Biography

Dr. Schatzkin received a B.A. from Yale University, an M.D. from the State University of New York, and an M.P.H. and Dr.P.H. from Columbia University. He is board certified in internal medicine and preventive medicine/public health. Dr. Schatzkin joined the NCI in 1984 as a senior staff fellow in the Cancer Prevention Studies Branch of the Division of Cancer Prevention and Control, and became a senior investigator in 1988. He joined the DCEG Nutritional Epidemiology Branch in 1997, and was appointed Branch Chief in 1999. Dr. Schatzkin's research focuses on the nutritional etiology and prevention of cancer. He received the NIH Merit Award for his work on the Polyp Prevention Trial.

Research Interests

Intervention, Cohort, and Methodologic Studies of Nutrition and Cancer

We are focusing on three areas of research to clarify the relation between nutrition and cancer: error in dietary assessment, dietary homogeneity within study populations, and uncontrolled or unmeasured confounding in observational studies. Intervention studies complement observational investigations by minimizing the potential for confounding and guaranteeing wide intake differences across randomization groups. Intensive biomarker studies, in conjunction with new mathematical modeling techniques, aid in quantifying dietary measurement error in a new generation of large prospective epidemiologic studies of diet and cancer.

Diet and Colorectal Neoplasia

The Polyp Prevention Trial was a multicenter intervention study of the effect of a low fat (20% calories from fat), high fiber (18 g dietary fiber/1,000 kcal), high fruit and vegetable (5-8 daily servings) eating plan on the recurrence of colorectal adenomatous polyps, precursor lesions for most large bowel cancers. Results after four years of follow-up showed no difference in recurrence of any or advanced adenomas between the intervention and control groups. Trial participants will be followed for an additional 5 years. Several observational epidemiologic studies are near completion, including those of NSAIDs use, blood carotenoids, and hormone replacement therapy in relation to subsequent adenoma recurrence. Additional observational studies of adenoma recurrence in this population are feasible.

The trial incorporated several interdisciplinary, biospecimen-based substudies. Serum lipids and carotenoids were used to monitor the intervention progress and provide a biochemical complement to self-reported dietary assessments. Integration of bromodeoxyuridine and proliferating cell nuclear antigen assays of rectal mucosal biopsy specimens into the main trial enables us to evaluate the effect of dietary change on cell proliferation, as well as the extent to which proliferation indices predict neoplasia. An initial report documents considerable "noise" in proliferation markers, emphasizing the need for gathering information on sources of biomarker variability, and suggesting that some of these markers may have limited utility in population studies. In the first of two molecular genetic studies, we are examining ras and p53 mutations in resected adenomas to determine whether dietary change alters somatic mutations. In the second study, using DNA extracted from white blood cells, we are investigating interactions of dietary factors and allelic variants of polymorphic genes encoding metabolizing enzymes, such as MTHFR and the NAT and GST gene families, in relation to adenoma formation.

NCI-AARP Diet and Health Study

The NCI-American Association of Retired Persons Diet and Health Study, comprising over 560,000 men and women in the United States, was designed to overcome specific methodologic limitations of previous epidemiologic studies of diet and cancer. The large cohort exhibits substantial dietary heterogeneity for major nutrients and foods, thereby circumventing the narrow intake range in many previous study populations. The prospective study design avoids the dietary recall bias that limits case-control studies. Also, its large size offsets, at least partially, the attenuation in relative risk resulting from dietary measurement error. The study uses a new food frequency questionnaire based on cognitive psychologic principles and extensive focus-group testing. In order to characterize qualitatively and quantitatively the error structure of the new dietary assessment instrument, we incorporated a calibration study of approximately 2,000 participants who completed multiple food records along with repeated food frequency questionnaires. The large study size also facilitates the evaluation of interactions among dietary factors and environmental exposures or host characteristics. The study is expected to yield nearly 4,000 incident breast, over 4,000 incident colorectal, over 10,000 incident prostate, approximately 900 pancreatic and over 400 ovarian cancers by 2003.

