The first cancer mortality survey in the 1970s in China［1］ showed that the very high mortality rate ［35.74 per 10（5） person-year］ of gastric cancer in Qixia County, Shandong Province, China was 12 times more than that of very-low mortality rate ［3.88 per 10（5） person-year］ in Changshan County in the same province. There was only 300 km between the two counties,［2］ and the characteristics of geography, weather, foods, and lifestyle were similar. The mainly dietary difference between the two counties' residents was consumption of garlic: the latter county is famous for growing garlic plants in China. On average, a resident living in Changshan took more than 6 kg of raw garlic each year, compared with only 0.5 kg for a resident living in Qixia. We also found the garlic growing in Changshan contains more selenium than in Qixia from an analysis of the compounds of garlics. Two case-control studies in the two counties also showed that taking more raw garlic had a lower risk associated with gastric cancer.［3］ You et al［4］ reported that consumption of raw garlic correlated with protective effect for stomach cancer after matching resident's life style in high-incidence area of Liqu and in low-incidence area of Changshan. Also he found the prevalence of Helicobacter pylori (H. pylori) infection was higher in Liqu than in Changshan. Results from our previous laboratory studies showed that garlic extracts and synthetic allitridum can inhibit growth of gastric cancer cells, inhibit the synthetic nitroamine in gastric fluid which may be a risk factor of cancer, and inhibit growth of H. pylori in vitro. In the present study, an intervention trial using large-dose allitridum and microdose of selenium to prevent stomach cancer was carried out in 7 communities of Qixia County over Nov ember 1989-December 1991 and all subjects were followed up until 2001.
An epidemiological experiment was selected for this study. Participants in the study were limited to age 34-74 years old, and who had a history of at least one of the followings: (1) stomach disorder, or (2) tumours in other family members, or (3) smoking and/or alcohol consumption. The sample size was determined by the arcsin transfer formula with the following parameters: significance level α=0.05, Zα =1.64, and power β=0.1, Zβ=1.28, P (incidence of gastric cancer)=0.02, and the protective ratio was 60%. The estimated minimum sample size for analysis was 16,600 person-years in each group.
All subjects were selected from 288 natural villages of 7 communities, for which the total population was 220,000, and randomly divided into an intervention group and control group. A total of 2526 people were enrolled in the intervention group and 2507 people in the control group. No significant differences were found in distribution of age, gender, and other risk factors between the two groups, except heavy smoking in men and heavy alcohol consumption in woman between the intervention group and control group ( Table 1 ). Doctors working at the village-clinic hygiene room, community hospitals, and the Gastric Cancer Institute of Qixia delivered the medicine to all subjects. Village doctors managed their village's subjects, and doctors of each community hospital monitored their whole community's subjects, and doctors of the Gastric Cancer Institute of Qixia conducted the whole research including the delivery of medicine, follow-up, registering newly diagnosed cancer cases and all deaths. The death certificates, issued by Qixia County Hospital, stated the underlying cause of death according to the International Classification of Diseases (ICD-9).
In the intervention group, each subject took orally two soft intestinal dissolving capsules that contains 200 mg synthetic allitridum every day and one table of sodium seletine which contains 100 μg selenium every other day for one month of each year during Nov.1989-Dec.1991, and at the same time each subject in control group was given two placebo capsules which contains corn oil and one starch table of identical appearance to that in intervention group. (Note: synthetic allitridum was made from in Dongfeng Pharmaceutical Factory, Lianyuguang, China, and the soft intestinal dissolving capsule of allitridum turned out in the Weihai Branch Factory of Shangdong Xinhua Pharmaceutical Factory, Zibo, China. The tables of sodium seletine came from the Office of Shangdong Endemic Diseases, and were manufactured by Hezhe Pharmaceutical Factory, Hezhe, China). The date of delivering the drugs was in November 1989, November 1990, and November 1991, respectively. The whole process of the interventional study was conducted by the double-blind method. All subjects were followed up until 2001.
The village doctors interviewed all subjects living in their village and recorded all side effects which occurred when the subjects took the medicine, and sent the information to the community hospital. The doctors of the Institute of Qixia Gastric Cancer and the members of the research group randomly sampled 1% of subjects to interview face to face.
Cumulative risk, relative risk, and 95 percent confidence limits (95%CL) were calculated by logistic regression model using SPSS 10.0.
The major side effect of the drugs was garlic-odour. A small number of subjects had heartburn in their stomachs. Forty to 60 days after the date of each drug delivery, we sampled at random 212 persons and interviewed them face to face. Of those, 180 persons complied, 22 took 15-25 days of dosage, 6 took 6-10 days of dosage and 3 took 1-4 days of dosage. Thiamine concentration in urine was measured for 110 persons in the intervention group and 108 persons in the control group from the random subsampling. The average values of thiamine were (356±80) μg and (320±50) μg for the two groups respectively. Mean compliance estimated for the taking of the medication was 93%.
