Preparing for the unexpected: Recommendations for returning secondary findings in late-stage cancer care

Abstract

Purpose:

We conducted qualitative interviews with patients with cancer and providers to identify gaps in clinical care and highlight care delivery solutions for the return of secondary germline findings.

Methods:

Twelve patients and 19 cancer providers from the United States were interviewed between January 2019 and May 2021. Interviews elicited feedback about patient information needs, emotional responses to secondary findings, and recommendations for improving pre-test education.

Results:

Patients’ responses ranged from gratitude to regret, depending on how much pre-test counseling they received before tumor testing. Providers cited insufficient clinic time as a major barrier to pretest education, favoring online support tools and standardized pre-test education models. Providers had differing perspectives on how pre-test education should be integrated into clinical workflows but agreed that it should include the differences between somatic and germline testing, the likelihood of medically actionable findings, and the possibility of being referred to a genetics provider.

Conclusion:

The spectrum of participants’ responses to their secondary findings underscores the importance of adequate pre-test discussions before somatic sequencing. Although educational interventions could address patients’ information needs and augment traditional pre-test counseling, health care systems, labs, and genetic providers may be called on to play greater roles in pre-test education.

Introduction

Genomic sequencing has transformed cancer care and improves outcomes for patients. Specifically, somatic (tumor) sequencing can identify driver DNA alterations that are predictive of response or resistance to genomically targeted therapy.^1–5 Somatic genomic testing is rapidly being incorporated into the care of hundreds of thousands of patients with cancer yearly and is included in care guidelines for patients with common cancer types.^6–11 However, because germline (inherited) variants are frequently retained in tumor tissue, somatic sequencing can also reveal unanticipated secondary or incidental information about cancer susceptibility.^12–14

Given that the widespread use of multigene panel somatic testing is a relatively recent phenomena, little is known about how clinicians incorporate unanticipated germline findings into clinical care or about patients’ experiences after test result disclosure. It is therefore essential to understand the views and challenges in the return of results process because they may shape how the process unfolds from a patient and provider perspective.

For patients and their families, secondary findings may have personal or medical value because they can impact medical care, lifestyle, and reproductive decisions, as well as physical and psychological well-being.^15,16 Patients may greatly benefit from unexpected germline information if confirmatory testing reveals a variant that can be used to guide their cancer care or facilitate “cascade” genetic testing for their blood relatives. However, such information may be harmful to patients if they are presented with unsolicited information about potential future health risks or cancer etiology. This may be particularly true for patients with cancer with metastatic disease, for whom the benefit of information about future cancer risk may be less relevant. Patients with advanced cancer may be overwhelmed by medical information related to their diagnosis and treatment and may be less likely to derive clinical benefit from their germline test results unless—as is increasingly common—the information identifies germline-targeted therapeutic options.^15–17 Interestingly, work to date suggests that a sizable majority of patients, ranging from 57% to 99% in several studies, desire secondary germline information from tumor sequencing.^16,18

Current oncology practice may not be optimally structured for cancer providers to deal with unanticipated germline findings from somatic sequencing.^19,20 First, somatic sequencing is typically ordered by an oncologist without extensive pre-test discussions about the types of results that can be identified.^21,22 Second, many providers report low to moderate confidence in their ability to explain genomic concepts to patients or discuss the benefits, limitations, and implications of sequencing, particularly in the context of germline information.^23–27 Third, many medical settings do not have ready access to genetic counselors who are more facile with pre-test counseling,^28–30 and genetic counselors are not traditionally trained to advise patients about somatic sequencing. Further, there are no standard informed consent processes for somatic sequencing, although best practices have been outlined by experts.^14,31,32

Given these limitations, we conducted an interview study with oncology patients and providers to better understand their unique knowledge needs and challenges surrounding the return of secondary germline information. Our goal was to identify gaps in clinical care and describe possible care delivery solutions that could prepare patients and providers for the return of unanticipated germline findings.

Methods

Study setting, participants, and activities

We used snowball sampling to identify patients referred through City of Hope (COH) clinicians and other professional networks (eg, the National Society of Genetic Counselors, laboratories). Study staff called and emailed eligible patients for study recruitment. Patients were eligible if they were as follows: (1) had a cancer diagnosis, (2) recalled having a secondary finding through somatic sequencing, (3) spoke English or Spanish, and (4) had an Eastern Cooperative Oncology Group Performance Status ≤2 (a scale from 0 to 5 that quantifies the capacity of patients with cancer for daily functioning).³³ Given the potential impact of germline information on family members, patients were allowed to invite 1 family member to the interview with them. Family members were free to comment on any question asked during the patient’s interview. After the interview, all patients completed a brief online survey.

