Factors that impact risk management decisions among women with pathogenic variants in moderate penetrance genes associated with hereditary breast cancer
Abstract
There is limited information known about how women with pathogenic variants (PV) in moderate penetrance genes make decisions to manage their increased risk of breast cancer. This study analyzed factors that may impact decision-making surrounding management for increased breast cancer risk. Women with a PV in a moderate penetrance gene associated with increased risk for breast cancer were identified from an institutional database. Semi-structured, qualitative interviews were conducted to analyze decision-making factors. Themes were developed using deductive codes based on previous literature and inductive codes based on interviewee responses. The 16 participants (mean age = 55.9 years) included 12 women with a breast cancer diagnosis. Six women (37.5%) chose bilateral mastectomy (BM), and 10 women (62.5%) chose surveillance as management. Of the 12 women with a personal history of breast cancer, four chose to have BM (33.3%). Two women without a personal history of breast cancer chose to have BM (50.0%). Transcriptions revealed seven comprehensive themes, as well as themes unique to affected and unaffected women (Cohen's kappa = 0.80). Physician opinion was the only factor present in all interviews reported to influence risk management decision-making. Several themes were consistent with prior BRCA1/BRCA2 research (family history, risk perception, sibling influence, and physician opinions). Autonomy and insurance/finances were also important factors to participants. There were certain differences in decision-making factors between affected and unaffected women, such as partner influence. Results indicate an opportunity for providers to engage their patients in a decision-making process.
1 INTRODUCTION
1.1 The genetics of breast cancer
It is estimated that 5%–10% of all breast cancer diagnoses are attributable to a pathogenic germline variant associated with a hereditary cancer predisposition syndrome (Garber & Offit, 2005). It is known that pathogenic or likely pathogenic variants (PV) in BRCA1 and BRCA2 account for around half of the diagnoses of hereditary breast cancer (Susswein et al., 2016). When panel testing became available in 2013, allowing for multiple genes to be sequenced at once, it was found that 20%–30% of identifiable causes of hereditary breast cancer are due to moderate penetrance genes. It is now estimated that 2%–5% of individuals who undergo cancer genetic testing have a PV in a gene other than BRCA1 and BRCA2 (Couch et al., 2015; Easton et al., 2015; Kurian et al., 2014; Lincoln et al., 2015; Maxwell et al., 2015; Tung et al., 2015).
1.2 Risk management options
Individuals who test positive for a PV in a known breast cancer susceptibility gene may use this information to decide how to manage their increased cancer risks (NCCN Guidelines V1.2020, 2019). Currently, the options to manage breast cancer risk include active surveillance, surgery, and chemoprevention (Jatoi, 2018). Surveillance aims to detect breast cancer as early as possible through strategies including a combination of breast imaging (often magnetic resonance imaging (MRI) and mammogram), self-awareness, and clinical exams ('NCCN Guidelines V1.2020', 2019). Risk-reducing surgery entails bilateral mastectomy (BM) to remove the majority of the at-risk breast tissue. Chemoprevention, a subcategory of endocrine therapy, is defined as the use of medications to inhibit, reverse, or delay carcinogenesis and may be considered under some circumstances (Benetou, Lagiou, & Lagiou, 2015).
1.3 Moderate penetrance genes
‘Moderate penetrance’ genes have been defined as conferring a twofold to fivefold increased lifetime risk to develop breast cancer (Cavaciuti et al., 2005; Easton et al., 2015; Maxwell et al., 2015; D. Thompson et al., 2005; West et al., 2018). Moderate penetrance genes currently include ATM, CDH1, CHEK2, NBN, NF1, PALB2, and STK11. Due to the higher breast cancer risk for individuals who are identified to have a PV in one of these genes, National Comprehensive Cancer Network (NCCN) guidelines for breast screening include annual mammogram (with consideration of tomosynthesis) and consideration of breast MRI with contrast starting at around age 25 (STK11), age 30 (CDH1, NF1, PALB2), and age 40 (ATM, CHEK2, NBN) (‘NCCN Guidelines V1.2020’, 2019). However, NCCN guidelines state there is insufficient evidence at this time to recommend discussion for risk-reducing mastectomy and to make management decisions based on family history. Due to this, there is limited information regarding a framework for clinical decision-making for individuals who have a PV in a moderate penetrance gene (Desmond et al., 2015; E. R. Thompson et al., 2016; Tung et al., 2016). Research and evidence are lacking with respect to the occurrence of second primary breast cancers in carriers of PV in moderate penetrance genes (Bernstein, Concannon, & Group, 2017; Weiss, Garber, & King, 2018). Previous studies have suggested that the decision to proceed with surgery must take into consideration the personal and family risk factors unique to each patient (Frey, Salibian, Schnabel, Choi, & Karp, 2017).
