S W Duffy

Publications of S W Duffy

• Explaining the difference in prognosis between screen-detected and symptomatic breast cancers.

  Authors: P C Allgood, S W Duffy, O Kearins, E O'Sullivan, N Tappenden, M G Wallis, G Lawrence

  British journal of cancer. 05/2011; 104(11):1680-5.

  We analysed 10-year survival data in 19,411 women aged 50-64 years diagnosed with invasive breast cancer in the West Midlands region of the United Kingdom. The aim was to estimate the survivalWe analysed 10-year survival data in 19,411 women aged 50-64 years diagnosed with invasive breast cancer in the West Midlands region of the United Kingdom. The aim was to estimate the survival advantage seen in cases that were screen detected compared with those diagnosed symptomatically and attribute this to shifts in prognostic variables or survival differences specific to prognostic categories. We studied tumour size, histological grade and the Nottingham Prognostic Index in very narrow categories and investigated the distribution of these prognostic factors within screen-detected and symptomatic tumours. We also adjusted for lead time bias. The unadjusted 10-year breast cancer survival in screen-detected cases was 85.5% and in symptomatic cases 62.8%; after adjustment for lead time bias, survival in the screen-detected cases was 79.3%. Within narrow categories of prognostic variables, survival differences were small, indicating that the majority of the survival advantage of screen detection is due to differences in the distributions of size and node status. Our results suggested that a combination of lead time with size and node status in 10 categories explained almost all (97%) of the survival advantage. Only a small proportion remained to be explained by biological differences, manifested as length bias or overdiagnosis.

• Polygenic susceptibility to prostate and breast cancer: implications for personalised screening.

  Authors: N Pashayan, S W Duffy, S Chowdhury, T Dent, H Burton, D E Neal, D F Easton, R Eeles, P Pharoah

  British journal of cancer. 04/2011; 104(10):1656-63.

  We modelled the efficiency of a personalised approach to screening for prostate and breast cancer based on age and polygenic risk-profile compared with the standard approach based on age alone. WeWe modelled the efficiency of a personalised approach to screening for prostate and breast cancer based on age and polygenic risk-profile compared with the standard approach based on age alone. We compared the number of cases potentially detectable by screening in a population undergoing personalised screening with a population undergoing screening based on age alone. Polygenic disease risk was assumed to have a log-normal relative risk distribution predicted for the currently known prostate or breast cancer susceptibility variants (N=31 and N=18, respectively). Compared with screening men based on age alone (aged 55-79: 10-year absolute risk ≥2%), personalised screening of men age 45-79 at the same risk threshold would result in 16% fewer men being eligible for screening at a cost of 3% fewer screen-detectable cases, but with added benefit of detecting additional cases in younger men at high risk. Similarly, compared with screening women based on age alone (aged 47-79: 10-year absolute risk ≥2.5%), personalised screening of women age 35-79 at the same risk threshold would result in 24% fewer women being eligible for screening at a cost of 14% fewer screen-detectable cases. Personalised screening approach could improve the efficiency of screening programmes. This has potential implications on informing public health policy on cancer screening.

• UK Lung Screen (UKLS) nodule management protocol: modelling of a single screen randomised controlled trial of low-dose CT screening for lung cancer.

  Authors: D R Baldwin, S W Duffy, N J Wald, R Page, D M Hansell, J K Field

  Thorax. 02/2011; 66(4):308-13.

  The UK Lung Screen (UKLS) is a randomised controlled trial of the use of low-dose multidetector CT for lung cancer screening. It completed the Health Technology Appraisal (HTA)-funded feasibilityThe UK Lung Screen (UKLS) is a randomised controlled trial of the use of low-dose multidetector CT for lung cancer screening. It completed the Health Technology Appraisal (HTA)-funded feasibility stage in October 2009 and the pilot UKLS will be initiated in early 2011. The pilot will randomise 4000 subjects to either low-dose CT screening or no screening. The full study, due to start in September 2012, if progression criteria are met, will randomise a further 28,000 subjects from seven centres in the UK. Subjects will be selected if they have sufficient risk of developing lung cancer according to the Liverpool Lung Project risk model. The UKLS employs the 'Wald Single Screen Design', which was modelled in the UKLS feasibility study. This paper describes the modelling of nodule management in UKLS by using volumetric analysis with a single initial screen design and follow-up period of 10 years. This modelling has resulted in the development and adoption of the UKLS care pathway, which will be implemented in the planned CT screening trial in the UK.