We used purposive sampling to ensure a range of provider types and treatment settings.³⁴ Cancer care providers (“providers”) were recruited through COH and an international network of cancer genomics providers (The COH Clinical Cancer Genomics Community of Practice). Eligible providers included cancer genetic counselors, medical geneticists, medical oncologists, and advanced practice oncology providers (eg, nurse practitioners) who (1) maintained an active clinical practice and (2) had previously cared for at least 1 patient with secondary findings. Once identified, eligible providers were approached by the study team via email and given the opportunity to opt into study participation.

Interviews were conducted via telephone and recorded in Zoom by trained interviewers (S.M., I.S., E.S., and V.P.-L.). Bilingual study staff conducted interviews with patients who spoke Spanish. Interviews in English and Spanish were transcribed verbatim. Spanish interviews were then translated into English. Participants were recruited for study interviews until thematic saturation was achieved.

Domains and measures

Interviews

Patient and provider interview guides were developed using a model from our prior work.³⁵ Provider guides were shortened for concision after the first 13 providers were interviewed.

Patient interviews explored their (1) recall and understanding of counseling before and after genetic testing, (2) emotional responses to learning secondary findings, and (3) attitudes toward sharing risk information with family members (Supplemental Material 1). Interview guides were informed by prior published research with patient focus groups.³⁵ Provider interviews focused on (1) the content and quality of counseling for tumor sequencing and confirmatory germline testing, (2) perceptions of patients’ emotional responses when receiving secondary findings, and (3) limitations and recommendations for the pre-test education process (Supplemental Material 2).

Patient surveys

To better describe our sample at the time of study enrollment, we administered a survey to assess recall of tumor and germline test results and genetic counseling³⁶ and demographic information (Supplemental Material 4).³⁷ We measured recall of sequencing results and genetic counseling with trichotomous “Yes/No/Don’t Know” questions.³⁶ We collected demographic information using questions from the Centers for Disease Control and Prevention’s Behavioral Risk Factor Questionnaire.³⁷ Surveys were administered by email via a secure REDCAp link.³⁸

We calculated the proportion of patients who correctly recalled sequencing results and genetic counseling by comparing patients’ survey responses with genetic results in the medical record. We reported the means and proportions of sex, age, language, race, and education across our cohort.

Analyses

Interviews

We used methods for thematic analysis outlined by Braun and Clark to identify descriptive and analytical themes for patient and provider interviews.³⁹ Three team members (M.C., S.M., and E.S.) independently coded interview transcripts using Atlas.ti software. Using a process of constant comparison, they met jointly to review coding and reconcile discrepancies, revising the coding list as they proceeded. Reading across codes they identified higher-order themes (Supplemental Material 3). Coders started the coding and analysis process as interviewing proceeded. We found redundancy (ie, thematic saturation) in participant responses after interviewing 12 patients and 19 providers.

Patient surveys

We generated descriptive statistics for patient characteristics.

Results

Patient characteristics

Twelve patients with 9 different cancer types (n for prostate = 3, lung = 2, and other cancer types = 7) were interviewed for the study. Most were non-Hispanic white males with a college degree and the majority had advanced or metastatic disease (Table 1). All patients were referred to genetic counseling for confirmatory germline testing after somatic sequencing suggested a secondary finding. Patient interviews were completed 2 weeks to 6 months after disclosure of secondary findings.

Table 1.

Participant demographics

Patient Characteristics	N = 12 (%)
Age (mean; range)	60 (57, 73)
Sex
Female	3 (25%)
Race
White	12 (100%)
Ethnicity
Hispanic	3 (25%)
Non-Hispanic	9 (75%)
Language
English	10 (83%)
Spanish	2 (17%)
Education
Grade 12 or below	2 (17%)
Some college	3 (25%)
College graduate and higher	7 (58%)
Clinician Characteristics	N = 19 (%)
Health care profession
Medical oncologist	8 (42%)
Genetic counselor	7 (37%)
Other advanced practice provider	4 (21.6%)
Clinical focus
Genetics	11 (58%)
Oncology	8 (42%)
Clinical setting
Academic institution	14 (74%)
Community hospital or outpatient clinic	5 (26%)
Years in practice (mean; IQR)	9 (4, 13)
Unknown	1
Age (mean; IQR)	41 (35, 50)
Unknown	2
Sex
Female	17 (89%)
Race
Asian	7 (37%)
White	12 (63%)

Half of the patients who completed the baseline survey (n = 5/10) correctly recalled their gene-specific secondary finding(s). Six of 10 respondents reported on their baseline survey that they received genetic counseling for their secondary finding(s).

