Risk-reducing surgery

Risk reducing surgery

• Evidence from a cost-utility analysis (Wei 2024) using modelling suggests that, for women with BRCA1 combined RRBM at 30 years and RRBSO at 35 years is likely to be cost-effective when compared to high-risk breast cancer surveillance and tamoxifen from age 30, RRBM at age 30, and RRBSO at age 35 with high-risk BC surveillance and tamoxifen from age 30. For women with BRCA2 combined RRBM at 35 years and RRBSO at 40 years is likely to be cost-effective when compared to high-risk breast cancer surveillance and tamoxifen from age 30, RRBM at age 35, and RRBSO at age 40 with high-risk breast cancer surveillance and tamoxifen from age 30. For women with PALB2, combined RRBM at 40 years and RRBSO at 45 years is the optimal strategy compared to high-risk breast cancer surveillance and tamoxifen from age 30, RRBSO at age 45 with high-risk breast cancer surveillance and tamoxifen from age 30, and RRBM at age 40. For women with RAD51C and RAD51D, RRBSO at 45 years with moderate-risk breast cancer surveillance and tamoxifen from age 40 is likely to be cost-effective when compared to moderate-risk breast cancer surveillance and tamoxifen from age 40 only. For women with BRIP1, RRBSO at 45 years is likely to be cost-effective compared to no surgery. The study is directly relevant to the NICE’s decision-making context and has minor limitations.
• Evidence from a cost-utility analysis (Bommer 2022) using modelling indicates that combined risk reducing bilateral mastectomy (RRBM) and risk reducing bilateral salpingo-oophorectomy (RRBSO) is likely to be dominant when compared to intensified surveillance, chemoprevention with Tamoxifen, RRBM alone and RRBSO alone in adult women with BRCA pathogenic variants in Switzerland. The study is partially applicable to NICE’s decision-making context and has minor limitations.
• Evidence from a cost-utility analysis (Müller 2018) using modelling suggests that combined RRBM and RRBSO at 30 years is likely to be the preferred option compared to intensified surveillance, RRBM alone, RRBSO alone, and RRBM and RRBSO at 40 years in adult women with BRCA pathogenic variants in Germany. The study is partially applicable to NICE’s decision-making context and has minor limitations.
• Evidence from a cost-utility analysis (Yamauchi 2018) using modelling suggests that combined RRBM at 35 years and RRBSO at 45 years is likely to be the preferred option compared to intensified surveillance from 35 years and RRBSO at 45 years, and RRBM only at 35 years in adult women with BRCA1 pathogenic variants in Japan. The study also found that in women with BRCA2 pathogenic variants, RRBM only was the preferred option compared to all the other options. The study is partially relevant to NICE’s decision-making context and it has potentially serious limitations.

Thresholds for risk reducing surgery

• Evidence from a cost-utility analysis using modelling (Manchanda 2016) in the UK indicates that offering RRBSO to premenopausal women aged over 40 with at least a 4% lifetime ovarian cancer risk may potentially be cost-effective compared to not offering RRBSO at this lifetime ovarian cancer risk. The study is directly relevant to NICE’s decision-making context and has minor limitations.
• Evidence from a cost-utility analysis using modelling (Manchanda 2015) in the UK suggests that offering RRBSO to low/intermediate risk postmenopausal women aged 50 or older with at least a 5% lifetime ovarian cancer risk may potentially be cost-effective compared to not offering RRBSO at this lifetime ovarian cancer risk. The study is directly relevant to NICE’s decision-making context and has minor limitations.

