology with known oncogenes revealed that the transforming sequence contained the c-raf-1 gene, a human homologue of v-raf. The latter is known to encode a serine/threonine-specific protein kinase. The transcript of the transforming c-raf-1 gene was estimated to be 2.8 kb. Analyses of cDNA clones of this gene revealed that the gene was generated by substitution of the 5'-sequence (exone 0-4) of the normal c-raf-1 gene with an unrelated human sequence. The substituted cDNA sequence was composed of an open reading frame which encodes an extremely hydrophobic polypeptide. Results suggested that the mechanism(s) of activation involves either altered regulation of expression, such as LTR insertion, or altered structure/function of the c-raf-1 gene product itself. The region where the rearrangement occurred has been identified. In these experiments it appeared that the transforming c-raf-1 gene was generated in the primary stomach cancer.
The molecular cloning of a novel oncogene from a human hepatocellular carcinoma was described by Dr. Kenichi Matsubara (Institute for Molecular and Cellular Biology, Osaka University, Osaka). A new transforming DNA, lca, was obtained from a primary human hepatocellular carcinoma (HCC). This DNA had a linkage to the Alu sequence. Restriction enzyme analyses revealed the minimal size of the lca transforming DNA to be about 10 kbp. Its cleavage profiles were found to be different from any previously reported human or retroviral oncogenes, nor did it cross-hybridize with known oncogenes. An independently obtained transforming DNA from another HCC exhibited identical restriction enzyme cleavage profiles, indicating transforming lca DNA may commonly be present in HCC.
Dr. Carlo M. Croce (Wistar Institute, Philadelphia, Pennyslvania) discussed c-myc deregulation by translocation of the alpha-chain locus of the T cell receptor in T cell leukemias. Two human T cell leukemias carrying a t(8;14)(q24;q11) chromosome translocation were studied for rearrangements and expression of the c-myc oncogene. In one, rearrangement in a region immediately distal (3') to the c-myc gene locus was detected. No rearrangements were detected in the second. Fresh leukemic cells from the second case were fused with BW5147 mouse T cell leukemic cells and the hybrids studied for the presence of the relevant human chromosomes, for the c-myc gene and for the genes for the variable and constant regions of the alpha chain of the T cell receptor. Results indicated that hybrids retaining the 8q+ chromosome contained the c-myc oncogene and the C
!!!locus of the T cell receptor, while hybrids retaining the 14q- chromosome retained the V!!!genes. Thus, in the leukemic cells, the breakpoint in chromosome 14 occurred between the V!!!and the C!!!genes of the involved chromosome 14. The C!!!translocated to a region more than 38 kb 3' to the involved c-myc oncogene. Since human c-myc transcripts were expressed only in hybrids carrying the 8q+ chromosome but not in hybrids containing the normal chromosome 8, it was concluded that the translocation of the C!!! locus 3' to the c-myc oncogene can result in its transcriptional deregulation, contributing to the development of T cell neoplasia. March 13, 1986- Morning Session (Drs. Croce and Shimizu, Chairmen)
Dr. George Vande Woude (Litton Bionetics, Inc., NCI-Frederick Cancer Research Facility, Frederick, Maryland), discussed the met oncogene, a member of the tyrosine kinase family, closely linked to the locus responsible for cystic fibrosis. This oncogene was isolated from a human cell line transformed by n-methyl-n'-nitronitrosoguanidine. Two transcripts, 11 kb (class I) and 12 kb (class II), were found to be expressed in noncoordinate fashion from the met proto-oncogene locus. In contrast, the activated met oncogene locus expressed an aberrant 6.5-kb fused hybrid RNA consisting of 5' sequences derived from the class 11 transcript and 3' sequences co-terminal with the class I transcript. Nucleotide sequence analysis of a 3' portion of met showed considerable sequence homology with the tyrosine kinase family of oncogenes. The kinase locus was mapped to position 7q21-q31 by in situ hybridization. Interstitial deletions were observed in this region in patients with acute nonlymphoblastic leukemia (ANLL). In addition, restriction fragment length polymorphisms present in the met locus were found to be closely linked to the hereditary human disease cystic fibrosis (CF). Studies are now underway to determine whether met is associated with either ANLL or CF and to ascertain its normal function.
Isolation and characterization of human genes related to the v-ros sequence of UR2 sarcoma virus was presented by Dr. Masabumi Shibuya (Institute of Medical Science, University of Tokyo, Tokyo). This group isolated a human gene, designated human c-ros-1, which was homologous to the v-ros sequence of the UR2 avian sarcoma virus. Ten exons (1417 bp) spanning 32 kilobases were detected in the human genome. Nine exone at the 3' region appeared to correspond to all stretches of the v-ros sequence and contained the entire kinase and transmembrane domains. In the deduced amino acid sequence of the kinase domain, human c-ros-1 carried a unique insertion of a short peptide at the v-ros position. These short stretches were not detected in products of any other genes of the src family or in the human insulin receptor. One exon just upstream of the transmembrane domain not acquired by the UR2 sarcoma virus genome was founed by cross hybridization between human and chicken c-ros genomic DNAs. In the predicted amino acid sequence of this exon, a possible N-glycosylation site was observed. Results strongly suggested that the human c-ros-1 gene encodes for a receptor for cell growth or differentiation located on the cell membrane.
During this study, another human DNA fragment which hybridized with the v-ros probe was obtained. This encoded an exon of 120 bases long whose deduced amino acid sequence was highly homologous to the middle portion of the tyrosine kinase domain in the src family. Since the overall amino acid sequence of this gene was distantly related to all members of the src family, it was designated human c-ros-2. Results of the chromosomal mapping of c-ros-1 and c-ros-2 were presented.