Analysis of morbidity data
During January 1990-December 2000, there were a total of 63 cases (42 men and 21 women) of malignant tumours diagnosed in the intervention group, and 79 cases (65 men and 14 women) in the control group. The diagnosed grade for all tumours is shown in Table 2. 60.9% and 60.0% of gastric cancer cases were diagnosed by histopathology for the two groups. Each patient's slide was read by two histopathologists, and each patient's film by two radiologists, independently. The morbidity rate of all malignant tumours was 236.85/10（5） person-year in the intervention group, and 295.27/10（5） person-year in the control group. For gastric cancer the morbidity rates were 82.71/10（5） person-year (intervention) and 112.13/10（5） person-year (control).In the first five-year follow-up after intervention (1992-1997), there were 35 and 51 cases of malignant tumours found in the intervention group and control group, respectively. The morbidity rates of all malignant tumours were 186.53/10（5） and 271.49/10（5） person-year, respectively. The morbidity rates of all cancers in the intervention group declined by 22% compared to the control group. For gastric cancer there were 10 and 19
cases and the incidences were 53.29/10（5） and 101.14/10（5） person-year in the two groups respectively: the rate of the intervention group declined by 47.3% compared with the control group.
Analysis of accumulative morbidity risk
The cumulative morbidity risk was a little higher in the control group than in the intervention group, and the biggest difference between the two groups was 0.49% at year 1997, as shown in Table 3 and Fig..
We used a binary logistic procedure to analyse the relationship between cancers detected during the period of 1992-1997, intervention drug and affective factors. The dependent variables were different tumours, and the independent variables were drug and affective factors including age, smoking, alcohol consumption, history of familial cancer ( Table 4 ). There were two significant factors ( intervention drug and age) in all four models.
After adjustment for age, gender, family history of cancer, smoking, alcohol consumption, and medical history of stomach illness, the RR for malignant tumours was 0.67 (95%CL: 0.43-1.03), and 0.48 (95%CL: 0.21-1.06) for gastric cancer during 1992-1997. After adjusted for age , history of cancer family cancer, smoking, alcohol conumption, and medical history of stomach illness for male subgroups, the RRs and 95%CL of malignant tumours and gastric cancer were 0.51 (0.30-0.85) and 0.36 (0.14-0.92) respectively. The parameters are shown in Table 5.
In this study we controlled some potential confounders and kept them in balance between the two groups. Although in the intervention group the number of male heavy-smokers and female heavy drinkers were a little higher than in the control group (which may cause more cancers in the intervention group), when analysing we adjusted for these factors by using logistic regression models. Because a double-blind method was used for the study and long-term follow up, the results of the research are reliable.
Garlic components exacted in water or oil have been used in clinical medicine as antibiotics for a long time. Allitridum, which is the major component inhibiting tumours, accounts for nearly 45% of garlic oil. This trial was the first use of large-dose allitridum to prevent stomach cancer in high-risk persons. The results showed that the large-dose allitridum can be accepted by complying subjects and has no harmful side effects. In a previous acute toxicity assay mice,［5］ the LD50 of allitridum was 228 mg/kg, and 95% confident limit (190-280) mg/kg. In the animal subacute toxicity experiment with doses of 1/8 and 1/4 of LD50 for three months, no toxicity damage was found.
The results of this study showed that five years after terminating the intervention, large-dose allitridum and selenium had some effect in gastric cancer prevention, which was consistent with the result of our previous research. In the past years many papers have shown that Allitridum or garlic extract can kill tumour cells;［6-8］ inhibit tumour growth by 50%, suppress synthetic nitroamine as a pre-carcinogen, inhibit H. pylori growth in stomach fluid,［9-11］ repair DNA synthesis and have antimutagenic effect,［12］ increase immunity,［13-17］ have antioxidant effect,［18］ and induce apoptosis of stomach cancer cells.［19］ Selenium can increase the anti-tumour effect of anti-cancer drugs.［20,21］
Our study also indicated that the large-dose allitridum offers protection from gastric cancer for males but not for females. The reason may be related to (1) the size of study (ie, observing person-years) for female was much smaller than the designed size required to show an effect; (2) the incidence of gastric cancer of male was higher than that of female, or the frequency of the risk factors in male was higher than that in female. Our experiment during the early 1980s showed that 24 μg/ml of allitridum killed 90% of MGC803 or SGC cells in vitro. Such a dose may be equivalent to 240 mg allitridum for a 60 kg human, corresponding to (100-240) g raw garlic for one person each day. In the present study the dose of allitridum may be equivalent to eating 100-200 g of raw garlic each day for an adult.