Clinician characteristics

Nineteen cancer care providers were interviewed, most of whom worked in cancer genetic specialties at academic institutions (Table 1). Advanced practice providers (ie, genetic counselors, nurse practitioners, a physician assistant, and a clinical nurse specialist) reported ordering and interpreting germline testing for 80% to 95% of patients in their practice, and tumor sequencing for 0% to 5% of patients. Medical oncologists reported ordering and interpreting tumor sequencing for 80% to 95% of patients in their practice (approximately 2 to 4 patients per week). Oncologists reported rarely ordering germline testing. None of the providers interviewed disclosed genetic test results to the patients participating in this study.

Patients’ affective responses to receiving secondary findings fell on a spectrum from positive to negative feelings

Patients reported feeling a range of responses from gratitude to regret about their secondary findings (Table 2). Most patients expressed gratitude for their genetic testing experience because they felt that their results helped them and their families to better prepare for the future. In the words of 1 patient:

Table 2.

Themes in patients’ affective responses to receiving secondary findings fell on a spectrum from positive to negative feelings

Themes in Patients’ Affective Response to Receiving Secondary Findings
Positive Affective Responses	Example Quotes
Gratitude when learning secondary findings helps to prepare for the future.	“Because my disease, well, it’s not mine anymore, you could also say that it’s my family’s, my children’s, all that. Because they are feeling all this too, right? So the genetic testing, it helps us a little to know, because they also have the thought of whether one day they are going to get it or not; so, with this genetic test, now we are more sure that, if it turns out they are positive, they’ll be able to – like I say, have those tests and be less worried, right?” (2028; Patient) “You can do the testing and if you don’t hear anything then you have nothing to worry about to a degree. But to me, it’s about the knowledge. So, I don’t see why someone wouldn’t want to. I know that they would be scared to, so I think educating them on the pros and the reasons why they should or helping them understand that it’s only to benefit them, it’s not to benefit anybody else or their kids or their grandkids or whoever is down their line in history of what they want to help, it’s only a benefit.” (2018; Patient)
Positive experience when patients receive sufficient information before genetic testing.	“I do think that preparing people for outcomes that they weren’t expecting has been very helpful. I think in part, the reason that I don’t have as many episodes that I can describe to you where people have been very emotional or very scared or totally swinging on emotional ends that I can remember is in part because of that preparatory, getting that preparation in advance.” (2319; Medical Oncologist) “I don’t know that any [patients] were ever super-upset. And I think that’s sort of the nice thing about the pretest counseling. Because I feel like if you do a good job of that, when you tell people, hey, this doesn’t guarantee you’re gonna get cancer, this is information, but now we can use it to protect you or prevent cancer, or prevent cancer in your family, or catch it early… I’ve had people say, oh, I’m so glad, I’ve been so anxious because all of my family members died from cancer. And I’m so glad now I can say to the health insurance company yay I get to have a mastectomy, or yay I can have increased screening, or whatever the case.” (2313; Nurse Practitioner)
Negative Affective Responses	Example Quotes
Stress and regret when testing create additional patient burden.	“I don’t remember. The whole experience has been pretty emotional for me, so [the provider] may have well said something and I just didn’t retain it. … I went through so much this past year with different tests and the treatments that it was just really one more test to me.” (2031; Patient) “And I also had to struggle with disseminating the information. I had to decide for myself, and also kind of like the representee of my family, if I want to get this test, which has repercussions for my other siblings. So, it felt like a big weight that was put on me. Because my brother … he’s having a kid, but he always wanted kids. So, it’s not just me. It impacted everyone.” (2007; Patient)
Distress when patients do not receive sufficient information before genetic testing.	“We’d actually see [cursory explanations of genetic testing] quite a bit…especially with these larger panels. Sometimes the doctors are ordering one of these larger panels but if they see a gene that they haven’t really heard of before, they don’t really tell the patient anything. They just say, okay, here’s the number for genetics. Go there. So, by the time they get into our office, they’re really panicked.” (2301; Genetic Counselor) “I’ve had a couple of patients who either were just confused or one was very angry that they were sent for essentially confirmation genetic germline testing without really understanding what that was or what the tests on their tumor really could have meant and for the most part I have patients that are I think just kind of told we’re gonna do some testing on their tumor to see what the best chemo is or whatever they’re told, but they are not really being told that it could raise suspicion of some inherited condition that they may or may not actually want to know about.” (2305; Genetic Counselor) “I think that shock and awe are definitely two of the emotions that I’ve seen a lot of. Anger is another one that people have experienced, and I had a patient recently – a man with prostate cancer – who was found to have a FANCA variant, which is so not anything, but he was so pissed off about whatever was found because he then had to apparently let his son know whose wife was going through IVF anyway, and then needed to tell them to have carrier testing…I guess why he was so dissatisfied with everything because he was thinking, “I’m getting this test to see if it’ll treat my prostate cancer or if they can treat my prostate cancer differently,” but now, there were all of these things that he had to deal with on top of his cancer diagnosis.” (2311; Medical Oncologist/Geneticist)