Dr. Tadashi Yamamoto (Institute of Medical Science, University of Tokyo, Tokyo), described the molecular structure and oncogenicity of the c-erb8-2/neu gene. The striking similarity of the sequence of v-erb B, an oncogene of avian erythroblastosis virus, to the carboxyl half of the human epidermal growth factor (EGF) receptor, indicates that it is derived from the chicken EGF receptor gene. This suggests that a mechanism of v-erb B oncogenicity is the subversion of transduction of the mitogenic signals generated by the EGF receptor complex. This, in turn, Ieads to the possibility that the EGF receptor plays an important role in the neoplastic process of human cancer, a presumption supported by the observation that the EGF receptor gene has, in fact, been found to be overexpressed in certain human tumors.
Using v-erb B as a hybridization probe, this group identified another v-erb B-related gene, c-erb B-2, which encodes a 138,000 dalton protein similar to, but distinct from, the EGF receptor. By virtue of its structural features, the c-erb B-2 protein is believed to be a novel growth factor receptor. Comparison of the amino acid sequences deduced from cDNA clones of c-erb B-2 and neu, the latter a v-erb B-related oncogene identified by its activation in ethylnitrosourea-induced rat neuro/glioblastomas, revealed them to be the same gene. Antibodies prepared against a peptide with an amino acid sequence identical to that predicted at the carboxyl terminus of the c-erb B-2 protein demonstrated the mature c-erb B-2 gene product to be a phosphorylated 185,000 dalton glycoprotein usually associated with protein tyrosine kinase activity. Although the c-erb B-2 gene was predicted to encode a protein similar to the EGF receptor, EGF did not stimulate kinase activity either in vivo or in vitro. However, EGF as well as TPA enhanced phosphorylation of the c-erb B-2 protein on serine and threonine residues, probably through activation of protein kinase C. Southern blots of DNAs from fresh human malignant tumors revealed amplification of the c-erb B-2 gene in 6 of 66 adenocarcinomas and none in 38 other types of tumors, suggesting that the protein product of the amplified c-erb B-2 gene may play a part in tumor progression of adenocarcinomas. Structural aspects of the c-erb B-2 protein in relation to its functions and EGF receptor were discussed.
March 13, 1986- Afternoon Session (Drs. Vande Woude and Nagao, Chairmen)
Dr. Takis S. Papas (National Cancer Institute, Bethesda, Maryland) described the cellular ets genes: molecular biology and clinical implications in human leukemias. The mammalian homologues of the ets region from the transforming gene of avian erythroblastosis virus, E26, consists of two distinct domains located on different chromosomes. The mammalian homolog of the 5' v-ets domain, ets-1, was mapped to chromosome 11 in man, to chromosome 9 in mouse, and to chromosome D1 in cat. The mammalian homolog of the3' v-ets domain, ets-2, was similarly mapped to human chromosome 21, to mouse chromosome 16, and to feline chromosome C2. To better define the human proto-ets domains, the genomic DNA was molecularly cloned and sequences analyzed. The ets-related sequences of human DNA on chromosome 11 and 21 were found to be discontiguous, unlike that of the chicken and avian E26 virus genome, except for a small overlap region. Thus, it was concluded that the ets sequence shared by the virus, chicken and man is likely to contain at least two functionally dissociable domains, ets-1 and ets-2. The human ets-1 locus is transcriptionally active and encodes a single mRNA of 6.8 kb, while the ets-2 encodes three mRNAs of 4.7, 3.2 and 2.7 kb. By contrast, the chicken homolog, having a contiguous ets-1 and ets-2 sequence, is primarily expressed in normal chicken cells as a single 7.5-kb mRNA.
Human probes with two panels of rodent-human cell hybrids were used to study specific chromosome translocations in acute myeloid leukemias (AML). The human ets-1 gene was found to translocate from chromosome 11 to 4 in t(4;11)(q21;q23) a translocation characteristic of a subtype of leukemia that represents an expansion of a myeloid/lymphoid precursor thought to be congenital in origin. Similarly, the human ets-2 gene was found to translocate from chromosome 21 to 8 in the t(8;21)(q223,q22), a nonrandom alteration commonly associated with acute myeloid leukemia with morphology M2. Both translocations, t(4;11) and t(8;21), are associated with the altered expression of the ets genes, raising the possibility that these genes may play a role in the pathogenesis of certain human leukemias.
Dr. Stuart A. Aaronson (National Cancer Institute, Bethesda, Maryland), discussed characterization of oncogenes of hematopoietic tumors: dbl and ras. A new oncogene isolated from a human B cell lymphoma was found to be unrelated to the ras gene family and 17 other known onc genes. This transforming gene, named dbl, was cloned in biologically active form as a 45-kb human DNA fragment in a cosmid vector. Specific fragments isolated from the cloned DNA were used as probes to identify dbl mRNA as well as to determine the gene structure. At least three exons had been identified and a 2.9-kb mRNA detected in transfectant cell lines. A third-cycle transfectant cell line was injected into NFS mice and sera from tumor-bearing animals examined for reactivity with any proteins induced specifically by the dbl transforming gene. One polypeptide precipitated from all dbl transfectant clones was not detectable in other transformed cells or in untransformed NIH 3T3. Analysis with the same antisera of a spontaneous mutant of dbl indicated that this polypeptide is the protein encoded by the dbl transforming gene.
Members of the ras gene family, frequently activated in human hematopoietic malignancies, were also studied. Important structural and functional features of normal and transforming ras gene products were presented.
A novel transforming gene, hst, from human stomach cancers and a non-tumorous portion of stomach mucosae was described by Dr. Masaaki Terada (National Cancer Center Research Institute, Tokyo). Three of 58 DNA samples from humans stomach cancers and non-cancerous portions of stomach obtained at surgery produced foci upon transfection into NIH 3T3 cells. One was derived from a primary cancer and the second from the noncancerous stomach mucosa of the same patient. The third was from a stomach cancer metastatic lesion of another patient. Primary and secondary transformants contained Alu sequences and were highly tumorigenic in nude mice. Portions of the transforming gene were cloned from one of the secondary transformants obtained by transfection of DNA from the primary stomach cancer. A11 of the transformants induced by transfection of the three different DNA samples were found to contain the same transforming gene. The gene product expressed as mRNA in the transformants was identified and a cDNA clone isolated. When inserted into an expression vector, this cDNA clone efficiently transformed NIH 3T3 cells upon transfection. Southern blot hybridization and sequence analyses of this cDNA revealed it to be distinct from the 24 known oncogenes. This novel gene was termed hst.
