Introduction

Deceased and living potential organ donors are selectively being tested for apolipoprotein L1 (ApoL1) variants G1 and G2 that are associated with a higher risk of kidney disease.1,2 Transplanted organs that carry these risk variants are associated with lower transplant survival3,4; whether living donors with two copies of the risk variants in any combination are at greater risk by donating a kidney is unknown.5–7

Individuals with recent sub-Saharan African ancestry are primarily at risk. The G1 and G2 variants that are associated with a higher risk of kidney disease risk may offer protection against African sleeping sickness.8

Although controversial, Black race is frequently used as a surrogate for African ancestry.9,10 There is considerable variation in who is tested. Some centers test all Black patients; others test only those with self-reported African ancestry or do not test any patients.2,11

Many genetic tests target people on the basis of race. Studies have targeted people when a genetic risk factor is known (e.g., sickle cell disease and glaucoma),12–14 as well as when genetic factors explain an increased disease burden (e.g., breast cancer and BRCA1/2 variants).15–17

A review article examining findings from 27 studies reports Black peoples' attitudes toward racially targeted genetic testing.18 Participants recognize potential benefits, such as supporting informed decision making or motivating healthier life choices.19–25 However, participants commonly mention concerns such as genetic testing might increase stigma or discrimination or that testing will occur without appropriate communication and education.19–21,25–29 An ApoL1 Long-term Kidney Transplantation Outcomes (APOLLO) network Community Advisory Council (CAC) article explores the importance of Black voices in shaping ApoL1 genetic testing policies from the research and implementation stages.30

Engaging with Black living organ donors, deceased donor family members, and transplant recipients, as well as the health professionals who serve patients—transplant physicians, surgeons, and genetic counselors—to seek a consensus on how to implement ApoL1 testing in clinical settings is crucial.

Building on the work of Mohan et al., we identified a series of policy and procedure questions to address as transplant programs translate ApoL1 testing into clinical practice31 (see Box 1). Transplant programs have highly variable ApoL1 testing practices, including no testing for clinical decision making.11 As the national APOLLO network has completed enrollment and initial findings are being published,2 transplant programs will likely face increasing pressure to consider how genetic risk variant information is being shared with recipients and thus, in many cases, with living donors. Programs will need policies on how to identify appropriate individuals for testing (e.g., by asking about race or ancestry or relying on health records), on whether testing is optional or required, on who receives results (e.g., the individual tested, deceased donor family members, or potential organ recipients), and how to use test results (e.g., to support shared decision making or automatically rule out donation).31 If the APOLLO study's hypothesis is confirmed that recipient ApoL1 risk variants negatively affect transplant outcomes,32 there will be an additional impetus to address these questions—especially given the prospect of effective therapies currently in phase 3 clinical trials.33,34

Box 1. Apolipoprotein L1 genetic testing policy questions. Who gets tested? How do we identify people? Who decides whether testing occurs? How are apolipoprotein L1 test results shared with individuals? Who receives apolipoprotein L1 test results? How are test results used? Methods

We received an APOLLO ancillary R01 grant from the National Institute on Minority Health and Health Disparities to conduct a Delphi consensus panel on ApoL1 clinical policies and practices (among other aims). Delphi panels help cultivate consensus among stakeholder groups on a topic or issue, such as prioritizing policies; Delphi panels also allow panelists to vote anonymously without excessive influence from other panelists.35,36 Delphi panels differ from other survey methods that measure dispersion of attitudes in several ways: They involve presentation of information, interaction among panelists, independent ratings, further interaction, and final ratings. We felt that this approach was necessary on the basis of our 4-year experience with the APOLLO project: Information about ApoL1 is easily misunderstood, making education and discussion with experts valuable before assessing policy preferences, and different groups of people have different interests at stake, making the exchange of views valuable prior to final ratings of support for policies. Accordingly, our Delphi panel process involved two rounds of educational webinars and three rounds of questionnaires administered to panelists with the purpose of converging on agreement about a policy option by a majority of panelists.37

The goal of the panel was to provide transplant centers with a range of acceptable policy options, as well as to rule out options that the panel viewed as unacceptable. We felt it was important to give transplant programs options where possible because transplant program resources and contexts vary tremendously.38 We felt it was equally important to provide transplant programs with guidance on which options to take off the table. This cannot be done simply by rank ordering Likert-scale results. Thus, in rounds that involved voting, we provided participants with choices derived from real-world community consensus-building projects: I support, I can live with, or I oppose a given policy.39–41

This research was approved by the Washington University School of Medicine Human Research Protection Office (IRB ID 202203095). Panelists were paid up to $600 for participating depending on the number of Delphi activities (i.e., webinars, surveys) they completed.

