Natural history of adenocarcinoma-large cell carcinoma of the lung: conclusions from screening programs in New York and Baltimore

Multivariate distributions of clinical covariates at the time of cancer detection

Many screening trials conducted in the past have generated a wealth of interesting data. These data represent an invaluable source of information for furthering our knowledge about the natural history of the disease. The traditional approach to modeling cancer screening tends to describe the process of tumor development in only one dimension, that is, the time natural history. A broader methodological idea is to construct a stochastic model of cancer development and detection that yields the multivariate distribution of observable variables at the time of diagnosis. By focusing on such multivariate observations, rather than just on the age of patients at diagnosis, this idea seeks to invoke an additional source of information (available only at the time of detection) in order to improve an estimation of unobservable quantitative parameters of cancer latency. In this article, we discuss modeling techniques that make the above-mentioned problems approachable. A special focus is placed on analytical tools for deriving joint distributions of clinical covariates at the time of cancer detection under an arbitrary screening protocol. In addition, some future research avenues and public health implications of the proposed approach are discussed.

Comparing the benefits of screening for breast cancer and lung cancer using a novel natural history model

To estimate the impact of early detection of cancer, knowledge of how quickly primary tumors grow and at what size they shed lethal metastases is critical. We developed a natural history model of cancer to estimate the probability of disease-specific cure as a function of tumor size, the tumor volume doubling time (TVDT), and disease-specific mortality reduction achievable by screening. The model was applied to non-small-cell lung carcinoma (NSCLC) and invasive ductal carcinoma (IDC), separately. Model parameter estimates were based on Surveillance Epidemiology and End Results (SEER) cancer registry datasets and validated on screening trials. Compared to IDC, NSCLC is estimated to have a lower probability of disease-specific cure at the same detected tumor size, shed lethal metastases at smaller sizes (median: 19 mm for IDC versus 8 mm for NSCLC), have a TVDT that is almost half as long (median: 252 days for IDC versus 134 days for NSCLC). Consequently, NSCLC is associated with a lower mortality reduction from screening at the same screen detection threshold and screening interval. In summary, using a similar natural history model of cancer, we quantify the disease-specific curability attributable to screening for breast cancer, and separately lung cancer, in terms of the TVDT and onset of lethal metastases.

Primary care radiography in the early diagnosis of lung cancer

Finding an abnormality on a plain chest radiograph is usually the first definite evidence of a lung cancer, so this investigation is currently pivotal in the diagnosis of the disease. Although the National Institute for Clinical Excellence (NICE) has produced guidance on when a chest radiograph should be done for putative lung cancer presentations, cancer will usually be only one of a number of possible diagnoses, so this is somewhat artificial. Neither is there any evidence that obtaining a chest radiograph for these features leads to an improved outcome. Another major concern is the poor public awareness of the symptoms for which a chest radiograph is recommended. This article discusses the role of the chest radiograph in the early diagnosis of lung cancer with particular emphasis on the limited value of a single negative result and on the potential implications of interventions to increase the number of chest radiographs done in primary care.

Randomized controlled trials of the efficacy of lung cancer screening by sputum cytology revisited: a combined mortality analysis from the Johns Hopkins Lung Project and the Memorial Sloan-Kettering Lung Study

BACKGROUND

: Two randomized controlled trials of lung cancer screening initiated in the 1970s, the Johns Hopkins Lung Project and the Memorial Sloan-Kettering Lung Study, compared 1 arm that received annual chest X-ray and 4-monthly sputum cytology (dual-screen) to a second arm that received annual chest X-ray only. Previous publications from these trials reported similar lung cancer mortality between the 2 groups. However, these findings were based on incomplete follow-up, and each trial on its own was underpowered to detect a modest mortality benefit.

METHODS

: The authors estimated the efficacy of lung cancer screening with sputum cytology in an intention-to-screen analysis of lung cancer mortality, using combined data from these trials (n = 20,426).

RESULTS

: Over (1/2) of squamous cell lung cancers diagnosed in the dual-screen group were identified by cytology; these cancers tended to be more localized than squamous cancers diagnosed in the X-ray only arm. After 9 years of follow-up, lung cancer mortality was slightly lower in the dual-screen than in the X-ray only arm (rate ratio [RR], 0.88; 95% confidence interval [CI], 0.74-1.05). Reductions were seen for squamous cell cancer deaths (RR, 0.79; 95% CI, 0.54-1.14) and in the heaviest smokers (RR, 0.81; 95% CI, 0.67-1.00). There were also fewer deaths from large cell carcinoma in the dual-screen group, although the reason for this is unclear.