Drs. Aaronson and Terada noted in their closing remarks the high quality of the scientific presentations and the excellent progress over the past year. They also noted the usefulness of the informal setting of the meeting in which the relatively small group of scientists were able to exchange information and ideas. There was unanimous accord among the participants that the interactions between United States and Japanese scientists at the meeting promoted better understanding and further collaborations among them.
March 14, 1986- Business Meeting (Drs. Terada and Aaronson)
At the business meeting, Drs. Terada and Aaronson reviewed the arrangements for the meeting and discussed ideas for future seminars. Both agreed that having the meeting facilities within the hotel at which the participants stayed made interactions easier and allowed more time for discussions. Having the meeting site in Hawaii rather than either the mainland United States or Japan was also considered the preferable alternative, in view of distances and travel times for the respective groups. The administrative arrangements for this meeting were, in general, considered to have been well worked out by the sponsoring institutions. Because of the rapid pace of scientific discovery in this critical area of cancer research, as well as the mutual interests of scientists in Japan and the United States, Drs. Terada and Aaronson agreed that it would be worthwhile to organize another seminar for the next year, if possible. The participation of a number of the same scientists from both countries would ensure continuity of the interactions and help to foster further collaborations. There being no further business, the meeting was adjourned at 12:00 noon.
(2) Seminar on "Modifying Factors in Gastrointestinal Carcinogenesis"
This conference was held on March 12-14, 1986 at the International Lecture Hall of the Foundation for Promotion of Cancer Research, Tokyo, Japan. The organizers were Dr. Shigeaki Sato, National Cancer Center Research Institute, Tokyo, Japan; Dr. Michihito Takahashi, National Institute of Hygienic Sciences, Tokyo, Japan; Dr. Gary M. Williams, Naylor Dana Institute, Valhalla, New York, and Dr. Lee Wattenberg, University of Minnesota, Minneapolis, Minnesota. There were seven speakers from Japan and six from the United States, as well as about 20 observers. The purpose of the conference was to exchange recent research findings on etiologic factors in digestive tract cancer and to make an appraisal of the strength of evidence for involvement of suspect factors. The conference consisted of two days of scientific presentations and discussions followed by a half day devoted to assessment of risk factors.
Summaries of the presentations of the participants and the general discussion on assessment of risk factors are presented below in the order in which they were given at the conference.
Esophagus and Stomach
Dr. William Lijinski (Litton Bionetics, Inc., NCI-Frederick Cancer Research Facility, Frederick, Maryland) presented studies on cancer of the esophagus. Edpidemiologic studies have shown that this cancer is common in the Linxian Province of China, Northern Iran, parts of Transkei, and the Calvados region of France. Elsewhere, the incidence of esophageal cancer in cigarette smokers is also relatively high. In experimental studies in animals, the only adequate model of esophageal cancer is induction of that cancer in rats by nitrosamines, especially when administered orally. of 140 N-nitrose compounds tested in rats, 47% induced esophageal tumors; many also induced tumors of the esophagus in mice, but of 41 tested in hamsters, none has induced this tumor. It seems likely that nitrosamines are contributing factors to human esophageal cancer, since they are major carcinogenic constituents of tobacco smoke and nitrosamines have been reported in foods eaten in Linxian, for example. Several of these nitrosamines induce esophageal tumors in rats and they are effective at very low doses; i.e., less than parts per million. Several of them also induce tumors in the oral cavity and tongue of rats. Sources of the nitrosamines associated with food in Linxian, etc., have included corn molds which produced amines and nitrate in water from which nitrite can arise. These combine in the food or in vivo to form nitrosamines. Nitrosamines in alcoholic beverages have been implicated in esophageal cancer in the Transkei and in France. Nutritional deficiencies, especially in riboflavin, other vitamins and in minerals, such as zinc, molybdenum and copper, have been implicated as accessory factors in esophageal cancer. Dietary deficiencies are prevalent in the areas with high esophageal cancer. Vitamin and mineral deficiencies have an influence on the activity of enzymes that activate nitrosamines, and this might increase the susceptibility of people to esophageal cancer induced by nitrosamines, although the existence of promoting factors in the diets is another possibility.
Dr. Shoji Fukushima (Nagoya City University Medical School, Nagoya) discussed modifications by butylated hydroxyanisole (BHA), butylated hydroxytoluene (BHT) and sodium L-ascorbate of carcinogenesis in the esophagus and forestomach. Rats were given the carcinogen N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) intragastrically at 1 50 mg/ kg body weight and from I week later were fed diets containing BHA (0.5%), BHT (1.0%), NaCl (5.0%), BHA (0.5%) plus NaCl (5.0%), BHT (1.O%) plus NaCl (5.0%), or no chemical (control group) for 51 weeks. After initiation with MNNG, administration of BHA significantly increased the incidences of both papillomas and squamous cell carcinomas in the forestomach, while BHT did not. NaCl increased only the incidence of papillomas relative to that in controls. Interestingly, when BHA or BHT was given together with NaCl after MNNG treatment, the incidences of papillomas and carcinomas were higher than those in groups given BHA or NaCl alone. The combination of BHA and NaCl was especially effective, raising the incidences of papillomas and carcinomas to I100% and 70%, respectively. Thus, this experiment showed that BHA and NaCl, but not BHT, have promoting activities on MNNG-induced forestomach carcinogenesis in rats and that when BHA and BHT were given with NaCl, promotion was greater, suggesting that these compounds had synergistic effects.
In another experiment, rats were initiated with 0.05% dibutylnitrosamine (DBN) in the diet for 4 weeks. Oral administration of 2.0% BHA for 32 weeks after DBN treatment significantly promoted induction of forestomach carcinomas, but not the induction of esophageal tumors (papillomas and carcinomas). BHT promoted the induction of esophageal tumors, although it did not promote the induction of forestomach tumors. Sodium L-ascorbate (SA) and ethoxyquin did not affect carcinogenesis of the esophagus or fore-stomach.