1.4 Decision-making factors
Many factors may impact individual decisions regarding management for an increased risk of breast cancer. Prior research has explored factors that impact risk management decisions for women who test positive for a PV in BRCA1 or BRCA2. Factors found to affect patient decision-making in BRCA1 and BRCA2 PV carriers include family history, risk perception and anxiety, relationships, culture, nationality, physician opinions, and surgeon availability (Antill et al., 2006; Bebbington Hatcher & Fallowfield, 2003; Borreani et al., 2014; Dean & Rauscher, 2017; Den Heijer et al., 2013; Gilbert et al., 2017; Howard, Bottorff, Balneaves, & Kim-Sing, 2010; Johns, Agarwal, Anderson, Ying, & Kohlmann, 2017; Julian-Reynier et al., 2001; Kardosh, Bar-Tal, & Barnoy, 2017; Metcalfe et al., 2008; Portnoy, Loud, Han, Mai, & Greene, 2015; Singh et al., 2013; St-Pierre, Bouchard, Gauthier, Chiquette, & Dorval, 2017; Yadav, Reeves, Campian, Sufka, & Zakalik, 2017). Previous studies regarding patient decision-making after a new breast cancer diagnosis report that choosing BM is associated with having a family history of cancer and anxiety regarding recurrence risk (Hawley et al., 2017; Mamtani & Morrow, 2017). Research looking at individuals unaffected by breast cancer who carry BRCA1/2 PV identified that patient age, family history, fear of cancer diagnosis, and being a parent influenced surgical decision-making (Hesse-Biber & An, 2016; Johns et al., 2017).
1.5 Objective
Limited literature exists to guide individuals who carry a PV in a moderate penetrance gene in making breast cancer risk management decisions. This study will investigate factors that impacted risk management decisions made by this cohort of women.
2 METHODS
This is a retrospective, qualitative, and interview-based study.
2.1 Participants
This study was approved by the Institutional Review Board (IRB) at Cincinnati Children's Hospital Medical Center (CCHMC). The target population was defined as women who have been diagnosed with a PV in a moderate risk gene associated with hereditary breast cancer. Women at least 18 years of age who were referred for a genetic risk assessment at CCHMC between 2013 and 2018 and found to have a PV in ATM, CDH1, CHEK2, NBN, NF1, PALB2, or STK11 were eligible to participate in this study. Individuals who have a PV in BRCA1, BRCA2, PTEN, or TP53, a rare high-risk variant associated with a moderate penetrance gene, a mosaic or potentially mosaic PV result, a diagnosis of more than one PV, identified as male, or needed language interpretation services were not included in the study. Invitations to participate were mailed to 65 eligible women. We received 16 responses, for a response rate of 24.6%.
2.2 Procedures
Three separate invitation mailings were sent to all patients who met inclusion criteria to participate in the study. An information sheet was mailed out with the interview invitation to gather patient demographics, personal medical history, and family history. A consent statement preceded all interviews, and consent was obtained verbally for in-person interviews or given over the phone and recorded. All in-person interviews were conducted in private meeting rooms within CCHMC. Qualitative semi-structured, in-depth interviews were conducted by a female genetic counseling student and a female genetic counselor. The genetic counseling student was trained in qualitative research methods through educational programming as part of her graduate program, and the genetic counselor has previously been a part of a qualitative research project in addition to receiving qualitative research education during her graduate program. No other persons were present for the interviews besides the participants and researchers. None of the participants received genetic counseling from the genetic counselor involved with the interviews. Relationship with participants was established over the phone preceding the interview, but no additional phone or in-person contact was made with participants after the phone interview was complete. Participants were informed that the research was conducted as part of a master's thesis project and that they would be interviewed by a graduate student. Participants were provided a $20 gift card as an incentive upon completion of the information sheet and interview. Two separate interview guides were developed: one for women with a personal history of breast cancer and one for women with no personal history of breast cancer. The interview guide and demographic information questionnaire (Data S1 and S2) were pretested by colleagues and volunteers from the community. All interviews conducted were less than one hour in duration and ranged from 17–48 min in length, with the average interview time being 28 min. No repeat interviews were carried out.