CONCLUSIONS

: These data are suggestive of a modest benefit of sputum cytology screening, although we cannot rule out chance as an explanation for these findings. Cancer 2009. (c) 2009 American Cancer Society.

Calibration methods used in cancer simulation models and suggested reporting guidelines

Increasingly, computer simulation models are used for economic and policy evaluation in cancer prevention and control. A model's predictions of key outcomes, such as screening effectiveness, depend on the values of unobservable natural history parameters. Calibration is the process of determining the values of unobservable parameters by constraining model output to replicate observed data. Because there are many approaches for model calibration and little consensus on best practices, we surveyed the literature to catalogue the use and reporting of these methods in cancer simulation models. We conducted a MEDLINE search (1980 through 2006) for articles on cancer-screening models and supplemented search results with articles from our personal reference databases. For each article, two authors independently abstracted pre-determined items using a standard form. Data items included cancer site, model type, methods used for determination of unobservable parameter values and description of any calibration protocol. All authors reached consensus on items of disagreement. Reviews and non-cancer models were excluded. Articles describing analytical models, which estimate parameters with statistical approaches (e.g. maximum likelihood) were catalogued separately. Models that included unobservable parameters were analysed and classified by whether calibration methods were reported and if so, the methods used. The review process yielded 154 articles that met our inclusion criteria and, of these, we concluded that 131 may have used calibration methods to determine model parameters. Although the term 'calibration' was not always used, descriptions of calibration or 'model fitting' were found in 50% (n = 66) of the articles, with an additional 16% (n = 21) providing a reference to methods. Calibration target data were identified in nearly all of these articles. Other methodological details, such as the goodness-of-fit metric, were discussed in 54% (n = 47 of 87) of the articles reporting calibration methods, while few details were provided on the algorithms used to search the parameter space. Our review shows that the use of cancer simulation modelling is increasing, although thorough descriptions of calibration procedures are rare in the published literature for these models. Calibration is a key component of model development and is central to the validity and credibility of subsequent analyses and inferences drawn from model predictions. To aid peer-review and facilitate discussion of modelling methods, we propose a standardized Calibration Reporting Checklist for model documentation.

Modeling of long-term screening for lung carcinoma

BACKGROUND

Results from the Mayo Lung Project (MLP), a randomized clinical trial for the early detection of lung carcinoma, were interpreted as proof that the early detection of lung carcinoma by chest X-ray does not reduce the mortality from this disease. Recent analysis of extended follow-up data from the MLP subjects found that after approximately 20 years there still was no apparent difference in lung carcinoma mortality between a study group and a control group.

METHODS

To view this result within context, the authors utilized a previously published simulation model of the MLP, with parametric values that were estimated at the time of the original publication based on the data collected by the MLP.

RESULTS

The model produced predictions of the extended follow-up statistics that were found to be consistent with the data published in the prior study. The authors believe this provides long-term validation for the model. Conversely, the same model demonstrated that had the study subjects been screened annually for the extended follow-up period, the difference in mortality would be noticeable, even with the low sensitivity of chest X-ray detection.

CONCLUSIONS

The results of current study strongly suggest that long-term screening with chest X-ray results in a reduction in lung carcinoma mortality. The limited extent of this benefit is the result of the low sensitivity of chest X-ray as a screening tool.

Design of the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial

The objectives of the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial are to determine in screenees ages 55-74 at entry whether screening with flexible sigmoidoscopy (60-cm sigmoidoscope) can reduce mortality from colorectal cancer, whether screening with chest X-ray can reduce mortality from lung cancer, whether screening men with digital rectal examination (DRE) plus serum prostate-specific antigen (PSA) can reduce mortality from prostate cancer, and whether screening women with CA125 and transvaginal ultrasound (TVU) can reduce mortality from ovarian cancer. Secondary objectives are to assess screening variables other than mortality for each of the interventions including sensitivity, specificity, and positive predictive value; to assess incidence, stage, and survival of cancer cases; and to investigate biologic and/or prognostic characterizations of tumor tissue and biochemical products as intermediate endpoints. The design is a multicenter, two-armed, randomized trial with 37,000 females and 37,000 males in each of the two arms. In the intervention arm, the PSA and CA125 tests are performed at entry, then annually for 5 years. The DRE, TVU, and chest X-ray exams are performed at entry and then annually for 3 years. Sigmoidoscopy is performed at entry and then at the 5-year point. Participants in the control arm follow their usual medical care practices. Participants will be followed for at least 13 years from randomization to ascertain all cancers of the prostate, lung, colorectum, and ovary, as well as deaths from all causes. A pilot phase was undertaken to assess the randomization, screening, and data collection procedures of the trial and to estimate design parameters such as compliance and contamination levels. This paper describes eligibility, consent, and other design features of the trial, randomization and screening procedures, and an outline of the follow-up procedures. Sample-size calculations are reported, and a data analysis plan is presented.