In a third experiment, rats were treated intraperitoneally with N-methyl-N-nitrosourea (MNU, 20mg/kg b.w.) twice a week for 4 weeks. Treatment with 2.0% BHA in the diet for 32 weeks after initiation with MNU significantly increased the induction of carcinomas in the forestomach, whereas treatment with BHT or SA did not. In addition, these three antioxidants did not promote esophageal carcinogenesis, although they promoted bladder carcinogenesis. In this experiment, the esophagus may not have been sufficiently initiated by MNU.
These studies show that antioxidants can modify the neoplastic process in two-stage carcinogenesis of the esophagus, forestomach and other organs, when,administered at high concentrations.
Dr. Martin Lipkin (Memorial Sloan-Kettering Cancer Center, New York) discussed the early identification of increased susceptibility to gastrointestinal cancer and approaches to cancer prevention. Attempts are being made to improve the early identification of individuals with increased susceptibility to cancer of the large intestine and other areas of the gastrointestinal tract. Recent findings have revealed modifications of cell proliferation and differentiation in individuals at increased risk for colorectal, stomach and esophageal cancers; indices characterizing affected and at-risk individuals in populations with increased susceptibility to cancer are being developed.
A common hyperproliferative abnormality was found in all preneoplastic gastrointestinal tissues studied: colon, esophagus and stomach. Accompanying this hyperproliferative disorder were other abnormalities of cell differentiation; they are being identified both by immunological and by cloned gene probes.
Thus in stomach, findings again indicated an expanded proliferative compartment and a hyperproliferative state in chronic atrophic gastritis. In the gastric antrum lesions that expressed a gamma fetal antigen also had an expanded proliferative compartment compared to antigen-negative lesions, similar to earlier findings in chronic ulcerative colitis. In esophagus, measurements also revealed an expansion of cell proliferation in high-risk subjects in Linxian, China, compared to lower-risk individuals in Jiaoxian.
Short-term assays based on these findings are now being used to determine the success of dietary intervention in subjects at increased risk for colonic and other gastrointestinal cancers. With this approach, it was recently demonstrated that oral supplemental calcium at only 1.5 times the recommended daily allowance modified the proliferation of colonic epithelial cells in subjects at increased risk for colon cancer, from a profile characteristic of high risk to one of low risk. Since bile acids and free fatty acids stimulate cell proliferation in the colonic mucosa, it was suggested that the calcium effect may stem from amelioration of the action of these lumenal constituents.
Dr. Shigeaki Sato (National Cancer Center Research Institute, Tokyo) presented data on the effects of sodium chloride on gastric carcinogenesis. Correlation between human gastric cancer development and salted food intake has been pointed out epidemiologically Sodium chloride is interpreted as a cocarcinogen and/or a tumor promoter in gastric carcinogenesis. Experiments in rats have shown that the administration of NaCl concomitant with MNNG increases the incidence of tumors in the stomach. Experimental results on the promoting activity of NaCl in gastric carcinogenesis have been inconclusive.
Two series of experiments were performed to elucidate the promoting activity of NaCl in gastric carcinogenesis. In the first experiment, MNNG was given at 50µ/ml in drinking water for 12 weeks to 80 male rats. Then 40 rats were administered 1 ml of saturated solution of NaCl by gavage once a week until the end of experimental week 65. To another 40 rats distilled water was given as above. Twenty rats which had not been administered MNNG were given NaCl for the last 53 weeks. The number of rats with gastric tumors, including adenocarcinomas and adenomas, in the MNNG-NaCl group was 10 out of 35 effective rats (29%), which was not significantly higher than 4 out of 38 effective rats (11%) in the MNNG group. However, the incidence of adenomatous hyperplasia, a preneoplastic lesion, in the former group was 23% and significantly higher than 3% in the latter group. No tumorous finding was obtained in rats given NaCl alone. These observations suggested that NaCl may also work as a tumor promoter.
In a second experiment, male rats were given 100 µg/ml of MNNG in drinking water for the first 12 weeks and then 1 ml of saturated NaCl solution by gavage once a week until the 93rd week. The numbers of rats administered MNNG alone and NaCl alone were 80 and 20, respectively. In the MNNG-NaCl group, the number of rats, with macroscopical tumors at the glandular stomach was 38 out of 66 effective rats (58%), which was higher than that in the group given MNNG (27 out of 78, 35%). No tumor was found in the group given NaCl alone. This is the first experimental result to show conclusively that NaCl is working as a tumor promoter in gastric carcinogenesis.
NaCl destroys the mucous barrier of the gastric mucosal surface and also the epithelial cells. Penetration of carcinogens to target cells in the stomach may become easier and continuous cell destruction and proliferation may work as a tumor promoting status. The presence of such agents in our environment, which enhance the tumor incidence probably through causing nonspecific tissue damage, should not be overlooked. Risk estimation of these chemicals and minimization of their intake are important for the primary cancer prevention.
Dr. Oichiro Kobori (University of Tokyo, Tokyo) reported on the enhancing effects of bile and bile acid on experimental stomach tumorigenesis. There is a dispute whether the frequency of carcinoma in the remnant stomach following surgery for benign lesions is higher than the frequency of stomach carcinoma in the general population. Authors who reported the increased incidence of remnant stomach cancer have shown data indicating that the incidence depends on the type of the operation, with an increased incidence in the Billroth II procedure. These data led them to suggest that refluxing intestinal juices are involved in carcinogenesis in the remnant stomach.
Several experimental studies on the carcinogenic or cocarcinogenic effects of reflux of intestinal juices have produced positive results. It has not been ascertained, however, on which factor in the intestinal juices this carcinogenic or cocarcinogenic effect most strongly depended.