2.3 Data analysis
Interviews were performed either in-person or over the phone, audio recorded, transcribed, and coded by two different coders. Field notes were created during and after interviews. Thematic analysis was used to analyze data (Braun & Clarke, 2014). A priori themes were selected based on existing literature about patients with BRCA1 and BRCA2 PV. These included how women made decisions regarding risk management, how family history influenced decisions, how relationships influenced decisions, anxiety about developing breast cancer, how cultural factors influenced decisions, care team members’ influence on patients through the care process, and effect of surgeon advice on decisions (Bebbington Hatcher & Fallowfield, 2003; Borreani et al., 2014; Dean & Rauscher, 2017; Den Heijer et al., 2013; Howard et al., 2010; Johns et al., 2017; Julian-Reynier et al., 2001; Kardosh et al., 2017; Singh et al., 2013). A codebook was developed to assist coding consistency (Data S3). Thematic analysis and coding of transcribed interviews occurred in parallel with data collection. Qualitative analysis software (ATLAS.ti) was utilized for coding, organization, and analysis. After interviews began, additional codes were added after inductive, novel themes emerged. Transcriptions contained 1,006 quotes tagged with 49 codes. After each coder coded individually, they resolved discrepancies through discussion. The final analysis of inter-coder reliability showed a Cohen's K = 0.80, which is defined as very good (Landis & Koch, 1977). Participants did not provide feedback on the findings.
Risk management decision-making factors were organized into thematic groups that included family history (with death, family member cancer diagnosis, and young age at diagnosis as subthemes), risk perception and anxiety, relationships, culture and nationality, physician-related factors, and emerging, inductive themes (autonomy, cosmetics, etc.). We defined a robust theme as a theme that had a representative code that occurred in at least 50% of interviews. Theme frequency was defined as the frequency with which a theme emerged (i.e., was topicalized with a representative comment at least once) across interviews. Theme density was defined as the average number of thematically representative comments per interview in which a theme was topicalized. Code co-occurrence was defined as two codes coding the same, or an overlapping, segment from a transcript. The c-coefficient was used to describe the strength of relationship between two co-occurring codes (Armborst, 2017).
3 RESULTS
3.1 Population characteristics
The patient population was relatively homogenous, and all were women who identified as white or Caucasian (Table 1). Of the 16 women, there were two with an ATM PV, 10 with a CHEK2 PV, and four with a PALB2 PV. Twelve of the women who interviewed had a personal history of breast cancer, and four women had no personal history of breast cancer. Of the women who reported age at diagnosis, the average age was 47 years. Regarding cancer history, 18.8% of the women reported a personal history of cancer besides breast cancer, 87.5% reported a family history of cancer in a first degree relative, and 93.8% reported a family history of cancer in a second degree relative.
