Abstract SP006: Weighing the benefits and harms of breast cancer screening

Abstract

Abstract Screening efficiency is the balance between health benefits and harms associated with screening. The balance sheet for breast screening needs to assess the benefits and harms associated with screening mammography, clinical breast examinations, adjunct imaging techniques (e.g., MRI, ultrasonography), biopsy procedures and therapies. Different groups have estimated the harm and benefit balance within the European and in the USA contexts. The contrasts in benefits estimated by different groups for the European context are glaringly obvious. The estimations of Loberg et al, 2015 and of the Research Council of Norway are comparable (Table). In these two reviews, the number of breast cancer death averted is derived from meta-analyses of randomised trials. Using pooled data from observational studies, the EuroScreen group obtained a 2 to 3-fold greater number of breast cancer deaths prevented by screening. Estimates for Europe are more convergent for false positive recalls and unnecessary biopsies. But estimates for overdiagnosis strongly differ because the EuroScreen group favoured results of studies that adjusted for lead-time, a controversial statistical manoeuvre that drastically reduces numbers of overdiagnosed breast cancer cases. The Table also highlights the salient contrasts in the harm-benefit balance between Europe and the USA. The substantial additional harm in the USA is documented by extensive data. In Europe, biennial screening between 50 and 69 years of age would result in 10 screening sessions. In the USA, women aged 40 to 74 years may attend up to 25 to 35 mammography screening sessions. As a consequence, it has been estimated that after 10 years of annual screening, one of two of screened US women have at least 1 false-positive mammogram result, whereas with triennial screening, one of eight 8 screened English women will have false positive result after 10 years. Of note, the cancer detection rates are similar in the European and in the US contexts, and declines in breast cancer mortality over time have been similar in the USA and high-income European countries, irrespective of the breast screening policies in place in European countries (e.g., high screening attendance since 1988 in the Netherlands vs. low screening attendance in Switzerland). So, the greater aggregated harm experienced by US women seems not compensated by a superior health benefit. A major limitation of harm-to-benefit balance studies done so far is the ignorance of the increasing potency of therapies. Randomised trials on breast screening were performed at a time effective therapies were non-existent or recently introduced. Because of the advent of effective therapies, more women need to be screened for preventing one death from breast cancer. In 2020, it could well be that breast cancer deaths averted thanks to screening displayed in the Table could be halved. However, the harms due to screening are not reduced by treatment improvement. Hence increasingly potent therapies steadily reduce the benefits from screening while keeping harms at the same level. This situation was similar for testis cancer, where the high effectiveness of cis-platinum based therapies, even for metastatic cancer, has rendered screening obsolete. Summary of harms and benefits of breast screeningEuropean contextUSA contextEuroscreen Paci et al, 2014 [3]Loberg et al, 2015 [1]Norway Research council of Norway, 2015 [4]Welch et al, 2014 [2]1,000 women screened every two years from 50 to 69 years, and followed until age 791,000 women screened every two years for 20 years starting at age 501,000 women offered 10 screening rounds starting a age 50 and followed until death1,000 women screened every year for 20 years starting at age 501,000 women screened every year for 30 years starting at age 40BenefitsAll-cause deaths avertedNR0NR00Breast cancer deaths averted7 to 92 †2.70.8 to 8.1 ‡0.9 to 9.7‡HarmsFalse positive recalls170200152880 to 12101390 to 1900Unnecessary biopsies303031130 to 170190 to 260Interval cancersNR3013--Overdiagnosed cancers needlessly treated41514.29 to 3410 to 45NR: not reported * Pooled analysis of IBM and of case-control studies † Breast cancer mortality reduction according to meta-analyses of randomised trials ‡ Taking results of the Swedish Two-county trial (Tabar et al, 1985 [16]) as the most optimistic scenario 1. Loberg M, Lousdal ML, Bretthauer M, Kalager M. Benefits and harms of mammography screening. Breast Cancer Res 2015;17:63. 2. Welch HG, Passow HJ. Quantifying the benefits and harms of screening mammography. JAMA Intern Med 2014;174(3):448-54. 3. Paci E, Broeders M, Hofvind S, Puliti D, Duffy SW, Group tEW. European Breast Cancer Service Screening Outcomes: A First Balance Sheet of the Benefits and Harms. Cancer Epidemiology Biomarkers & Prevention 2014;23(7):1159-1163. 4. The Research Council of Norway. Research-based evaluation of the Norwegian Breast Cancer Screening Program, Final report: Evaluation Division for Society and Health, www.forskningsradet.no/publikasjoner; 2015. Citation Format: P Autier. Weighing the benefits and harms of breast cancer screening [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr SP006.