Factors to take into account when considering risk-reducing surgery

The committee discussed that there are a number of general factors that need to be considered in relation to risk-reducing surgery. They based their recommendations on the effectiveness evidence of improved outcomes such as disease-free survival and cancer incidence which showed an important benefit of surgery as well as economic evidence. The quality of the effectiveness evidence was mainly high and the majority of the economic evidence had only minor limitations. They also noted the fact that ovarian cancer starts in the organs that are removed and so the committee agreed that surgery is clearly the most effective risk-reduction option (and clearly more effective than surveillance – see evidence review K for details). Based on experience they noted that it does not completely remove the risk of cancer because there is a small risk of peritoneal cancer. They discussed that bilateral salpingo-oophorectomy has direct consequences, for example the person can no longer become pregnant and enters menopause. On the balance of benefits and risks the committee decided that completion of family should be one of the deciding factors when risk-reducing surgery is offered because the incidence of ovarian cancer in people younger than 35 is relatively small (which is consistent with the findings of the economic model). Due to surgically induced menopause as a life changing consequence of salpingo-oophorectomy the committee also agreed that the risk level would need to be high enough to balance risks and benefits. They considered lifetime risk and noted that the economic evidence (such analyses weigh up the benefits, risks and costs) showed that a threshold level of 4% lifetime risk in people who are premenopausal would be cost-effective and 5% cost effective for people post menopause. The difference in lifetime risk is due to the risk of ovarian cancer decreasing after menopause due to hormonal changes and also that postmenopausal people can no longer through natural conception pass genetic risk on to their children. Such lifetime risk calculations would depend on whether they have a pathogenic variant or whether there is a verified family history of ovarian cancer for them or a family member (for example verified via the Cancer Registry or other medical documents). The committee agreed that this level of risk would minimise people having unnecessary surgery. The committee reflected on these different 4% and 5% lifetime risk thresholds and discussed that this would be difficult to implement and may result in potential inequalities and other unintended consequences (for example people may feel pressured into premenopausal surgery to avoid having to meet a higher risk threshold). They therefore decided to set a lifetime risk threshold of 5%.

The committee recognised, based on experience, that decisions around risk-reducing surgery can be distressing for people because for premenopausal women it would mean that they would become menopausal and can no longer have children and for postmenopausal women it is a surgical procedure associated with some risks. This could influence their ability to come to a decision about having surgery which could potentially be lifesaving for them and the committee emphasised that psychological factors (such as distress and anxiety) should be taken into account, including what psychological support may be available. The committee also noted, based on experience, that sometimes a referral for psychological support may be needed (because of the level of distress and anxiety and the level of the person’s risk) so that the person is supported in decision making and psychological distress is addressed.

The committee discussed early menopause as a consequence of risk-reducing surgery for premenopausal women. They decided that it was important that the person would receive specialist menopause counselling before (to be prepared for what to expect in relation to the menopause), and after surgery (to discuss potential menopause symptoms and associated treatments). They also recommended that information is provided (see section below on information provision).

The committee noted, based on their knowledge and experience, that decisions about risk-reducing surgery for people who are carriers of bi-allelic pathogenic variants in mismatch repair genes (for example, homozygous PMS2) are complex. However, they are also very rare so the committee agreed that a referral to a specialist multidisciplinary team would be needed for discussions about potential risk-reducing surgery.

Types of risk-reducing surgery and timing in relation to the person’s specific pathogenic variant

The committee discussed the evidence of an important benefit of bilateral salpingo-oophorectomy in terms, that is that bilateral salpingo-oophorectomy improves disease-free survival as well as the detection rate of early-stage ovarian cancer. They noted that most of the evidence came from studies with carriers of the BRCA1 or BRCA2 variants. Based on the evidence, they recommended bilateral salpingo-oophorectomy for people at increased risk of ovarian cancer with BRCA1 and BRCA2, and also RAD51C, RAD51D, BRIP1 or PALB2, which are also associated with an increased risk of ovarian cancer.

The MLH1, MSH2 or MSH6 pathogenic variants are associated with Lynch syndrome, which is associated with an increased risk of endometrial as well as ovarian cancer. Although there was no evidence identified related to different types of surgery within this specific group, the committee decided that total hysterectomy as well as bilateral salpingo-oophorectomy should be recommended to prevent both of these types of cancers. In terms of the specific criteria related to pathogenic variant and age, the committee recommended it based on the UK Cancer Genetics Group and the economic analysis. The UK Cancer Genetics Group (UKCGG) base their age ranges for each pathogenic variant on the difference between the general population risk of cancer (which they took from Cancer Research UK) and the risk of cancer for the specific variant (ascertained from specific related publications – see relevant UKCGG information). For example, for BRCA1 the risk increases to above population risk from age 31 onwards and then increases at a faster rate from that age onwards. The economic model presented to the committee by an expert witness (which was specifically designed to address variant and age) used the UKCGG data and started from age 30 to clarify at which age risk-reducing surgery would be most cost effective. This was done for each pathogenic variant most associated with ovarian cancer. The model was set up in this way to avoid risk-reducing surgeries taking place earlier than necessary given a particular risk level (see ‘cost effectiveness and resource use’ below).

