: Strong Science and Real Medicine. An overview.

Scott M. Lippman, Bernard Levin

Department of Clinical Cancer Prevention and Division of Cancer Prevention and Population Sciences, The University of Texas M. D. Anderson Cancer Center, Houston, TX

Many of us working in cancer prevention over the past two decades have begun our primary papers with something resembling the following remarks: “It is estimated that x cancer will account for y new cancer diagnoses and z deaths in the United States alone this year. These grim statistics persist despite advances in cancer therapy, arguing forcibly for a new approach, such as in cancer prevention, to help in controlling this devastating disease.” Alas, this general rationale for cancer prevention is still relevant. In 2004, it is estimated that cancers of the breast, prostate, bladder, colon/rectum, lung/bronchus, and oral cavity/pharynx will account for 856,760 new cancer diagnoses and 307,590 deaths in the United States alone. 1

Cancer prevention has advanced on several important fronts, including clinical practice and research in many sites of epithelial neoplasia; the molecular study of neoplasia, risk and drug activity; classical and molecular epidemiology; the biology of tobacco- and obesity-related neoplasia; and the behavioral and nutritional sciences. 2 Striking practical advances in cancer prevention came largely on the heels of decisive National Cancer Institute–funded randomized controlled trials (RCTs). Three such trials significantly reduced cancer risk in the breast, prostate, or skin. 3 – 5 The Breast Cancer Prevention Trial of tamoxifen led the US Food and Drug Administration (FDA) to make tamoxifen the first US FDA-approved agent for reducing cancer risk. 3 Although not evaluated by the US FDA for cancer risk reduction, finasteride reduced prostate cancer risk in the Prostate Cancer Prevention Trial, 4 which has major public health implications (since this drug is marketed for benign prostatic hypertrophy and hair loss) and stimulated wide-reaching research of the risk, and reduction in the risk, of prostate cancer, especially high grade. Another high-profile prevention RCT led the US FDA to approve celecoxib as a pharmacologic adjunct in the management of familial adenomatous polyposis (FAP). 6 RCTs also have established the activity of calcium and aspirin in sporadic colorectal adenomas and sulindac in FAP. 6 – 8 Further advances in practical cancer prevention result from the growing acceptance of cancer risk-reducing procedures related to the screening of cancer precursors, or intraepithelial neoplasia (IEN), 9 such as colonoscopy and polypectomy, or even prophylactic organ resection in extremely high-risk settings such as FAP and BRCA1 mutation carriers. 10 Integrating molecular techniques that can help in evaluating IEN screening, drug efficacy, or surgical margin widths into clinical practice is an ongoing challenge of practical clinical cancer prevention. 11 , 12

Prevention RCTs not only help define the clinical role of agents such as tamoxifen, celecoxib, and finasteride, but also generate hypotheses based on important secondary analyses for new trials, as did the Alpha-Tocopherol and Beta-Carotene trial (involving vitamin E) and the Nutritional Prevention of Cancer trial (involving selenium) in generating the primary hypotheses of the Selenium and Vitamin E Cancer Prevention Trial (in the prostate). 13 Other important sources of hypotheses-generating data are classical epidemiologic studies of populations at risk, etiologic factors, and potential interventions to reduce cancer risk. Cancer epidemiology data on the influence of non-steroidal anti-inflammatory drugs on the incidence of colorectal cancer generated hypotheses for RCTs that confirmed the epidemiologic data. 14 RCTs do not always confirm their epidemiologic forebears, however, as was shockingly illustrated by studies of beta-carotene. After strong epidemiologic evidence that beta-carotene could reduce lung cancer risk, RCTs indicated that beta-carotene actually increased lung cancer risk and mortality in heavy smokers. 15 , 16

Tobacco use is the major global cause of many serious diseases of the heart and lung and a large number of cancers. Four to 5 million people die from tobacco-related diseases every year. If current smoking patterns persist, it is estimated that annual tobacco-related mortality will increase to 10 million by the year 2030 (7 million in developing countries) and that tobacco will cause over 1 billion deaths in the 21st century. 17 Tobacco is the greatest preventable cause of cancer morbidity and mortality and causes profound worldwide economic losses every year—over $150 billion in the United States alone. Intensive efforts in behavioral counseling and pharmacologic tobacco dependence therapy have had limited long-term success in controlling tobacco-related diseases. Controlling the use of tobacco will require novel therapy for nicotine dependence and other approaches. These approaches will benefit greatly from advances in research of the genetic etiology of smoking behavior, including the discovery of constitutional variations in genes in the dopamine, serotonin, and nicotine metabolism pathways. 18 Reducing tobacco use also will rely heavily on new legislative and political initiatives, such as the WHO Framework Convention on Tobacco Control.

Obesity or overweight is the largest avoidable non-smoking cause of cancer mortality, accounting for 90,000 cancer deaths annually in the United States. After relative stability for decades, the prevalence of obesity or overweight shot up in the 1980s and 1990s and continues to rise, including in children and adolescents, in many parts of the world. This rise in children (a doubling in the last two to three decades) foreshadows the continued rise of the cancer consequences of obesity since overweight children frequently become overweight adults. By the definition of the WHO, nearly two-thirds of U.S. adults are considered obese or overweight. Researchers are attempting to identify the genetic factors (eg, obesity susceptibility genes) and other biologic factors underlying energy regulation and obesity risk, such as factors relating to unhealthful cravings for both food and tobacco. The study of interventions to modify patterns of diet and physical activity, and thus control obesity and its cancer consequences, will benefit from integrating behavioral research with advances in our understanding of the genetics, molecular epidemiology, metabolic consequences, and basic biology of obesity. 19 As with tobacco control, obesity control will benefit greatly from public education and health policies both discouraging unhealthful and encouraging healthful dietary practices.

Infections are another important global cause of cancer. Preventing or treating cancer-causing infections already has produced important advances in clinical cancer prevention. Vaccinating children against hepatitis B virus has dramatically reduced the incidence and mortality of liver cancer in Taiwan. 20 Treating chronic hepatitis B also has reduced the incidence of liver cancer. 21 Vaccines targeting human papilloma viruses to prevent cervical IEN and cervical (and potentially anal, penile, and oropharyngeal) cancers are being developed. 22 , 23 Controlling H pylori may reduce the risks of atrophic gastritis and gastric cancer. 24 – 26

Cancer screening and early detection is an ever-evolving, complex field. This complexity is illustrated by cancer screening with prostate-specific antigen (PSA), which, although established and widely employed, is coming under closer scrutiny. Questions and controversy surround the ability of PSA screening to reduce prostate cancer mortality, the optimal PSA level for recommending a prostate biopsy, 27 and PSA as a reflection of disease status. 27 – 29 The potential benefits of cancer screening in many sites may