Fleminger, Inc. 160 Hawley Lane, Suite 205, Trumbull CT 06611
Office of Sin Hang Lee, M.D.
Tel. 203 385-3836 Fax 203 385-3832
July 5, 2005
Mr. Michael M. Landa
Deputy Director for Regulations
C/o Mr. Vincent deJesus
Center Food Safety and Applied Nutrition
Food and Drug Administration (HFS-830)
5100 Paint Branch Parkway
College Park, MD 20740-3835
FedEx 842881988990
Re: FDA News June 30, 2005 Health Claim Petition:
Green Tea and Reduced Risk of Cancer Health Claim (Docket No. 2004Q-0083)
Dear Mr. Landa:
This letter requests that the FDA publish a supplement to the FDA News P05-36, June 30, 2005 titled “FDA Issues Information for Consumers about Claim for Green Tea and Certain Cancers” (FDA News). A supplementary statement to the above FDA News is necessary in order to not confuse those consumers who have been drinking green tea for the potential benefits of reducing the risk of breast or prostate cancers. Based on the letter that you faxed to me on June 30, 2005, at least some Asian American women in Los Angeles are getting the benefits of green tea consumption according to Wu et al. 2003 (1). It would be a tragedy if some of these women should discontinue their regular potentially beneficial healthy dietary program including green tea because of the FDA News.
Out of the totality of 372 scientific publications on “green tea and cancer prevention” available in the public domain (see PubMed search printed on 7/2/2005, enclosed) the FDA chose five studies for review in making its decision on my petition under docket no. 2004Q-0083 which has proposed a qualified health claim linking daily consumption of high-EGCG green tea in a sufficient amount to potential reduction of the risk of certain forms of cancer, based on a comprehensive review of the world’s scientific literature.
On the relationship between green tea and breast cancer, the FDA News stated “Two studies do not show that drinking green tea reduces the risk of breast cancer in women, but one weaker, more limited study suggests that drinking green tea may reduce the risk. Based on these studies, FDA concludes that it is highly unlikely that green tea reduces the risk of breast cancer”. This statement needs to be clarified for the following reasons.
(I) In your letter supporting the FDA’s decision you stated that the first two studies referenced above are summarized in an article published by Suzuki et al. 2004 (2) which is a Short Communication. The third study labeled as “weaker, more limited study” in the FDA News is identified to be a report by Wu et al. 2003 (1). In your letter, you stated that all “three studies were considered to be of high methodological quality”. Therefore, the language used to degrade Wu et al.’s study in the FDA News to give more weight to the first two studies contradicts the FDA scientific evaluation as stated in your letter. When your letter addressed to me is posted on the FDA Internet, some Asian American participants in the Los Angeles study will know that they had been recruited in an inferior Wu et al.’s study according to the FDA News. But in fact it is not the case according to your letter.
(II) Suzuki et al.’s Short Communication pools two cohort studies, namely study cohort 1 pointing to a report by Tsubono et al. entitled “Green tea and the risk of gastric cancer in Japan” published in 2001 in N Eng J Med (3). Cohort 2 refers to an article by Nakaya et al., entitled “Personality and Risk of Cancer” (4). Green tea was one of the sideline items included in these two cohort studies among numerous risk factors listed in a self-administered questionnaires validated for food frequency questionnaire survey in rural Japan. The main purposes of these two cohort studies were for survey of gastric cancer and for personality and cancer risk, respectively. Through population-based cancer registries, 103 cases of breast cancer were identified in cohort 1 and 119 cases in cohort 2. In comparison, Wu et al.’s study entitled “Green Tea and Risk of Breast Cancer in Asian Americans” included 501 cases of breast cancer patients and 509 control subjects in the Los Angeles County. Wu et al.’s cases were identified through the Los Angeles County Cancer Surveillance Program, the population-based cancer registry that is a member of the National Cancer Institute’s Surveillance, Epidemiology and End Results (SEER) program and the statewide California Cancer Registry. In-person interviews were conducted using a standardized, structured questionnaire that covered demographic characteristics and migration history, menstrual and reproductive history, body size physical activity and diet history. The diet questionnaire was developed by Dr. J . Hankin (University of Hawaii) and modeled after validated diet instrument used in Multiethnic Cohort Study being conducted in Hawaii and Los Angeles. Wu et al.’s study has more cases, involves in-person interview with the study subjects and is part of the NCI SEER program. Referring to Wu et al.’s study as a “weaker” one in the FDA News release of June 30, 2005 in comparison with Suzuki et al.’s for the purpose of giving a higher credibility to the latter is unfortunate and misleading.
