Streamlined Genetic Testing in Prostate Cancer

Genetic testing is part of clinical care for women with breast and ovarian cancer. Guidelines have led to increasing genetic testing participation among eligible women. Identifying pathogenic variants (PVs) in cancer susceptibility genes has important treatment, management and risk reduction implications for patients and their family members. Men are just as likely as women to carry a PV in a cancer risk gene and men with prostate cancer are at particularly high risk for carrying a PV. NCCN guidelines recommend germline genetic testing in men with metastatic prostate cancer or with Gleason 7+ prostate cancer. These guidelines are likely to expand given recent studies suggesting that 17% of prostate cancer patients may harbor a cancer susceptibility PV and that current referral criteria miss many of these patients. Prostate cancer patients with a PV in BRCA1 or BRCA2 (BRCA) have more aggressive prostate cancers, are at risk for other primary cancers (e.g., male breast and pancreatic) and their family members are at risk for a variety of cancers. Thus, genetic testing of prostate cancer patients has implications for treatment, management, and risk reduction for the patient and his family members.

Despite these benefits and existing referral guidelines, few men with prostate cancer are tested. This reflects low genetic counseling referral and participation, since men who attend genetic counseling get tested at the same high rate as women. Low participation is likely due to underestimation of the personal health relevance of testing and lack of physician genetic counseling referral. Thus, the requirement to obtain individual genetic counseling prior to genetic testing may be a barrier to the receipt of guideline consistent genetic testing. Recent evidence suggests that traditional comprehensive, patient-centered, educationally focused pre-test genetic counseling often does not match the needs of patients. Further, as genetic referral guidelines continue to expand, demand for genetic counseling will outstrip delivery capacity. Thus, alternative approaches that raise awareness and facilitate access to genetic testing are needed to maximize our ability to extend the benefits of testing to prostate cancer patients and their family members while at the same time accommodating increased demand and conserving scarce genetic counseling resources.

We will test a proactive and streamlined pre-test genetic education (ST) print intervention designed to speed, simplify and target genetic testing delivery for prostate cancer patients. Participants randomized to ST will have the option to proceed directly to genetic testing bypassing traditional pre-test genetic counseling. Usual care (UC) participants will be informed that they meet guidelines for genetic counseling referral and will be provided with contact information to schedule a standard telephone pre-test genetic counseling session. All participants with a PV or variant of uncertain significance (VUS) will be scheduled for a telephone genetic counseling disclosure session. ST participants who are found not to carry a PV or VUS will have this result disclosed via clinical letter while disclosure for all UC participants will be per standard clinical care by the genetic counselor by phone. By proactively identifying eligible patients, providing streamlined information and facilitating genetic testing, the ST intervention will increase awareness, facilitate informed testing decisions, remove the barrier of pre-test genetic counseling and improve genetic testing delivery capacity.

We will enroll eligible prostate patients diagnosed from 2010-2019 who are being followed within radiation and medical oncology clinics at the MedStar Georgetown University Hospital and MedStar Washington Hospital Center. Our specific aims are:

Evaluate the impact of ST vs. UC on genetic testing uptake. H1.1: Patients randomized to ST will be more likely to complete genetic testing compared to UC. H1.2: Participants randomized to ST will make better informed genetic testing decisions (characterized by high knowledge, risk comprehension and concordance with attitudes) compared to UC. H1.3: Few patients in the ST arm will opt for traditional genetic counseling prior to testing.
Evaluate patient satisfaction and psychosocial outcomes in ST vs. UC. H2.1: Compared to UC, those randomized to ST will be more satisfied with their genetic testing decision and will have less decisional regret.
Evaluate the impact of ST vs. UC on uptake of cascade testing in unaffected family members. In exploratory analyses designed to provide effect size estimates, we will compare the arms on rates of genetic testing in relatives of patients found to carry a PV.

Gender and Minorities. Participants will be adult men. Given the racial/ethnic breakdown of the clinics we are recruiting from. We expect that at least 30% of participants will be members of racial/ethnic minority groups.

Recruitment. Working with our clinical collaborators and the Survey, Recruitment and Biospecimen Collection Shared Resource (SRBSR) will obtain contact information of potentially eligible patients from clinical databases and/or the EMR. We will mail an introductory letter describing the study (from the study MPIs and the patient's physician), opt-out postcard/phone number, informed consent document and a print version of the baseline survey to all potentially eligible participants. We will also email electronic versions of these materials. Two-weeks following this mailing, a research assistant (RA) will call patients who have neither opted out nor completed the baseline survey. For men who have not returned the consent document, we will use an IRB-approved verbal consent for completing the baseline survey and request return of the written consent prior to randomization. Following completion of print or electronic consent and the baseline survey, the RA will randomize participants. Men randomized to ST will be sent a priority mail packet with detailed genetic testing educational material. Men randomized to UC will be mailed a letter informing them that they meet guidelines for genetic counseling and provided with contact information to schedule a genetic counseling appointment with a LCCC genetic counselor.