CT screening for lung cancer

Recommendations against screening for lung cancer were based on the lack of a reduction in mortality of the screened group as compared with the control group in randomized control trials. These results were interpreted as showing that early detection of lung cancer as a result of screening did not decrease the mortality rate compared with detection after presentation of symptoms for the populations being screened. Evidence, however, shows that earlier-stage intervention leads to substantially higher rates of survival. Screening, therefore, is an effective means to prevent deaths from this otherwise fatal disease. This article discusses the evidence of both CT and chest radiograph screening.

Screening for lung cancer

Screening for lung cancer has remained controversial since the completion, more than two decades ago, of the three large randomized controlled trials, sponsored by the National Cancer Institute, which led to the recommendation against screening by major medical organizations. Details of the controversy are given, which include concerns about the study design, implementation, and analysis. New evidence about the potential benefit of screening with chest radiography that has emerged since the completion of those trials is reviewed, as well as the results of studies of CT screening for lung cancer.

Neural networks for the analysis of small pulmonary nodules

PURPOSE

Small pulmonary nodules can be readily detected by computed tomography (CT). The goal of this detection is to diagnose early lung cancer as the five year survival at this early stage is over 70% in contradistinction to the overall 5-year survival of around 10%. Critical to the efficacy of CT for early lung cancer detection is the ability to distinguish between benign and malignant nodules. We explored the usefulness of neural networks (NNs) to help in this differentiation.

METHODS

CT images of 28 pulmonary nodules, 14 benign and 14 malignant, each having a diameter less than 3 cm were selected. All were sufficiently malignant in appearance to require needle biopsy and surgery. The statistical-multiple object detection and location system (S-MODALS) NN technique developed for automatic target recognition (ATR) was used to differentiate between these benign and malignant nodules.

RESULTS

S-MODALS was able to correctly identify all but three benign nodules. S-MODALS classified a nodule as malignant because it looked similar to other malignant nodules. It identified the most similar nodules to display them to the radiologist. The specific features of the nodule that determined its classification were also shown, so that S-MODALS is not simply a "black box" technique but gives insight into the NN diagnostics.

CONCLUSION

This initial evaluation of S-MODALS NNs using pulmonary nodules whose CT features were very suspicious for lung cancer demonstrated the potential to reduce the number of biopsies without missing malignant nodules. S-MODALS performed well, but additional optimization of the techniques specifically for CT images would further enhance its performance.

Early detection in lung cancer. Case finding and screening

There is general agreement that the most effective approach to lung cancer is primary prevention--stop smoking. Richards has proposed the MVROCST--the Monosyllabic Verbal Response Office Cancer Screening Test: "Do you smoke?" If "yes," intervene. If "no," move on. Ample evidence exists that a clear message from a physician to a patient about the importance of stopping smoking makes a difference. In contrast to the maze of arguments and data on early detection, this is something that each physician clearly can and should do. A reduced risk for lung cancer may begin as early as 5 years after cessation of cigarette use. Huuskonen has proposed conceptualizing screening as a coordinated intervention with the goal of identifying populations at risk and working to modify that risk. Primary prevention should be central to any efforts to reduce mortality from lung cancer, and attention to this area needs to increase despite the difficulties and frustration. Despite declining percentages of smokers in the population as a whole, it is estimated that more than 3000 teenagers become regular smokers each day in the United States. In this environment, the question of whether to recommend a CXR or sputum for early detection is not going to disappear in the near future. The NCI has recognized the persistent and important nature of this debate and is currently funding the Prostate, Lung, Colon and Ovary Cancer Screening Trial. This is a large and powerful randomized study of men and women aged 60 to 74. The lung cancer arm is designed to look at the usefulness of a yearly CXR intervention in reducing cancer-specific mortality. The overall power of the study (based on national mortality data) is 0.99 for a 15% reduction in lung cancer mortality and 0.89 for a 10% reduction, with differentially better sensitivity in men than women. The study is currently in progress at multiple sites and will be completed over the next 12 to 14 years. In the meantime, what is the right approach? It is useful in considering this question to return to the concepts of early detection, screening, and case finding. 1. Early detection in lung cancer remains a concept of uncertain applicability because of the unknowns and variability in the natural history of the disease. The available, accessible, and acceptable detection tools appear to be inadequate by current evidence. This is not a static field, however, and new work in the area of biomarkers carries promise for significantly more sensitive and specific techniques. Tockman and colleagues conclude that early detection is conceptually sound, although not currently practical, and further research may expand the role of intervention. In the end, a judgement on early detection in lung cancer must be linked to the proposed setting--screening or case finding. 2. Screening, defined as the application of a test to the general population to define disease risk further with the implied benefit of improved treatment and outcome, cannot be recommended for lung cancer. This is the perspective of the major organizations cited previously, and it is based on admittedly imperfect but nonetheless convincing data. 3. Case finding, the situation of the patient who seeks care and is available for informed discussion and negotiation on possible testing, is a potentially different situation.(ABSTRACT TRUNCATED)