PMS2 is a pathogenic variant that is also associated with Lynch syndrome, but it is not associated with ovarian cancer compared to MLH1, MLH2 and MSH2 but with endometrial cancer only. They decided to not include it in the table of types of risk-reducing surgery alongside the other Lynch pathogenic variants, because PMS2 increases the risk of endometrial cancer alone rather than endometrial as well as ovarian cancer. The committee decided that it should be mentioned because of its connection to Lynch syndrome which is included in the scope of the guideline and because it is on the gene panel for Lynch syndrome. Therefore, the committee agreed, base on expertise that total hysterectomy can be considered (weaker recommendation) in people with this pathogenic variant (no earlier than age 45). This is in line with UKCGG but was not something that was specifically modelled in the economic analysis because of it being linked to endometrial rather than ovarian cancer. When a person with a PMS2 pathogenic variant also has a family history of ovarian cancer the committee decided that a total hysterectomy as well as a bilateral salpingo oophorectomy should be considered because both the risk of endometrial and ovarian cancer would be increased.

Whilst the committee agreed that the earliest ages they selected for risk-reducing surgery were those with the best balance of risks and benefits, they discussed that there could be exceptional circumstances where risk-reducing surgery may be relevant and appropriate at a younger age (for example when the risk is very high).

The committee discussed that delayed oophorectomy would avoid surgical menopause and could therefore be a preferred option. They noted that some of the evidence related to this showed promise, for example, moderate quality evidence showed that women who had salpingectomy with delayed salpingo-oophorectomy reported fewer climacteric symptoms 12 months after surgery as compared to women who had bilateral salpingo-oophorectomy. However, the evidence for this comparison mainly relates to quality of life and patient satisfaction outcomes, and there was no evidence identified for the critical outcomes such as disease-free survival and ovarian cancer detection. They therefore only recommended this in the context of a clinical trial. They did not recommend research into this because they were aware that a trial was currently in progress which was large enough and with a long enough follow-up to address this (the PROTECTOR trial).

They noted that for most pathogenic variants associated with ovarian cancer (apart from those associated with Lynch syndrome) the risk of endometrial cancer was not significantly increased above population level, so they recommended against total hysterectomy unless a personalised risk assessment shows a high risk of endometrial cancer (due to other reasons) or there is another gynaecological indication for hysterectomy.

Tests before risk-reducing surgery, referral to the gynaecology oncology multidisciplinary team, and what to consider during surgery

Based on experience and expertise, the committee, decided that transvaginal ultrasound and a serum CA125 tests should be performed before risk-reducing salpingo-oophorectomy surgery because they are tests that can identify asymptomatic tubal or ovarian cancer. If only a total hysterectomy is planned, then the test should be an endometrial biopsy which can detect asymptomatic cancer in the womb. Whilst this was not part of the evidence that was looked for, the committee based on expertise, agreed that it is crucial to do this because the type of management would be different if a person is shown to have cancer.

There was high quality evidence that bilateral salpingo-oophorectomy improves detection rates for asymptomatic cancer. Based on this evidence the committee recommended referral to the gynaecology oncology multidisciplinary team if asymptomatic cancer is identified so that cancer treatment can be planned.