Randomization. Following completion of the baseline survey and consent document, an RA will randomize participants via computer-generated random numbers in blocks of 8 and a ratio of 1:1. Participants will be notified of their random assignment via letter and email.

Assessments. We will administer baseline (T0) a 1- to 2-month (T1) follow-up telephone/electronic surveys. Genetic testing uptake will be assessed via clinical records and basic clinical and demographic information abstracted from the EMR.

Streamlined Testing (ST) Intervention. Men randomized to ST will be sent a priority mail packet with the ST print education materials and information on how to proceed with genetic testing. ST participants will have the option of proceeding directly to genetic testing or scheduling a telephone genetic counseling session. The content of the ST intervention is described in section 5.0 above.

Usual Care (UC). After completion of the baseline survey, UC participants will be sent a letter from their physician indicating that they meet eligibility criteria for genetic testing, recommending that they schedule genetic counseling and providing a contact telephone number to schedule their session. Patients may opt for free telephone genetic counseling.

Genetic Testing. Participants who opt for genetic testing will be offered a standard multigene panel of at least 40 genes, including BRCA and genes associated with potential differential diagnoses (e.g., Lynch syndrome). All testing will be performed by Invitae, a CLIA certified clinical lab not affiliated with the GLCCC. Invitae offers genetic testing free of charge to eligible prostate cancer patients. Prior to proceeding with genetic testing, participants will be required to sign and return a clinical genetic testing consent form, after which we will mail them an at-home DNA kit for saliva collection which will be returned directly to Invitae (shipping costs are pre-paid).

Data Analysis: We will characterize participants by comparing them to decliners on key demographic and clinical variables derived from the EMR. After performing univariate analysis on dependent variables, we will apply any needed normalizing or variance stabilizing transformations. We will follow Consort guidelines for Intent-to-Treat analyses (ITT). We will compare groups at baseline with χ2 and t-tests and will include covariates that are associated with group (p <0.10) in our final models.

Aim 1: Evaluate the impact of Proactive Streamlined Education and Testing (ST) vs. Usual Care (UC) on genetic testing. Analysis: H1.1: Patients randomized to ST will be more likely to complete genetic testing compared to UC. We will use logistic regression to test this hypothesis. After entering significant covariates, we will add intervention arm to test the overall intervention effect. H1.2: Participants randomized to ST will make better informed genetic testing decisions (characterized by high knowledge, risk comprehension and positive attitudes) compared to UC. Using multiple regression, we will enter control/confounding variables and intervention to test the impact of group on our informed decision making outcomes (knowledge, testing attitudes, risk comprehension). H1.3: Few patients in the ST arm will opt for traditional genetic counseling prior to testing. We will conduct descriptive analyses describing the proportion of ST participants who opt for genetic counseling vs. proceed directly to testing vs. opt against testing.

Aim 2: Evaluate patient satisfaction and psychosocial outcomes in ST vs. UC. Analysis: H2.1: Compared to UC, those randomized to ST will be more satisfied with their genetic testing decision and will have less decisional regret. After identifying baseline confounders of each outcome, we will generate linear regression models in which we enter: 1) confounders; 2) genetic test result (dummy coded to generate comparisons of carriers vs. non-carriers and non-carriers vs. untested); 3) group assignment. In the event that we have an insufficient number of PV carriers for these analyses, we will categorize test result into two levels (tested vs. untested) and evaluate results among PV carriers more descriptively. Results of Aim 2 analyses will be primarily focused on generating effect size estimates for our planned trial of newly diagnosed prostate cancer patients.

Aim 3: Evaluate the impact of ST vs. UC on uptake of cascade testing in unaffected family members. In exploratory analyses designed to provide effect size estimates for our planned R01, we will use t-test to compare the ST to UC on the number of relatives of PV carriers who undergo cascade genetic testing.

Power: Power estimates are for 2-tailed tests (α=.05). The study is powered based on our primary outcome of patient genetic testing uptake. We expect no attrition since data will be abstracted from clinic records. We will enroll 120 participants in order to attain 80% power to detect a clinically meaningful increase of 25% in test uptake from 20% in UC to 45% in ST. For secondary outcomes, assuming 10% attrition, we will have 80% power to detect modest effect sizes of d=.54 SDs.