| Characteristic | n = 16 | |
|---|---|---|
| Mean (Range) | N (%) | |
| Age (years) | 55.9 (35–77) | |
| Race | ||
| White, non-Hispanic | 16 (100.0) | |
| Other | 0 (0.0) | |
| Number of biological children | 1.6 (0–5) | |
| Education | ||
| Some college | 3 (18.8) | |
| College degree | 4 (25.0) | |
| Advanced degree | 9 (56.3) | |
| Employment Status | ||
| Full time | 11 (68.8) | |
| Retired | 4 (25.0) | |
| Part time | 1 (6.3) | |
| Household income (14/16 reported) | ||
| <$49,999 | 2 (14.3) | |
| $49,999–$99,999 | 8 (57.1) | |
| >$100,000 | 4 (28.6) | |
| Marital status | ||
| Single (including divorced and widowed) | 3 (18.8) | |
| Married | 13 (81.3) | |
| Gene with PV | ||
| ATM | 2 (12.5) | |
| CHEK2 | 10 (62.5) | |
| PALB2 | 4 (25.0) | |
| Personal breast cancer diagnosis | ||
| Yes | 12 (75.0) | |
| No | 4 (25.0) | |
| Age of breast cancer onset (11/12 reported) | 47 (31–62) | |
| Family history of cancer | ||
| Self (non-breast cancer) | 3 (18.8) | |
| First degree relative | 14 (87.5) | |
| Second degree relative | 15 (93.8) | |
| Third degree relative or further | 10 (62.5) | |
| Breast cancer in family (including self) | ||
| Yes | 16 (100.0) | |
| No | 0 (0.0) | |
| Breast cancer in family (excluding self) | ||
| Yes | 12 (75.0) | |
| No | 4 (25.0) | |
| Medical Management Decision | ||
| Surgery (BM) | 6 (37.5) | |
| Surveillance | 10 (62.5) | |
3.2 Risk management decision outcomes
Risk management outcomes were condensed into two possible decisions: BM or surveillance (Table 1). We investigated whether women had genetic testing before or after they chose their risk management option and explored if this had an impact on their ultimate decision. Of the 12 women interviewed with a personal history of breast cancer, four chose to have a BM (33.3%). Three of the four women who had a personal history of breast cancer and a BM chose to have surgery beyond what was necessary as part of cancer treatment in light of their genetic testing results. One woman, with a personal history of cancer and a BM, had BM recommended as part of cancer treatment. This individual reported the final decision to proceed with BM was ultimately a result of shared decision-making between herself, her husband, and her doctors (Table 2). Two women without a personal history of breast cancer chose to have a BM (50.0%), and two chose surveillance (50.0%). All 16 women are still following their original risk management plan.
| Participant | Genetic Testing Timing | Was BM medically necessary as part of cancer treatment? | Was cosmetic result a factor that impacted the decision to have BM? |
|---|---|---|---|
| P1, affected | After surgery (19 years) | Recommended | No |
| P7, affected | Before surgery | No | Yes |
| P8, affected | Before surgery | No | No |
| P13, affected | Before surgery | No | No |
Six women (five with a personal history of breast cancer and one without) had genetic testing before they made their risk management decision, and ten women (seven with a personal history of breast cancer and three without) had testing after they had already decided their risk management plan. Of the six women who had genetic testing prior to deciding, four chose to have BM (66.7%). Two of the ten women who had genetic testing after their risk management decision chose BM (20.0%).
P4, affected: But we have talked about perhaps doing a mastectomy. But neither one of us is at that point just yet. I think for the PALB2 there’s not as much data as there is on the BRCA. Because it hasn’t been around, they haven’t discovered it for that long. So, I don’t, you know, for me a mastectomy is something, major and I don’t know, I’ve tried to find studies where people who’ve had breast cancer and have the PALB2.
P11, affected: And my kids and my husband are okay with me doing the removal and that kind of stuff. But I wasn’t ready for that yet… Just knowing that it was a newer gene mutation that they didn’t have a lot of data to support.
3.3 Decision-making factor results
3.3.1 Comprehensive themes
Data produced seven condensed, robust themes among all the interviews (Table 3).
