11-25 Lecture on Clinical Research in Oncology PDF

Additional types of clinical research in Oncology Robert Wieder, MD, PhD Division of Medical Oncology/Hematology Rutgers New Jersey Medical School  Types of clinical trials  One way of classifying clinical trials is by the way the researchers behave.  In an observ...

Additional types of clinical research in Oncology Robert Wieder, MD, PhD Division of Medical Oncology/Hematology Rutgers New Jersey Medical School  Types of clinical trials  One way of classifying clinical trials is by the way the researchers behave.  In an observational study, the investigators observe the subjects and measure their outcomes. The researchers do not actively manage the experiment. This is also called a natural experiment. An example is the Nurses' Health Study.  In an interventional study, the investigators give the research subjects a particular medicine or other intervention. Usually, they compare the treated subjects to subjects who receive no treatment or standard treatment. Then the researchers measure how the subjects' health changes.  Study Designs A fundamental distinction in evidence-based medicine is the method for deriving conclusions: either from observational studies or randomized controlled trials. Types of observational studies in epidemiology such as the cohort study and the case-control study provide less compelling evidence than the randomized controlled trial. In observational studies, the investigators only observe associations (correlations) between the treatments experienced by participants and their health status or diseases.  Case-control studies - Case-control studies can identify factors that may contribute to a condition by comparing subjects who have that condition (the 'cases') with patients who do not have the condition but are otherwise similar (the 'controls'). - Case-control studies are a relatively inexpensive and frequently- used type of epidemiological study, can be carried out by small teams or individual researchers in single facilities in a way that more structured experimental studies often cannot be. - They pointed the way to a number of important discoveries and advances, but their retrospective, non-randomized nature limits the conclusions that can be drawn from them.  Case-control studies (cont.) -The great triumph of the case-control study was the demonstration of the link between tobacco smoking and lung cancer, by Sir Richard Doll and others after him. Doll was able to show a statistically significant association between the two in a large case control study. - Opponents, usually backed by the tobacco industry, argued (correctly) for many years that this type of study cannot prove causation, but the eventual results of cohort studies confirmed the causal link which the case-control studies suggested, and it is now accepted that tobacco smoking is the cause of about 87% of all lung cancer mortality in the US.  Cohort Studies - A cohort is a group of people who share a common characteristic or experience within a defined period (e.g., are born, leave school, lose their job, are exposed to a drug or a vaccine, etc.). - The comparison group may be the general population from which the cohort is drawn, or it may be another cohort of persons thought to have had little or no exposure to the substance under investigation, but otherwise similar. - Subgroups within the cohort may be compared with each other - A cohort study is often undertaken to obtain evidence to try to refute the existence of a suspected association between cause and disease; failure to refute a hypothesis strengthens confidence in it.  Cohort Studies (cont.) - Crucially, the cohort is identified before the appearance of the disease under investigation. The study groups, so defined, are observed over a period of time to determine the frequency of new incidence of the studied disease among them. - The cohort cannot therefore be defined as a group of people who already have the disease. Distinguishing causality from mere correlation cannot usually be done with results of a cohort study alone. - The advantage of cohort study data is the longitudinal observation of the individual through time, and the collection of data at regular intervals, so recall error is reduced. - Cohort studies are expensive to conduct, are sensitive to attrition and take a long time to generate useful data. The Nurses’ Health Study – an observational cohort study - To avoid potential biases in retrospective studies and establish a base for the evaluation of risks and benefits of oral contraceptives after they became widely used, Frank Speizer, of Harvard Medical School, initiated the Nurses’ Health Study (NHS) in 1976 - Speizer and colleagues enrolled 121,700 women aged 30–55 with the primary goal of investigating the long-term health consequences of oral-contraceptive use. -The Nurses’ Health Study has grown from a simple questionnaire- based study initiated in 1976 to a rich resource of information collected over 29 years. - Important details about lifestyle were collected throughout the study - as the study has progressed, blood samples and DNA from buccal cells have been collected and stored. Tumor samples have also been collected from participants who developed cancer. - Through analyses that integrate information from these various sources we are advancing our understanding of the causes of cancer and the potential for prevention.  