UK consensus recommendations for clinical management of cancer risk for women with germline pathogenic variants in cancer predisposition genes: RAD51C, RAD51D, BRIP1 and PALB2

Germline pathogenic variants (GPVs) in the cancer predisposition genes BRCA1, BRCA2, MLH1, MSH2, MSH6, BRIP1, PALB2, RAD51D and RAD51C are identified in approximately 15% of patients with ovarian cancer (OC). While there are clear guidelines around clinical management of cancer risk in patients with GPV in BRCA1, BRCA2, MLH1, MSH2 and MSH6, there are few guidelines on how to manage the more moderate OC risk in patients with GPV in BRIP1, PALB2, RAD51D and RAD51C, with clinical questions about appropriateness and timing of risk-reducing gynaecological surgery. Furthermore, while recognition of RAD51C and RAD51D as OC predisposition genes has been established for several years, an association with breast cancer (BC) has only more recently been described and clinical management of this risk has been unclear. With expansion of genetic testing of these genes to all patients with non-mucinous OC, new data on BC risk and improved estimates of OC risk, the UK Cancer Genetics Group and CanGene-CanVar project convened a 2-day meeting to reach a national consensus on clinical management of BRIP1, PALB2, RAD51D and RAD51C carriers in clinical practice. In this paper, we present a summary of the processes used to reach and agree on a consensus, as well as the key recommendations from the meeting.


BACKGROUND
Tubo-ovarian cancer (OC) is the sixth most common cancer in women in the UK with over 7500 new diagnoses per year. 1 The risk of OC in first-degree relatives of patients with OC has been estimated to be threefold greater compared with the population risk. Most of this excess familial risk is due to germline pathogenic variants (GPVs) in the cancer predisposition genes BRCA1 and BRCA2, which are associated with high lifetime risks of breast cancer (BC) and OC. 2 However, other genes which have lower lifetime risks of OC have been identified. These include MLH1, MSH2, MSH6, BRIP1, PALB2, RAD51D and RAD51C as well as common, low-risk OC genetic susceptibility variants identified through genome-wide association studies. 3 4 While guidelines around clinical management of cancer risk in patients with GPV in BRCA1 and BRCA2 5 6 and the mismatch repair genes; MLH1, MSH2, MSH6 7 have been published, there are few guidelines on how to manage patients with GPV in genes associated with more moderate risks of OC: BRIP1, PALB2, RAD51D and RAD51C.

Position statement
Health Service (NHS) Very High Risk (VHR) Breast Screening programme. 12 While some guidelines have addressed management of OC in individuals with RAD51C, RAD51D or BRIP1 GPV, 13 there are no UK guidelines available for the management of all cancer risk in this patient group.
A previous UK Cancer Genetics Group (UKCGG) Consensus meeting held in 2018 agreed that PALB2 should be included on a BC predisposition gene panel and RAD51C, RAD51D and BRIP1 on an OC predisposition gene panel. 14 The national genomic test directory (NHS England) was first published in August 2020 and included PALB2 in the panel for inherited BC and OC (R208) and RAD51C, RAD51D and BRIP1 in the inherited familial OC panel (R207). The test directory is updated each year with input from the clinical and scientific community incorporating changes to both eligibility criteria and genes on a panel (https:// www.england.nhs.uk/wp-content/uploads/2018/08/Rare-andinherited-disease-eligibility-criteria-version-3.1-August-2022. pdf). Similar panel testing is available in Wales, Scotland and Northern Ireland. Recently, evidence substantiating the role of RAD51C and RAD51D in BC predisposition was published, 8 which suggests further amendments to the test directory may be required (see table 1 for a summary of studies assessing BC risk for RAD51C and RAD51D).
On the morning of Thursday, 30 September and the morning of Friday, 1 October 2021, a virtual consensus meeting was hosted as a collaboration between CanGene-CanVar (Cancer Research UK-funded catalyst project) and the UKCGG. The aim of the meeting was to develop guidelines around clinical management of patients with GPV in moderate-risk OC genes PALB2, RAD51C, RAD51D and BRIP1 (henceforth referred to as carriers) to clarify surveillance and risk-reducing options.