In terms of surgical techniques, the committee noted that most of the studies used minimal access surgery. Whilst there was no direct comparison between minimal access and open surgery the committee agreed, based on experience, that this is generally the preferred and safer option. They also discussed that some of the evidence included peritoneal washing, but the study included this in both arms of the comparison. It was therefore unclear whether this would be more effective than not using it. Despite this uncertainty in the evidence, the committee were aware that cancerous cells can spread to the peritoneal cavity and recommended to take peritoneal washings to prevent missing cancerous cells which could be spreading. In their knowledge and experience, the committee, were aware that up to 5% of incidental cancers could be missed if ultrasound alone is used, and that ultrasound is also particularly unreliable in Lynch syndrome. They therefore recommended that any lesions noticed during surgery should be investigated – even if they are found outside the organs that are being removed (such as in the peritoneal cavity) – to increase the likelihood of finding any asymptomatic cancers. The committee noted, based on expertise, that early detection of cancerous cells and timely intervention are essential to improving outcomes.

The committee noted that it is general good practice to investigate any lesions that are noticed during surgery even if they are found outside the organs that are being removed, to increase the likelihood of finding any asymptomatic cancers.

Information about risk-reducing surgery

The committee agreed that, when discussing a potential risk-reducing surgery, there are some key issues that the woman will need to know about to be able to make an informed decision. They acknowledged that people affected by this condition reported that they were not always satisfied with the information that they were receiving (see evidence review A) and that it would therefore be important to list the minimum information that should be given related to risk-reducing surgery so that this is standard practice.

Not all people may be aware of what risk-reducing surgery is and how it would be carried out so in the shared decision-making process this information should form the starting point for the discussion. Based on the clinical evidence and reasons described above, advice should be given about the effectiveness of risk-reducing surgery as the most reliable way to reduce the likelihood of developing ovarian cancer. The committee noted, based on experience, that there is a misconception that risk-reducing surgery would eliminate the risk completely and they therefore recommended that it should be explained that there will still be a small risk that remains.

There is information to be provided about risk levels associated with different pathogenic variants and the timing around risk-reducing surgery that would be important for the woman to know about.

As described above there could be psychological distress and symptoms of the menopause that may have an impact on the person’s sex life (genitourinary symptoms) and any other ways that an early menopause could affect them.

There are some pathogenic variants that also increase the risk of other cancers, such as increased risk of breast cancer associated with BRCA1 and BRCA2 and to be able to make informed choices the person needs to be aware of these risks.

It was discussed that people may not know which local or national organisations could support them and may also not know that there are peer support groups. They discussed that there are a number of support organisation and that people ought to be made aware that they exist (for example The Eve Appeal, BRCA Umbrella and ovarian cancer action).

Other factors the committee took into account

The committee acknowledged the BRCA1 and BRCA2 not only increase the risk of ovarian cancer but also the risk of breast cancer. Risk-reducing surgery for breast cancer therefore also needs to be considered. The committee therefore cross referred to the NICE guideline on familiar breast cancer so that the relevant recommendations on risk reducing mastectomy are taken into account.

As part of the considerations around risk-reducing surgery the issue of surgery as part of gender affirming care for trans men and non-binary people registered female at birth was discussed. No evidence matching the review protocol was identified for these groups of people but the committee was aware of some recently published guideline that was making reference to this. They noted that anyone who is high risk may have surgery at a younger age if that is appropriate and advised by the specialist for gender affirming care. That is the context for having the procedure at that time point and that would be independent of risk reduction. They emphasised that rationale for earlier surgery cannot be risk reduction as the risk is not high enough to reduce at that time point. Therefore, they concluded that this type of surgery at a younger age is outside the scope of this guideline and did not comment on this.

Cost effectiveness and resource use

There was UK-based evidence on the cost-utility of risk-reducing surgery in individuals with pathogenic variants that increase ovarian cancer risk. The committee discussed the findings which indicated that risk reducing bilateral mastectomy at age 30 and risk reducing bilateral salpingo-oophorectomy at age 35 was the optimal strategy for BRCA1. For BRCA2 risk reducing bilateral mastectomy at age 35 and risk reducing bilateral salpingo-oophorectomy at age 40 was the optimal strategy. For PALB2, combined risk reducing bilateral mastectomy at age 40 and risk reducing bilateral salpingo-oophorectomy at age 45 was deemed optimal, while risk reducing bilateral salpingo-oophorectomy at age 45 with moderate-risk breast cancer surveillance and tamoxifen from age 40 was optimal for RAD51C and RAD51D. For BRIP1 risk reducing bilateral salpingo-oophorectomy at age 45 was the optimal strategy.