Male rats were given a solution of N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) at a concentration of 80 µg/ml for 7 or 13 weeks in drinking water. After the cessation of MNNG administration for at least 3 weeks, bile or bile acid was introduced into the stomach by choledochogastrostomy or control diet mixed with 0.25% sodium taurocholate. The animals that received MNNG plus bile or sodium taurocholate had a significantly higher incidence of hyperplastic and neoplastic lesions in the stomach mucosa than did the relevant MNNG-treated controls. Neither hyperplastic nor neoplastic lesions were found in control animals that received bile or sodium taurocholate without previous administration of MNNG. The result, therefore, suggested an enhancing effect of bile and sodium taurocholate in stomach tumorigenesis. Dr. Gary M. Williams (Naylor Dana Institute for Disease Prevention, American Health Foundation, Valhalla, New York) discussed studies of risk factors for esophageal and gastric cancer. In the U.S., the rates of cancer of the esophagus in the general population have not changed appreciably in the past 40 years, suggesting that no significant new risk factors for this cancer have been introduced into the environment. For example, the use of organochlorine pesticides has increased considerably with an increase in human exposures. Two of these, chlordane and heptachlor, were studied for enhancement of carcinogenesis in mice when given after diethylnitrosamine. No evidence was found for a promoting effect in the esophagus, although liver tumors were increased.
The major risk factors for esophageal cancer in the U.S. are chemical abuse in the form of cigarette smoking and heavy alcohol consumption. Cancer of the esophagus is induced in experimental animals by many nitrosamines. Therefore, in collaboration with Dr. Stephen Hecht, rat esophagus was studied in organ culture for its ability to biotransform nitrosonornicotine. It was found that the esophageal explants preferentially carried out 2'-hydroxylation which is believed to be an activation reaction for this nitrosamine.
Ethanol has been shown to modulate the carcinogenicity of some nitrosamines, including tobacco-specific nitrosamines, but the effect has not been uniform and, as yet, does not explain the epidemiological findings.
In contrast to all other cancers, gastric cancer has declined sharply in the U.S. during the last 40 years. Again, this is in spite of the introduction of many new chemicals, including food additives. Studies with the food additive, butylated hydroxyanisole, showed an enhancing effect on forestomach cancer in rats at extremely high doses (12,000 ppm, 6,000 ppm), but not at doses in the range of human exposures (60 ppm). Studies by Weisburger and coworkers have shown that nitrite and dietary constituents combine to form a geno-toxic gastric carcinogen(s), probably a reactive diazo compound. Reduced exposure to these elements and increased consumption of micronutrients in fruits and vegetables are suggested to account for the decline in stomach cancer.
Dr. Hiroko Ohgaki (National Cancer Center Research Institute, Tokyo) described studies on genetic control of susceptibility of rats to gastric carcinogenesis by N-methyl-N'-nitro-N-nitrosoguanidine (MNNG). Carcinogenesis by MNNG was studied in susceptible ACI, resistant Buffalo and Fl and F2 offsprings between the two strains. Rats were given 83 µg/ml of MNNG in the drinking water for 32 weeks and were killed at experimental week 72. The incidences of gastric adenocarcinomas were as follows: ACI males (80%), females (47%); Buffalo males (18%),females (0%); F1 males (17%), females (8%); F2 males (36%), females (14%). These results suggested that resistance to MNNG in Buffalo rats was genetically determined and a dominant characteristic.
Histological changes of gastric epithelium of both strains of rats administered MNNG were compared sequentially under the same experimental protocol of MNNG treatment as the genetic study. In ACI rats, erosion, regenerative changes, focal and slightly typical changes, diffuse and severe typical changes were observed sequentially during the period of MNNG administration and adenocarcinomas developed after the cessation of MNNG treatment. The main histological changes induced by MNNG in Buffalo rats were erosion and hyperplasia of the pyloric glands at the margin of erosions. After the cessation of MNNG treatment, areas of hyperplasia of the pyloric glands in Buffalo rats decreased and areas of atrophic pyloric glands increased. The results suggested that responses of the gastric mucosa to MNNG in ACI and Buffalo rats were qualitatively different right from the early stage of gastric carcinogenesis.
Cell kinetics of gastric epithelium in male ACI and Buffalo rats treated with or without 83 µg/ml of MNNG for 5 and 10 weeks were studied by the incorporation of [3H] thymidine. In nontreated ACI rats, average number of cells/pit column, average labeled cells/pit column and labeling index were 22.7, 1.1 and 4.9%, respectively. In nontreated Buffalo rats, they were 22.6, 1.4 and 6.0%, respectively. Thus there was no significant difference in these parameters between the two nontreated rat strains. In treated ACI rats average number of cells/pit column, average labeled cells/pit column and labeling index were 1.7, 4.7 and 2.5 times higher than those in nontreated rats. In treated Buffalo rats they were 1.3, 2.5 and 2.0 times higher than those in nontreated rats. Thus, these experimental results showed a higher increase in generative cells of gastric epithelium in ACI rats than in Buffalo rats by MNNG administration.
Location of gastric epithelial cells methylated by MNNG in the pit column in ACI and Buffalo rats was compared by the histoautoradiography using [3H-methyl] MNNG. Male ACI and Buffalo rats, which were pretreated or nontreated with 83 µg/ml of MNNG for 7 weeks, were given [3H-methyl] MNNG solution (1.9 mCi/0.9 mg/12 ml/kg) in the drinking water and 1.5 h after administration, the stomach was removed and autoradiography was done with paraffin sections. Average numbers of cells with grains in the nucleus in non-treated ACI and Buffalo rats were 1 1 and 7 cells from the luminal surface, respectively. Those in treated ACI and Buffalo rats were 1 6 and 11 cells, respectively. There was a tendency for epithelial cells at the deeper position to become exposed to MNNG after treatment with it in both strains, and the numbers of cells exposed to MNNG were higher in ACI rats than in Buffalo rats. The mechanism of the difference in the susceptibility in the two strains of rats to gastric carcinogenesis by MNNG has not been clarified. However, changes of the position and numbers of proliferating cells in the pit column by MNNG treatment itself seem to be related to the determination of susceptibility to gastric carcinogenesis in rats.