(III) In your letter addressed to me, the FDA had placed a high value on Suzuki et al.’s Short Communication (2) for two cohort studies on breast cancer. However, the latter report refers to a cohort 1 study relying on self-administered questionnaires returned from 13,992 men and 17,353 women collected by Tsubono et al. primarily for gastric cancer analysis, and to a cohort 2 study in Nakaya et al.’s original report which also relied on collecting self-administered questionnaires returned by 14473 men and 15133 women as the raw materials for analysis, especially for the analysis of personality and cancer risk. In Nakaya et al.’s paper, it listed only 87 cases of breast cancer for cancer risk estimates [page 801, left column, line 18; page 803, Table 3, Cancer site Breast (n = 87)]. But in Suzuki et al.’s Short Communication, on which the FDA had placed a high value, it states that there were 14,409 subjects and 103 breast cancer cases in cohort 1, and that there were 20,595 subjects and 119 breast cancer cases in cohort 2, as quoted on page 8 in your letter. There is no explanation on how these numbers were changed by exclusion and inclusion from those listed in the two original reports to those in a pooled analysis reported in the table format of a Short Communication. As a pathologist, I am most concerned about the accuracy of the diagnosis of each breast cancer before the case is accepted for statistical analysis. In Tsubono et al.’s report (3) which forms the cohort 1 of the Suzuki et al.’s study which was relied upon by the FDA to reach the current decision on the green tea health claim, the authors stated “The diagnosis of adenocarcinoma had been histologically confirmed in 80% of the cases.” (Reference 3, page 633, left column, line 39). That means 20% of cancers had not been confirmed by histological examination before they were entered for statistical analysis. It is unreasonable to expect that all Japanese women living in a rural region of Miyagi prefecture to give an accurate diagnosis of a tumor found in or removed from her breast as cancer or non-cancerous. As emphasized in your letter from page 9 to page 16, validation of methodology and research protocols is the most important element in evaluating the data collected for statistical analysis. If the cancer diagnosis of each patient is not validated by a pathologist, all the data listed on the self-administered questionnaires are of little value for analysis linking food and cancer risk. Therefore, based on the FDA internal regulations, the Suzuki et al.’s Short Communication is of little value and certainly cannot be relied upon for making an FDA statement. In contrast, Wu et al.’s study confirmed the diagnoses of their cases through the NCI SEER program and the statewide California Cancer Registry whose data, I know for sure, are based on tissue diagnosis reports of pathologists certified by the American Board of Pathology.