| Theme density | Frequency | Example quote | |
|---|---|---|---|
| A priori themes | |||
| Physician opinions | 7.8 | 100.0 | P13, affected: ‘I remember saying like, I feel like I should just do prophylactic. And [my oncologist] was like, yeah, I would recommend that too. Why not? I mean that was basically what it came down to’. |
| Physician opinion encourage | 3.5 | 81.3 | |
| Physician/patient concordance | 3.0 | 50.0 | |
| Family history | 5.8 | 81.3 | P2, unaffected: ‘I felt like it would be hard to not feel like you had a gene mutation when almost every woman in my family has had breast cancer. I presumed that I did even though everybody's test had been negative. You know, up until my sister's. So, I had a mastectomy years ago’ |
| First degree relative | 2.7 | 68.8 | |
| Personal history | 3.8 | 81.3 | P10, affected: ‘I mean it's obviously a personal decision whether to have a mastectomy versus doing the route that I did (surveillance), but I think because I was in early stages of breast cancer that was easier probably than stage 4…I probably would have had a mastectomy then’ |
| Risk perception | 3.0 | 62.5 | P12, affected: ‘And because of my age and you know my daughter's age, she has a lot more time ahead of her, you know…. Age wise. So that's why she [had a mastectomy] and my being older, I have less years to have a reoccurrence or a new kind of cancer. So, I still feel pretty good about that. I mean if anything should happen, I would have to go that route (have a mastectomy) I think’. |
| Recurrence | 3.4 | 56.3 | |
| Future cancer risk | 2.6 | 50.0 | |
| Sibling influence | 4.2 | 62.5 | P4, affected: ‘My sister is also PALB2 positive. We've kind of discussed it. She had ovarian cancer and I had the breast cancer so it's a little bit different. She still has her breasts. She isn't doing MRIs at this point. But we have talked about perhaps doing a mastectomy. But neither one of us is at that point just yet’. |
| Emerging themes | |||
| Patient autonomy | 3.4 | 56.3 | P15, unaffected: ‘I’ve actually been very lucky with the doctors that I’ve had. And I think they've done a wonderful job of presenting options very honestly and very openly, but not necessarily…giving their opinion but not necessarily saying you must do this or that. They provide the information, they might make a recommendation, but they allowed me as a patient to make my own decision’. |
| Insurance and finances | 2.6 | 50.0 | P7, affected: ‘I have a friend from high school who had a mastectomy, but she had an augmentation. And I was really impressed with the [results]... because she's a very active person, I’m very impressed with her post augmentation appearance. And the insurance will cover this also, so I’ll do the double’. |
The themes are listed in the order of frequency in which they occurred across transcribed interviews. The interviews indicated that patient medical management decisions are influenced by physician opinions, family history, personal cancer history, risk perception, sibling influence, patient autonomy, and insurance/finances.
Physician opinions theme: Every participant spoke about physician opinions being a decision-making factor (n = 16). Related to general physician opinions, it was reported that physicians encouraged patients to choose a certain outcome (n = 13). Physician/patient concordance in respect to the decision that was made was also a frequent theme (n = 8). Physician opinions and physicians encouraging the patient to choose a certain outcome frequently co-occurred (c-coefficient = 0.42).
Family history theme: Family history included any cancer diagnosis for any degree of relative. Personal history theme: All women with a personal history of breast cancer and one woman with a personal history of ovarian cancer spoke of how their own diagnosis had influenced their risk management outcome (n = 13).
Risk perception theme: Risk perception was defined as how each participant perceived her own risk due to her PV. Breast cancer recurrence risk or future cancer risk was typically discussed alongside general risk perception. Perception of breast cancer recurrence risk and perception of future risk of any type of cancer co-occurred (c-coefficient = 0.37).
Sibling influence theme: Siblings’ genetic testing results, PV status, cancer history, and opinions were reported to influence decision-making. Participants in our cohort specifically mentioned their sister(s) as having an impact on their risk management decision.
Patient autonomy theme: Patient autonomy was defined as participants stating they felt they had a say in their healthcare and/or their voice was heard when making a medical management decision. Participant account of instances of patient autonomy often involved a shared decision-making process with physicians.
Insurance/finance theme: Participants discussed ability to pay for and/or cover risk management decisions as an influencing factor.
3.3.2 Robust themes among women with a personal history of breast cancer
There were nine robust decision-making factor themes among the group of 12 women who had been diagnosed with breast cancer. Physician opinions (including physician encouragement to choose a certain outcome), sibling influence, family history (including first degree relative cancer diagnosis), personal history, autonomy, and risk perception (including perception of recurrence risk) were important decision-making factors similar to the overall comprehensive themes. Unique to the women affected by breast cancer was the inclusion of patient/physician concordance as a subcategory of physician opinions, partner influence, lack of information available, and information gathering (Table 4). Lack of information available was defined as the participant reporting that the lack of information available about their PV influenced decision-making. Information gathering was defined as the participant stating that they gathered information to help them make a risk management decision.
