JOURNAL TRANSCRIPT

---

Accepted Manuscript Title: Systematic review of the empirical investigation of resources to support decision-making regarding BRCA1 and BRCA2 genetic testing in women with breast cancer Authors: Chloe Grimmett, Karen Pickett, Jonathan Shepherd, Karen Welch, Alejandra Recio-Saucedo, Elke Streit, Helen Seers, Anne Armstrong, Ramsey I. Cutress, D. Gareth Evans, Ellen Copson, Bettina Meiser, Diana Eccles, Claire Foster PII: DOI: Reference:

S0738-3991(17)30639-0 https://doi.org/10.1016/j.pec.2017.11.016 PEC 5837

To appear in:

Patient Education and Counseling

Received date: Revised date: Accepted date:

8-6-2017 9-11-2017 24-11-2017

Please cite this article as: Grimmett Chloe, Pickett Karen, Shepherd Jonathan, Welch Karen, Recio-Saucedo Alejandra, Streit Elke, Seers Helen, Armstrong Anne, Cutress Ramsey I, Evans D Gareth, Copson Ellen, Meiser Bettina, Eccles Diana, Foster Claire.Systematic review of the empirical investigation of resources to support decision-making regarding BRCA1 and BRCA2 genetic testing in women with breast cancer.Patient Education and Counseling https://doi.org/10.1016/j.pec.2017.11.016 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Systematic review of the empirical investigation of resources to support decision-making regarding BRCA1 and BRCA2 genetic testing in women with breast cancer

Short title: Review of resources to support decisions regarding BRCA1 and BRCA2 genetic testing

IP T

Chloe Grimmetta ([email protected]), Karen Pickettb ([email protected]),

Jonathan Shepherdb ([email protected]), Karen Welchb ([email protected]),

SC R

Alejandra Recio-Saucedoc ([email protected]), Elke Streitb ([email protected]), Helen Seersa, Anne Armstrongd ([email protected]), Ramsey I Cutresse

([email protected]), D Gareth Evansf ([email protected]), Ellen Copsone

U

([email protected]), Bettina Meiserg ([email protected]), Diana Ecclese

A

N

([email protected]) & Claire Fostera ([email protected])

M

a. Faculty of Health Sciences, University of Southampton, Southampton, UK

Southampton, UK

ED

b. Southampton Health Technology Assessments Centre, University of Southampton,

PT

c. National Institute for Health Research (NIHR) Collaboration for Applied Health Research and Care (CLAHRC) Wessex, University of Southampton, Southampton, UK

CC E

d. Christie Hospital NHS Foundation Trust and Faculty of Biology, Medicine and Health, The University of Manchester, Manchester, UK

A

e. University of Southampton and University Hospital Southampton, Somers Cancer Research Building, Southampton, UK

f.

Centre for Genomic Medicine, Division of Evolution and Genomic Sciences, University of Manchester, Manchester, UK

g. Faculty of Medicine, University of New South Wales, New South Wales 2033, Australia

1

Corresponding authors at 1. Professor Claire Foster, Faculty of Health Sciences, University of Southampton, Southampton, UK. Email: [email protected], Telephone: +44 (0)23 8059 6885, Fax: +44 (0)23 8059 7967 2. Professor Diana Eccles, Somers Cancer Research, University of Southampton, Southampton General Hospital, Southampton, UK. Email: [email protected], Telephone: +44 (0)23 8120

IP T

8537, Fax: +44 (0)23 8120 4346 3.

U

Few resources available to support genetic testing decision-making in breast cancer Resources to support decision-making for BRCA testing are endorsed by patients Existing tools increase knowledge/understanding of hereditary breast cancer New tools should include functionalities to support the decision-making process

N

   

SC R

Highlights

A

Abstract

M

Objective: Identify existing resources developed and/or evaluated empirically in the published

for BRCA1/2 mutations.