Additional Prospective Studies

The Breast Cancer Detection Demonstration Project Follow-up Study is a prospective cohort study of approximately 50,000 older women established from a large NCI-sponsored breast cancer screening program. A decade ago, we integrated a food frequency questionnaire into the study, along with queries on supplement use, body size, and physical activity. Over 1,000 breast cancer cases and over 450 colorectal cancer cases subsequently developed in about 40,000 women who completed the questionnaire satisfactorily. One analysis showed a direct association between adult body mass index and increased risk of breast cancer, especially among older women. Other investigations have shown no overall association between dietary fat and breast cancer, an analysis that considered multiple methods of energy adjustment; an inverse relation between physical activity and breast cancer; and little association between fruit and vegetable intake and large bowel cancer. Ongoing studies include dietary patterns in relation to breast and colorectal cancers, dietary fiber vis-a-vis colorectal cancer, folate and one-carbon metabolism in colorectal cancer, and meat consumption and large bowel cancer.

We collaborated with investigators at the National Heart, Lung, and Blood Institute and Boston University to generate the cancer outcomes component of the original Framingham Heart Study cohort and the Framingham Offspring Study cohort. Using these prospective studies, we examined the relation between alcohol and breast cancer (no association); physical activity and colorectal cancer (protective in men); physical activity and breast cancer (no association); central to peripheral body fat ratio and breast cancer (direct association); and bone mineral density, a possible proxy for cumulative exposure to exogenous estrogens, and breast cancer (direct association).

We established an interagency agreement with the Department of Veterans Affairs to create a biologic specimen bank within a unique Veterans Affairs study of adenoma prevalence and recurrence among asymptomatic persons undergoing full colonoscopy. In addition to analyzing blood micronutrients in relation to prevalent and recurrent adenomas, we are using DNA extracted from banked white cells to explore diet-gene interactions in the study population.

In a study similar to the VA study, we have established an interagency agreement to study adenoma prevalence among asymptomatic women undergoing full colonoscopy at regional Naval hospitals in the U.S. (the CONCeRN study). We will be able to analyze blood and colonic tissue micronutrients, in conjunction with information on dietary and genetic polymorphisms, in relation to prevalent adenomas.

Methodologic Research

In a large collaborative study with investigators from NCI's Division of Cancer Control and Population Sciences, we are assessing energy expenditure (using the doubly labeled water technique) and protein intake (via urinary nitrogen excretion) to estimate the error (generally under-reporting) in a food frequency questionnaire and multiple 24-hour recalls. Previous studies, even those "adjusting" for measurement error, took into account biases in the food frequency questionnaire but did not fully considered biases associated with the "reference" instrument (24-hour recall), especially the potential correlation of biases in the two instruments. The information on measurement error from this study can be used in interpreting and adjusting data from the NCI-American Association of Retired Persons Diet Study and other epidemiologic investigations.

In conjunction with NCI biostatisticians, we investigated the utility of intermediate and surrogate end point markers in cancer research. We showed that a priori validity for a potential cancer surrogate requires a marker to be virtually a necessary component in the carcinogenesis pathway. A marker in one but not another causal pathway to cancer-thereby being non-necessary and having an attributable fraction less than one-may give misleading answers about the effect of an intervention on cancer. Moreover, a marker may be a reasonable cancer surrogate for one intervention (or exposure) but not for another. Even reasonably valid surrogates, like colorectal adenomas, cannot be considered definitive in themselves and must be evaluated in conjunction with observational and experimental epidemiologic findings as well as other types of evidence.

Keywords

adenoma, alcohol, anthropometry, biomarkers, breast cancer, clinical trials, colonoscopy, colorectal cancer, dietary factors (fat, fiber, fruits, vegetables), dietary intervention studies, dietary measurement error, dietary patterns, diet-gene interactions, epithelial cell proliferation, Framingham Heart Study, Polyp Prevention Trial, prostate cancer, somatic mutations, surrogate end points

Selected Publications

• Schatzkin A, et al. "Lack of effect of a low-fat, high-fiber diet on the recurrence of colorectal adenomas." Polyp Prevention Trial Study Group. N Engl J Med 2000; 342:1149-1155.
• Kant AK, et al. "A prospective study of diet quality and mortality in women." JAMA 2000; 283:2109-2115.
• Schatzkin A. "Intermediate markers as surrogate endpoints in cancer research." Hem/Onc Clin N Am 2000; 14:887-905.
• Kipnis V, et al. "Empirical evidence of correlated biases in dietary assessment instruments and its implications." Am J Epidemiol 2001; 153:394-403.