| Theme density | Frequency | Example quote | |
|---|---|---|---|
| Women with a personal history of cancer | |||
| Patient/physician concordance | 3.1 | 58.3 |
P6, affected: ‘[Surveillance] was the recommendation of my surgical oncologist… I did not hesitate to follow through with what was recommended’. |
| Partner influence | 3.8 | 50.0 | P1, affected: ‘I asked [my husband] would it bother you if I have reconstruction surgery? And he said, it's your decision. It's not going to bother me; I just want you alive. So, he was a very big support’. |
| Lack of information | 2.8 | 50.0 |
P11, affected: ‘I mean I think I discussed [surveillance] with my husband and the doctor, you know, about how my thoughts on it, and they also discussed too…[PALB2] is still a newer gene and they're still gathering information’. |
| Information gathering | 2.0 | 50.0 | P12, affected: ‘I did a lot of research myself. My breast surgeon, you know, we talked about [CHEK2] and I talked about it with my oncologist, but at the time it was very new to many physicians’. |
| Women without a personal history of cancer | |||
|
Children or grandchildren |
5.0 | 75.0 | P2, unaffected: ‘I was very concerned that if I waited too long to have a mastectomy, I would end up getting diagnosed with cancer. And that would affect my life expectancy…I needed to live a long life to take care of our kids’ |
3.3.3 Robust themes among women with no personal history of breast cancer
Interviews with the four unaffected women produced six robust themes. Due to the small sample size, a robust theme was defined for the unaffected women as a theme that had a representative code that occurred in over 50% of interviews. Similar to the overall robust themes, family history (including first degree relative cancer diagnosis), physician opinions (including physician encouragement to choose a certain outcome), risk perception (including perception of future risk of any type of cancer), sibling influence, and insurance, and/or finances were all robust themes in the unaffected group. Of note, every unaffected woman mentioned both physician opinions and family history in their interview. Distinctive to the unaffected women, having children or grandchildren was an important decision-making factor (Table 4).
3.4 Physician recommendations
P8, affected: ‘But it’s very important that you have somebody that you feel you trust, and that will listen to you. I was fortunate that I had that right from the beginning. Never second guessing anything that had been presented to me and my husband’
P9, affected: ‘I felt like sometimes I was viewed by healthcare providers like once they had the diagnosis, they were just managing a project. Step 1, this. Step 2, this. Step 3, step 4. And feeling more like a process. It’s like okay, got it, I’m your project and you’re working on it. I guess just to remember that it is a human being there not just someone who’s been diagnosed and now you do the steps that you have to do because it’s what you’ve done with everyone’
P5, affected: ‘Just making sure [physicians] have all of the information. My oncologist wanted to do the double mastectomy right away and…I mean I would have gone along with that, but then he came back and said something different. I’m glad he did some more research and I didn’t have to have it done. But it would be helpful if the doctors have all the information ready for you’.
P14, unaffected: ‘When I went to certain appointments, [physicians] always assumed I had BRCA1 or BRCA2. And I’d tell them no, CHEK2. I wish they could give out more information to the health providers and have more available…. It could become more common information for both patients and medical providers. I think it would help tremendously. You have BRCA1 or BRCA2? They’re like yeah, yeah, sure hold on here’s what we’re going to do. Like Angelina Jolie. We’re going to do surgery and remove portions of your body since your breast cancer risk is so great. I know it’s not as high as BRCA1 or BRCA2, but it’s still there’
4 DISCUSSION
4.1 Comprehensive theme discussion
Similar to previous research conducted on women who carry a PV in BRCA1 and BRCA2, family history, risk perception, sibling influence, and physician opinions were all found to be factors that impact risk management decisions for women who carry a PV in a moderate penetrance gene associated with hereditary breast cancer (Antill et al., 2006; Bebbington Hatcher & Fallowfield, 2003; Borreani et al., 2014; Dean & Rauscher, 2017; Gilbert et al., 2017; Johns et al., 2017; Singh et al., 2013; St-Pierre et al., 2017). In our cohort, patient autonomy and insurance/finances were among the most frequently discussed decision-making factors. This could be because unlike BRCA1 and BRCA2, where there is a known risk for a second primary breast cancer, there is no defined risk for a second primary in individuals who carry PV in moderate penetrance genes associated with breast cancer (Bernstein et al., 2017; Weiss et al., 2018). Additionally, NCCN guidelines are not clear on discussion of risk-reducing management. This may lead the patient and provider to engage in more conversation and shared decision-making surrounding risk management decisions. Due to these same factors, insurance coverage is not guaranteed for genes that have not been researched and studied as extensively as BRCA1 and BRCA2 (Aetna, 2019). Participants spoke of their insurance coverage of different management options, financial capability to choose different options, and ability to have enough time off at work while not sacrificing income as specific factors pertaining to insurance/finances.