• Measuring the accuracy of diagnostic imaging in symptomatic breast patients: team and individual performance.

  Authors: P Britton, J Warwick, M G Wallis, S O'Keeffe, K Taylor, R Sinnatamby, S Barter, M Gaskarth, S W Duffy, G C Wishart

  The British journal of radiology. 01/2011;

  Objectives: The combination of mammography and/or ultrasound remains the mainstay in current breast cancer diagnosis. The aims of this study were to evaluate the reliability of standard breastObjectives: The combination of mammography and/or ultrasound remains the mainstay in current breast cancer diagnosis. The aims of this study were to evaluate the reliability of standard breast imaging and individual radiologist performance and to explore ways that this can be improved. Methods: A total of 16 603 separate assessment episodes were undertaken on 13 958 patients referred to a specialist symptomatic breast clinic over a 6 year period. Each mammogram and ultrasound was reported prospectively using a five-point reporting scale and compared with final outcome. Results: Mammographic sensitivity, specificity and receiver operating curve (ROC) area was 66.6%, 99.7% and 0.83 respectively. The sensitivity of mammography improved dramatically from 47.6% to 86.7% with increasing age. Overall ultrasound sensitivity, specificity and ROC area was 82.0%, 99.3% and 0.91 respectively. The sensitivity of ultrasound also improved dramatically with increasing age from 66.7% to 97.1%. Breast density also had a profound effect on imaging performance, with mammographic sensitivity falling from 90.1% to 45.9% and ultrasound sensitivity reducing from 95.2% to 72.0% with increasing breast density. Conclusion: The sensitivity ranges widely between radiologists (53.1-74.1% for mammography and 67.1-87.0% for ultrasound). Reporting sensitivity was strongly correlated with radiologist experience. Those radiologists with less experience (and lower sensitivity) were relatively more likely to report a cancer as indeterminate/uncertain. To improve radiology reporting performance, the sensitivity of cancer reporting should be closely monitored; there should be regular feedback from needle biopsy results and discussion of reporting classification with colleagues.

• Mammographic density and its interaction with other breast cancer risk factors in an Asian population.

  Authors: C S Wong, G H Lim, F Gao, R W Jakes, J Offman, K S Chia, S W Duffy

  British journal of cancer. 01/2011; 104(5):871-4.

  Joint effects of mammographic density and other risk factors on breast cancer risk remain unclear. From The Singapore Breast Screening Project, we selected 491 cases and 982 controls. MammographicJoint effects of mammographic density and other risk factors on breast cancer risk remain unclear. From The Singapore Breast Screening Project, we selected 491 cases and 982 controls. Mammographic density was measured quantitatively. Data analysis was by conditional logistic regression. Density was a significant risk factor, adjusting for other factors. Density of 76-100% had an odds ratio of 5.54 (95% CI 2.38-12.90) compared with 0-10%. Density had significant interactions with body mass index and oral contraceptive use (P=0.02). Percent density increases breast cancer risk in addition to effects of other risk factors, and modifies the effects of BMI and OCs.

• Comparison of discriminatory power and accuracy of three lung cancer risk models.

  Authors: A M D'Amelio, A Cassidy, K Asomaning, O Y Raji, S W Duffy, J K Field, M R Spitz, D Christiani, C J Etzel

  British journal of cancer. 07/2010; 103(3):423-9.