CONSENSUS MEETING
The meeting was held virtually via Zoom and was moderated by HH, FL, AK and MT on behalf of CanGene-CanVar and UKCGG. There were 65 attendees: clinical geneticists (31), genetic counsellors (9), gynaecologists (10), menopause specialists (1), clinical nurse specialists (1), breast surgeons and oncologists (5), radiologists (2), clinical scientists (1), primary care (1) and patient representatives (4). There was representation from the four devolved nations of the UK. The meeting also included commissioners (individuals involved in funding decisions at both NHSE (National Health Service England) and the devolved nations), although they did not take part in clinical discussions.
Prior to the meeting, a review of relevant literature concerning GPV in PALB2, RAD51C, RAD51D and BRIP1 was undertaken, and a background document circulated summarising published evidence for associated cancers, cancer risks, surveillance and risk-reducing surgery (background document available via https://www.ukcgg.org/information-education/ukcgg-consensusmeetings/). Participants were also asked to complete a survey (via SurveyMonkey) prior to the meeting, assessing opinions on genetic testing for GPV in moderate-risk OC genes, surveillance that should be offered to carriers and options around riskreducing surgery.
The two sessions of the meeting were divided into consideration of BC risks and management on the first morning and OC risks and management on the second morning. At the start of each session, the results from the premeeting survey were presented followed by short lectures covering risk assessment, surveillance and risk-reducing surgery by expert speakers. Following these lectures, open discussions were held around specific statements (as discussed further) to inform the clinical recommendations. Voting on each statement was undertaken using Slido, which allows real-time voting online. Statements were displayed and five options for answers from 'strongly agree' to 'strongly disagree' were proposed. Within this process, it is important to set a consensus level at the beginning . 15 A consensus was taken as 80% of respondents agreeing (voting agree and strongly agree). Alternative statements were debated until consensus was reached. The numbers of participants voting for each statement varied, depending on the expertise of the attendees. Notes were taken throughout the meeting and a draft document of the meeting outcomes was written and edited by FL and HH and then circulated via the core group for further input and agreement. Question 1: should RAD51C and RAD51D be included on a BC predisposition panel?
GPV in both RAD51C and RAD51D has been identified in BC and OC families. 16 17 For both genes, the association appears to be strongest with triple negative or ER (estrogen receptor) -negative BC. 8 18 The largest study to date 16 8 The cumulative lifetime risk of BC associated with truncating GPV in these genes is approximately 20% for both genes. 8 16 However, lifetime BC risk can be significantly modified by family history with a risk as high as 44%-46% for carriers, with two first-degree relatives diagnosed with BC. 16 Poll statement: RAD51C and RAD51D should be included on a BC predisposition panel.
Question 2: should BRIP1 be included on a BC predisposition panel?
In 2006 BRIP1 was reported as a low-penetrance BC gene. 19 However, subsequent larger and more comprehensive studies have suggested that there is not a significant association with BC predisposition. 18  Traditionally, diagnostic genetic testing has taken place within regional genetic services with the likelihood of identification of a GPV in a cancer predisposition gene calculated based on the family history of cancer. The main driver of testing was to facilitate predictive testing and subsequent surveillance and risk reduction strategies for unaffected family members. However, it is known that a high proportion BRCA1 and BRCA2 carriers do not have a significant family history of BC or OC. 22 More recent developments in personalised cancer management for individuals with GPV have resulted in lowered thresholds for germline testing with increasing numbers of individuals now eligible. As a result, mainstreaming testing (genetic testing through a non-genetics specialty at the time of new cancer diagnosis) has been widely adopted. Predictive genetic testing in an unaffected family member for a known familial GPV in a cancer predisposition gene is undertaken through Clinical Genetics, which can provide pretest counselling and assessment of residual risk based on family history and other risk factors if an individual has not inherited a familial variant. There have been a number of studies assessing mainstreaming pathways both from the UK 23 24 and internationally. 25 26 While there have been some initial concerns from non-genetic specialists around offering genomic testing, 27 many studies have found that the mainstreaming of testing is acceptable to both patients and healthcare professionals. 10 28-31 The group considered whether the addition of lower-risk genes to a BC predisposition panel would be of concern to mainstreaming clinicians due to unfamiliarity or uncertainty about associated cancer risks. This was weighed against the practicality of having multiple different panels, along with the consideration that all patients identified with a GPV would be referred into a genetics service for detailed discussions regarding cancer risk and surveillance/risk-reducing strategies alongside discussion of cascade testing for other at-risk family members. Their study also demonstrated that family history may modify this risk. With two first-degree relatives affected with BC, these lifetime risks increase to 44% and 40%, respectively. 17 Similarly, the BCAC study suggested higher OR estimates when the comparison was made between cases with BC family history and controls. One of the more commonly used risk algorithms, CanRisk (www.canrisk.org), incorporates both family history and carrier status into individual risk assessments, and this or similar models should be used to provide an individualised risk assessment. While clinical judgement can also be used to assess the extent of family history, CanRisk provides a more detailed risk assessment and estimated 5-year, 10-year and lifetime risks, which are helpful in counselling patients and in shared decision making. The CanRisk tool can also include questionnaire-based factors (eg, hormonal and lifestyle factors), polygenic risk scores and mammographic density, although the latter two are not currently routinely assessed in standard clinical practice.