Based on the result of cumulative risk analysis we can see that the preventive effect for stomach cancer is not related to the etiological prevention mechanism, ie, inhibition of the synthetic nitroamine and the growth of H. pylori in stomach, but may be related to inducing apoptosis in tumour cells. Allitridum may enhance the effect of chemotherapy for patients with cancer. However, the allitridum has not the severe toxicity of such as chemotherapeutic drugs. Allitridum may be of benefit to patients after surgery. The number of females was too small in this study and a larger trial is needed. Garlic oil only accounts for 0.3% of raw garlic, so many health food products made from garlic in supermarket have little effect for the prevention of cancer.
1.Xu HX, Xu DF, Zhang XM, et al. A death survey of malignant tumour and non-cancer in Shandong during 1970-1974. Compiling of the Office of Tumor Treatment and Prevention in Shandong (Chin) 1979;25-26.
2.Xu HX, Chen AG, Sun R, et al. A five years cohort study of stomach cancer association with garlic. Tumor (Chin) 1989;9:3-5.
3.Li HQ, Jin SK, Yuan XR, et al. A epidemiological study of tumour related to garlic Tabloid of the Four（th） Tumour Academia Conference of China (Chin) 1990;307.
4.Gail MH, You WC, Chang YS, et al. Factorial trial of three interventions to reduce the progression of precancerous gastric lesions in Shandong, China: design issues and initial data. Controlled Clin Trials 1998;19:352-369.
5.Li HQ, Jin SK, Zhang YK, et al. An experimental study of the effect of anti-atherosis by combing galic oil and selenium. J Chinese and Western Medicine (Chin) 1996;16(Suppl):39-40.
6.Milner JA. A historical perspective on garlic and cancer. J Nutr 2001;131(3s):1027S-1031S.
7.Arivazhagan S, Balasenthil S, Nagini S. Garlic and neem leaf extracts enhance hepatic glutathione and glutathione dependent enzymes during N-methyl-N'-nitro-N-nitrosoguanidine (MNNG)-induced gastric carcinogenesis in rats. Phytother Res 2000;14:291-293.
8.Hu X, Benson PJ, Srivastava SK, et al. Induction of glutathione S-transferase pi as a bioassay for the evaluation of potency of inhibitors of benzo (a) pyrene-induced cancer in a murine model. Int J Cancer 1997;73:897-902.
9.Sivam GP. Protection against Helicobacter pylori and other bacterial infections by garlic. J Nutr 2001;131(3s):1106S-1108S.
10.You WC, Zhang L, Gail MH, et al. Helicobacter pylori infection, garlic intake and precancerous lesions in a Chinese population at low risk of gastric cancer. Int J Epidemiol 1998;27:941-944.
11.You WC, Chang YS, Heinrich J, et al. An intervention trial to inhibit the progression of precancerous gastric lesions: compliance, serum micronutrients and S-allyl cysteine levels, and toxicity. Eur J Cancer Prev 2001;10:257-263.
12.Tang F, Zhou J, Gu L. In vivo and in vitro effects of selenium-enriched garlic on growth of human gastric carcinoma cells. Chin J Oncology (Chin) 2001;23:461-464.
13.Lu J, Pei H, Ip C, et al. Effect on an aqueous extract of selenium-enriched garlic on in vitro markers and in vivo efficacy in cancer prevention. Carcinogenesis 1996;17:1903-1907.
14.L'vova GN, Zasukhina GD. Modification of repair DNA synthesis in mutagen-treated human fibroblasts during adaptive response and the antimutagenic effect of garlic extract. Genetika 2002;38:306-309.
15.Kyo E, Uda N, Kasuga S, et al. Immunomodulatory effects of aged garlic extract. J Nutr 2001;131(3S):107S-109S.
16.Lamm DL and Riggs DR. Enhanced immunocompetence by garlic: role in bladder cancer and other malignancies. J Nutr 2001;131(3S):1067S-1070S.
17.Moon DG, Cheon J, Yoon DH, et al. Allium sativum potentiates suicide gene therapy for murine transitional cell carcinoma. Nutr Cancer 2000;38:98-105.
18.Lu J, Pei H, Ip C, et al. Effect on an aqueous extract of selenium-enriched garlic on in vitro markers and in vivo efficacy in cancer prevention. Carcinogenesis 1996;17:1903-1907.
19.Lamm DL, Allaway M. Current trends in bladder cancer treatment. Ann Chir Gynaecol 2000;89:234-241.
20.Borek C. Antioxidant health effects of aged garlic extract. J Nutr 2001;131(3S):1010S-1015S.
21.Li XG, Xie JY, Fang M, et al. An experiment study of garlic oil induced apoptosis for HL-60 cell. J Tradit Chin Basic Med (Chin) 1997;3:23-25.