ED

literature designed to support women with breast cancer making decisions regarding genetic testing

PT

Methods: Systematic review of seven electronic databases. Studies were included if they described

CC E

or evaluated resources that were designed to support women with breast cancer in making a decision to have genetic counselling or testing for familial breast cancer. Outcome and process evaluations, using any type of study design, as well as articles reporting the development of decision

A

aids, were eligible for inclusion. Results: Total of 8 publications, describing 6 resources were identified. Resources were effective at increasing knowledge or understanding of hereditary breast cancer. Satisfaction with resources was high. There was no evidence that any resource increased distress, worry or decisional conflict. Few

2

resources included active functionalities for example, values-based exercises, to support decisionmaking. Conclusion: Tailored resources supporting decision-making may be helpful and valued by patients and increase knowledge of hereditary breast cancer, without causing additional distress.

IP T

Practice implications: Clinicians should provide supportive written information to patients where it is available. However, there is a need for robustly developed decision tools to support decision-making

SC R

around genetic testing in women with breast cancer.

U

Keywords:

A

N

BRCA1; BRCA2; breast cancer; decision aid; decision support; treatment-focused genetic testing

M

1. Introduction

Traditionally, genetic testing for women diagnosed with breast cancer has been offered within

ED

specialist services. The focus of provision has been on those with a family history and usually after

PT

completion of active treatment. Several factors are influencing the increasing number and nature of referrals for genetic testing. Technological advances mean testing is becoming cheaper and faster. It

CC E

is also recognised that a substantial proportion of women with no family history of breast cancer, but who have other increased-risk features (including younger age at breast cancer diagnosis, certain ethnicities, tumour characteristics) may carry a BRCA1 or BRCA2 (hereafter BRCA1/2)

A

mutation. This has resulted in changes to the threshold for referral to genetic services [1]. Additionally, while considerable international variation exists, testing for germline mutations at the time of cancer diagnosis to inform treatment decisions (treatment-focused genetic testing - TFGT) is becoming more common. Knowledge of BRCA mutation status is increasingly used or requested to support decision-making between breast conserving surgery versus mastectomy and/or 3

contralateral mastectomy. The advantages and disadvantages of combining treatment of the primary cancer with that of future risk of the developing a second new primary can then be considered There is also evidence of increased public awareness of inherited predisposition to breast cancer and more referrals for genetic counselling following wide media coverage of preventative surgeries of

IP T

actress Angelina Jolie (BRCA1 positive and with a family history of breast and ovarian cancer) [2, 3]. Finally, the advent of targeted drug therapies [4] means more breast cancer specialists are

SC R

recommending genetic testing to their patients to inform the choice of chemotherapy regimens and inclusion in clinical trials.

U

The result is a trend towards ‘mainstreaming’ of genetic testing; with the ambition of offering testing

N

for BRCA1/2 as part of routine patient care for young women diagnosed with breast cancer. It is

A

therefore imperative that women are supported to make an informed choice about genetic testing

M

given the likely short timeframes following diagnosis [5]. Women diagnosed at a younger age (≤50

ED

years) represent a group for whom treatment decisions can be particularly complex; for example, they may consider risk-reducing surgery in the context of concerns about adverse impacts on

PT

fertility, sexual functioning, body image and self-esteem [6]. In a busy oncology clinic, the time and knowledge base to adequately inform women about the pros and cons of genetic testing is often

CC E

limited.

Empirically evaluated decision support interventions (or decision aids) have been found to increase

A

knowledge and result in patients feeling better informed and with clarity as to clearer about what matters most to them when making medical treatment and screening choices [7]. Decision aids aim to supplement clinical consultations and help patients make considered choices. They clearly state the decision to be made and the options available with associated benefits and harms. They also include components that allow patients to recognise their values associated with the decision being 4

made [8]. This is different to educational materials that describe the health condition and options but do not attempt to support patient contributions to the decision-making process [7]. Similar resources have been developed for women with breast cancer choosing between breast conserving surgery and mastectomy [9] and women at high-risk of developing breast and/or ovarian cancer making decisions about genetic testing [10]. Decision support tools could be particularly valuable in

IP T

the context of decision-making for TFGT, where there is increased pressure to compress specialist

SC R

genetic counselling into the timeframe required for the treatment of the primary cancer.