Position statement
years from age 50). These guidelines also suggest that patients with a lifetime risk of BC between 30% and 40% (10-year risk of 8%-12% aged 40-50 years) (high risk) should undergo annual mammography until the age of 59 years and then revert to population screening. As previously mentioned, BC risk of RAD51C and RAD51D carriers can be modified by family history and other BC risk factors. Based on lifetime risk calculations undertaken in the CanRisk web tool which incorporates BOADICEA V.6 (www. canrisk.org) and considering the multifactorial model (including lifestyle/hormonal risk factors, mammographic density and polygenic risk scores), carriers can be classified into different risk categories. For an 'average' 20-year-old woman with a RAD51C GPV (without considering cancer family history), based on the multifactorial model, approximately 38% of carriers would fall into a population risk category; 43% of carriers would fall into a moderate-risk category; and 19% would fall into a high-risk category. However, for a 20-year-old RAD51C carrier and a mother affected with BC at age 50, based on the multifactorial model, 15% of carriers would fall into a population risk category; 42% of carriers would fall into a moderate-risk category; and 43% would fall into a high-risk category. 32 It was noted that, in some areas of the UK, despite NICE guidelines, there remains patchy provision of moderate-risk breast screening 33 and that access to this needs to be improved, as well as more standardised quality of reporting, as this screening currently lies outside the NHSBSP.
It was recognised that, at present, risk assessment is based predominantly on family history, but that other factors such as mammographic density and polygenic risk score could also modify risk and consequently recommendations for surveillance. The group commented that future work should focus on developing and implementing new clinical pathways that incorporate these additional risk factors. The NHS VHR screening programme offers a combination of annual mammography and MRI screening to patients at VHR of BC between the ages of 25-30 years and 70 years. 12 This is defined as 'women with a lifetime risk of 40% or greater due to a specific genetic abnormality in the woman or her family'. To access the VHR screening programme, an individualised risk assessment using an NHS endorsed computer risk modelling software program such as CanRisk needs to be undertaken to demonstrate that 10-year BC risks are greater than 8% between the ages of 25 years and 29 years, 8% between the ages of 30 years and 39 years, or 12% between the ages of 40 years and 49 years. While most RAD51C and RAD51D carriers are unlikely to reach this level of BC risk, a small number of patients may reach this level of risk based on the strength of their family history and/ or other modifying factors. 16 Poll NICE guidelines include RRM as part of the pathway for managing patients with a VHR of BC. There is a strong body of evidence demonstrating a greater than 90% risk reduction associated with RRM in patients with GPV in BRCA1 and BRCA2. 34 There is also emerging evidence that surgery will increase survival. [35][36][37] However, no formal study of RAD51D or RAD51C carriers has been undertaken. The group discussed that as for patients with a strong family history of BC who meet a 30% or greater lifetime risk of BC, discussion of the option of RRM with a patient is appropriate. However, discussion with the patient should be based on individual circumstance and shared decision making. Given the lack of studies for RAD51C and RAD51D carriers, detailed counselling with patients should include, but not be restricted to, individualised cancer risk assessment, personal circumstance and preferences of the counsellee. Nongenetic risk factors such as dense breast tissue, hormonal/lifestyle modifiers and other pre-existing medical conditions should also be considered. Importantly, age-specific risksand 5 and 10 year risks, should be communicated to the patient to help in their decision making. It was noted that some patients may fall into a level of risk where RRM, but not surveillance within the VHR surveillance programme, is offered (lifetime risk 30%-39% inclusive). In this situation, particular consideration should be paid to detailed discussion of age-specific risks.