4.2 Women with a personal history of breast cancer theme discussion
The women who had a personal diagnosis of breast cancer spoke of more concordance regarding risk management decision-making between themselves and their physicians than the women without a personal history of breast cancer. This could be because women with a cancer diagnosis are more likely to regularly see their physicians, have a closer relationship with them and thus feel more comfortable when discussing risk management options.
Partner influence was a significant decision-making factor within the group. Based on the interviews, partners may have been a greater influence among the woman with a personal diagnosis of breast cancer because of the support they provide and value of their opinion during the cancer diagnosis and treatment process. For several women, their risk management decision conversation was not the first multifaceted medical conversation they had had with their partner.
Information was a topic spoken of frequently by women with a personal history of breast cancer, both in acknowledgment that there was not enough information currently available to make the decision to have a BM and in the context of the women taking it upon themselves to collect information from various sources to help them make a risk management decision. Many of the women spoke of the need to be logical and collect information rather than be emotional when making decisions regarding their moderate penetrance PV. This could be because many of the women had already gone through cancer treatment when they learned of their PV, and they had a previous opportunity to ascertain how to make risk management decisions in an emotionally distressing situation.
4.3 Women without a personal diagnosis of breast cancer theme discussion
Women in our study who had not been diagnosed with breast cancer chose surgery more frequently than women with a personal history of breast cancer. This could show that knowing the status of a hereditary cancer predisposition syndrome could influence surgical decision-making despite NCCN guidelines recommendation that there is insufficient evidence for the discussion of a risk-reducing mastectomy. In this cohort, children or grandchildren was the only distinctive factor that influenced decision-making in women without a personal history of breast cancer. Three of the four unaffected women spoke of their children as an important determinant of how they chose to manage their risk related to their PV. Of the three women who discussed their children as a factor, two ultimately had BM to manage risk, both cited desiring longevity to be around for their children as the reason. This is consistent with prior research on unaffected women with BRCA1/2 PV (Hesse-Biber & An, 2016). Unaffected women all mentioned both physician opinions and family history as decision-making factors. Since the unaffected women did not have a personal breast cancer journey to reference and factor into risk perceptions, it is reasonable they would look to the experiences of family members.
4.4 Anxiety
Similar qualitative studies on BRCA1/BRCA2 carriers noted fear and/or anxiety as a common theme. (Bebbington Hatcher & Fallowfield, 2003; Dean & Rauscher, 2017; Glassey et al., 2018; Howard et al., 2010). Anxiety was only a factor for two women in our study and was regarding cancer specific anxiety. This could be due to evidence-based literature currently indicating there is a lower risk for cancer for individuals who carry a PV in a moderate penetrance gene than those who carry a PV in higher risk genes such as BRCA1 and BRCA2. Further, our patient population had a higher percentage of women who had already been diagnosed with breast cancer and literature does not currently show there is an increased risk for a second primary breast cancer in individuals with a PV in a moderate penetrance gene (Bernstein et al., 2017; Weiss et al., 2018).
4.5 Clinical impact
With few published guidelines on how to manage women diagnosed with a PV in a moderate penetrance gene, the final decision comes down to both patient and physician opinions and patient personal factors. It can be intimidating to navigate a gray area where two patients in similar situations might choose different paths, neither one more correct than the other. When providers encounter a patient with a moderate penetrance PV in clinic, they should evaluate the patient's perception of his/her own cancer risk, their family history, their support-system, and their financial resources/ socio-economic status and use that information to engage in a decision-making process. Providers can better navigate this dialogue with patients by entering conversations with an awareness of the many factors that may influence risk management decision-making.