Panelists

Panelists were recruited between April and May 2022 via email using a script approved by the Washington University School of Medicine Human Research Protection Office. As is typical with Delphi panels, we used criterion-based sampling to identify panelists with relevant expertise and experience who live in the United States35,42 We intentionally recruited diverse stakeholders who might be viewed as having competing interests with respect to ApoL1 testing policies and practices. For example, organ recipients might want to know ApoL1 results while living donors and their families might prioritize confidentiality; transplant centers might want to use ApoL1 results to exclude a living donor with two risk alleles while a donor may wish to proceed with donation; genetic counselors may believe they are uniquely qualified to disclose ApoL1 results while transplant physicians may feel competent to share results and may lack adequate access to genetic counseling services.43 Accordingly, we recruited several panelists for each of seven stakeholder groups: (1) living kidney donors, (2) deceased donor family members, (3) recipients of a deceased donor kidney, (4) recipients of a living donor kidney, (5) nephrologists, (6) transplant surgeons, and (7) genetic counselors. We recruited panelists from the APOLLO steering committee and CAC, our R01 advisory committee, our professional networks, and participants from our team's other studies on ApoL1 genetic testing.

Procedures

All surveys were administered in Qualtrics survey software, and links were distributed via email. The complete round 1 survey was reviewed in cognitive interviews with representatives from each stakeholder group.44Figure 1 presents an overview of the Delphi panel process.

Figure 1:

Overview of Delphi panel procedure.

Review of Background Materials

Before completing the round 1 survey, panelists were sent informational materials to read, including an infographic about how ApoL1 genetic variants function and journal articles describing ApoL1 genetic testing, as well as a scoping review about attitudes and beliefs regarding race-targeted genetic testing of Black people.18

Round 1 Survey

The panelists were presented with key information about ApoL1 genetic testing, answered five knowledge check questions to confirm their understanding of key information, answered brief demographic questions, and read through each of 41 different ApoL1 genetic testing policy options or practices developed and refined by the research team as part of an earlier research effort. These policy options explore both living and deceased donor contexts.

Panelists were asked to add any important policies or practices they thought were missing that should be considered in the Delphi panel and to suggest edits to current options. The research team met to discuss the proposed additions from panelists.

Orientation Webinar

After the round 1 survey, the research team held an hour-long educational orientation webinar with all panelists via videoconferencing. The webinar was held at two time points to accommodate panelist schedules and was recorded for viewing by those panelists unable to attend either webinar times. During the orientation webinar, the research team presented preliminary data gathered by the team in an in-progress survey of APOLLO network participants and team members on ApoL1 policies and reviewed the materials sent ahead of the round 1 survey to demonstrate and reiterate the project's purpose and significance.11,18 The orientation webinar was also designed to foster informed engagement by giving panelists the opportunity to ask questions and providing panelists with information about what they would be expected to do.

Round 2 Survey

Panelists were asked to read through 42 different policies related to ApoL1 genetic testing and indicate whether they: (1) support, (2) could live with, or (3) oppose each policy. Following best practices for Delphi consensus panels, we operationally defined consensus before voting as approximately 77% saying they either support/could live with a policy or oppose a policy.45

Deliberation Webinar

Panelists were sent a document summarizing the results of the round 2 survey ahead of the deliberation webinar. During the hour-long deliberation webinar, panelists were presented with the percentage of panelists who supported, could live with, or opposed each of the policy options that did not reach consensus in the round 2 survey. In addition, two policy options that reached consensus during the round 2 survey were revisited because they were especially complex and required providing additional information to panelists about the genetics of ApoL1 so that the policy options could be interpreted correctly. Panelists were asked to share their views about policy options in light of other panelists' ratings to inform deliberations and voting for the round 3 survey. The webinar was recorded so that panelists unable to attend the webinar could listen to the points made for and against a policy option before completing the round 3 survey.