Pulmonary cancer from the general practitioner's point of view. Experience from the health centre area of Kungsbacka, Sweden

OBJECTIVE

To study the incidence of pulmonary cancer in a community with special reference to the diagnostic process and the role of the general practitioner.

DESIGN

Study of the records of all patients within the community with pulmonary cancer reported to the Swedish Cancer Registry during the years 1980-1984.

SETTING

The community of Kungsbacka in southwestern Sweden with about 48,000 inhabitants.

PARTICIPANTS

40 subjects with pulmonary cancer.

OUTCOME MEASURES

Incidence, main symptoms, level of care, doctor delay, survival rate.

RESULTS

The incidence was 16 per 100,000 per year. Most patients first visited a general practitioner. Most common initial symptom was cough. Mean doctor delay was 12.5 weeks. The five-year mortality rate was 95%.

CONCLUSION

The high mortality emphasizes the importance of an early diagnosis. The general practitioners are very important in the diagnostic process.

Radiographic screening for cancer. Proposed paradigm for requisite research

Computed tomography (CT) imaging as an excellent approach to the detection and characterization of small solitary pulmonary nodules (SSPN) raises three questions: (1) How often does CT imaging lead to detection of SSPN? (2) How often is such an SSPN malignant? (3) If malignant, how curable is it? The first question pertains to decisions about screening use of CT (clinical or mass screening), the second to decisions about screening for SSPN and diagnosis of malignancy given SSPN, and the third--in the context of known curability at ordinary clinical diagnosis--to decisions about screening for SSPN, diagnosis given SSPN and intervention given malignant SSPN. We present a three component study design that addresses these questions. The first is directed primarily to the first question. Some 1000 persons at high risk for lung cancer will be screened for SSPN using screening-type CT. The primary aim is to determine the prevalence of CT-detectable SSPN as a joint function of risk-relevant aspects of the person. The second component addresses the prevalence of malignancy among the detected cases of SSPN. To develop the prevalence function, a larger series of CT-detected SSPN will be obtained by developing a multi-center SSPN "registry." A subsequent, third component will focus on the registered cases of malignant SSPN screening incidentally detected and address their curability on the basis of long-term follow-up. This design, in lieu of a randomized trial, may represent a new paradigm for applied research on radiologic technologies in cancer screening, given its advantages in terms of research efficiency and implications to decisions about diagnostic workup and therapeutic intervention.

Limitation of annual screening chest radiography for the diagnosis of lung cancer. A retrospective study

BACKGROUND

The incidence and mortality rates of lung cancer have been increasing in many countries. However, the true effectiveness of screening for lung cancer is still controversial. This study aimed to examine the growth pattern of lung cancer and to evaluate the efficacy of screening.

METHODS

The authors linked the records of annual radiologic screening to cancer registry data and conducted a retrospective follow-up study of radiographs in all patients with lung cancer arising in a population undergoing screening.

RESULTS

Among a total of 305,934 participants, screening detected 206 lung cancers, 103 of which were Stage I disease. Seventy-one of the 131 adenocarcinomas were Stage I, and 58% of them showed evidence of cancer for 2 years on a retrospective review of radiographs. Presentation as small faint lesions overlapping the normal chest structures delayed the early detection of adenocarcinoma. The overall sensitivity of screening was 70%, 52% for squamous cell carcinoma and 50% for small cell carcinoma. Rapidly growing Stage II-IV tumors without retrospective evidence of cancer on previous radiographs accounted for most of the cancers detected during the intervals between screening.