4.6 Practice implications
Physician opinions were discussed by all participants as a decision-making factor (Dean & Rauscher, 2017). With this in mind, it is imperative that healthcare providers learn and communicate accurate, current information about the cancer risks associated with moderate penetrance genes. It is also important for providers to be aware of current guidelines as our knowledge about cancer risks continues to grow. The American Society of Breast Surgeons (ASBrS) consensus guideline on genetic testing for hereditary breast cancer acknowledges personalized risks for patients with a moderate penetrance PV cannot always be predicted and states risk should be mitigated according to age, family history, and sometimes, the specific mutations (Manahan et al., 2019). Our participants provided feedback regarding their experience with providers who were not as knowledgeable about moderate penetrance PV cancer risks or how to apply those risks to make risk management recommendations. The anecdotes from the women point to the need for continuing education of healthcare providers as genetic tests expand and evolve.
4.7 Study limitations
One study limitation is the inability to account for how risk management decisions may change over time and why. Considering the affected women who chose BM as medical management, it is not possible to precisely parse out when BM was part of cancer treatment or when BM was chosen primarily to optimize cosmetic results (when lumpectomy or unilateral mastectomy was part of cancer treatment). Possible biases include volunteer bias as individuals who participated in this study are more likely self-motivated and therefore may have been more autonomous and information seeking in their decision-making; selection bias as participants did not equally represent all races and income levels; and hindsight bias due to instances when genetic testing occurred after a risk management decision had already been made. Since the women all had access to medical care and surgeons, surgeon availability as a factor could not be sufficiently investigated. Coding bias may have also occurred due to the nature of the tools used with ATLAS.ti. Due to small sample size, especially considering the number of unaffected women in the study, generalizations cannot be made about medical management decision-making factors.
4.8 Research recommendations
This study was from a small sample size in one geographical area. It would be beneficial to create a larger study nationwide of women who carry pathogenic variants in moderate penetrance genes to see if themes are consistent. While this study was able to investigate risk management decisions from a patient perspective, it would also be valuable to study surgeons’ perspectives on the same topic. It may be important to explore physician knowledge and counseling patterns related to moderate penetrance genes associated with an increased risk of breast cancer.
4.9 Conclusions
Family history, risk perception, sibling influence, and physician opinions were consistent with decision-making factor themes previously revealed with research on BRCA1/BRCA2 patient populations. Patient autonomy and insurance/finances were also important to this cohort of women. The one theme present in every participant interview was physician opinions. Some decision-making factors differed in priority between women with and without a personal history of breast cancer, and several were unique to each group. Participant commentary offered insight into ways in which we could improve the discussion surrounding risk management with our patients. Results point to an opportunity for providers to assess patients’ knowledge, feelings, and perceptions about their cancer risk based on family history, genetic testing results, and personal factors. This information can be used by providers to engage patients in a decision-making process.
AUTHOR CONTRIBUTIONS
Each author, Melissa Napoli, Kimberly Widmeyer, Jennifer Hopper, and Dr. Jaime Lewis, substantially contributed to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafted the work or revised it critically for important intellectual content; approved the final version to be published; and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
ACKNOWLEDGEMENTS
This study was conducted when the first author was enrolled in the Genetic Counseling Graduate Program, College of Medicine, University of Cincinnati and Division of Human Genetics, Cincinnati Children's Hospital Medical Center, Cincinnati, OH. This publication was supported by a Research Award from the Division of Human Genetics at Cincinnati Children's Hospital Medical Center. This work has been supported by the Jane Engelberg Memorial Fellowship Student Research Award, provided by the Engelberg Foundation to the National Society of Genetic Counselors, Inc. We are very grateful to each of the participants who offered their time and stories.
COMPLIANCE WITH ETHICAL STANDARDS
Conflict of interest
Melissa Napoli, Kimberly Widmeyer, Jennifer Hopper, and Dr. Jaime Lewis declare that they have no conflict of interest.
Human studies and informed consent
This study was approved by the Institutional Review Board (IRB) at Cincinnati Children's Hospital Medical Center (CCHMC). A consent statement preceded all interviews, and consent was obtained verbally for in-person interviews or given over the phone and recorded. Any identifying information was securely stored and only accessed by the authors when necessary.
Animal studies
No non-human animal studies were carried out by the authors for this article.