  Three lung cancer (LC) models have recently been constructed to predict an individual's absolute risk of LC within a defined period. Given their potential application in prevention strategies, aThree lung cancer (LC) models have recently been constructed to predict an individual's absolute risk of LC within a defined period. Given their potential application in prevention strategies, a comparison of their accuracy in an independent population is important. We used data for 3197 patients with LC and 1703 cancer-free controls recruited to an ongoing case-control study at the Harvard School of Public Health and Massachusetts General Hospital. We estimated the 5-year LC risk for each risk model and compared the discriminatory power, accuracy, and clinical utility of these models. Overall, the Liverpool Lung Project (LLP) and Spitz models had comparable discriminatory power (0.69), whereas the Bach model had significantly lower power (0.66; P=0.02). Positive predictive values were highest with the Spitz models, whereas negative predictive values were highest with the LLP model. The Spitz and Bach models had lower sensitivity but better specificity than did the LLP model. We observed modest differences in discriminatory power among the three LC risk models, but discriminatory powers were moderate at best, highlighting the difficulty in developing effective risk models.

• Cancer mortality in the 50-69 year age group before and after screening.

  Authors: S W Duffy, L Tabar, A H Olsen, B Vitak, P C Allgood, T H Chen, A M Yen, R A Smith

  Journal of medical screening. 01/2010; 17(3):159-60.

• Duffy and colleagues reply.

  Authors: S W Duffy, L Tabar, A H Olsen, B Vitak, P C Allgood, T H-H Chen, A M-F Yen, R A Smith

  Journal of medical screening. 01/2010; 17(2):103.

• Reply to Explaining variations in survival in breast cancer in the Eastern Region of England.

  Authors: S W Duffy, D C Greenberg, P Chou, C H Brown, A D Purushotham, G C Wishart

  Annals of oncology : official journal of the European Society for Medical Oncology / ESMO. 01/2010;

• Molecular characteristics of screen-detected vs symptomatic breast cancers and their impact on survival.

  Authors: S J Dawson, S W Duffy, F M Blows, K E Driver, E Provenzano, J Lequesne, D C Greenberg, P Pharoah, C Caldas, G C Wishart

  British journal of cancer. 09/2009;

  Background:Several recent studies have shown that screen detection remains an independent prognostic factor after adjusting for disease stage at presentation. This study compares the molecularBackground:Several recent studies have shown that screen detection remains an independent prognostic factor after adjusting for disease stage at presentation. This study compares the molecular characteristics of screen-detected with symptomatic breast cancers to identify if differences in tumour biology may explain some of the survival benefit conferred by screen detection.Methods:A total of 1379 women (aged 50-70 years) with invasive breast cancer from a large population-based case-control study were included in the analysis. Individual patient data included tumour size, grade, lymph node status, adjuvant therapy, mammographic screening status and mortality. Immunohistochemistry was performed on tumour samples using 11 primary antibodies to define five molecular subtypes. The effect of screen detection compared with symptomatic diagnosis on survival was estimated after adjustment for grade, nodal status, Nottingham Prognostic Index (NPI) and the molecular markers.Results:Fifty-six per cent of the survival benefit associated with screen-detected breast cancer was accounted for by a shift in the NPI, a further 3-10% was explained by the biological variables and more than 30% of the effect remained unexplained.Conclusion:Currently known biomarkers remain limited in their ability to explain the heterogeneity of breast cancer fully. A more complete understanding of the biological profile of breast tumours will be necessary to assess the true impact of tumour biology on the improvement in survival seen with screen detection.British Journal of Cancer advance online publication, 22 September 2009; doi:10.1038/sj.bjc.6605317 www.bjcancer.com.

• One-stop diagnostic breast clinics: how often are breast cancers missed?

