Abstract P1-22-04: Decision aids and risk prediction models to support decision making about DCIS treatment: A systematic literature review

Abstract

Abstract Background: Although most low-risk ductal carcinoma in situ (DCIS) lesions will not progress to invasive breast cancer if left untreated, clinical guidelines advise surgery with/without radiotherapy for all women diagnosed with DCIS. There is therefore increasing concern about the possible overtreatment of DCIS. Currently, clinical trials are being conducted to investigate the safety of active surveillance in low-risk DCIS patients. It is hypothesized that, in future, both surgery and active surveillance will be accepted treatment strategies. Active surveillance is offered to women in the ongoing trials and is expected to become a standard DCIS management option in the future. Choosing whether to undergo surgery for DCIS or to opt for active surveillance can be a difficult decision fraught with uncertainty for both patients and oncologists. Good quality decision support tools such as prediction models and patient decision aids to guide decision making about DCIS management, including the option of active surveillance, are therefore urgently needed. The aim of this study is to identify and evaluate the quality of published decision aids and prediction models aiming to support decision making about DCIS treatment. Methods: A systematic literature review was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement criteria. The databases Medline(ovid), Embase (ovid), Scopus, and TRIP were searched to identify published manuscripts describing the development and/or evaluation of DCIS decision aids and prediction models. The protocol was published in the PROSPERO database (ID CRD42020212297). The CHARMS (Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies) checklist was used to evaluate the methodological quality of prediction models and the IPDAS (International Patient Decision Aid Standards) checklist was used to evaluate the quality of decision aids. Data extraction was performed by two researchers with discrepancies resolved through consensus. Results: The review identified 10,636 publications, 33 describing the development and/or validation of four decision aids and seven clinical prediction models were selected (Table 1). The decision aids identified met at least 50% of the IPDAS quality criteria. However, most decision aids lacked tools to help patients reflect on the information received and to facilitate discussion of the information with their family and healthcare providers. Most prediction models were designed to predict the risk of a subsequent ipsilateral breast event after a primary DCIS. No models included the option of active surveillance. Sufficient, good quality, external validation was lacking for all prediction models identified. Conclusions: There are only a few decision aids available that can be used to support patients diagnosed with DCIS. These decision aids could be improved to facilitate the processing of information by patients and enhance communication between patients and their support system and healthcare providers. There is no prediction model that considers active surveillance as a management option for DCIS, and based on the available evidence, there is no prediction model that can be recommended for use in clinical practice. More and qualitatively better validations are required in the future. Table 1.Overview of DCIS decision aids and prediction models identifiedDCIS DECISION AIDSDecision aid by Berger-Hoger et al.(2014)Communication aid by De Morgan et al.(2009)onlineDeCISion.org by Ozanne et al.(2016)DCISoptions.org by COMET trial team(SABCS 2020)Target audience:Women with DCISCliniciansClinicians and women with DCISWomen with DCISLanguage:GermanEnglishEnglishEnglishEvaluation study conducted:YesYesNot reportedNot reportedDesign evaluation study:RCTInterviewNot applicableNot applicableSample size evaluation study:6425Not applicableNot applicableMain finding evaluation study:More active patient involvementCommunication tool assists shared decision makingNot applicableNot applicableImplementation study conducted:None retrievedNone retrievedNone retrievedNone retrieved% IPDAS criteria met regarding:Content87%57%65%78%Development process71%59%67%42%Effectiveness100%50%75%75%DCIS PREDICTION MODELSOncotype DCIS(Solin et al. (2013))DCISionRT/PreludeDX(Bremer et al. (2018))Van Nuysprognostic index(Silverstein et al. (1995))MSKCC DCIS nomogram(Rudlof et al. (2010))Patient prognostic score(Sagara et al. (2016))PredictCBC(Giardello et al. (2019))CBC Risk model(Chowdhury et al. (2017))Predicted outcome:Ipsilateral breast eventIpsilateral breast eventIpsilateral breast eventIpsilateral breast eventIpsilateral breast eventContralateral breast cancerContralateral breast cancerTool based on:Multigene assayBiomarkers + clinico-pathological factorsClinicopatho-logical factors onlyClinicopatho-logical factors onlyClinicopatho-logical factors onlyClinicopatho-logical factors onlyClinicopatho-logical factors onlyIntended to support decision making about:Need for adjuvant radiotherapyNeed for adjuvant radiotherapyType of surgery and need for radiotherapyNeed for adjuvant radiotherapyNeed for adjuvant radiotherapyScreening or prophylactic mastectomyScreening or prophylactic mastectomyRisk of bias based on CHARMS:ModerateModerateModerate/HighModerateLowLowLowNumber (external) validations:3193001Reported C-index/AUC0.68None reportedNone reported0.61-0.68None reported0.52None reportedThis work was supported by Cancer Research UK and by KWF Dutch Cancer Society (ref.C38317/A24043) Citation Format: Renée SJM Schmitz, Erica Wilthagen, Frederieke van Duijnhoven, Marja van Oirsouw, Ellen Verschuur, Thomas Lynch, Rinaa S Punglia, Shelley Hwang, Jelle Wesseling, Marjanka K Schmidt, Eveline Bleiker, Ellen G Engelhardt, Grand Challenge PRECISION consortium. Decision aids and risk prediction models to support decision making about DCIS treatment: A systematic literature review [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P1-22-04.