CONCLUSIONS

Both the low detectability of Stage I adenocarcinoma and the late recognition of rapidly growing small cell and squamous cell carcinomas reduced the effectiveness of screening. More effective imaging methods and an antismoking campaign are required to reduce lung cancer mortality.

Screening for lung cancer. The Mayo Lung Project revisited

BACKGROUND

The Mayo Lung Project (MLP) reported lung cancer incidence and mortality in a population offered chest radiographs and sputum cytologic screening examinations every 4 months and a population offered only the Mayo Clinic advice to undergo annual examination. No mortality benefit attributable to screening was observed after 6 years of observation and at least 1 year of follow-up.

METHODS

The authors describe a simulation study designed to estimate from Mayo data the parameters in a mathematical model of the natural history of lung cancer and to estimate the potential benefit associated with periodic screening of high-risk people starting at 45 years of age.

RESULTS

It was found that the mean duration of Stage I non-small cell lung cancer is at least 4 years and that rates of Stage I detectability and curability are less than 25% and 35%, respectively.

CONCLUSIONS

A trial of the magnitude, duration, and contamination of the MLP would have a less than 20% probability of showing significant benefit from screening; however, long-term annual screening might result in a modest decrease in lung cancer mortality, ranging from 0% to 13%. A greater benefit would accrue from improved detection and treatment.

Early lung cancer as a potential target for chemoprevention

Carcinoma of the lung is the most common cause of death from cancer in the United States. In considering lung cancer for possible chemoprevention trials, we have analyzed the data collected by the collaborative NCI program on early lung cancer. The data indicate that at least 12 years of study of 80,000 people at risk for lung cancer (adult male cigarette smokers) would be required to establish a 25% reduction in squamous carcinoma of the lung. No intermediate markers of developing lung cancer are presently available to shorten the observation period. It is concluded that a study of the magnitude required is not feasible at the present time.

Innovative therapies in hematology and oncology

As one can see, there are a number of new and exciting advances in the prevention, detection, and treatment of various malignancies. We anxiously await the future to see the exact integration of the tremendous advances taking place in the cellular and molecular biopsy of disease and its application to the therapy of patients.

The effect of surgical treatment on survival from early lung cancer. Implications for screening

We assessed the effect of surgery on survival from stage I non-small-cell lung cancer based on data collected in these screening programs. The majority of patients diagnosed in each program were treated by surgical resection, but 5 percent of the Sloan-Kettering group, 21 percent of the Hopkins group and 11 percent of the Mayo group failed to receive surgical treatment. Approximately 70 percent of the stage I patients in each program who were treated surgically survived more than five years, but there were only two five-year survivors among those who did not have surgery. We conclude that patients with lung cancers detected in stage I by chest x-ray film and treated surgically have a good chance of remaining free of disease for many years. Those stage I lung cancers which are not resected progress and lead to death within five years. Therefore, every effort should be made to detect and treat lung cancer early in high-risk populations.

Evaluation of cancer prevention strategies by computerized simulation model: an approach to lung cancer

A computerized simulation model was developed to evaluate the potential impact of primary and secondary prevention on lung cancer mortality in Japan. The natural history of lung cancer was modeled as a Markovian stochastic process from cancer-free to preclinical, clinical, and finally to terminal states. The increase in mortality rate of lung cancer among males aged 75 to 79 years has been the major force of increase in the total number of lung cancer deaths in Japan. The simulation showed that this tendency would continue until the late 1990s, presumably due to the increase in the proportion of ever-smokers in that cohort, who started smoking at an earlier age than did prior generations. It was shown that the number of lung cancer deaths can be reduced either by smoking cessation or screening programs, and that the reduction is proportional to the increase in the annual smoking-cessation rate and to the annual increment in the screening rate. However, only two to three percent reduction of lung cancer deaths in the year 2001 can be expected when the annual smoking-cessation rate is raised from the current value of 0.46 percent to five percent during the period from 1991 to 2000 or when the screening rate is increased by three percent annually for the same period.

Screening for lung cancer. Is it worthwhile?

Prevention of lung cancer remains the best method of decreasing lung cancer mortality. Patients who smoke should be urged to quit, and children, teenagers, and young adults must not begin smoking. At high risk are smokers, especially those under 40 years of age who may have smoked two to four packs of cigarettes per day for 20 years; persons who have had a previous lung cancer; patients with bullous emphysema; patients with asbestosis; and patients with evidence of chronic airflow obstruction. Although radiographic screening may detect lung cancer earlier and lead to increased 5-year survival rates, it does not reduce lung cancer mortality rates.