  Authors: P Britton, S W Duffy, R Sinnatamby, M G Wallis, S Barter, M Gaskarth, A O'Neill, C Caldas, J D Brenton, P Forouhi, G C Wishart

  British journal of cancer. 06/2009;

  The aim of this study was to estimate the number of patients discharged from a symptomatic breast clinic who subsequently develop breast cancer and to determine how many of these cancers had beenThe aim of this study was to estimate the number of patients discharged from a symptomatic breast clinic who subsequently develop breast cancer and to determine how many of these cancers had been 'missed' at initial assessment. Over a 3-year period, 7004 patients were discharged with a nonmalignant diagnosis. Twenty-nine patients were subsequently diagnosed with breast cancer over the next 36 months. This equates to a symptomatic 'interval' cancer rate of 4.1 per 1000 women in the 36 months after initial assessment (0.9 per 1000 women within 12 months, 2.6 per 1000 women within 24 months). The lowest sensitivity of initial assessment was seen in patients of 40-49 years of age, and these patients present the greatest imaging and diagnostic challenge. Following multidisciplinary review, a consensus was reached on whether a cancer had been missed or not. No delay occurred in 10 patients (35%) and probably no delay in 7 patients (24%). Possible delay occurred in three patients (10%) and definite delay in diagnosis (i.e., a 'missed' cancer) occurred in only nine patients (31%). The overall diagnostic accuracy of 'triple' assessment is 99.6% and the 'missed' cancer rate is 1.7 per 1000 women discharged.British Journal of Cancer advance online publication, 19 May 2009; doi:10.1038/sj.bjc.6605082 www.bjcancer.com.

• Mean sojourn time, overdiagnosis, and reduction in advanced stage prostate cancer due to screening with PSA: implications of sojourn time on screening.

  Authors: N Pashayan, S W Duffy, P Pharoah, D Greenberg, J Donovan, R. M. Martin, F Hamdy, D E Neal

  British journal of cancer. 05/2009; 100(7):1198-204.

  This study aimed to assess the mean sojourn time (MST) of prostate cancer, to estimate the probability of overdiagnosis, and to predict the potential reduction in advanced stage disease due toThis study aimed to assess the mean sojourn time (MST) of prostate cancer, to estimate the probability of overdiagnosis, and to predict the potential reduction in advanced stage disease due to screening with PSA. The MST of prostate cancer was derived from detection rates at PSA prevalence testing in 43,842 men, aged 50-69 years, as part of the ProtecT study, from the incidence of non-screen-detected cases obtained from the English population-based cancer registry database, and from PSA sensitivity obtained from the medical literature. The relative reduction in advanced stage disease was derived from the expected and observed incidences of advanced stage prostate cancer. The age-specific MST for men aged 50-59 and 60-69 years were 11.3 and 12.6 years, respectively. Overdiagnosis estimates increased with age; 10-31% of the PSA-detected cases were estimated to be overdiagnosed. An interscreening interval of 2 years was predicted to result in 37 and 63% reduction in advanced stage disease in men 65-69 and 50-54 years, respectively. If the overdiagnosed cases were excluded, the estimated reductions were 9 and 54%, respectively. Thus, the benefit of screening in reducing advanced stage disease is limited by overdiagnosis, which is greater in older men.

• Differences in endpoints between the Swedish W-E (two county) trial of mammographic screening and the Swedish overview: methodological consequences.

  Authors: L Holmberg, S W Duffy, A M-F Yen, L Tabár, B Vitak, L Nyström, J Frisell

  Journal of medical screening. 02/2009; 16(2):73-80.

  OBJECTIVES: To characterize and quantify the differences in the number of cases and breast cancer deaths in the Swedish W-E Trial compared with the Swedish Overview Committee (OVC) summaries and toOBJECTIVES: To characterize and quantify the differences in the number of cases and breast cancer deaths in the Swedish W-E Trial compared with the Swedish Overview Committee (OVC) summaries and to study methodological issues related to trials in secondary prevention. SETTING: The study population of the W-E Trial of mammography screening was included in the first (W and E county) and the second (E-county) OVC summary of all Swedish randomized mammography screening trials. The OVC and the W-E Trial used different criteria for case definition and causes of death determination. METHOD: A Review Committee compared the original data files from W and E county and the first and second OVC. The reason for a discrepancy was determined individually for all non-concordant cases or breast cancer deaths. RESULTS: Of the 2615 cases included by the W-E Trial or the OVC, there were 478 (18%) disagreements. Of the disagreements 82% were due to inclusion/exclusion criteria, and 18% to disagreement with respect to cause of death or vital status at ascertainment. For E-County, the OVC inclusion rules and register based determination of cause of death (second OVC) rather than individual case review (W-E Trial and 1st OVC) resulted in a reduction of the estimate of the effect of screening, but for W-County the difference between the original trial and the OVC was modest. CONCLUSIONS: The conclusion that invitation to mammography screening reduces breast cancer mortality remains robust. Disagreements were mainly due to study design issues, while disagreements about cause of death were a minority. When secondary research does not adhere to the protocols of the primary research projects, the consequences of such design differences should be investigated and reported. Register linkage of trials can add follow-up information. The precision of trials with modest size is enhanced by individual monitoring of case status and outcome status such as determination of cause of death.

