The LLP risk model: An individual risk prediction model for lung cancer

Roy Castle Lung Cancer Research Programme, University of Liverpool Cancer Research Centre, Liverpool, L3 9TA, UK.
British Journal of Cancer (Impact Factor: 4.84). 02/2008; 98(2):270-6. DOI: 10.1038/sj.bjc.6604158
Source: PubMed


Using a model-based approach, we estimated the probability that an individual, with a specified combination of risk factors, would develop lung cancer within a 5-year period.
Data from 579 lung cancer cases and 1157 age- and sex-matched population-based controls were available for this analysis. Significant risk factors were fitted into multivariate conditional logistic regression models. The final multivariate model was combined with age-standardised lung cancer incidence data to calculate absolute risk estimates.
Combinations of lifestyle risk factors were modelled to create risk profiles. For example, a 77-year-old male non-smoker, with a family history of lung cancer (early onset) and occupational exposure to asbestos has an absolute risk of 3.17% (95% CI, 1.67–5.95). Choosing a 2.5% cutoff to trigger increased surveillance, gave a sensitivity of 0.62 and specificity of 0.70, while a 6.0% cutoff gave a sensitivity of 0.34 and specificity of 0.90. A 10-fold cross validation produced an AUC statistic of 0.70, indicating good discrimination.
If independent validation studies confirm these results, the LLP risk models' application as the first stage in an early detection strategy is a logical evolution in patient care.

Download full-text


Available from: Martie van Tongeren,
  • Source
    • "These models include the Bach model, Spize model, and Liverpool Lung Project (LLP) model44–47 as well as the improvement models based on LLP.48,49 The LLP risk model,45 developed from the LLP case–control study, provides a single unified model for smokers (current and former) and nonsmokers, whereas the Bach model was developed for predicting risk only in smokers and the Spitz model46 requires three separate models for predicting risk in current smokers, former smokers, or nonsmokers. In addition, the LLP model also accounts for important lung cancer risk factors in addition to age, sex, and smoking duration. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Cancer risk prediction models are important in identifying individuals at high risk of developing cancer, which could result in targeted screening and interventions to maximize the treatment benefit and minimize the burden of cancer. The cancer-associated genetic variants identified in genome-wide or candidate gene association studies have been shown to collectively enhance cancer risk prediction, improve our understanding of carcinogenesis, and possibly result in the development of targeted treatments for patients. In this article, we review the cancer risk prediction models that have been developed for popular cancers and assess their applicability, strengths, and weaknesses. We also discuss the factors to be considered for future development and improvement of models for cancer risk prediction.
    Cancer informatics 09/2014; 13(Suppl 2):19-28. DOI:10.4137/CIN.S13788
  • Source
    • "But the UKLS trial used a different approach for participants. It recruited the population according to the Liverpool Lung Project risk model27. UKLS selected participants with a 5% risk of developing lung cancer in 5 years28. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The US National Lung Screening Trial (NLST) demonstrated a 20% reduction in lung cancer mortality and a 6.7% decrease in all-cause mortality. The NLST is the only trial showing positive results in a high-risk population, such as in patients with old age and heavy ever smokers. Lung cancer screening using a low-dose chest computed tomography might be beneficial for the high-risk group. However, there may also be potential adverse outcomes in terms of over diagnosis, bias and cost-effectiveness. Until now, lung cancer screening remains controversial. In this review, we wish to discuss the evolution of lung cancer screening and summarize existing evidences and recommendations.
    Tuberculosis and Respiratory Diseases 08/2014; 77(2):55-9. DOI:10.4046/trd.2014.77.2.55
  • Source
    • "The scientific community of risk prediction for lung cancer has established several risk models. Bach's and the EPIC models mainly used smoking variables as predictors (Bach et al. 2003; Hoggart et al. 2012), while Spitz's, the LLP and the PLCO models used more non-smoking variables (Tammemagi et al. 2013; Cassidy et al. 2008; Spitz et al. 2007). Smoking is the major risk factor for lung cancer, but evidence from epidemiological studies have also implicated environmental exposures (e.g. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Lung cancer is among the most common and fatal malignancies worldwide, and is often diagnosed at an advanced stage. Self-assessment tools for individualised lung cancer risk could potentially refine our understanding of risk factors, and contribute to disease prevention and earlier detection, thus reducing the disease burden. The originally published algorithm of the validated Liverpool Lung Project risk prediction model for lung cancer was transformed into a web-based questionnaire called MyLungRisk (MLR, This provides a user-friendly interface for individuals aged 50–79 years to estimate their risk of developing lung cancer in the next five years. It incorporates epidemiological risk factors such as age, gender, smoking duration, family history of lung cancer, previous history of pneumonia, previous diagnosis of cancer (other than lung cancer) and exposure to asbestos. Upon the completion of the web-based questionnaire, the individual is given an estimated lung cancer risk score for the next five years with an explanation of their risk, as per 100 average people in the population. In addition, advice is provided if the individual completing the questionnaire has any current symptoms (as per UK National Institute for Health and Care guidelines). MLR is a user-friendly interface for lung cancer risk assessment, based on validated scientific modelling. It is the first of its kind to be put on a widely accessible platform to the public.
    International Journal of Health Promotion and Education 02/2014; 52(3). DOI:10.1080/14635240.2014.888814
Show more