Round 3 Survey

Panelists were presented with the percentage of all panelists who supported, could live with, or opposed each of the no-consensus policy options and were asked to consider the deliberations from the previous webinar and rerate these policy options. For two policy options that reached consensus in round 2 but were revisited, panelists were asked to consider the points raised during the deliberation webinar, rerate both policies, and provide a rationale for their ratings.

The entire Delphi panel process ran from September 2022 through January 2023.

Results

Twenty-seven panelists participated in round 1, 26 panelists attended an orientation webinar and one panelist watched a recording, 26 participated in round 2, 24 attended the deliberation webinar and two watched a recording, and 26 participated in round 3. Of the 27 panelists who participated, four were living kidney donors, three were deceased donor family members, five were recipients of a deceased donor kidney, five were recipients of a living donor kidney, four were nephrologists, four were transplant surgeons, and four were genetic counselors. One panelist received kidney transplants from a deceased donor and later a living donor. They were also a deceased donor family member. Supplemental Table 1 presents the names, affiliations, and roles of all panelists. Fifty-two percent of panelists were male. Panelists were age between 20 and 69 years. Seventy percent of panelists identified as Black, 19% as White, 11% as Asian, and 4% as American Indian or Alaska Native. Table 1 presents full demographic information for panelists.

Table 1 - Apolipoprotein L1 Long-Term Kidney Transplantation Outcomes Delphi panel demographics Characteristic No. (%) Self-identified racea 27 (100)  American Indian or Alaska Native 1 (4)  Asian 3 (11)  Black 19 (70)  Native Hawaiian or other Pacific Islander 0 (0)  White 5 (19)  Prefer not to answer 0 (0) Ethnicity 27 (100)  Hispanic or Latino 2 (7)  Not Hispanic or Latino 24 (89)  Prefer not to answer 1 (4) Age, yr 27 (100)  20–29 4 (15)  30–39 6 (22)  40–49 6 (22)  50–59 4 (15)  60–69 7 (26)  70 or older 0 (0)  Prefer not to answer 0 (0) Sex 27 (100)  Male 14 (52)  Female 13 (48)  Prefer not to answer 0 (0)

aNote: One panelist selected more than one race. On the survey, panelists were able to select all options that applied to them.

Following round 1, one new policy option was added: When feasible, give patients options on how to receive results. The team also made some wording changes to existing instructions and policy options to add clarity. A consensus was reached on 24 policy options during the round 2 survey. The remaining 18 policies that did not reach consensus were discussed by panelists during the deliberation webinar and included in the final round 3 survey.

Across rounds 2 and 3, the panel reached a consensus in support of 18 policy options and against 15; it failed to reach a consensus on nine policy options. Tables 2–4 present rank-ordered consensus information for all policy options corresponding to each overarching policy question. Specifically, Table 2 presents consensus information for policy options with consensus in support, Table 3 presents consensus information for policy options with consensus in opposition, and Table 4 presents policy options that failed to reach consensus.