• Reply: Estimation of lead-time and overdiagnosis in breast cancer screening.

  Authors: S W Duffy, E Lynge, H Jonsson

  British journal of cancer. 02/2009; 100(1):220.

• Complexities in the estimation of overdiagnosis in breast cancer screening.

  Authors: S W Duffy, E Lynge, H Jonsson, S Ayyaz, A H Olsen

  British journal of cancer. 10/2008;

  There is interest in estimating and attributing temporal changes in incidence of breast cancer in relation to the initiation of screening programmes, in particular to estimation of overdiagnosis ofThere is interest in estimating and attributing temporal changes in incidence of breast cancer in relation to the initiation of screening programmes, in particular to estimation of overdiagnosis of breast cancer as a result of screening. In this paper, we show how screening introduces complexities of analysis and interpretation of incidence data. For example, lead time brings forward time- and age-related increases in incidence. In addition, risk factors such as hormone replacement therapy use have been changing contemporaneously with the introduction of screening. Although we do not indicate exactly how such complexities should be corrected for, we use some simple informal adjustments to show how they may account for a substantial proportion of increased incidence, which might otherwise erroneously have been attributed to overdiagnosis. We illustrate this using an example of analysis of breast cancer incidence data from Sweden.British Journal of Cancer advance online publication, 2 September 2008; doi:10.1038/sj.bjc.6604638 www.bjcancer.com.

• Lung cancer screening: the way forward.

  Authors: J K Field, S W Duffy

  British journal of cancer. 09/2008; 99(4):557-62.

  To take lung cancer screening into national programmes, we first have to answer the question whether low-dose computed tomography (LDCT) screening and treatment of early lesions will decrease lungTo take lung cancer screening into national programmes, we first have to answer the question whether low-dose computed tomography (LDCT) screening and treatment of early lesions will decrease lung cancer mortality compared with a control group, to accurately estimate the balance of benefits and harms, and to determine the cost-effectiveness of the intervention.

• Screen-detected vs symptomatic breast cancer: is improved survival due to stage migration alone?

  Authors: G C Wishart, D C Greenberg, P D Britton, P Chou, C H Brown, A D Purushotham, S W Duffy

  British journal of cancer. 07/2008; 98(11):1741-4.

  This paper examines whether screen-detected breast cancer confers additional prognostic benefit to the patient, over and above that expected by any shift in stage at presentation. In all, 5604 womenThis paper examines whether screen-detected breast cancer confers additional prognostic benefit to the patient, over and above that expected by any shift in stage at presentation. In all, 5604 women (aged 50-70 years) diagnosed with invasive breast cancer between 1998 and 2003 were identified by the Eastern Cancer Registration and Information Centre (ECRIC) and mammographic screening status was determined. Using proportional hazards regression, we estimated the effect of screen detection compared with symptomatic diagnosis on 5-year survival unadjusted, then adjusted for age and Nottingham Prognostic Index (NPI). A total of 72% of the survival benefit associated with screen-detected breast cancer can be accounted for by age and shift in NPI. Survival analysis by continuous NPI showed a small but systematic survival benefit for screen-detected cancers at each NPI value. These data show that although most of the screen-detected survival advantage is due to a shift in NPI, the mode of detection does impact on survival in patients with equivalent NPI scores. This residual survival benefit is small but significant, and is likely to be due to differences in tumour biology. Current prognostication tools may, therefore, overestimate the benefit of systemic treatments in screen-detected cancers and lead to overtreatment of these patients.