Collaborators

DCEG Collaborators

• Demetrius Albanes, M.D.; Louise Brinton, Ph.D.; Neil Caporaso, M.D.; Nilanjin Chatterjee, Ph.D.; Wong-Ho Chow, Ph.D.; Andrew Flood, Ph.D.; Mitchell Gail, M.D., Ph.D.; Montserrat Garcia-Closas, M.D., Dr.P.H.; Alisa Goldstein, Ph.D.; Barry Graubard, Ph.D.; Richard Hayes, D.D.S., Ph.D.; Allan Hildesheim, Ph.D.; James Lacey, Ph.D.; Volker Mai, Ph.D.; Dominique Michaud, Sc.D.; Ulrike Peters, Ph.D.; Ruth Pfeiffer, Ph.D.; Nathaniel Rothman, M.D.; Catherine Schairer, Ph.D.; Mark Schiffman, M.D.; Usha Singh, M.P.H.; Rashmi Sinha, Ph.D.; Tara Vogt, Ph.D.; Sholom Wacholder, Ph.D.

Other NCI Collaborators

• Rachel Ballard-Barbash, M.D., M.P.H.; Cynthia Brown, Ph.D.; Carolyn Clifford, Ph.D.; Lisa Colbert, Ph.D., Donald Corle, M.S.; Sandy Dawsey, M.D.; Linda Harlan, Ph.D.; Anne Hartman, M.S.; Victor Kipnis, Ph.D.; Elaine Lanza, Ph.D.; Lisa McShane, Ph.D.; Douglas Midthune, M.S.; Nancy Potischman, Ph.D.; Luke Ratnasinghe, Ph.D.; Amy Subar, Ph.D.; Joseph Tangrea, Ph.D.; Philip Taylor, M.D., Sc.D.; Fran Thompson, Ph.D.; Richard Troiano, Ph.D.; Karen Woodson, Ph.D.

Other Scientific Collaborators

• Henry Appelman, M.D., University of Michigan, Ann Arbor, MI
• Dennis Ahnen, M.D., University of Colorado, CO
• Douglas Bauer, M.D., University of California School of Medicine, San Francisco, CA
• Colin Begg, Ph.D.; Moshe Shike, M.D., Memorial Sloan-Kettering Cancer Center, New York, NY
• Sheila Bingham, Ph.D., University of Cambridge, Cambridge, England
• Randall Burt, M.D.; Martha Slattery, Ph.D, M.P.H., R.D., University of Utah School of Medicine, Salt Lake City, UT
• Bette Caan, Dr.P.H., Kaiser Foundation Research Institute, Oakland, CA
• Ray Carroll, Ph.D., Texas A&M University, College Station, TX
• M. Robert Cooper, M.D., Wake Forest University School of Medicine, Winston-Salem, NC
• Joanne Dorgan, Ph.D., Fox Chase Cancer Center, Philadelphia, PA
• Curtis Ellison, M.D.; Yuqing Zhang, Ph.D., Boston University, MA
• Laurence Freedman, M.S., Bar Ilan University, Ramat Gan, Israel
• Stanley Hamilton, M.D., M.D. Anderson Cancer Center, Houston, TX
• Frank Iber, M.D., Edward Hines Jr. Veterans Administration Hospital, Chicago, IL
• Rudolf Kaaks, Ph.D.; Elio Riboli, M.D. International Agency for Research on Cancer, Lyon, France
• Ashima Kant, Ph.D., Queens College of the City University of New York, Flushing, NY
• J. Walter Kikendall, M.D., Walter Reed Army Medical Center, Washington, D.C.
• Martin Kulldorf, Ph.D., University of Connecticut, CT
• Peter Lance, M.D., State University of New York, Buffalo, NY
• Klaus Lewin, M.D., UCLA Medical Center, Los Angeles, CA
• David Lieberman, M.D., Portland Veterans Administration Hospital, Portland, OR
• James Marshall, Ph.D., University of Arizona, AZ
• Sheila Prindiville, M.D., M.P.H., University of Colorado, CO
• Robert Schoen, M.D.; Joel Weissfeld, M.D., University of Pittsburgh, Pittsburgh, PA
• Philip Schoenfeld, M.D., University of Michigan, MI
• Michael Wargovich, Ph.D., South Carolina Cancer Center, Columbia, SC