The committee found it encouraging that probabilistic sensitivity analysis demonstrated that, at the NICE cost-effectiveness threshold of £20,000 per QALY, the combined risk reducing bilateral mastectomy and risk reducing bilateral salpingo-oophorectomy strategy was the most cost-effective in a high percentage of simulations: 96.5% for BRCA1, 89.2% for BRCA2 and 84.8% for PALB2. Risk reducing bilateral salpingo-oophorectomy at age 45 was the optimal strategy in 100% of simulations for RAD51C/RAD51D/BRIP1.

Furthermore, the committee found it reassuring that even when varying parameters at the extremes of their confidence intervals or ranges, the ICERs for risk-reducing surgeries remained below the lower NICE cost-effectiveness threshold of £20,000 per QALY gained. Similarly, the committee acknowledged that the conclusions were unchanged in various scenario analyses. These analyses included varying ages of risk-reducing surgeries, modelling lower hormone replacement therapy adherence, changing overall mortality after RRBSO assumptions, and including PARP-i treatment costs.

The committee acknowledged the direct applicability of this evidence to NICE’s decision-making process, noting only minor methodological limitations. They explained that the findings were as expected and aligned with the current practice.

The committee also considered other existing economic evidence, comprising three non-UK studies focusing on BRCA carriers. All these studies evaluated slightly different risk-reducing strategies and age thresholds for risk-reducing surgeries. Three studies concluded that risk reducing bilateral mastectomy and risk reducing bilateral salpingo-oophorectomy were optimal for individuals with BRCA, with varying risk-reducing surgery initiation ages ranging from 30 to 45 years.

The committee noted that this non-UK evidence was partially applicable to the NICE decision-making context. Also, even though these studies were well conducted and had only minor methodological limitations the committee discussed the difficulty of generalising from these studies due to potential differences in cost inputs. For example, cancer management and risk-reducing surgery costs in the NHS are likely to be different.

The committee highlighted that before risk-reducing surgery, information provision and support are crucial and recommendations reflect good practice that should be already undertaken by services. The decision to undergo risk-reducing surgery is complex and psychological support is essential, which should already be available. However, they recognised the potential strain on specialist psychological services due to the lack of such services.

Risk-reducing surgery can induce surgical menopause in premenopausal people. Therefore, comprehensive menopause counselling is essential to ensure people understand the surgery’s implications and their treatment options, including associated risks and benefits. The committee noted that these recommendations reflect current practice across services. Furthermore, they acknowledged the complexity of managing risk-reducing surgery decisions in people with bi-allelic pathogenic variants in mismatch repair genes, such as homozygous PMS2, and expect such decisions to be currently undertaken by specialist tertiary teams.

The committee explained that hysterectomy is standard practice for endometrial cancer. In people over 45 with a confirmed family history of ovarian cancer, it would be rare to leave the ovaries if a hysterectomy is being performed. Undertaking these procedures simultaneously could lead to cost savings due to reduced need for separate pre- and post-operative care, shorter overall hospital stays and earlier quality of life improvements. The recommendation not to perform hysterectomies in people with certain pathogenic variants unless, for example, there is a high endometrial cancer risk should align with most services’ current practices. However, making this explicit could potentially reduce the number of unnecessary risk-reducing hysterectomies.

All other recommendations reinforce current practice, including preoperative testing before risk-reducing surgery, referring asymptomatic individuals to the gynaecology oncology multidisciplinary team if cancer is, for example, detected during preoperative investigation, and procedures during risk-reducing surgery. However, it was acknowledged that where such care is currently suboptimal, there could be some additional resource implications.

The committee also noted that widening the genetic testing criteria may lead to an increase in the number of people undergoing risk-reducing surgery, requiring expansion of services. However, they highlighted that any additional costs associated with this expansion will be outweighed by a decrease in cancer risk and its associated costs.