Table 2 - Policy options with consensus in support Policy Option Support Live With Oppose Round Consensus Achieved Policy question 1: Who should get tested?  Ask donors or their families if they have African ancestry and offer testing to donors who identify as having African ancestry 7 18 1 2 Policy question 2: Who should make decisions about donor testing?  Educate living kidney donors about the option of ApoL1 testing, and offer the test if the donor wants it 20 6 0 2  Educate living kidney donors about the option of ApoL1 testing, and offer and encourage people to take the test 11 13 2 2 Policy question 3: Who should give test results to living donors?  They can get their results during a meeting with a genetic counselor. The meeting could be face-to-face in person or online during a telemedicine appointment. They can talk with the genetic counselor about the results 21 5 0 2  They can get their results during a meeting with their doctor. The meeting could be face-to-face or online during a telemedicine appointment. They can talk to their doctor about the results 19 7 0 2  They can get their results from a genetic counselor over the phone. They can talk with the genetic counselor about the results 17 8 1 2  When feasible, give patients options on how to receive results 16 9 1 2  They can read their results online. If they want to discuss the results, they can request a call with a genetic counselor 15 9 2 2  They can get their results from a doctor over the phone. They can talk to their doctor about the results 12 13 1 2 Policy question 4: Who should have access to test results?  Transplant centers should encourage living donors to give their immediate family members their results 11 14 1 2  OPOs should offer to share results with the deceased donor's immediate family members. They should share the results only if the immediate family members want the results 18 6 2 2  Transplant centers should encourage living donors to give the person who needs a kidney their results 9 15 2 2 Policy question 5: What should be done with test results?  Living kidney donors and transplant doctors decide together whether to move forward with donation 18 8 0 2  People who need a kidney and transplant doctors discuss the deceased donors ApoL1 results and then decide together whether to accept the kidney 17 8 1 2  People who need a kidney get information about the living donors ApoL1 results. After considering the results, people who need a kidney decide whether to receive the kidney 12 12 2 2  People who need a kidney get information about a deceased donors ApoL1 results. After considering the results, people who need a kidney decide whether to take the kidney 11 13 2 2  Living kidney donors, the person who needs a kidney, and transplant doctors all get the donors ApoL1 results. They decide together whether to transplant the kidney 11 10 5 2  Living kidney donors get their ApoL1 results and then decide whether to donate 10 11 5 2

N=26. Consensus=at least 20 of 26 panelists who either oppose or support/could live with a policy. ApoL1, apolipoprotein L1; OPO, Organ Procurement Organization.

Table 3 - Policy options with consensus in opposition Policy Option Support Live With Oppose Round Consensus Achieved Policy question 1: Who should get tested?  Do not offer testing to any kidney donors, no matter what their race or ancestry is 0 3 23 2  Offer testing to all kidney donors if the transplant center doctors think the donor is Black or has African ancestry 0 2 24 3 Policy question 2: Who should make decisions about donor testing?  Do not provide ApoL1 testing to living kidney donors at all 0 0 26 2  Automatically do ApoL1 testing on all living kidney donors. Do not educate them about the test, and do not ask for their permission to do the test 0 1 25 2  Educate living kidney donors about ApoL1 testing, but do not encourage or offer the test 0 4 22 2  Educate living kidney donors about ApoL1 testing and require them to get the test 0 6 20 3 Policy question 3: Who should give test results to living donors?  They can read their results online without help from a genetic counselor or physician 1 2 23 3 Policy question 4: Who should have access to test results?  OPOs should never share results with the deceased donor's immediate family members 0 3 23 3  Transplant doctors should never share a deceased donors ApoL1 results with the person who needs a kidney 0 4 22 2 Policy question 5: What should be done with test results?  Transplant doctors get information about a living donors ApoL1 results and then decide alone whether to accept the donors kidney for transplant 0 2 24 3  Transplant doctors get information about a deceased donors ApoL1 results and then decide alone whether to accept the donors kidneys for transplant 0 3 23 3  Never use ApoL1 results to make decisions about whether living donors can donate their kidney 4 0 22 3  Do not allow kidneys from deceased donors who have the risky forms of ApoL1 to be donated 1 4 21 2  Never use ApoL1 results to make decisions about whether deceased donor kidneys should be transplanted 3 2 21 3  Do not allow living kidney donors who have the risky forms of ApoL1 to donate 1 5 20 3

N=26. Consensus=at least 20 of 26 panelists who either oppose or support/could live with a policy. ApoL1, apolipoprotein L1; OPO, Organ Procurement Organization.

Table 4 - Policy options that failed to reach consensus Policy Option Support Live With Oppose Policy question 1: Who should get tested?  Offer testing to all kidney donors, no matter what their race or ancestry is 11 5 10  Ask donors or their family members if they identify as Black, and offer testing to donors who identify as Black 4 13 9  Offer testing to all kidney donors whose medical record says that they are Black 3 12 11 Policy question 2: Who should make decisions about donor testing? Not applicable Policy question 3: Who should give test results to living donors? Not applicable Policy question 4: Who should have access to test results?  Transplant centers should require living donors to give the person who needs a kidney their results 5 13 8  Transplant doctors should automatically share the deceased donor's results with the person who needs a kidney 10 6 10  Transplant doctors should offer to share a deceased donor's ApoL1 results with the person who needs a kidney. They should share the results only if (1) the person who needs a kidney wants the results and (2) if the deceased donor's family members give permission 5 7 14  OPOs should automatically share results with the deceased donor's immediate family members 2 9 15 Policy question 5: What should be done with test results?  Discourage living kidney donors from donating if they have the risky forms of ApoL1 1 8 17  A living kidney donor and the person who needs a kidney get the donors ApoL1 results. Then they decide together whether to transplant the kidney 5 5 16

N=26. Consensus=at least 20 of 26 panelists who either oppose or support/could live with a policy. ApoL1, apolipoprotein L1; OPO, Organ Procurement Organization.