“I think it’s a good thing that they did the testing to find out. I think it’s great you have this ability to test for these genes that might be indicators of potential cancer risks. I thought that was a good thing. That’s really all I can say about that. I’m not happy I have this gene but it’s good to know...That’s going to be important information not only for me but possibly family members.” (2035; Patient)

Another patient shared a sense of regret after testing because her results created additional burden in her cancer care.

“Looking back, I kinda wish I didn’t do it, to be honest, because now I need more surveillance…And it’s just kinda piled on another element that I’m not too happy about.” (2007; Patient)

Some patients described their experience receiving secondary finding as one that was “stressful” and “emotional.” Some also said that they felt burdened with the knowledge of how secondary findings could affect family and how they would share these results with family members. Fourteen providers from multiple advanced practice disciplines also observed how some patients felt distressed in their genetic testing experience, specifically when oncologists gave patients limited or no information about genetic testing and potential implications for secondary findings (Table 2). All 8 of the medical oncologists interviewed reported that they talked to patients in their practice about the possibility of germline information before somatic testing and subsequently observed that those patients were not surprised by their secondary findings.

Patients’ recall and understanding of secondary findings fell on a spectrum from informed to unaware

Of the 12 patients who were referred to genetic counseling for confirmatory germline testing, 7 recalled being told by their provider that confirmatory testing could reveal information relevant to family members. These 7 participants stated that they were not surprised to learn about their secondary findings because they had been prepared for the possibility that their test results could reveal family cancer risk. One patient shared the following:

“I was kind of expecting [the results]. They kind of prepared me to where I was expecting it because the tumor had the BRCA1 gene, so it was more than likely I would be, too. So, I wasn’t really surprised.” (2029; Patient)

Patients had varying levels of test result recall after disclosure. Only some respondents accurately recalled their secondary finding and/or meeting with a genetic counselor. Several patients also confused the counseling they received for somatic sequencing versus germline testing (Table 3). Clinicians similarly observed that most patients in their practice did not understand the differences between somatic and germline testing, which often caused misunderstandings about the different implications of each test, particularly the possibility of learning information relevant to family members through germline testing (Table 3).

Table 3.

Themes in patients’ cognitive experiences receiving secondary findings fell on a spectrum from informed to unaware

