Estimating distribution of age of the onset of detectable asymptomatic cancer

How much can screening reduce colorectal cancer mortality in Japan? Scenario-based estimation by microsimulation

BACKGROUND

Screening is one of the effective interventions for the reduction of colorectal cancer mortality. Though the Japanese government recommends faecal occult blood test and colonoscopy as a follow-up examination following a diagnosis, both participation rates have not been so high and the national mortality rate has not shown a clear decreasing trend.

METHODS

Microsimulation models simulate the life histories of a large population of individuals under various scenarios. In this study, we applied a microsimulation model to estimate the reduction of colorectal cancer mortality based on screening scenarios.

RESULTS

The effect of reducing the age-standardized mortality rate for colorectal cancer was estimated at 9.4% for men and 6.0% for women under the scenario which calls for 50% participation in faecal occult blood test and 90% participation of follow-up examination. This scenario corresponds to the goal setting for screening in the third-term of the Basic Plan to Promote Cancer Control Programs in Japan.

CONCLUSIONS

Our microsimulation model was found to be useful in estimating the mortality reduction effect of cancer control policy. Such modelling techniques can be utilized to develop effective and optimal cancer control programs.

A multistate survival model of the natural history of cancer using data from screened and unscreened population

One of the main aims of models using cancer screening data is to determine the time between the onset of preclinical screen-detectable cancer and the onset of the clinical state of the cancer. This time is called the sojourn time. One problem in using screening data is that an individual can be observed in preclinical phase or clinically diagnosed but not both. Multistate survival models provide a method of modeling the natural history of cancer. The natural history model allows for the calculation of the sojourn time. We developed a continuous-time Markov model and the corresponding likelihood function. The model allows for the use of interval-censored, left-truncated and right-censored data. The model uses data of clinically diagnosed cancers from both screened and nonscreened individuals. Parameters of age-varying hazards and age-varying misclassification are estimated simultaneously. The mean sojourn time is calculated from a micro-simulation using model parameters. The model is applied to data from a prostate screening trial. The simulation study showed that the model parameters could be estimated accurately.

Estimating the frequency of indolent breast cancer in screening trials

Cancer screening can detect cancer that would not have been detected in a patient's lifetime without screening. Standard methods for analyzing screening data do not explicitly account for the possibility that a fraction of tumors may remain latent indefinitely. We extend these methods by representing cancers as a mixture of those that progress to symptoms (progressive) and those that remain latent (indolent). Given sensitivity of the screening test, we derive likelihood expressions to simultaneously estimate (1) the rate of onset of preclinical cancer, (2) the average preclinical duration of progressive cancers, and (3) the fraction of preclinical cancers that are indolent. Simulations demonstrate satisfactory performance of the estimation approach to identify model parameters subject to precise specifications of input parameters and adequate numbers of interval cancers. In application to four breast cancer screening trials, the estimated indolent fraction among preclinical cancers varies between 2% and 35% when assuming 80% test sensitivity and varying specifications for the earliest time that participants could plausibly have developed cancer. We conclude that standard methods for analyzing screening data can be extended to allow some indolent cancers, but accurate estimation depends on correctly specifying key inputs that may be difficult to determine precisely in practice.

Nonparametric estimation of asymptomatic duration from a randomized prospective cancer screening trial

We propose a nonparametric estimation of preclinical duration distribution in cancer based on data from a randomized early detection trial. In cancer screening studies, the preclinical duration of a disease is of great interest for better understanding the natural history of the disease, and for developing optimal screening strategies. To estimate the sojourn time distribution nonparametrically, we first estimate the distribution of the age at onset of preclinical disease nonparametrically using data from the screening arm in a randomized screening trial, and the distribution for the age at onset of clinical disease from the control arm of the randomized screening trial. Finally, by using deconvolution the two estimated distributions lead to a nonparametric estimate of the distribution for the gap time between the onset of preclinical disease and the onset of clinical disease. We illustrate the methodology using data from a randomized breast cancer screening trial.

A Model-Based Comparison of Breast Cancer Screening Strategies: Mammograms and Clinical Breast Examinations

In screening for secondary prevention of breast cancer, clinical breast examination (CBE) combined with mammography may improve overall screening sensitivity compared with mammography alone. A systematic evaluation of the relative expenses and projected benefit of combining these two screening modalities is not presently available. We addressed this issue using a microsimulation model incorporating age-specific preclinical duration of the disease, age-specific sensitivities of the two modalities, age-specific incidence of the disease, screening strategy, and competing causes of mortality. We examined a total of 48 screening strategies, depending on the age range, the examination interval, and whether mammography or CBE is given at every one or two exam. Our results indicate that a biennial mammography can be cost-effective if coupled with annual CBE. For each screening interval and starting age, giving mammography every two exams and CBE at every exam has the lowest marginal cost per year of quality-adjusted life saved, whereas giving both at every exam has the highest. Comparing annual mammography and CBE to biennial mammography and annual CBE from 50 to 79, the total cost was reduced by 35%, whereas the marginal quality-adjusted life years only decreased by 12%. Similar reductions are observed for other starting ages. It is cost-effective to have a biennial mammography if coupled with an annual CBE. Annual mammography combined with CBE every 6 months will lead to a 41% increase in the quality-adjusted life years compared with annual mammography and CBE from 50 to 79, whereas the total cost increases by 30%.

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.