This systematic review therefore aimed to identify existing resources developed and/or evaluated empirically in the published literature designed to support women’s decision-making regarding

U

genetic testing for BRCA1/2 mutations. This was the first phase of a study, which set out to develop

N

such a decision support tool for young women recently diagnosed with breast cancer. This is in line

A

with the Medical Research Council guidelines for developing and evaluating complex interventions,

M

which recommends identification of the relevant existing evidence base, ideally by way of a

ED

systematic review as a first stage of intervention development [11]. The objectives of the review are threefold: to characterise published resources, assess their acceptability and evaluate their impact.

PT

As well as informing intervention development, this evidence synthesis will serve to generate new research questions in line with clinical priorities and make practice recommendations on the basis of

CC E

current evidence.

2. Methods

A

Guidance from the Centre for Reviews and Dissemination (2009) [12] and the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) statement [13] informed the methods for conducting and reporting this review.

5

2.1 Literature searching: The following bibliographic databases were searched: MEDLINE, MEDLINE In-Process and other non-indexed citations, Embase, Web of Science (Conference Proceedings Citations Index) and Science Citation Index Expanded, Cochrane Library, PsychInfo, and Delphis University of Southampton Resource Search. Databases were searched from 1990 (to capture early studies to identify BRCA1/2 in women with a family history) to February 2015. The exception was

IP T

Web of Science CPCI, which was used to identify recent conference abstracts which may not have resulted yet in manuscript publications. Searches in this database were conducted from 2013 to

SC R

February 2015. No language restrictions were applied to the searches. Other potentially relevant

references were also identified by examining the bibliographies of the studies included in the review. One additional reference (reporting in full a study detailed in a conference abstract found through

A

N

supplementary appendix 1 for Medline search strategy.

U

the other searches) was identified by one of the co-authors (BM) and included in the review. See

M

2.2 Study selection: The eligibility criteria for study inclusion were; studies including women who had

ED

a personal diagnosis of breast cancer or studies including women with and without a personal diagnosis of breast cancer provided results were given separately for participants diagnosed with

PT

breast cancer on at least one outcome of interest to the review. The inclusion criteria are detailed in Table 1. In brief, studies were included if they described or evaluated resources that were designed

CC E

to support women in making a decision to have genetic counselling or testing for familial breast cancer. Outcome and process evaluations, using any type of study design, as well as articles

A

reporting the development of decision aids, were eligible for inclusion.

Study selection followed a two-stage process. First, one of five reviewers (KP, JS, CG, A R-S, CF), assessed the title and abstracts of all the references identified in the searches for potential eligibility for inclusion in the review. Second, full texts were obtained of publications identified to be potentially relevant and these were screened by one of five reviewers (KP, JS, CG, A R-S, CF) against 6

the inclusion criteria. Publications where inclusion was uncertain were considered by a second reviewer and discussion took place until consensus was reached. Of the non-English language publications identified, full text references were retrieved for Dutch, Spanish and German studies as individuals within the research team were able to translate these. One Polish study was not

IP T

retrieved as translation was not available.

2.3 Data extraction and quality assessment: Data extraction from each study was carried out using a

SC R

piloted and standardised form (see supplementary appendix 2) by one of three reviewers (KP, CG,

ES). A fourth reviewer performed a second, independent data extraction of all papers (A R-S). Data were extracted about the study aims, design and methodology, the country and setting, the

U

participant inclusion criteria and participants’ baseline characteristics. We also extracted detailed

N

information about the intervention (including information about any theoretical models

A

underpinning decision aid interventions), outcome measures, full results for all outcomes assessed,

ED

M

and study limitations.