Themes in Patients’ Cognitive Experiences Receiving Secondary Findings	Example Quotes
Most patients were informed that genetic testing could reveal family cancer risk.	“[The ordering provider said] that’s what the saliva test had to do with, to see if – yes, if it was – if it was hereditary or family– or hereditary, family-related…I don’t exactly remember if they gave me the option [to have genomic testing] or not; but what I can tell you is that I did want to have [testing] to protect my children, to see if they have it or not or what chance they have that they could get it, and once they can get it, if it turns out that they have the same gene, then, start preventing it with tests.” (2027; Patient) “I think they ran a whole battery [of genetic testing panels]. I kinda got a little bit of everything. So, the more the merrier if they track things down and it can help me or help family members; that was the plus side.” (2003; Patient)
Some patients had trouble recalling their genetic testing care experience.	“The doctor […] told me that she was going to give me some tests; I don’t remember exactly if it was her or they spoke to me…I don’t know. I don’t remember…I don’t know exactly who it was that recommended that I have them, if it was the doctor or the one who gave me – they gave me a test for – to see exactly where I had the cancer, and I don’t remember which of the two of them recommended this to me.” (2028; Patient) “And before the thyroidectomy, I have a vague recollection of – I think before, like the day of the procedure, reading through something that talked about genetics. And it sounded like it was for research. It didn’t really – I wasn’t aware that it was for me. Or, I know – I thought they would ask me if I wanted to know before telling me. I don’t think I was in a position that day to really absorb anything.” (2007; Patient) “I’ve even seen patients who have actually had germline testing and they cannot recall anything about it – what was told to them. They remember doing a swab from the mouth but that’s all they remember, and they don’t really know what that was.” (2312; Nurse Practitioner)
Most patients misunderstood the different implications of somatic and germline testing.	“I think [patients] need to know the purpose of [somatic testing] is really to form some sort of target for treatment decisions. But I think they also should have the informed consent that this could pick up potentially germline findings but that they would need to be confirmed because there definitely are patients that just are very confused by that possibility and really don’t feel properly informed about that part until they’re in our office after the fact.” (2301; Genetic Counselor) “Well, when you look at the notes that come over with the patient, it’ll say something like – they’ll say, “Discuss the value of somatic testing with the patient for treatment decisions. Patient concurs testing order.” But to me, that’s not really informed consent. And my take is, when they arrive, that they had no idea. A lot of them seem to understand what might help pick a treatment. Most of them have at least – I think most of them are kind of there, but I don’t think many of them realize the difference between a somatic and a germline and that there could be overlap. I don’t think they get that at all. I can’t think of any patients that came who were very savvy about that.” (2310; Clinical Nurse Specialist)

“[Oncologists are] ordering [tumor sequencing] on all sorts of patients, and then it’s a kind of by the way, after it comes back. [For] most of the patients, we’ll ask them point blank, “Did you know that this could have implications for possible germline?” And most of them have not a clue, so they’re pretty along the way when they get sent to see me because they didn’t see it coming.” (2310; Clinical Nurse Specialist)

Provider challenges to pre-test education about secondary germline findings

Several genetic counselors attributed patients’ confusion to receiving inaccurate or insufficient information about genetic testing from their oncologist, which could exacerbate patients’ post-test anxiety and frustration. As a result, these genetic counselors felt like they were “more often mopping up messes rather than getting things all straight and organized to start with” (2305; Genetic Counselor).

When asked about common challenges to oncologists providing information to patients before ordering tumor sequencing, oncologist participants mentioned discussing sequencing in the context of patients’ treatment but not always discussing the possibility of secondary findings because of limited clinic time. Given the complexity of genetics, a few oncologists were concerned that patients might not understand the genetic information provided within the short span of a clinic visit. Several oncologists also stated that they ordered sequencing as part of a patient’s routine care and would only provide additional genetic education if results indicated a secondary finding (Table 4).

Table 4.

Themes in provider-reported challenges to pre-test education about secondary germline findings