(IV) Suzuki et al.’s two cohort studies were conducted among residents living in Miyagi prefecture of northern rural Japan where no green tea is produced. Enclosed herewith is a copy downloaded from the Internet “Japanese Green Tea Museum” listing all the tea-producing prefectures in Japan (5) and a map showing the Tea Producing Prefectures (6). All tea plantations are located in southern Japan. The climate in Miyagi is not suitable for growing tea plants. Good green tea is traditionally an expensive food item in Japan and in China. Japan generally produces less than 100,000 tons of green tea per year, but consumes more than 125,000 tons of tea. Most quality green teas, or Sencha, are consumed in southern Japan and in big cities. In general, only lower quality green teas, or a grade of Sencha called Hojicha, would be sold to the people living in northern rural regions. Many lower-income Japanese people drink imported cheap tea. As a result, sending self-administered questionnaires for survey to the residents living in Miyagi prefecture would likely be filled out and returned by those who could only afford to drink the left-over green tea from past years or a lower grade green teas or Hojicha. Several Japanese scientists, one of them Professor Hirota Fujiki, have dealt with the question of the quality of green tea consumed in different locations of Japan and its effects on cancer prevention. The Saitama experience of using green tea as a chemopreventive agent for cancer prevention was developed in the high quality tea-producing region of Saitama (7, 8). In Los Angeles, the situation is different since America is not a tea producing country. Most of the early Asian immigrants in the U. S. were from Guangdong province’s rural regions. To this date, the descendants of these Chinese peasants still shun green tea as a beverage for a historical reason which is to counter starvation and is no longer valid. The green tea consumers in Los Angeles are under the influence of the old family tradition transplanted from the Middle Kingdom of China. They are usually from the provinces where green teas have been cultivated and consumed for centuries and know where to get quality green teas, for example, from the tea plantations in China directly or from specialty tea merchants, usually in the spring of each year. To illustrate the value that has been put on quality green tea by the Middle Kingdom Chinese, I am enclosing a photograph (9) taken about two months ago in Hangzhou, China, showing a tea peddler selling new green teas for about US$ 200 to 350/lb. (The monthly salary for an upper-income earner is about US$ 300 in Hangzhou)
Based on the above analyses, I propose that the FDA issue a supplementary statement on the breast cancer green tea health claim in the following language:
Upon reviewing additional materials submitted by the green tea health claim petitioner, the FDA issues the following statement to further clarify the green tea qualified health claim on breast cancer risk:
“Two studies conducted in northern Japan do not show that drinking green tea up to 5 cups a day reduces the risk of breast cancer in women, but one study conducted in Los Angeles, California, USA suggests that drinking green tea may reduce the risk. Based on these studies, FDA concludes that it is highly unlikely that most green teas reduce the risk of breast cancer ”.
On the relationship between green tea and prostate cancer, the FDA News stated “One weak and limited study does not show that drinking green tea reduces the risk of prostate cancer, but another weak and limited study suggests that drinking green tea may reduce the risk. Based on these studies, FDA concludes that it is highly unlikely that green tea reduces the risk of prostate cancer”. This statement needs to be clarified for the following reasons.
(I) In your letter supporting the FDA’s decision you identified the Sonoda et al.’s study (10) as the basis for the first part of the statement and the Jian et al.’s 2004 study (11) as the basis for the second part of the statement. Careful reading of Sonoda et al.’s paper reveals that the surveys on diet in this case control study were performed mainly in Ibaraki and Hokkaido prefectures, including 84 cases from Ibaraki and 56 from Hokkaido. While Ibaraki is located just north of the famous green tea-producing prefecture of Saitama, Hokkaido is the most northern rural part of the Japanese islands where no tea trees can survive. As a result, the qualities of the green tea consumed by the study subjects living these two locations are quite different. The flaw in the research design of the Sonoda et al.’s study is that the authors pooled the diet data of the subjects collected from these two greatly diverse regions as one homogenous mass for analysis. As a result, the odds ratios (ORs) of green tea intake and prostate cancer risk were found to be dose-dependent showing an inverse relationship, namely 1.00 for or = 10 cups/day. However, the 95% CIs spread so widely to 0.48-2.03, 0.38-1.63 and 0.27-1.64, respectively. The lack of statistical significance in the reduced ORs is probably due to combination of an insignificant risk reduction in the tea-drinking subjects surveyed in Hokkaido and a highly significant reduction in prostate cancer incidence in the heavy green tea drinkers living in Ibaraki for calculation. The bias is caused by the difference in quality of the green teas consumed in two populations.