Poll statement: RRM should be discussed with RAD51C and RAD51D carriers with a lifetime BC risk of ≥30%. Currently, population screening for OC is not recommended in the UK due to lack of evidence of a mortality benefit. 38  Given the lack of specific studies of the utility of OC surveillance in RAD51C, RAD51D, BRIP1 and PALB2 carriers, the group discussed whether surveillance could or should be offered within the context of a research study. While the majority of the group agreed with this approach, there was also concern that if OC surveillance in BRCA1 and BRCA2 carriers was supported in the future, then it would be difficult not to extend this to moderate-risk gene carriers, as OC in RAD51C, RAD51D, BRIP1 and PALB2 carriers is similarly, most likely to be highgrade serous OC, compared with other OC pathologies. 41 While there is no formal analysis, it is likely that the cost per OC case detected would be higher for the moderate-risk genes compared with BRCA carriers, given the overall incidence of OC is lower. However, the absolute cost of adding RAD51C, RAD51D, BRIP1 and PALB2 carriers to any pre-existing high-risk programme would be low due to relatively small numbers. 41 43 calculated a cumulative risk of OC to the age of 80 years of 5.8% (95% CI 3.6 to 9.1%). This study gave a larger range of OR for risks which may be indicative of the influence of family history of OC on risk.
Presence of a GPV in PALB2 was demonstrated in a study of 852 carriers 44 to give an RR of OC of 2.91 (95% CI 1.4 to 6.04). Age-specific risks were then calculated with the lifetime risk to 80 years being quoted as 5% (95% CI 2% to 10%). A further study of risks estimated the cumulative lifetime risk to age 80 years to be 3.2% (95% CI 1.8% to 5.7%). 45 As with RAD51C and RAD51D, the risk of OC associated with GPV in PALB2 appears to be modified by family history of OC. For example, female carriers with a mother and sister with OC diagnosed at 50 years have a lifetime risk of 16% (95% CI 8% to 28%). 44 In addition to family history, it is recognised that other hormonal and lifestyle factors can modify OC risk, including use of oral contraception, hormonal replacement therapy (HRT), parity, body mass index, tubal ligation and endometriosis. All these factors can be incorporated into the CanRisk model alongside family history to provide an individualised risk assessment. In addition, polygenic risk scores may also modify risk in either direction but at present are not available in routine clinical practice.
Poll statement: Lifetime risk of OC for a RAD51C, RAD51D, BRIP1 and PALB2 carrier should be based on an individualised risk assessment taking family history into consideration. Poll results: 35% strongly agree; 63% agree (98% consensus, n=51). Recommendation: consensus reached that lifetime risk of OC of a RAD51C, RAD51D, BRIP1 and PALB2 carrier should be based on an individualised risk assessment taking family history into consideration.

Question 12: when discussing risk-reducing bilateral salpingooophorectomy (RRSO) with a patient, what risks should be considered?
The risks of OC for RAD51D and RAD51D carriers are largely conferred after the age of 50 years. The highest 10-year risk with GPV in RAD51C is between the ages of 50 and 60 with the highest risks between 50 and 70 years in RAD51D carriers (Yang, personal communication, 2021). These risks as previously described vary with family history. 16 To discuss RRSO, both lifetime and 5-10 year OC risks should be considered, balancing risks versus benefits of surgery, with particular consideration to the average age of menopause in the population and individual menopause status of the patient. Other issues such as fertility and the impact of premature menopause may also affect timing of surgery and patient decision making. For BRIP1 carriers, the majority of OC risk also occurs after the age of 50 years. The average age of diagnosis for a BRIP1 carrier was at age 63.8 years compared with 58 years in non-carriers. 43 This

Position statement
was replicated in a study assessing 222 patients with OC and BRIP1 GPV. 46 In this study, 90% of cases occurred after the age of 50 years with a median age of 65 years. A recent study suggested that the BRIP1 variant c.1045G>C is a higher-risk allele, 47 although this still demonstrated a mean age of diagnosis of 62.5 years.