Themes in Provider-Reported Challenges	Example Quotes
Insufficient time for pretest education	“And I think another challenge for genetics is because of that it’s very time consuming. Even for the people who don’t really ask for much. It’s still time consuming for other doctors because we go into it trying to prepare and make sure that we have at least a little bit of expected to know about what we’re trying to talk about. So, a lot of our nurse practitioners do. The doctors may order it, but we leave a lot of the counselling to nurse practitioners because it does take so much time.” (2320; Medical Oncologist)
Ordering providers discuss the therapeutic impact of testing rather than its potential impact on family.	“Not a really long time because from the business perspective, all they are thinking about is whether or not to have a treatment. So, what I’m basically telling them is that they have 1 percent to 2 percent chance of having microsatellite instable pancreatic cancer...I’m just kind of highlighting the variants that we typically see. Probably five minutes or so, to really kind of go over and talk about those cons, what variants I’m looking for, and the rarity of the variants.” (2315; Medical Oncologist)
Testing is ordered as part of routine care and post-test education provided only when necessary.	“I think a lot of [oncologists] feel like, ‘Well, if it comes back and there’s nothing…then we didn’t need to really tell the patient that anyway.’ So, I think [oncologists] feel like, ‘Well, I’ll put that fire out when it happens.’” (2310; Clinical Nurse Specialist)
Patients can receive inaccurate information from their ordering provider.	“So, they just seem a little bit blindsided or overwhelmed by the results and learning about cancer risks and maybe in some of those scenarios without really getting adequate counseling at the time of result disclosure because – in those instances at least in my experience – sometimes when the person who’s doing the testing isn’t necessarily specialized in genetic testing or focused or have a lot of background in it and so that’s why maybe the consent process wasn’t as informed as it would have been with someone like myself.” (2306; Medical Oncologist) “I’ve had many patients who have gotten completely inaccurate information [for their ordering provider]. They have had a VUS identified from another provider, and then that provider has counseled them that they need to have surgical intervention over big, big life changes because the providers don’t even know themselves what they’re dealing with. And then, patients that had no idea that this might affect their kids, they don’t know if they would have done the test if they had known that, and so many different things when they haven’t had the counseling.” (2312; Nurse Practitioner)
Provider perceptions of patients’ genetic literacy influences amount of time spent on pre-test education.	“I think depending how the patient understands – their education level. If they do understand what’s a gene and a variant, [pre-test education] will be [sic] easier, right? So it will be like five minutes.” (2316; Medical Oncologist) “I don’t think we have a consent process for somatic. I mean I do let patients know that I’m doing it, but most people…seem a little bit confused going over it, trying to approach that subject. So, I keep it very general that the cancer has pieces that we can try to target and try to find different treatments for.” (2320: Medical Oncologist)

Patient and provider recommendations for the content and provision of information before somatic testing

Overall, most clinicians advised that patient education before somatic sequencing must include a discussion of 3 points:

1. Ordering providers should discuss the different purposes of tumor sequencing and germline testing to minimize potential confusion for patients.

2. Ordering providers need to spend more time discussing the medical decisions that patients might need to make if their tumor findings are actionable (eg, starting a new cancer therapy or undergoing surgery). Providers can also prevent unrealistic expectations about testing by mentioning that tumor sequencing might not reveal any actionable findings.

3. Ordering providers should mention the possibility that tumor sequencing may reveal information relevant to family members and result in a referral to a genetics provider.

Several patients wished they had received more counseling about the implications of germline findings and the likelihood that their results would affect family members. Patients also recommended that clinicians mention the lengthy time between sample collection and results disclosure, because the prolonged wait caused additional anxiety (Table 5).

Table 5.

Themes in patient recommendations for the content and provision of information before somatic testing

Themes in Patient Recommendations for Somatic Testing Care Delivery	Example Quotes
Tell patients about the possible result types and the potential impact on blood relatives.	“I think that before [consenting to testing], a little information about what, what the procedure is and what this genetic test is about and what chances there are as well that – the chances that my children have of me passing it on, or what percent [chance] I had also that my parents had passed it down to us, and what is it that – those kinds of genes, what kind of disease do they usually find, whether in men like with the prostate or colon; where does the gene usually stick, so to speak, it’s a cancer gene, where is it that it shows up in men more and in women more.” (2028; Patient)
Tell patients about the medical decisions they may face if their test results are actionable.	“But [the doctor] never asked me before, like, ‘We have results that might tell you” – he never really gave an introduction, like, “Do you wanna hear this? Are you interested in hearing about any genetic things? And this is what it would be like.’ Which I kinda wish it had. Or he told me that I had that. And then it’s kinda nuts that you’re already in the – like, your foot’s already there. How can you not take the second step? You just feel like you’re gonna fall.” (2007; Patient)
Tell patients how long confirmatory germline testing will take.	“The only thing that I would change personally, for me, is the long delay from the time of drawing the blood samples to getting the results. And so, the longer it [inaudible], the more we were certain there wasn’t going to be anything negative in the report.” (2023; Patient) “I wish I would’ve known right – when they saw that I had the variant…So, while I was waiting, that’s what I was thinking the whole time. But not until they got the results back, which wasn’t their fault, did they tell me it wasn’t that variant. So, it was just the anticipation of waiting….” (2031; Patient)

Clinicians suggested various methods for delivering this information before testing (Table 6). Overall, oncologists preferred letting genetic counselors provide all necessary informed consent for patients because of their greater knowledge of genetics and longer clinic time with patients. Some advanced practice providers maintained that oncologists should provide explicit informed consent before tumor sequencing, whereas genetics providers (ie, genetic counselors and geneticists) facilitate informed consent before germline testing. Additionally, clinicians recommended streamlining the referral workflow between oncology clinicians and genetics providers so that patients with secondary findings receive confirmatory germline testing and appropriate support in a timely manner.