Four reviewers critically appraised each study independently (KP, ES, CG, A R-S). To appraise

PT

publications describing outcome evaluations, we used a modified version of a tool by Reisch, Tyson & Mize (1989) [14] that was used in a previous related systematic review [6] and based on the work

CC E

of Deeks et al. (2003) [15]. The tool includes 32 questions asking reviewers to assess aspects of the study’s methodology and reporting, such as if aims are clearly stated, if the randomisation method was appropriate (for randomised controlled trials), if the participants were likely to be

A

representative of the population and reporting of dropouts. Based on the responses, an overall global rating of the quality of the study (strong, moderate or weak) was calculated. The calculation of the global rating was carried out independently by two reviewers (CG, A R-S) for each study as a quality assurance measure.

7

Process evaluations were critically appraised using quality assessment criteria for process evaluations [16]. Reviewers (KP, ES, CG, A R-S) judged the extent to which steps were taken in the study to reduce bias, how grounded the findings were in the data and breadth and depth of findings. Reviewers then provided a weighting of a) the reliability of the study findings (low, medium or high) and b) the usefulness of the study findings (low, medium or high). If a study included both outcome

IP T

and process evaluations, both quality assessment tools were applied to the study.

SC R

2.4 Data synthesis: The findings were narratively synthesised and tabulated, describing the study

design, resource developed/evaluated, target population and outcomes (where appropriate). It was not possible to conduct meta-analysis for any outcomes due to the heterogeneity of the included

N

U

studies.

M

interest and structured the synthesis:

A

In line with the aims of this review, three broad areas of information from the studies were of

ED

1. Description of the format and components of the resources. 2. Assessment of the impact of using the resources on the decision-making process. Elwyn et al.

PT

(2012) [17] developed a model for shared decision-making in clinical practice which is divided into 3 steps; a. Introducing choice (patients are made aware of the options that are available); b.

CC E

Describing options (knowledge is checked and additional educational materials may be used) c. Decision talk (with an emphasis of personal preferences and values). Therefore outcomes relevant to these categories were captured, as well as outcomes measured after a decided action was taken, i.e.

A

psychological morbidity and decisional conflict. 3. Evaluation of acceptability of the resources.

3. Results 8

Figure 1 shows the flow of studies in this review. A total of 3,598 publications were identified (after removal of duplications). On inspection of titles and abstracts, 3,434 publications were excluded, and 2 full texts were unobtainable. Full texts were examined for 162 articles and 9 met the criteria for final inclusion [18-25]. Rahman et al. (2012) [21] was a conference abstract detailing a study reported in full in Quinn et al. (2017) [26]. The 9 references identified described 6 different

IP T

resources. Two publications describe development and testing of a resource, five describe the evaluation of a resource and two describe the development of a resource and process evaluation

SC R

(see Table 2). The majority of studies were conducted in the United States, followed by Australia and the Netherlands.

U

3.1 Quality of studies: Of the studies that included a process evaluation, Meiser et al. (2012) [18],

N

Permuth-Wey et al. (2010) [20] and Vadaparampil et al. (2014) [24] were rated as ‘medium’ and

A

Quinn et al. (2017) [26] as strong for reliability and usefulness of findings. The main limitations of

M

these studies were lack of transparency of participant selection processes [18], recruitment of an

ED

atypical population for part of the study, i.e. delegates at a conference [24], and limited breadth and depth of qualitative data [20]. Thompson et al. (2004) [23] was rated as ‘low’ for usefulness of

PT

findings and reliability of findings due to an unrepresentative sample, the use of rapid focus groups

CC E

and no audio recording of transcriptions of the focus groups.

Four of the studies reporting an outcome evaluation were rated as ‘moderate’. Key weaknesses include descriptions of samples and response rates [18, 19, 21, 25] and a lack of validated outcome

A

measures [18, 19]. Furthermore there was no evidence of checks of intervention fidelity in Venne & Hamann’s (2007) [25] study. Sie et al. (2014) [22] was rated as strong in terms of quality.

3.2 Target population:

9

The target audience and purpose of the resources varied. Two [18, 20, 26, 27] targeted young women. Meiser et al.’s pamphlet was designed specifically for young women (