For policy options about who should get ApoL1 genetic testing and under what conditions, panelists reached consensus in support of one option: asking kidney donors or their immediate family members if they have African ancestry and offering ApoL1 testing to those who identify as having African ancestry. By contrast, panelists reached consensus against two policy options: testing based on the physician's guess about race and not offering ApoL1 testing to anyone regardless of race.

For policy options about who should make decisions about donor ApoL1 genetic testing, consensus was reached for two policy options. These policy options were focused on providing education to living donors about the option of ApoL1 testing and giving them autonomy in their decision to receive testing. Actively encouraging testing was acceptable, but not requiring testing. The four remaining policy options were opposed by panelists.

For policy options about how living donors should receive ApoL1 test results, all but one policy option reached consensus in support. Most supported options focused on donor education and consultation with health care professionals.

For policy options describing who should have access to a living or deceased donor's ApoL1 test results, three policy options reached consensus in support. These policy options were focused on encouraging sharing results. The two policy options about never sharing ApoL1 test results were opposed by consensus.

For policy options about what should be done with ApoL1 test results for both living and deceased donors, six policy options received consensus in support; these largely focused on shared decision making. For example, our most strongly supported option in the realm of living organ donation was living kidney donors and transplant doctors decide together whether to move forward with donation, and in the realm of deceased donation it was people who need a kidney and transplant doctors discuss the deceased donors ApoL1 results and then decide together whether to accept the kidney. Six policy options were opposed by consensus; they focused on unilateral decision making and prohibiting donation altogether.

Discussion

The CAC for the APOLLO network has challenged a number of common clinical research assumptions in its effort to ensure that Black voices are appropriately influential in ApoL1 research.30 We have viewed the mission of this ancillary consensus project as an extension of this: We facilitated a Delphi consensus process with panelists who nearly evenly represented seven different stakeholder groups and were predominantly Black (70%) and diverse in terms of age and sex (52% male). Given the risks of racially targeted genetic testing and a history of medicine and research that has engendered mistrust within the Black community, we believe the Black community should be gatekeepers of ApoL1 genetic testing policy.

Four authors of this article served on the steering committee or the CAC of the APOLLO network. One thing has become clear across the 7 years of the project: The influence of ApoL1 gene variants on kidney health is challenging to understand. People sometimes misunderstand what recent African ancestry means (e.g., in the past 4000 years), that one copy alone of a risk variant is benign, and that individuals with two risk alleles are still more likely than not to avoid ESKD. Our team's previous work also indicated that increased knowledge of ApoL1 testing in the context of organ donation does affect levels of support for testing, although the effect is variable, with some becoming more supportive and some less supportive after education on ApoL1 testing.43 For this reason, our project provided education on ApoL1 before our first meeting and then met to discuss issues before the first round of ratings. The discussion was rich because different stakeholder groups brought different kinds of expertise to the table.

Race-targeted genetic testing can be controversial and generates some concerns.19–21,25–29 Our panel supported focusing on people of African ancestry when offering testing while rejecting an approach that focuses on race, particularly when race is determined by the impressions of the transplant team. Overall, the question for our panel was not whether but how best to proceed with ApoL1 testing. This position is consistent with earlier work on Black attitudes toward ApoL1 testing.20,21,28,46 Our project succeeded in achieving a consensus on one or more acceptable option for each clinical policy question. In most cases, our panel offers several acceptable options. Equally valuable, the panel achieved a consensus on many options that should be taken off the table; these are options that >77% said they cannot live with—a strong statement of disapproval.

In general, the panel supported options that involve discussion with patients, donors, and family stakeholders: Do not make assumptions about race, ask about A