Colon
Dr. Norman D. Nigro (University Health Center, Wayne State University School of Medicine, Detroit, Michigan) (discussed colon carcinogenesis. Epidemiological, laboratory, and animal studies have established dietary factors as major contributors (at least 90%) to its etiology. A major factor that enhances colorectal cancer formation is a diet containing large amounts of fat associated with minimal amounts of fiber. The effect of fat appears to be mediated, at least in part, through excretion of bile acids, which increase ornithine de-carboxylase and DNA synthesis in the colon.
Certain minor dietary components also appear to play an important role in modulating the disease process. Many of the compounds that have been shown to inhibit carcinogenesis are too toxic when given in amounts required to inhibit cancer. A possible solution to this problem is the use of small, nontoxic amounts of several chemicals whose additive effect will be potent enough to reduce cancer risk even when associated with relatively minor alterations in the fat-fiber content of the diet. As an example,!!!
-SIT, selenium and 13-cis-retinoic acid together reduced colon cancer, whereas none did so alone. The effect of the interaction between the various dietary factors is an important area that requires more investigation. Nonetheless, epidemiological and experimental evidences now available suggest that effective dietary guidelines to prevent colorectal cancer can be developed with relatively little additional information.
Dr. Tomoyuki Shirai (Nagoya City University Medical School, Nagoya) discussed modifications of colon carcinogenesis in rats. Four groups of rats were given weekly sub-cutaneous injections of 1,2-dimethylhydrazine (DMH) at a dose of 20 mg/kg body weight for 2, 4, 6 and 8 weeks. The total number of colonic carcinomas increased with increase in the dose of DMH. The number of tubular adenocarcinomas, which were mostly located in the descending colon, was clearly dose-related, whereas the number of signet ring cell carcinomas, which developed in the ascending colon was apparently not dose-related. Thus, the modes of incidence of tubular adenocarcinomas and signet ring cell carcinomas were different.
The effects of pathological states of the liver and kidney on DMH-Colon carcinogenesis were investigated. Liver injury was induced by addition of 0.1% 4,4'-diaminodiphenyl-methane (DDPM) to the diet, and kidney injury by addition of 0.5% N-(3,5-dichlorophenyl) succinimide (NDPS) to the diet, both for 8 weeks. Then, the rats were treated with DMH (12 weekly injections of 10 mg/kg s.c.). Treatment with DDPM or NDPS before DMH resulted in no significant difference in the incidences of tumors, but there were slight differences in the histological patterns of the tumors and their sizes and extents of invasion.
The effects of antioxidants given in the post-initiation phase of colon tumor development were investigated in rats treated with DMH. Animals were given s.c. injections of DMH at a dose of 20 mg/kg once a week for four consecutive weeks, and from one week after the last injection, they were given diet containing 5% sodium L-ascorbate (SA), 0.5% butylated hydroxyanisole (BHA, 0.5% butylated hydroxytoluene (BHT), 0.8% ethoxyquin (EQ) or 1.0% propyl gallate (PG) for 36 weeks. A control group was given basal diet not containing antioxidants. The experiment was terminated 40 weeks after the first injection of DMH. SA significantly increased the incidence of adenomas in the colon and the number of these tumors per rat, especially in the distal colon. EQ and BHT did not affect the number of rats with colon tumors, but in the distal colon EQ significantly increased the number of tumors per rat, whereas BHT decreased their number. BHA and PG caused no apparent modification of tumor development. Thus, of the antioxidants tested SA, EQ and BHT apparently modified tumor development, mainly in the distal colon.
The effects of sex hormones on colon carcinogenesis were investigated using a wide spectrum carcinogen, 3,2'-dimethyl-4-aminobiphenyl (DMAB). Rats were given 20 consecutive s.c. injections of 50 mg/kg of DMAB and then diet containing 300 ppm methyl-testosterone or 2.5 ppm ethinyl estradiol (EE) for 40 weeks. These hormones had no apparent effect on the incidence or number of colon tumors, but EE increased the incidence of liver tumors and mesothelimas.
Dr. Anne Caignard (National Cancer Center Research Institute, Tokyo) gave a brief presentation on oncogene activation in two cell lines established from a rat colon tumor induced by dimethylhydrazine. One of these cell lines (DHDK12/TRb) is resistant to trypsin and forms progressively growing tumors in the syngeneic rats and another cell line (DHDK12/Tsb) is sensitive to trypsin and the tumors once formed regress. DNAs were isolated from these two cell lines and transfected into NIH 3T3 cells. DNAs from both cell lines at 60 µg induced six primary transformants each. Further, five and four secondary transformants were obtained by 60 µg of DNA isolated from the primary transformants induced by DNAs from TRb and Tsb cells, respectively. DNAs isolated from these secondary transformants induced by both TRb and Tsb cell DNAs contained rat repetitive sequence and rat Ki-ras oncogene. No gross structural change in Ki-ras oncogene was observed. Expression of p21, a product of Ki-ras oncogene detected by a monoclonal anti-body RAP-5, was observed in these transformants to the same extent. Thus, a clear difference in the oncogene activation between these tumorigenic and nontumorigenic cell lines was not established, but the importance of this sort of study in the elucidation of molecular mechanisms of tumor growth or metastasis was emphasized.