• A model for predicting non-sentinel lymph node metastatic disease when the sentinel lymph node is positive.

  Authors: A Pal, E Provenzano, S W Duffy, S E Pinder, A D Purushotham

  The British journal of surgery. 04/2008; 95(3):302-9.

  BACKGROUND: Women with axillary sentinel lymph node (SLN)-positive breast cancer usually undergo completion axillary lymph node dissection (ALND). However, not all patients with positive SLNs haveBACKGROUND: Women with axillary sentinel lymph node (SLN)-positive breast cancer usually undergo completion axillary lymph node dissection (ALND). However, not all patients with positive SLNs have further axillary nodal disease. Therefore, in the patients with low risk of further disease, completion ALND could be avoided. The Memorial Sloan-Kettering Cancer Center (MSKCC) developed a nomogram to estimate the risk of non-SLN disease. This study critically appraised the nomogram and refined the model to improve predictive accuracy. METHODS: The MSKCC nomogram was applied to 118 patients with a positive axillary SLN biopsy who subsequently had completion ALND. Predictive accuracy was assessed by calculating the area under the receiver-operator characteristic (ROC) curve. A further predictive model was developed using more detailed pathological information. Backward stepwise multiple logistic regression was used to develop the predictive model for further axillary lymph node disease. This was then converted to a probability score. After k-fold cross-validation within the data, an inverse variance weighted mean ROC curve and area below the ROC curve was calculated. RESULTS: The MSKCC nomogram had an area under the ROC curve of 68 per cent. The revised predictive model showed the weighted mean area under the ROC curve to be 84 per cent. CONCLUSION: The modified predictive model, which incorporated size of SLN metastasis, improved predictive accuracy, although further testing on an independent data set is desirable.

• The LLP risk model: an individual risk prediction model for lung cancer.

  Authors: A Cassidy, J P Myles, M van Tongeren, R D Page, T Liloglou, S W Duffy, J K Field

  British journal of cancer. 02/2008; 98(2):270-6.

  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 cancerUsing 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.

• A case-control study of the impact of the East Anglian breast screening programme on breast cancer mortality.

  Authors: P C Allgood, J Warwick, R M L Warren, N E Day, S W Duffy

  British journal of cancer. 02/2008; 98(1):206-9.

  Although breast cancer screening has been shown to work in randomised trials, there is a need to evaluate service screening programmes to ensure that they are delivering the benefit indicated by theAlthough breast cancer screening has been shown to work in randomised trials, there is a need to evaluate service screening programmes to ensure that they are delivering the benefit indicated by the trials. We carried out a case-control study to investigate the effect of mammography service screening, in the NHS breast screening programme, on breast cancer mortality in the East Anglian region of the UK. Cases were deaths from breast cancer in women diagnosed between the ages of 50 and 70 years, following the instigation of the East Anglia Breast Screening Programme in 1989. The controls were women (two per case) who had not died of breast cancer, from the same area, matched by date of birth to the cases. Each control was known to be alive at the time of death of her matched case. All women were known to the breast screening programme and were invited, at least once, to be screened. There were 284 cases and 568 controls. The odds ratio (OR) for risk of death from breast cancer in women who attended at least one routine screen compared to those who did not attend was 0.35 (CI: 0.24, 0.50). Adjusting for self-selection bias gave an estimate of the breast cancer mortality reduction associated with invitation to screening of 35% (OR=0.65, 95% CI: 0.48, 0.88). The effect of actually being screened was a 48% breast cancer mortality reduction (OR=0.52, 95% CI: 0.32, 0.84). The results suggest that the National Breast Screening Programme in East Anglia is achieving a reduction in breast cancer deaths, which is at least consistent with the results from the randomised controlled trials of mammographic screening.