Yang et al, 44 in a study of 852 female PALB2 carriers, demonstrated that the majority of OC risk is over the age of 50 years, with an estimated cumulative risk below this age of less than 1%. This was supported by a further study 45 with a cumulative risk of OC of less than 1% under the age of 50.
Poll statement: Discussion of RRSO should consider both lifetime risk and risks of 5-10 years.
It is recognised that in young women undergoing surgical oophorectomy, there is an impact on both morbidity and mortality. 48 The sequelae can include vasomotor symptoms, decrease in sexual function, osteoporosis, increased risk of cardiovascular disease, depression, anxiety, dementia and cognitive decline, and multimorbidity. [49][50][51][52][53] While HRT is routinely recommended and a number of these outcomes are attenuated by the use of HRT, not all of the sequelae are fully mitigated by this. 51 53-62 Other issues that need to be considered include fertility. RRSO is only recommended once childbearing is complete.
Studies have demonstrated need for information about the postsurgical effects of RRSO. 63 64 The group considered the need for detailed discussions with patients to ensure they are aware of the potential side effects so that these can be balanced with individualised discussion of risk.
Poll statement: Discussion of premenopausal RRSO should include a full detailed discussion of OC risk versus potential sequelae of early menopause. Poll results: 72% strongly agree, 28% agree (100 consensus, n=53). Recommendation: consensus reached that discussion of premenopausal RRSO should include a full detailed discussion of OC risk versus potential sequelae of early menopause.

Question 14: at what level of risk should RRSO be offered?
Historically, in UK practice, a lifetime OC risk of 10% has been used as the threshold of risk for discussion of RRSO. However, prior to comprehensive risk assessment models for OC, both with and without a recognised causative GPV in a family, calculation of individualised risk has been complex. As a result, most typically, discussion of RRSO has been based on either a genetic diagnosis in a family or clinically based criteria.For example, two or more cases of OC in a family. Recent studies in the UK 65 66 have demonstrated the cost-effectiveness of RRSO above a threshold of 4%-5% lifetime risk. These studies suggest that patients with a lifetime risk above this threshold should be offered the opportunity to discuss RRSO. However, while surgery will decrease OC risk, there are potentially long-term sequelae associated with surgical oophorectomy, as described previously.
Considering the aforementioned, the risk threshold for offering RRSO was discussed. The group considered that counselling of patients should include an individualised risk assessment with discussion of both lifetime risk and 5-10 year risks (see question 12) that takes genetic test results and family history as a minimum into the risk assessment. Discussion should also include counselling on the possible sequelae of an early menopause (see question 13).  (table 2A). The risk-to-benefit ratio for surgery therefore changes over the age of 50 years. However, risk can also be modified by family history, and the risk classifications presented earlier, based on the multifactorial model, will also be family history specific (table 2B). Therefore, the group considered that an individualised risk assessment, as well as assessment of menopausal status, should be undertaken for all RAD51C carriers. The initial poll statement 'For RAD51C carriers RRSO should only rarely be considered <50 years and should include individualised risk assessment and shared decision making' was reworded following detailed group discussion to reflect the importance of these considerations to 'For RAD51C carriers with a 5% or greater lifetime risk, RRSO should be considered at 50. It can be considered in patients younger than 50 following individualised risk assessment including assessment of menopausal symptoms and shared decision making'. Poll question: For RAD51C carriers with a 5% or greater lifetime risk, RRSO should be considered at 50. It can be considered in carriers younger than 50 following individualised risk assessment, including assessment of menopausal symptoms and shared decision making.
Poll results: 14% strongly agree; 86% agree (100% consensus, n=43). Recommendation: consensus reached that for RAD51C carriers with a 5% or greater lifetime risk, RRSO should be considered at 50 years. It can be considered in carriers younger than 50 years following individualised risk assessment including assessment of menopausal symptoms and shared decision making.

Question 16: at what age should RRSO be considered for RAD51D carriers?