Dr. Lee W. Wattenberg (Medical School, University of Minnesota, Minneapolis, Minnesota) reported on inhibition of neoplasia by minor dietary constituents. He reviewed the growing body of evidence showing that the diet contains a number of minor dietary constituents that can inhibit the occurrence of cancer. There is a suprising variety of such com-pounds. They include selenium salts, phenols, indoles, aromatic isothiocyanates, methylated flavones, coumarins, terpenes, aliphatic sulfides, dithiolthiones, protease inhibitors, ascorbic acid, tocopherols, retinol, and plant sterols. Not only is the number of inhibitors substantial but they are highly diverse chemically. This diversity, coupled with the widespread occurrence of these compounds in food, make it virtually impossible to consume a diet that does not contain inhibitory substances. One of the implications of the wide variety of chemical structures amongst inhibitors is that others not yet identified almost certainly exist. They should be identified in order to fully evaluate the impact of diet on occurrence of cancer. In addition, more information is needed as to the mechanism of action of many of the inhibitory compounds already identified. One means of classifying inhibitors is on the basis of the stage in the carcinogenic process at which they exert their inhibitory effects. Some prevent formation of carcinogens. Others, termed "blocking agents," prevent carcinogens from reaching or reacting with critical target sites; thus, they exert a barrier function. An example is ellagic acid which binds benzo(a)pyrene diol epoxide. A third group called "suppressing agents" are effective when given subsequent to administration of carcinogens. For example, feeding of cabbage inhibited mammary tumor development in rats when started 7 days after carcinogen. Suppressing agents prevent evolution of the neo-plastic process in cells that would otherwise become malignant. Some compounds are capable of inhibition at more than one time point.
The identification of a compound as an inhibitor of experimental carcinogenesis should be understood to be only a beginning step for evaluating its potential role in modulating human cancer. A great deal of further work is required to assess potency, to examine the range and variety of carcinogenic agents inhibited, to define the conditions under which inhibition occurs and to demonstrate any possible adverse effects. The task of evaluating the role that minor dietary constituents of food play on the occurrence of cancer is clearly a very large one. Yet until this evaluation is accomplished, not only will there be a serious void in the understanding of dietary effects on the occurrence of cancer but opportunities for fully optimizing the role of diet in preventing cancer may also be lost.
Liver and Pancreas
Dr. Gary M. Williams (Naylor Dana Institute for Disease Prevention, American Health Foundation, Valhalla, New York) presented data on modulation of liver carcinogenesis. Experimental liver cancer is enhanced by numerous xenobiotics including drugs, pesticides and food additives. However, there is no evidence linking these chemicals to human liver cancer. Among experimental animals, major species differences have been found in response to liver tumor promoters; rats and mice were responsive to phenobarbital, but not hamsters. Studies on the mechanisms of promotion in the liver suggest that effects on the cell membrane are critical. These findings could indicate that low or intermittent human exposures to experimental promoters should not enhance liver cancer.
Several factors have been shown to inhibit experimental liver carcinogenesis. Studies on the inhibition of liver cancer by the food additives, butylated hydroxyanisole (BHA) and butylated hydroxytolune (BHT), revealed complexities. When BHT was fed together with 2-acetylaminofluorene, it inhibited liver carcinogenesis, but enhanced bladder tumor formation. Feeding of BHT or BHA at the time of administration of aflatoxin B1 was inhibitory to liver carcinogenesis without increasing cancer elsewhere. In fact, 6000 ppm BHT reduced the incidence of liver tumors from 63% after aflatoxin alone to 0. These latter experiments indicate a definite potential for cancer prevention in areas where populations have high aflatoxin exposure.
Dr. Yoichi Konishi (Cancer Center, Nara Medical University, Nara), discussed modulation of hepato-biliary and pancreatic carcinogenesis of nitrosamines in hamsters. N-Nitro-sobis(2-hydroxypropyl)amine (BHP) and N-nitrosobis(oxopropyl)amine (BOP) initiate hepato-biliary and pancreatic carcinogenesis in hamsters. The first experiments were conducted to study the effects of dietary administration of phenobarbital (PB) or the second-ary bile acids, deoxycholic acid (DCA) and lithocholic acid (LCA), on tumorigenesis in the liver, gallbladder, and pancreas were studied in male Syrian golden hamsters after initiation by BHP. BHP (500 mg/kg body weight) was injected subcutaneously once weekly for 5 weeks. The animals were then maintained on a basal diet or a diet containing either 0.05% PB, 0.1% and 0.5% DCA, or 0.5% LCA for 30 weeks. DCA enhanced the development of cholangiocarcinomas without influencing that of hepatocellular lesions. PB promoted the induction of hepatocellular carcinomas but not cholangiocarcinomas. DCA at a dose of 0.5% enhanced the induction of polyps in the gallbladder. These data document that different tumors may be differentially promoted following initiation with a common carcinogen. Secondly, the effects of cholecystectomy and/or LCA on the composition of biliary bile acid and on pancreatic carcinogenesis by BHP were examined in male Syrian golden hamsters. Cholecystectomy was performed one week before initiation. BHP (250 mg/kg body weight) was injected subcutaneously once a week for 5 weeks. A diet containing 0.5% LCA was begun one week after the final BHP injection. All hamsters were sacrificed 36 weeks after cholecystectomy. Only the LCA treatment, but no other treatment, influenced the bile acid composition, i.e., the increase in LCA and decrease in cholic acid. The incidence of pancreatic carcinoma was 23130 (77%) in hamsters receiving cholecystectomy plus BHP followed by LCA diet. The tumor incidence was 5/18 (28%) with BHP followed by basal diet, 10/18 (56%) with cholecystectomy plus BHP followed by basal diet, and 6/18 (33%) with BHP followed by LCA diet, respectively. These results indicate that combined treatments of cholecystectomy and dietary LCA enhanced BHP inducing pancreatic carcinogenesis in hamsters. Thirdly, common bile-duct obstruction by pancreatic head carcinoma induced by BOP in female hamsters with or without cholecystectomy was examined. Cholecystectomy did not influence BOP-induced pancreatic carcinogenesis; however, microadenocarcinomas which are hardly detectable macroscopically and compressed the common bile-duct, are more frequently developed in hamsters that received cholecystectomy. These results might give information concerning obstructive jaundice by pancreatic head carcinoma in humans who received cholecystectomy.
Dr. Michihito Takahashi (National Institute of Hygienic Sciences, Tokyo), reported on modulation of experimental pancreatic carcinogenesis in hamsters. The effects of caffeine, L-methionine, inhibitors of prostaglandin synthesis, and other compounds on the incidence of N-nitrosobis (2-oxopropyl) amine (BOP)-initiated pancreatic tumor development in female Syrian golden hamsters were investigated.