Table 6.

Themes in provider recommendations for the content and provision of information before somatic testing

Themes in Provider Recommendations for Somatic Testing Care Delivery	Example Quotes
New delivery models are needed to facilitate pre-test education and pre-test counseling.	“The genetic counselor had created all these information sheets for us to share with our patients, we had weekly or monthly gene reviews that we would go over what are the clinical implications. If I was in a practice that didn’t have those things, I would say all of those things would be very helpful.” (2313; Nurse Practitioner) “And so, a tool…is fantastic for a busy oncology practice where the oncologist’s goals are to know which tumors are present and which variants are present in the tumor and/or germline. And it balances sort of the principles of having informed consent. I think it’s very useful. It just has to be fast. It has to be fast.” (2311; Medical Oncologist)
Explicit pre-test education should occur before consenting to both tumor sequencing and germline testing.	“Before consenting to somatic testing, patients should understand that there could be other implications to testing or information that isn’t what we’re seeking to learn that could be found out and that might warrant further follow up, such as an increased risk or explanation for why they developed that cancer.” (2306; Medical Oncologist) “I think patients should be required to sign a consent stating that they understand they’re having somatic testing, why they’re having it done, that there can be possible implications, including something unexpected like a germline variant [sic] being found. I think they should have an awareness that that can happen. I think they should be counseled, kind of like they do with germline testing, only for somatic testing.” (2307; Genetic Counselor)
Ordering providers should seek guidance from genetics providers as soon as a secondary finding is suspected.	“[Providers] need to be involving genetics…earlier rather than later, later, later. Sometimes, they’ll send me and they’ll say, “Well, the patient has expired, and we want you to work with the family,” where I really needed to test that person. So, it’s kinda like not informative to the family if they test negative because they don’t really know what I’m dealing with. And so, they need to have earlier input from genetics, not wait until the last moment.” (2310; Genetic Counselor) “I guess at the end of the day, if you do present a patient a problem or a medical issue is seen that you have this particular germline or somatic variant, then as a clinician, my job is to make sure that I have solutions or options available. And if I don’t know the answer, which happens quite often because that’s not my expertise sometimes, then I need to have the right people available that can give people the right answers that they’re looking for. So, one thing I feel like that I need to work on at our institution is establishing better referral workflow so that people can get seen as quick as possible…make sure that we connect them to the right experts as soon as possible.” (2314; Medical Oncologist) “I think, my personal preference, is…genetic testing has to be done by a genetic counselor because it requires appropriate counseling. And appropriate background to be honest with you. You know, you need to have the history to interpret these results.” (2317; Medical Oncologist)

Both medical oncologists and nurse practitioners suggested new delivery models to facilitate pre-test education, such as education sheets or online tools (Table 6). Most patient respondents similarly favored the use of an online support tool to make informed decisions about testing and share accurate information with family members (N = 8/12; 66%).

Discussion

Patients

Patients displayed a variety of emotional responses to their secondary findings, with most feeling gratitude and some experiencing distress. As in prior studies, responses of stress, confusion, and anxiety were often displayed by patients who received minimal pre-test education about germline testing.^29,40–42 An additional burden existed for patients who were concerned that disclosing their result(s) would cause family members worry or guilt. For many patients, secondary findings bring the responsibility of disseminating accurate information to family members, making difficult medical decisions, possibly causing psychological distress to loved ones, and navigating strained relationships within their family.^43–46 These findings indicate a need for increased genetic interventions and clinician guidance to facilitate genetic information sharing within families.^43–46

Although all patients were referred for confirmatory testing of suspected secondary germline findings, only some of our patients recalled their personal test results and/or recalled having genetic counseling. This aligns with previously published work on testing recall and could be due to several factors.^36,47 First, testing recall could be hindered by the overwhelming physical, mental, and emotional challenges of cancer, such that secondary findings and genetic counseling sessions were less central to their cancer care; therefore, the information was less frequently retained.^15–17 Alternatively, if patients have low levels of distress from the result, they may be less likely to remember the results. Third, active anti-cancer therapy has been shown to alter patients’ cognitive processing and understanding of their care.^48–51 Finally, late-stage patients with cancer may also be coping with external stressors that clinicians are not always privy to, making it even more important to provide psychosocial resources when patients appear to have little external support. Future research with a larger patient cohort could explore how cognitive and emotional responses differed when patients were seen by genetics versus non-genetics providers for secondary findings.