Yang et al estimated that the cumulative risk of OC for RAD51D carriers to 50 years is 0.8% (95% CI 0.5% to 2.0%) and 13% (95% CI 7% to 23%) to age 80 years. CanRisk data presented in the meeting by AA consider that the average OC risk for a RAD51D carrier is 13% to age 80 years. Based on the multifactorial model, 2% of RAD51D carriers fall below a lifetime risk of 5%. However, when considering risk to age 50 years, the average risk is 0.9%, and based on the multifactorial model, 99.9% of carriers fall below a 3% risk before age 50 years (table 2A). This risk can be influenced by family history, with risk rising with increasing number of first-degree relatives affected with OC and risk classification also dependent on cancer family history (table 2B). Like the discussions for RAD51C carriers, the group felt that the age at which to consider RRSO should reflect both an individualised risk assessment and also menopausal status, and rewording of the poll question for voting was undertaken in real time during the meeting to reflect this. Poll question: For RAD51D carriers with a 5% or greater lifetime risk, RRSO should be considered at 50. It can be considered in carriers younger than 50 following individualised risk assessment including assessment of menopausal symptoms and shared decision making. Poll results: 14% strongly agree; 86% agree (100% consensus, n=44).
Recommendations: consensus reached that for RAD51D carriers with a 5% or greater lifetime risk, RRSO should be considered at 50 years. It can be considered in carriers younger than 50 years following individualised risk assessment, including assessment of menopausal symptoms and shared decision making.

Question 17: at what age should RRSO be considered for BRIP1 carriers?
The cumulative OC risk associated with pathogenic variants in BRIP1 is 5.8% (95% CI 3.6% to 9.1%). It would appear that the risk is highest after the age of 50 years, with the average diagnosis being at 63 years. 43 A further study of BRIP1 carriers demonstrated that 90% developed OC after the age of 50, with a median age of diagnosis of 65 years. 46 CanRisk data presented in the meeting model the average OC risk for a BRIP1 carrier to be ~6% to age 80 years, and based on the multifactorial model, 47% carriers fall below a lifetime risk of 5%. When considering risk to age 50 years, the average risk is 1.0%, with 99.7% of carriers falling below a 3% risk before age 50 years when using the multifactorial model (table 2A). Again, reflecting the previous discussions for RAD51C and RAD51D carriers, the group felt that the age at which to consider RRSO for BRIP1 carriers should include both an individualised risk assessment and also menopausal status, and rewording of the poll question for voting was undertaken in real time during the meeting to reflect this. It was also noted that unlike RAD51C and RAD51D, many BRIP1 carriers, in the absence of OC family history, may not reach a lifetime OC risk of 5%, and calculation of an individualised risk assessment was therefore fundamental to patient discussions ( A study of 852 PALB2 carriers calculated age-specific OC risks. It was estimated that there is 0.6% risk (95% CI 0.3% to 1.0%) to 50 years and 5% (95% CI 2% to 10%) to age 80 years. 44 A further study suggested the risk is 3.2% until 80 years. 45 Family history can modify this risk. Given low cumulative risk under the age of 50 years, recent guidelines from the American College of Medical Genetics and Genomics suggested that RRSO be discussed only from 50 years onwards. 11 CanRisk data presented in the meeting model the average OC risk for a PALB2 carrier to be ~5% to age 80 years, with many PALB2 (62%) carriers falling below a lifetime risk of 5%, based on the multifactorial model and absence of OC family history (

Position statement
Recommendation: consensus reached that for PALB2 carriers with a 5% or greater lifetime risk, RRSO should be considered at 50 years. It can be considered in carriers younger than 50 years following individualised risk assessment, including assessment of menopausal symptoms and shared decision making.
Question 19: should the option of risk-reducing early salpingectomy with delayed oophorectomy be considered for RAD51C, RAD51D, BRIP1 and PALB2 carriers?
There is now evidence that OC can arise from the fallopian tubes and that removal of the fallopian tubes may therefore decrease the risk of OC. [68][69][70] This raises the possibility of using salpingectomy with delayed oophorectomy to decrease OC risk while delaying surgical menopause in carriers at increased risk of OC. A recent study from the Netherlands assessed quality of life after risk-reducing salpingectomy versus RRSO in 577 BRCA1 and BRCA2 carriers. This demonstrated a better quality of life after risk-reducing salpingectomy irrespective of HRT. 71 Within the UK, the PROTECTOR study is evaluating the option of risk-reducing early salpingectomy with delayed oophorectomy in patients at high risk of OC. BRCA1, BRCA2, PALB2, BRIP1, RAD51C and RAD51D carriers are eligible for this study. 72 While there are several ongoing studies about the acceptability of this approach, there are still no long-term data on outcomes, in particular around OC diagnoses in these cohorts. Recent reviews by Boerner et al and Gaba et al suggested that this surgery should be offered only in the context of a clinical trial. 72 73 Poll question: Risk-reducing early salpingectomy with delayed oophorectomy should be offered only to RAD51C, RAD51D, BRIP1 and PALB2 carriers within the context of a research study.