The Nurses’ Health Study (cont.) - The Nurses' Health Study is the first large prospective cohort study of women with updated exposure assessment for a broad range of lifestyle factors, endogenous hormones and DNA, in relation to risk of cancer. - It assesses overall risks and benefits of lifestyle factors, evaluates them in relation to a range of cancers and chronic conditions, and mortality. -Using diet data, which was first collected in 1980, it showed that alcohol intake is related to an increased risk of breast cancer independent of other dietary factors. - The long-term use of multivitamins containing folate is associated with a reduced risk of colon cancer. This association was confirmed by findings from several studies, including the discovery that a functional polymorphism in the methylene-tetrahydrofolate reductase gene was associated with increased colon cancer risk and that plasma levels of folate were inversely related to colon cancer risk. - Both estrogen and testosterone levels in postmenopausal women are related to an increased risk of breast cancer. Because of this, plasma estrogen and testosterone levels are now being added to risk-prediction models for breast cancer. Evolution of the Nurses’ Health Study Starting with the questionaire of 1976 > developed a diet based questionaire > blood and urine collection > growing number of participants. Implications for prevention from the Women’s Health Study - A range of basic lifestyle choices, based on findings from the Nurses' Health Studies and other epidemiological investigations, indicate that more than 50% of cancer cases can be prevented by adopting a healthy lifestyle. - In addition to lowering cancer risk, these lifestyle habits enhance overall health. There are six key lifestyle choices that can be made to reduce cancer risk:  not smoking  being more physically active  maintaining a healthy weight and avoiding weight gain as an adult  eating a diet rich in fruits, vegetables, whole grains and fiber, and low in saturated and trans-fats  taking a multivitamin every day  avoiding long-term use of postmenopausal hormone therapy. The Women’s Health Study is the source of knowledge of the estrogen-associated risk factors for breast cancer - The timing of first birth (late) is associated with both short-term increase (5– 10 years following first birth) and a long-term decrease in breast cancer risk. - the timing of subsequent births modifies risk — the closer additional births are to the first, the lower the ultimate risk of breast cancer - The cumulative risk of developing breast cancer, up to the age of 70, has been determined for various patient profiles, and confidence intervals for these cumulative risks have been estimated. For example, history of benign breast disease, family history of breast cancer, and use of estrogen plus progestin therapies for 10 years or more all provide comparable increases in breast cancer risk, up to the age of 70. -Follow up analyses demonstrated that higher parity is protective against developing estrogen receptor-positive tumors but not estrogen-receptor- negative tumors, whereas - family history of disease equally increases the risk of developing any tumor type.  Interventional Trials  Another way of classifying trials is by their purpose. The U.S. National Institutes of Health (NIH) organizes trials into five (5) different types: 1. Prevention trials: look for better ways to prevent disease in people who have never had the disease or to prevent a disease from returning. These approaches may include medicines, vitamins, vaccines, minerals, or lifestyle changes. 2. Screening trials: test the best way to detect certain diseases or health conditions. 3. Diagnostic trials: conducted to find better tests or procedures for diagnosing a particular disease or condition. 4. Treatment trials: test experimental treatments, new combinations of drugs, or new approaches to surgery or radiation therapy. 5. Quality of life trials: explore ways to improve comfort and the quality of life for individuals with a chronic illness (a.k.a. Supportive Care trials). Other types of trials: Compassionate use trials: provide experimental therapeutics prior to final FDA approval to patients whose options with other remedies have been unsuccessful. Usually, case by case approval must be granted by the FDA Retrospective studies, population studies, quality of life and symptom management, prevention studies  Prevention trials  Cancer prevention remains the ideal strategy for reducing the burden of cancer on society.  Cancer prevention clinical trials represent the maturation of decades of epidemiologic and laboratory investigations, resulting in the identification of exogenous and endogenous factors that influence cancer risk.  Strategies have included lifestyle and medical approaches, as associated cancer risk factors are identified.  The NCI has supported cancer prevention clinical trials for the past 30 years using a systematic, phased approach to investigate natural and synthetic agents.  Prevention trials (cont.)  The phased approach included phase I dose finding studies, phase II studies of efficacy in a small population at high risk for either specific cancers or the presence of biomarkers and large, randomized, double-blinded, placebo-controlled trials conducted in a large population (phase III).  Those agents, including both food constituents and drugs, that proceed successfully through each of the first 2 phases represent the best hope for use in large phase III prevention clinical trials.  These trials aim to prevent, arrest, or reverse either the initiation phase of carcinogenesis or the progression of premalignant cells.  Prevention trials (cont.)  Epidemiologic evidence (cohort studies) suggested that lifestyle factors, such as tobacco smoking and nutrition, might account for 70% of cancer risk. More recent studies indicated that overweight and obesity contribute to cancer mortality—as much as 20% of cancer deaths in women and 14% of cancer deaths in men are associated with overweight and obesity.  Understanding the effect of positive lifestyle changes to reduce cancer risk from association studies has provided impetus to the NCI in designing and conducting interventional clinical studies and trials for cancer prevention.  For example, the inverse relationship between the intake of vegetables and fruits and cancer risk led to a search for bioactive natural and synthetic bioactive food components, hormonal factors, or other anticarcinogenic compounds for evaluation in clinical studies that could account for this finding).  The resulting cancer prevention clinical trials, represent a medical approach to address results from lifestyle studies and have become a significant strategic approach for reducing cancer risk.  Selected NCI phase I, II, and III chemoprevention trials for breast and prostate cancer Dietary factor Target organ No. patients Phase I Soy isoflavones Prostate 12 Soy isoflavones Prostate 24 Lycopene (tomato paste), 3 trials Prostate 75 Selenomethionine Prostate 32 Indole-3-carbinol Breast 18 Perillyl alcohol Breast 30 Perillyl alcohol Breast 24 Soy isoflavones Breast 24 Phase II Soy isoflavones Prostate (presurgical) 80 Doxercalciferol Prostate 100 Selenomethionine Prostate 60 1 -hydroxyvitamin D2 Prostate 60 Polyphenol E (green tea extract) Breast 288 Celecoxib Breast 110 Soy isoflavones Breast 45 Phase III Soy isoflavones and green tea Prostate 154 (2 arms) Selenomethionine Prostate 465 (2 arms) Selenium/vitamin E Prostate 32,000 (4 arms) Tamoxifen/raloxifene Breast 19,000 (2 arms)  Colorectal Neoplasia Prevention Trials Demonstrating.Preventive Effect in Cohorts at Risk for Sporadic CRC Study Intervention Cohort* Primary Results Baron et Calcium 3.0 gm/d v placebo x 4 930 patients with Patients with adenoma incidence: al, 199982 years prior adenoma 19% reduction ; adenoma number: 24% reduction Faivre and Calcium 2.0 gm/d v 3.8 gm/d 655 patients with Adenoma incidence: 34% Bonithon- ispaghula husk v placebo x 3 years prior adenoma; reduction (NS) with calcium; 67% Kopp, 35-75 years of increase with fiber 200296 age Clark et al, Selenium 200 µg/d v placebo x 6.4 1,312 patients CRC incidence: 58% reduction 199697 years with prior skin cancers Rossouw Conjugated equine estrogen 0.625 16,608 healthy CRC incidence: 37% reduction et al, mg/d + medroxyprogesterone women; 50-79 200270 acetate 2.5 mg/d v placebo x 5.2 years of age years Baron et Aspirin 81 v 325 mg/d v placebo x 1,211 patients 81 mg/d, adenoma incidence: 19% al, 200385 3 years with prior reduction ; advanced adenoma adenomas incidence: 41% reduction ; 325 mg/d, adenoma incidence: 4% reduction (NS); advanced adenoma incidence: 17% reduction (NS) Benamouz Lysine acetylsalicylate 160-300 291 patients with Persons with adenomas: 27% ig et al, mg/d v placebo x 4 years prior adenomas reduction (borderline significance, 200398 P =.08); adenoma > 1 cm in diameter: 83% reduction Abbreviations: NS, not statistically significant; CRC, colorectal cancer. * Number randomized. Hawk and Levin, Journal of Clinical Oncology. 23:378-91, 2005 All referenced trials were double-blind, randomized, and controlled. Statistically significant (P <.05). Interim result at 1 year; intended duration of intervention is 4 years  Tamoxifen and raloxifene trials  Early breast cancer prevention trials investigated tamoxifen to reduce breast cancer risk. Results were encouraging, but some adverse effects occurred.  The Breast Cancer Prevention Trial (BCPT) conducted by the National Surgical Adjuvant Bowel and Breast Project (NSABP), was the first large-scale prevention trial conducted by NCI.  BCPT was stopped after only 3.6 years when data showed a statistically significant benefit for tamoxifen: a 49% reduction in invasive breast cancer compared with controls and a 69% reduction in the occurrence of invasive ER breast tumors compared with no difference in the occurrence of ER tumors.  There was, however, a 2.5-fold increase in endometrial cancer in women—predominantly in women over age 50—taking tamoxifen, which raised concern in the medical community.  Tamoxifen and raloxifene trials (cont.)  A Study of Tamoxifen and Raloxifene (STAR) to compare raloxifene and tamoxifen in 19,000 high-risk postmenopausal women, ages 35 and older, in a randomized, double-blind, placebo trial.  STAR was funded by the NCI and supported by the Community Clinical Oncology Program (CCOP), an NCI mechanism to improve accrual to NCI-sponsored clinical trials with the involvement of community-based physicians.  STAR assessed the occurrence of noninvasive breast cancer, endometrial cancer, and cardiovascular events, with potential side effects of raloxifene and tamoxifen targeted as secondary end points.  Schema of the STAR trial, National Surgical Adjuvant Breast and Bowel Project P-2 Because the STAR experiment was double blind, the patients and Researchers could not know what Groups were taking Tamoxifen or Raloxifene. Both tamoxifen and raloxifen Were chosen because they are look  Participant Characteristics--NSABP STAR Trial Patient characteristics Were very similar. Vogel, V. G. et al. JAMA 2006;295:2727-2741. Copyright restrictions may apply.  Rates of Noninvasive Breast Cancer and Uterine Disease/Hysterectomy--NSABP STAR Trial Vogel, V. G. et al. JAMA 2006;295:2727-2741.  Cumulative Incidence of Invasive Uterine Cancer and Thromboembolic Events Vogel, V. G. et al. JAMA 2006;295:2727-2741. Uterine cancer looked at using raloxifene and tamoxifen. Raloxifene Decrease side effect and still prevented cancer. Copyright restrictions may apply.  Adherence to hormone therapy  In breast cancer prevention trials, tamoxifen was prematurely discontinued by 20–46% of the participants.  In clinical practice settings, 2 reports addressed longer duration (> 4 years) adherence to adjuvant tamoxifen use. In these, tamoxifen was prematurely discontinued by 30–50% of the patients.  Long-term hormone therapy adherence may represent an area limiting optimal breast cancer patient treatment. Chlebowski, Oncology 2006;71:1–9  Screening trials  Randomized screening trials are now accepted as the only unbiased method to evaluate the efficacy of screening.  The first randomized screening trial employed mammography and breast physical examination to screen for breast cancer and was largely responsible for the widespread introduction of mammography screening among women 50 to 69 years of age.  Screening trials (cont.)  In contrast, trials of lung cancer screening were largely responsible for the nonadoption of lung cancer screening in the general population worldwide.  That experience should have taught us a lesson to prevent the introduction of screening for other cancers without evidence from a properly designed randomized trial; but, unfortunately, based on evidence known to be subjected to bias (cancer detection, improved survival of screen-detected cases), widespread screening for prostate cancer using PSA blood testing has occurred, with an increase in prostate cancer incidence but no definitive evidence of benefit in terms of prostate cancer mortality. reduction.  Basic criteria for screening Aspect Requirement  Disorder Well defined  Prevalence Known  Natural history Medically important disorder for which there is an effective remedy available  Test choice Simple and safe  Test performance Distributions of test values in affected and unaffected individuals known.  Financial Cost-effective  Facilities Available or easily installed  Acceptability Procedures following a positive result are generally agreed and acceptable both to the screening authorities and to those screened  Equity Equity of access to screening services Strong et al. J Med Screen 2005;2:12–19  Design of a randomized screening trial to evaluate the sensitivity of a screening test for cervical cancer. Miller AB. Journal of Surgery. 30:1152-62, 2006  A meta-analysis of randomized and nonrandomized controlled trials of Hemoccult screening programs as an intervention for reducing mortality from colorectal cancer. Towler, et al, The Cochrane Library, Issue 4. Chichester UK, John Wiley & Sons, Ltd, 2004.  Prevalence of disease: 1/10, Screen 100,000 people. Test 90% sensitive and 90% specific Presence of disease + - 9,000 9,000 9,000/18,000 Positive predictive value = % of + = 50% people with positive test who have disease Result of test 1,000 81,000 81,000/82,000 Negative predictive value = % of - = 98.8 % people with negative test who don’t have disease 10,000 90,000 100,000 Sensitivity = Specificity = Total Percent of people Percent of people screened with disease without disease testing positive testing negative 9,000/10,000 81,000/90,000 Have to have high incidence of disease in order to have a reasonable screening test. If there is not a high incidence of disease in the screened population, there will be false positives. Population of 100,000 people, the prevalence of the disease being screened for is 1/10. Screening test is very good (90% sensitive and 90% specific).  Prevalence of disease: 1/100, Screen 100,000 people. Test 90% sensitive and 90% specific Presence of disease + - + 900 9,900 900/10,800 Positive predictive value = % of = 8.3% people with positive test who have disease Result of test - 100 89,100 89,100/89,200 Negative predictive value = % of = 99.9 % people with negative test who don’t have disease Prevalence of disease: 1/100, Screen 100,000 people. Test 90% sensitive and 90% specific 1,000 99,000 100,000 Sensitivity = Specificity = Total Percent of Percent of screened people with people disease testing without positive disease 900/1,000 testing negative 89,100/99,000  Prevalence of CT-detected nodules Author Cou No. No. noncalcified nodule (% CT No. ntry screened of patients) collimation detectors Kaneko et al. Japan 1,369 588* (17%) 10 1 (5) Nawa et al. Japan 7,956 2,099 (26%) 10 1 (11) Henschke et USA 1,000 233 (23%) 10 1 al. (6) Pastorino et Italy 1,035 284 (27%) 10 1 al. (12) Swensen et USA 1,520 782 (51%) 5 4 al. (10) McWilliams Canada 561 259 (46%) 7 (1.25) 1 (4) et al. (13) Diederich et Germany 817 350 (43%) 5 1 al. (9) * CT findings described as abnormal shadows rather than Jett JR. Clinical Cancer Research. 11(13 Pt 2):4988s- noncalcified nodule; 17% is the proportion of screened 4992s, 2005 participants given in the published report. Invasive procedures for benign lesions detected by CT screening Study n (%) Japan (Tokyo) 27 of 49 (55) Japan (Nagano) 9 of 43 (21) USA (Rochester, MN) 8 of 40 (20) Germany (Munster) 3 of 13 (23) Italy (Milan) 6 of 28 (21) Jett JR. Clinical Cancer Research. 11(13 Pt 2):4988s-4992s, 2005  Risk of cancer from diagnostic X-rays Attributable risk (%) Cases per year United Kingdom 0.6 700 Germany 1.5 2,049 USA 0.9 5,695 Japan 3.2 7,587 Jett JR. Clinical Cancer Research. 11(13 Pt 2):4988s-4992s, 2005  Quality of life trials  Symptom management trials have primary endpoints of symptom relief, which generally involve patient-reported outcomes  These are in contrast to treatment trials where primary endpoints are tumor response, survival and/or progression-free survival.  There are 3 key challenges to advance the state of the science of health-related quality of life (HRQOL) assessment in cancer symptom management trials.  1. Must have strong rationales for including HRQOL assessment in symptom management trials: what is to be gained to addition to the primary endpoint of symptom relief?  2. Models must delineate the relations among symptoms, function, dimensions/domains (eg, physical, emotional, social, etc) with global quality of life QOL.  3. Studies need to pay attention to measurement issues and analytic strategies for assessing HRQOL, specifically, issues of effect sizes, sample sizes, and power analyses for hypotheses relating to secondary HRQOL endpoints. Buchanan, CANCER October 1, 2007 / Volume 110 / Number 7  Flow of participants after randomisation Randomization will decrease the starting number of participants over time duration of the study. Welton, A. J et al. BMJ 2008;337:a1190 Copyright ©2008 BMJ Publishing Group Ltd.  Mean (SE) scores on health related quality of life as measured with women’s health questionnaire by treatment group Component Baseline One year Adjusted difference P value at one year (95% CI) Combined HRT Placebo Combined HRT Placebo (n=1043*) (n=1087*) (n=1043*) (n=1087)* Depression 0.803 (0.004) 0.797 (0.004) 0.803 (0.004) 0.805 (0.004) 0.00 (–0.01 to 0.01) 0.39 Somatic 0.755 (0.006) 0.764 (0.006) 0.775 (0.006) 0.781 (0.006) 0.00 (–0.02 to 0.01) 0.80 Memory 0.730 (0.010) 0.721 (0.009) 0.759 (0.009) 0.757 (0.009) 0.00 (–0.02 to 0.02) 0.88 Vasomotor 0.766 (0.012) 0.771 (0.011) 0.926 (0.007) 0.833 (0.010) 0.09 (0.07 to 0.12)