(II) In contrast, Jian et al.’s study was conducted in the city of Hangzhou, Zhejiang Province, known as the capital of green tea in China for the past 1,000 years. In the late 12th century the Zen priest Eisai brought the seeds of the green tea plant as a source of medicine to Japan from Hangzhou, the then capital of the Southern Song Dynasty. Therefore, it is safe to say that the tea drink consumed by people living in the major tea producing regions of southern Japan is similar to that consumed in Hangzhou. The high-grade Japanese green tea typically contains more than 7% EGCG in dry weight, similar to those sold in Hangzhou. The NCI adopted the definition of typical green tea, containing 710 microgram/ml EGCG, for cancer research according the traditional green tea prepared in Hangzhou. Jian et al. selected their prostate cancer patients from the urology wards of the 8 public hospitals in the city of Hangzhou, who had been residents of Zhejiang Province for at least 10 years for their case-control study. In addition to pathology diagnosis confirmation in every case, all patients and controls were interviewed in-person. The questionnaire was patterned after those used in Hawaii Cancer Survey, Australian Health Survey and the United States food survey. All were validated according to standard methods. Although this study appears to be small in number, including only 130 cases of prostate cancer patients, it is in fact substantial in a country where the incidence rate of prostate cancer per 100,000 is 1.74, compared to 104.33 per 100,000 in the U.S.A. It is unfortunate that the FDA had classified such a valuable study as another weak and limited study.
Based on the above analyses, I propose that the FDA issue a supplementary statement on the prostate cancer green tea qualified health claim in the following language:
Upon reviewing additional materials submitted by the green tea health claim petitioner, the FDA issues the following statement to further clarify the green tea qualified health claim on prostate cancer risk:
“One two-centered study conducted in part in northern rural Japan does not show that drinking green tea reduces the risk of prostate cancer, but another study conducted in Hangzhou, China suggests that drinking green tea may reduce the risk. Based on these studies, FDA concludes that it is highly unlikely that most green teas reduce the risk of prostate cancer”
As always, the applicant for this qualified green tea health claim is prepared and ready to offer additional information to facilitate the FDA review of all matters related to this petition. Please contact the undersigned if you have any questions at my request.
As part of the consumer education programs of the FDA, I hereby request that this letter be posted on the FDA docket management website under Docket No. 2004Q-0083 for ready viewing of the public.
Thank you for your attention and cooperation.
Sincerely,
Sin Hang Lee, M.D.
Enclosed:
Search PubMed for “green tea and cancer prevention” National Library of Medicine 7/2/2005
References
(1) Wu AH, Mini CY, Tseng C-C et al. Green tea and risk of breast cancer in Asian Americans. Int J Cancer 2003;106: 574-579.
(2) Suzuki Y, Tsubono Y, Nakaya N et al. Short Communication Green tea and the risk of breast cancer: pooled analysis of two prospective studies in Japan. British J Cancer 2004;90:1361-1362.
(3) Tsubono Y, Nishino Y, Komatsu S et al. Green tea and the risk of gastric cancer in Japan. N Engl J Med 2001;344:632-636.
(4) Nakaya N, Tsubono Y, Hosokawa T et al. Personality and the risk of cancer. J NCI 2003; 95:799-805.
(5) Japanese green tea museum: Japanese green tea production areas.
(6) Tea Harvest in Japan: Tea Producing Prefectures.
(7) Fujiki H, Suganuma M, Imai K et al. Green tea:cancer preventive beverage and/or drug. Cancer letters 2002;188:9-13.
(8) Fujiki H. Green tea: health benefits as cancer preventive for humans. Chem Rec 2005;5:119-132.
(9) Sonoda T, Nagata Y, Mori M et al. A case-control study of diet and prostate cancer in Japan:possible protective effect of traditional Japanese diet. Cancer Sci 2004;85:238-242.
(10) Jian L, Xie LP, Lee AH et al.Protective effect of green tea against prostate cancer: a case- control study in southeast China. Int J Cancer 2004;108:130-135.
(11) A photograph: The value of tea in the East! Dr.Lee’s TeaForHealth.
cc. Senator Chris Dodd (Attention: Ryan 860 258-6958); Senator Joe Lieberman; Congresswoman Rosa L. DeLauro; Congressman Christopher Shays (Tom McCabe 579-0771) ; Congresswoman Nancy L. Johnson; Congressman John B. Larson; Congressman Rob Simmons.
Congressman Henry A. Waxman (Los Angeles, CA)