Clinicians

Overall, most clinicians agreed that a pre-test education model had significant benefit for patients undergoing tumor sequencing but expressed different perspectives and recommendations depending on their role in patient care.^52–54 Consistent with prior work, oncologists were inclined to discuss testing only as it pertained to treatment, citing minimal clinic time as a primary obstacle to formal pre-test counseling.⁵⁵ Several oncologists were concerned that patients might not understand additional genetic information within the short span of a clinic visit.^56,57 Consequently, some oncologists felt that education about somatic and germline testing would be best provided by genetic counselors to support patient understanding.⁵⁸ Genetic counselor-mediated testing models have been associated with higher patient knowledge, positive patient experience, and greater understanding of genetic testing.⁵⁹ Thus, one solution could include greater genetic counselor engagement and support during tumor sequencing to address patient knowledge and emotional needs.^60–62 This option would necessitate increasing the number and role of genetic counselors in the tumor sequencing process. Although this solution could optimize patient understanding, it may be much less feasible than non-person-mediated interventions, given the shortage of genetic counselors nationally.⁶³

Recommendations

Insights from patients and providers in our study align with current literature recommendations for the content and provision of patient education before tumor sequencing.¹⁵ Namely, ordering providers should discuss (1) the different purposes of somatic and germline testing and (2) the possibility that tumor testing may reveal unanticipated germline findings or no actionable findings at all.¹⁵ Because a patient’s health condition may affect their preferences and intentions for genetic testing, it is important that clinicians reconfirm patients’ interest in receiving secondary findings while also timing disclosure appropriately.¹⁷ These preferences should be thoroughly documented in patients’ medical records for continuity of quality care.^14,18

Moreover, investigators should continue to develop scalable and effective educational interventions that facilitate pretest education and informed consent.¹⁶ Digital genetics platforms for decision support and results disclosure are correlated with decreases in counseling time, risk perception, patient anxiety, and decisional conflict and a greater intent to undergo testing and increase in general genomics knowledge.^35,64,65 These tools could be used with providers during clinic visits, or in the presence of family members, when patients have cognitive or physical limitations. Digital platforms may also ease the burden of disseminating accurate genetic information to loved ones.

Limitations and strengths

Whereas prior research has primarily focused on secondary findings from a clinician perspective, this study highlights the unique emotional challenges and perspectives of advanced-stage patients with cancer. Few studies also offer parallel insights from patients and advanced cancer providers, which is key to understanding the attitudes and challenges of both parties in the return of results process.

However, similar to many qualitative studies, our sample size was small, with limited diversity in race, age, or language. Most patients interviewed were white males and treated at well-resourced academic centers where they met with a genetic counselor and a medical geneticist before and after confirmatory germline testing. Given that our study was qualitative and confined to late-stage patients with cancer, our results are not generalizable to early stage patients or patients with cancer at large. Moreover, despite the use of back translation of Spanish interviews by 2 team members, it is possible that meaning could have been lost during translation. Lastly, patients who were interviewed 2 weeks after disclosure may have had better recall than those 6 months out, but we did not find compelling data to infer a significant difference.

Conclusion

Insights from clinicians and patients in our study underscore the importance of preparing patients for unanticipated germline findings through quality pretest education before tumor sequencing. Despite agreeing on the difficulties of implementing pre-test education within current clinical workflows, providers had differing perspectives on who should be responsible for providing this information. Although there is no simple solution to providing pre-test education for the growing number of patients receiving tumor sequencing, digital educational interventions may address patients’ information needs and offset the clinical demands of counseling. Although institutions, health care systems and clinicians continue to grapple with this issue, this work serves as a first step toward creating an empiric foundation on which to develop future solutions and interventions.

Data Availability

Data for this project consist of qualitative interview data. De-identified interview data may be available upon request if approved by the Institutional Review Board.