In the first experiment, hamsters were given 3 weekly subcutaneous injections of BOP (10 mg/kg body weight) during the first 3 weeks and then given 0.2% caffeine or one of several other compounds in the drinking water for 37 weeks. Forty weeks after the start of the experiment, the animals were killed and the number, size, and distribution of the pancreatic tumors were examined. Although incidence of pancreatic carcinoma was not affected by the treatment with caffeine, number of the carcinomas per animal was significantly increased (1.00/hamster) compared with the control (0.43/hamster). No promoting effects of the other compounds were found.
In the second experiment, hamsters were given 5 weekly injections of BOP during the first 5 weeks and then given 20 ppm indomethacin in diet, 0.25% phenylbutazone in drinking water, 1% aspirin in drinking water, or 2% L-methionine in diet. Thirty-two weeks after the start of the experiment, the animals were killed and the pancreas was examined. Pancreatic cancer incidence was significantly lower in animals receiving phenylbutazone (37%) or L-methionine (37%) compared with the control group (71%), and number of carcinomas per animal was also significantly decreased by indomethacin, phenylbutazone or L-methionine treatment. Aspirin also showed a tendency to decrease the tumor incidence (53%) and the number of carcinoma per animal, but not significantly.
There are epidemiological reports of an association between coffee consumption and pancreatic cancer, although this is still controversial. In the present experiment, caffeine showed a promoting effect in the hamster pancreas. The inhibitors of prostaglandin synthesis inhibited BOP-induced pancreatic carcinoma development, suggesting that prostaglandins play some important role in the promotion stage of carcinogenesis of this organ. L-Methionine induces liver and pancreas injury in rodents when given at high doses, and this toxicity may relate to the inhibitory effect of L-methionine on the BOP pancreas carcinogenesis. The mechanism of promotion or inhibition of the pancreatic carcinogenesis by the chemicals tested here is still unknown. However, the present observations may have some suggestion for future protective studies in human high-risk groups of pancreatic cancer.
Dr. Daniel S. Longnecker (Dartmouth Medical School, Hanover, New Hampshire), discussed risk factors and modulation of pancreatic cancer. He expressed the view that epidemiologic studies have provided only a few conclusions regarding the causes of pancreatic cancer. Cigarette smoking is the most widely accepted exogenous risk factor, increasing the relative risk by a factor of two in chronic smokers. Other factors that have been implicated as increasing the risk for pancreatic cancer are less well supported. These include occupational exposure to specific chemical carcinogens (!!!
-naphthylamine, benzidine) and consumption of high fat diets. Currently available dietary studies in humans probably do not distinguish between the effect of diets that are high in animal proteins and high in fat. Factors that are controversial in regard to increasing the relative risk for pancreatic cancer and, perhaps, worthy of further evaluation include consumption of coffee and alcoholic beverages. In animal studies, more than 20 chemicals have been shown to induce pancreatic carcinoma in experimental animals. Most of these chemicals are N-nitroso compounds. There are strong genetic differences in the response of experimental animals to pancreatic carcinogens; in particular, hamsters are more sensitive than rats to the induction of pancreatic carcinoma by several derivatives of dipropyl-nitrosamine. Various strains of rats have different sensitivity to the induction of neoplastic lesions in the pancreas by azaserine. Sex hormones influence the response to experimental pancreatic carcinogens in rats. Males develop a high incidence of pancreatic cancers than females in several experimental models, and castration of males has inhibited the development of carcinogen-induced lesions.
Dietary factors strongly modify the response of animals to pancreatic carcinogens. A dietary content of 20% unsaturated fat enhanced pancreatic carcinogenesis in rats and hamsters over that seen with 5% unsaturated fat. Adding soy bean trypsin inhibitor to the diet enhanced pancreatic carcinogenesis in rats. This enhancement is apparently mediated by increased secretion of cholecystokinin in response to the absence of active trypsin in the proximal small intestine. Feeding a high protein diet enhanced pancreatic carcinogenesis in hamsters. Reducing caloric intake inhibited pancreatic carcinogenesis in rats. Adding high levels of retinoids to the diet inhibited pancreatic carcinogenesis in rats.
The effects of alcohol and coffee on pancreatic carcinogenesis in rats and hamsters have been inconsistent.
Further evaluation of the effect of specific hormones and additional dietary factors is needed, and the mechanisms of hormonal and dietary effects should be examined. Finally, little is known of the role of specific viral or cellular oncogenes in pancreatic carcinogenesis.
General Discussion on the Role of Modifying Factors in Human Cancer Development
One of the subjects in this session was the risk assessment of tumor promoters. The presence or absence of a threshold for the action of tumor promoters was discussed. Some experiments have suggested its presence although the threshold may vary between individuals and population groups. The effect of combined administration of multiple tumor promoters was also a subject of discussion. In bladder carcinogenesis, BHA, BHT and ascorbic acid seemed to have additive effects in tumor promotion; and in the forestomach, too, BHA and NaCl seemed to enhance the tumor induction. High fat and trypsin inhibitor worked also as an additive in enhancing the pancreatic tumor induction. However, in the liver carcinogenesis combined administration of polychlorinated biphenyl and DDT was inhibitory. It is generally agreed that it is better not to have large exposures to agents that have been proven to be tumor promoters. However, it is not yet clear to what extent their intake should be lowered. Precise risk assessment of tumor promoters is a subject of further study. In doing so, studies in various organs other than mouse skin need to be expanded.
Dr. Takashi Sugimura (National Cancer Center, Tokyo) emphasized the importance of primary prevention of cancer. Gastrointestinal tract cancers represent about two-thirds of total cancers in the world and control of these cancers is of utmost importance. In this effort, the search for tumor-suppressive agents is necessary. Moreover, based on the fact that most of human cancer development is dependent on life style, Dr. Sugimura introduced "Twelve Points for Cancer Prevention," developed by him and now being distributed to the Japanese public in the form of a pamphlet through the Foundation for Promotion of Cancer Research. These points are: