The Dana-Farber CISNET Model for Breast Cancer Screening Strategies: An Update.

BACKGROUND: We present updated features to a model developed by Dana-Farber investigators within the Cancer Intervention and Surveillance Modeling Network (CISNET). The initial model was developed to evaluate the impact of mammography screening strategies. METHODS: This major update includes the incorporation of ductal carcinoma in situ (DCIS) as part of the natural history of breast cancer. The updated model allows DCIS in the pre-clinical state to regress to undetectable early-stage DCIS, or to transition to invasive breast cancer, or to clinical DCIS. We summarize model assumptions for DCIS natural history and model parameters. Another new development is the derivation of analytical expressions for overdiagnosis. Overdiagnosis refers to mammographic identification of breast cancer that would never have resulted in disease symptoms in the patient's remaining lifetime (i.e., lead time longer than residual survival time). This is an inevitable consequence of early detection. Our model uniquely assesses overdiagnosis using an analytical formulation. We derive the lead time distribution resulting from the early detection of invasive breast cancer and DCIS, and formulate the analytical expression for overdiagnosis. RESULTS: This formulation was applied to assess overdiagnosis from mammography screening. Other model updates involve implementing common model input parameters with updated treatment dissemination and effectiveness, and improved mammography performance. Lastly, the model was expanded to incorporate subgroups by breast density and molecular subtypes. CONCLUSIONS: The incorporation of DCIS and subgroups and the derivation of an overdiagnosis estimation procedure improve the model for evaluating mammography screening programs.

Modeling the impact of population screening on breast cancer mortality in the United States.

OBJECTIVE: Optimal US screening strategies remain controversial. We use six simulation models to evaluate screening outcomes under varying strategies. METHODS: The models incorporate common data on incidence, mammography characteristics, and treatment effects. We evaluate varying initiation and cessation ages applied annually or biennially and calculate mammograms, mortality reduction (vs. no screening), false-positives, unnecessary biopsies and over-diagnosis. RESULTS: The lifetime risk of breast cancer death starting at age 40 is 3% and is reduced by screening. Screening biennially maintains 81% (range 67% to 99%) of annual screening benefits with fewer false-positives. Biennial screening from 50-74 reduces the probability of breast cancer death from 3% to 2.3%. Screening annually from 40 to 84 only lowers mortality an additional one-half of one percent to 1.8% but requires substantially more mammograms and yields more false-positives and over-diagnosed cases. CONCLUSION: Decisions about screening strategy depend on preferences for benefits vs. potential harms and resource considerations.

Effect of screening mammography on breast-cancer mortality in Norway.

BACKGROUND: A challenge in quantifying the effect of screening mammography on breast-cancer mortality is to provide valid comparison groups. The use of historical control subjects does not take into account chronologic trends associated with advances in breast-cancer awareness and treatment. METHODS: The Norwegian breast-cancer screening program was started in 1996 and expanded geographically during the subsequent 9 years. Women between the ages of 50 and 69 years were offered screening mammography every 2 years. We compared the incidence-based rates of death from breast cancer in four groups: two groups of women who from 1996 through 2005 were living in counties with screening (screening group) or without screening (nonscreening group); and two historical-comparison groups that from 1986 through 1995 mirrored the current groups. RESULTS: We analyzed data from 40,075 women with breast cancer. The rate of death was reduced by 7.2 deaths per 100,000 person-years in the screening group as compared with the historical screening group (rate ratio, 0.72; 95% confidence interval [CI], 0.63 to 0.81) and by 4.8 deaths per 100,000 person-years in the nonscreening group as compared with the historical nonscreening group (rate ratio, 0.82; 95% CI, 0.71 to 0.93; P<0.001 for both comparisons), for a relative reduction in mortality of 10% in the screening group (P=0.13). Thus, the difference in the reduction in mortality between the current and historical groups that could be attributed to screening alone was 2.4 deaths per 100,000 person-years, or a third of the total reduction of 7.2 deaths. CONCLUSIONS: The availability of screening mammography was associated with a reduction in the rate of death from breast cancer, but the screening itself accounted for only about a third of the total reduction. (Funded by the Cancer Registry of Norway and the Research Council of Norway.)

Effects of mammography screening under different screening schedules: model estimates of potential benefits and harms.

BACKGROUND: Despite trials of mammography and widespread use, optimal screening policy is controversial. OBJECTIVE: To evaluate U.S. breast cancer screening strategies. DESIGN: 6 models using common data elements. DATA SOURCES: National data on age-specific incidence, competing mortality, mammography characteristics, and treatment effects. TARGET POPULATION: A contemporary population cohort. TIME HORIZON: Lifetime. PERSPECTIVE: Societal. INTERVENTIONS: 20 screening strategies with varying initiation and cessation ages applied annually or biennially. OUTCOME MEASURES: Number of mammograms, reduction in deaths from breast cancer or life-years gained (vs. no screening), false-positive results, unnecessary biopsies, and overdiagnosis. RESULTS OF BASE-CASE ANALYSIS: The 6 models produced consistent rankings of screening strategies. Screening biennially maintained an average of 81% (range across strategies and models, 67% to 99%) of the benefit of annual screening with almost half the number of false-positive results. Screening biennially from ages 50 to 69 years achieved a median 16.5% (range, 15% to 23%) reduction in breast cancer deaths versus no screening. Initiating biennial screening at age 40 years (vs. 50 years) reduced mortality by an additional 3% (range, 1% to 6%), consumed more resources, and yielded more false-positive results. Biennial screening after age 69 years yielded some additional mortality reduction in all models, but overdiagnosis increased most substantially at older ages. RESULTS OF SENSITIVITY ANALYSIS: Varying test sensitivity or treatment patterns did not change conclusions. LIMITATION: Results do not include morbidity from false-positive results, patient knowledge of earlier diagnosis, or unnecessary treatment. CONCLUSION: Biennial screening achieves most of the benefit of annual screening with less harm. Decisions about the best strategy depend on program and individual objectives and the weight placed on benefits, harms, and resource considerations. PRIMARY FUNDING SOURCE: National Cancer Institute.

MULTI-CENTER CLINICAL TRIALS: RANDOMIZATION AND ANCILLARY STATISTICS.

The purpose of this paper is to investigate and develop methods for analysis of multi-center randomized clinical trials which only rely on the randomization process as a basis of inference. Our motivation is prompted by the fact that most current statistical procedures used in the analysis of randomized multi-center studies are model based. The randomization feature of the trials is usually ignored. An important characteristic of model based analysis is that it is straightforward to model covariates. Nevertheless in nearly all model based analyses, the effects due to different centers and, in general, the design of the clinical trials are ignored. An alternative to a model based analysis is to have analyses guided by the design of the trial. Our development of design based methods allows the incorporation of centers as well as other features of the trial design. The methods make use of conditioning on the ancillary statistics in the sample space generated by the randomization process. We have investigated the power of the methods and have found that, in the presence of center variation, there is a significant increase in power. The methods have been extended to group sequential trials with similar increases in power.

Mortality modeling of early detection programs.

Consider a group of subjects who are offered an opportunity to receive a sequence of periodic special examinations for the purpose of diagnosing a chronic disease earlier relative to usual care. The mortality for the early detection group is to be compared with a group receiving usual care. Benefit is reflected in a potential reduction in mortality. This article develops a general probability model that can be used to predict cumulative mortality for each of these groups. The elements of the model assume (i) a four-state progressive disease model in which a subject may be in a disease-free state (or a disease state that cannot be detected), preclinical disease state (capable of being diagnosed by a special exam), clinical state (diagnosis by usual care), and a death state; (ii) age-dependent transitions into the states; (iii) age-dependent examination sensitivity; (iv) age-dependent sojourn time in each state; and (v) the distribution of disease stages on diagnosis conditional on modality of detection. The model may be used to (i) compare mortality rates for different screening schedules; (ii) explore potential benefit of subpopulations; and (iii) compare relative reductions in disease-specific mortality due to advances and dissemination of both treatment and early detection screening programs.

A stochastic model for predicting the mortality of breast cancer.

Consider a cohort of women, identified by year of birth, some of whom will eventually be diagnosed with breast cancer. A stochastic model is developed for predicting the U.S. breast cancer mortality that depends on advances in therapy and dissemination of mammographic screening. The predicted mortality can be compared with the same cohort having usual care with no screening program and absence of modern therapy, or a cohort in which only a proportion participate in a screening program and have modern therapy. The model envisions that a woman may be in four health states: i.e., 1) no disease or breast cancer that cannot be diagnosed (S0), 2) preclinical state (Sp), 3) clinical state (Sc), and 4) disease-specific death (Sd). The preclinical disease refers to breast cancer that is asymptomatic but that may be diagnosed with a special exam. The clinical state refers to symptomatic disease diagnosed under usual care. One of the basic assumptions of the model is that the disease is progressive; i.e., the transitions for the first three states are S0-->Sp-->Sc. The other basic assumption is that any reduction in mortality associated with earlier diagnosis is due to a stage shift in diagnosis; i.e., early diagnosis results in a larger proportion of earlier stage patients. The model is used to predict changes in female breast cancer mortality in the U.S. women for 1975-2000. The model is general and may predict mortality for other chronic diseases that satisfy the two basic assumptions.

Biostatisticians, biostatistical science and the future.

Biostatistical Science is in a 'golden period'. The definition of Biostatistical Science is the application of statistics, probability, mathematics and computing to advance our understanding of the subject matter in the biomedical sciences. Our field is experiencing unparallel developments due of the advances in communication and computing. We are becoming more global, we have resources which can expand educational opportunities for distant learning; the growth of quantitative methods in the biomedical sciences has made biostatistical science a key component in many research areas. What about the future? Are we receptive to change as many new scientific areas expand? Will the interaction between academia and industry likely to grow-especially in the training of future practitioners of biostatistical science. This paper discusses some of challenges facing our profession if we are to continue to be relevant in the biomedical sciences.

Effect of screening and adjuvant therapy on mortality from breast cancer.

BACKGROUND: We used modeling techniques to assess the relative and absolute contributions of screening mammography and adjuvant treatment to the reduction in breast-cancer mortality in the United States from 1975 to 2000. METHODS: A consortium of investigators developed seven independent statistical models of breast-cancer incidence and mortality. All seven groups used the same sources to obtain data on the use of screening mammography, adjuvant treatment, and benefits of treatment with respect to the rate of death from breast cancer. RESULTS: The proportion of the total reduction in the rate of death from breast cancer attributed to screening varied in the seven models from 28 to 65 percent (median, 46 percent), with adjuvant treatment contributing the rest. The variability across models in the absolute contribution of screening was larger than it was for treatment, reflecting the greater uncertainty associated with estimating the benefit of screening. CONCLUSIONS: Seven statistical models showed that both screening mammography and treatment have helped reduce the rate of death from breast cancer in the United States.

Robust modeling in screening studies: estimation of sensitivity and preclinical sojourn time distribution.

In early-detection clinical trials, quantities such as the sensitivity of the screening modality and the preclinical duration of the disease are important to describe the natural history of the disease and its interaction with a screening program. Assume that the schedule of a screening program is periodic and that the sojourn time in the preclinical state has a piecewise density function. Modeling the preclinical sojourn time distribution as a piecewise density function results in robust estimation of the distribution function. Our aim is to estimate the piecewise density function and the examination sensitivity using both generalized least squares and maximum likelihood methods. We carried out extensive simulations to evaluate the performance of the methods of estimation. The different estimation methods provide complimentary tools to obtain the unknown parameters. The methods are applied to three breast cancer early-detection trials.

Early detection of disease and scheduling of screening examinations.

Special examinations exist for many chronic diseases, which can diagnose the disease while it is asymptomatic, with no signs or symptoms. The earlier detection of disease may lead to more cures or longer survival. This possibility has led to public health programs which recommend populations to have periodic screening examinations for detecting specific chronic diseases, for example, cancer, diabetes, cardiovascular disease and so on. Such examination schedules when embedded in a public health program are invariably costly and are ordinarily not chosen on the basis of possible trade-offs in costs and benefits for different screening schedules. The possible candidate number of examination schedules is so large that it is not feasible to carry out clinical trials to compare different schedules. Instead, this problem can be investigated by developing a theoretical model which can predict the eventual disease specific mortality for different examination schedules. We have developed such a model. It is a stochastic model which assumes that i) the natural history of the disease is progressive and ii) any benefit from earlier diagnosis is due to a change in the distribution of disease stages at diagnosis (stage shift). The model is general and can be applied to any chronic disease which satisfies our two basic assumptions. We discuss the basic ideas of schedule sensitivity and lifetime schedule sensitivity and its relation to the reduction in disease specific mortality. Our theory is illustrated by applications to breast cancer screening. The investigation of schedules compares not only examination schedules with equal intervals between examinations but also staggered schedules using the threshold method. (Examinations are carried out when an individual's risk status reaches a preassigned threshold value.).

Planning of randomized early detection trials.

Consider a randomized clinical trial to evaluate the benefit of screening an asymptomatic population. Suppose that the subjects are randomized into a usual care and a study group. The study group receives one or more periodic early detection examinations aimed at diagnosing disease early, when there are no signs or symptoms. Early detection clinical trials differ from therapeutic trials in that power is affected by: i) the number of exams, ii) the time between exams and iii) the ages at which exams will be given. These design options do not exist in therapeutic trials. Furthermore; long-term follow-up may result in a reduction of power. In general, power increases with number of examinations, and the optimal follow-up time is dependent on the spacing between examinations. Clinical trials in which the usual care group receives benefit are also discussed. Two designs are discussed, for example the 'up-front design' in which all subjects receive an initial exam and then are randomized to the usual care and study groups and the 'close-out design' in which the usual care group receives an exam which is timed to be given at the same time as the last exam in the study group. Both families of designs significantly reduce the power. Power calculations are made for two clinical trials, which actually used these two designs.

Overdiagnosis in early detection programs.

Overdiagnosis refers to the situation where a screening exam detects a disease that would have otherwise been undetected in a person's lifetime. The disease would have not have been diagnosed because the individual would have died of other causes prior to its clinical onset. Although the probability of overdiagnosis is an important quantity for understanding early detection programs it has not been rigorously studied. We analyze an idealized early detection program and derive the mathematical expression for the probability of overdiagnosis. The results are studied numerically for prostate cancer and applied to a variety of screening schedules. Our investigation indicates that the probability of overdiagnosis is remarkably high.

Forward and backward recurrence times and length biased sampling: age specific models.

Consider a chronic disease process which is beginning to be observed at a point in chronological time. The backward recurrence and forward recurrence times are defined for prevalent cases as the time with disease and the time to leave the disease state, respectively, where the reference point is the point in time at which the disease process is being observed. In this setting the incidence of disease affects the recurrence time distributions. In addition, the survival of prevalent cases will tend to be greater than the population with disease due to length biased sampling. A similar problem arises in models for the early detection of disease. In this case the backward recurrence time is how long an individual has had disease before detection and the forward recurrence time is the time gained by early diagnosis, i.e., until the disease becomes clinical by exhibiting signs or symptoms. In these examples the incidence of disease may be age related resulting in a non-stationary process. The resulting recurrence time distributions are derived as well as some generalization of length-biased sampling.

The theory of case-control studies for early detection programs.

Although case-control studies are widely used for evaluating the benefit of early detection programs, the theoretical basis underlying this application has not been well developed. In this paper the properties of chronic disease case-control studies for evaluating early detection programs are investigated. An idealized case-control study is analyzed and the theoretical expression for the odds ratio associated with the benefit of screening is derived. The odds ratio is related to the natural history of disease and the screening program. Our results indicate that case-control studies result in odds ratios that are surprisingly close to unity and consequently have low power.

Modeling and optimization in early detection programs with a single exam.

The choice of timing of screening examinations is an important element in determining the efficacy of strategies for the early detection of occult disease. In this article, we describe a flexible decision-making framework for the design of early detection programs, and we investigate the choice of timing when each individual in the screening program is examined only once. We focus on the theoretical relation between the optimal examination time and the distributions of sojourn times in health-related states. Specifically, we derive closed-form solutions of the optimal age using two specifications of utility functions, discuss the effects of natural history and utility specifications on the optimal solution, and present an application to early detection of colorectal cancer by once-only sigmoidoscopy or colonoscopy.