DISCUSSION
Clear management guidelines exist for BRCA1 and BRCA2 carriers with guidance around surveillance and risk-reducing surgery for BC and OC. While the contribution of GPV in PALB2, BRIP1, RAD51C and RAD51D is smaller than for BRCA1 and BRCA2, the inclusion of these genes on breast and OC gene panels has resulted in the need for similar guidance around management of carriers for GPV in these genes. However, there are no guidelines setting out surveillance or encompassing all cancer risk management available in the UK. As with much of clinical genetics, clear evidence of the optimal management of affected individuals is scarce due to the rarity of the disease and requirement for very long-term follow-up studies to generate data on which to base guidelines. A multidisciplinary workshop was therefore convened to draw on expert clinical experience. By the end of the two-session workshop, a consensus (over 80% agreement) had been obtained for a majority of recommendations for best clinical practice for carriers of GPV in RAD51C, RAD51D, PALB2 and BRIP1. A summary is presented in Box 1.
In summary, carriers of GPV in these genes should have a detailed discussion about their family history, individualised risk assessment and offered RRSO at the appropriate level of risk and age. RAD51C, RAD51D and PALB2 carriers should also have an individualised risk assessment for BC and be entered into the appropriate breast screening programme.
While these guidelines suggest 'best practice' management, there are a number of issues which impact on the implementation of the guidelines, including issues around resources and geographical differences in the delivery of care. It is suggested that RAD51C or RAD51D carriers should have enhanced BC risk screening as defined in the NICE CG164 familial BC guidelines. 6 However, from discussions in the workshop, it is apparent that, in line with a previous publication, 33 provision of services for moderaterisk and high-risk breast surveillance varies around the UK. This needs to be addressed, not least in response to the NHS Long Term Plan (2019), which aims to increase the number of early cancer diagnosis by screening. Populations at increased risk of malignancy, such carriers of GPV in these moderate-risk genes, are populations for targeted screening, fulfilling criteria set by the National Screening Committee (https://www.gov.uk/government/publications/ evidence-review-criteria-national-screening-programmes/ criteria-for-a-targeted-screening-programme).
VHR breast screening is included within the NHSBSP for England, and there is now agreement to offer colonoscopy surveillance for individuals with Lynch syndrome within the NHS Bowel Screening programme with roll-out anticipated in 2023. One possibility is that moderate-risk and high-risk screening could be incorporated into the NHSBSP (as for VHR) so current inequities for access and reporting can be improved and standardised. However, there are issues with the funding and services offered to the devolved nations resulting in disparate provision across the UK.
The workshop has produced clear guidance around recommendations of risk levels at which to consider RRSO. However, it should be remembered that the risk estimates based on multifactorial risk models may be associated with wide CIs based on the uncertainty related to input parameters. Risk estimates are also dependent on the accuracy, quality, validity and extent of the information input into the tool. External, independent validation studies provide useful guidance on these aspects. 74 As such, discussions with patients will require explanation of the variation in risk estimations along with detailed discussions about timing of surgery and potential sequelae. This will then facilitate shared decision making with each individual patient to optimise care as per NICE guidance 197. 75 These sequelae may include menopausal symptoms, depending on the time of the surgery. Women should have access to discussions around HRT or alternative therapies when appropriate. However, it was highlighted in the workshop that while some centres of excellence exist, access to specialised menopausal care is variable around the UK.
Overall, with increased numbers of RAD51C, RAD51D, BRIP1 and PALB2 carriers likely to be identified imminently with updates to the national test directory within all four UK nations, we believe that these guidelines represent a framework for consistent and best practice based on the current evidence. It is likely that new relevant information will be published in the next 5 years, both from larger studies of carriers and results from studies addressing ovarian surveillance and risk-reducing surgery; therefore, regular review and updates will be required and discussions with patients should also include the potential for clinical recommendations to change over time, as and when new evidence becomes available.