Genetic Testing in Young Adults With Cancer Study

Overview

The overarching goal of our research is to define an evidence-based, sustainable approach to identifying and managing genetic risk among young adults with cancer and their relatives. Conventional practice leaves referral and testing decisions to mostly non-expert clinicians implementing complex guidelines at the point of care, leading to substantial under-utilization. The investigators hypothesize that panel-based universal screening coupled with electronic medical record- (EMR-) based algorithms can improve ascertainment of genetic risk by functioning as an automated, radically simplified default practice in place of repeated single decisions requiring clinician cognitive effort and action. A secondary goal is to explore differences in ascertainment of genetic risk among first-degree relatives of probands.

The investigators will conduct a randomized trial among young adult (YA) cancer patients to rigorously test the hypothesis that universal testing identifies many YA patients with germline cancer risk who would be missed by conventional testing strategies, to improve adherence to risk reduction interventions among YAs with cancer susceptibility, and to increase ascertainment among relatives of these patients at increased genetic risk. The primary outcome of the randomized trial will be ascertainment of probands, but in a companion study, the investigators will also explore the incremental utility of universal testing with respect to ascertainment of risk among relatives. Consenting patients will be immediately randomized using a computer-generated randomization algorithm using permuted variable block sizes with a concealed sequence. Randomization will be stratified by gender, National Comprehensive Cancer Network/American College of Medical Genetics and Genomics (NCCN/ACMG) referral criteria (meets/doesn't meet), and hospital (i.e., HUP vs. other Penn Medicine hospitals). Note that because of demographic differences between hospitals, the investigators are using hospital type as a proxy for race/ethnicity and so will not stratify for it separately. Randomization will occur in a 2:1 ratio (experimental: control). Patient enrollment for genetic testing will be completed within the first 12 months. If patients receive a pathogenic or likely pathogenic result then they will be followed for up to five years to evaluate the impact of the clinical decision support. Recruitment will end when approximately 1238 subjects are enrolled. Subjects will be recruited across five Penn Medicine sites, including Hospital of the University of Pennsylvania (HUP), Penn Presbyterian Medical Center (PPMC), Pennsylvania Hospital (PAH), Chester County Hospital (CCH), and Lancaster General Hospital (LGH). Patients who are notified of their eligibility status and indicate an interest in study participation will be directed to an educational website about hereditary cancer risk and the implications of genetic testing for patients and family members. They will also be provided information about the study. If they wish to consent after viewing the educational module, patients will be automatically sent to REDCap where, following consent, they also will complete a brief survey to obtain demographics and family history information. Patients will also have the option of reviewing information, providing consent, and completing sociodemographic and family history data collection via paper forms. Upon provision of signed informed consent and determination of high/low risk (based on ACMG/NCCN guidelines and determined by either the study research coordinator or clinical genetic counselor as outlined above), patients will be randomized in a 2:1 ratio (experimental: control) into one of the two study arms stratified by gender, meeting NCCN/ACMG criteria, and hospital type (HUP vs. other). Study enrollees will be encouraged to sign up to use their local patient portal, if they haven't done so already, in order to receive study communications, health information, and heath maintenance alerts. Management of patients randomized to the phenotype-based (standard) arm will depend on whether they are identified as being at high or low risk of having a genetic predisposition. Patients who do not meet high-risk criteria will have routine clinical care (i.e., no study-directed referral to a genetic counselor). Patients who do meet high-risk criteria will be referred to a genetic counselor at their treating Penn Medicine institution for a standard cancer genetics evaluation and testing based on genetic counselor recommendation and patient preference. Participants on this arm who receive genetic testing will be tracked and results will be manually entered into the study database, although if they are tested at Ambry Genetics (the lab used in the universal-testing intervention arm), positive results will be directly imported into PennChart (the investigators' local instance of Epic) via a secure interface, which already has been built and is operational. Patients randomized to the universal-testing arm will provide either a saliva or blood sample to have sequencing of a broad cancer risk gene panel conducted by Ambry Genetics. Genetic test results will go directly into PennChart from Ambry via a secure interface (HL7). Patients identified as being at high risk according to NCCN/ACMG criteria will be referred to a genetic counselor at their treating institution, who will be asked to obtain testing using the Ambry broad cancer risk gene panel rather than by usual clinical tests. If they do not follow-up for clinical testing, as outlined above, they will be tested by the study team, using mailed genetic testing kits (with pre-paid mailers to Ambry Genetics). Patients identified as being at low risk will undergo testing using the Ambry broad cancer risk gene panel via mailed genetic testing kits (as above). Participants seen by a genetic counselor will have their results disclosed to them by their care team. Participants who enroll remotely will be notified that their results are available via their preferred method (patient portal, mail, or e-mail). Those patients with a negative result will be provided the result via the preferred method and given the option to speak with a genetic counselor. Those patients with positive results will be provided their results over the phone by either the study or their clinical genetic counselors or physician. Patients tested at Ambry will have their results directly imported into their charts via HL7; release of results via MyPennMedicine (MPM) will only occur only after genetic counselor or physician has spoken with the patient (manual release). As part of this study, genetic testing is being done for variants in genes known to be associated with increased cancer risk. However, it is possible that the investigators do not know of all such genes, and the investigators are interested in studies that may help to identify them in the future. As an optional part of the study, the investigators will seek consent from participants to use their DNA sample to help understand the development of cancer and other diseases. Possible future research may include other research on inherited cancer risk or the link between DNA and medical outcomes, including response to treatment or contributions to other disease research. Some research may include whole genome sequencing, a method that determines the exact sequence of a person's DNA. Participants may decline participation in this aspect of the study and only have their samples used for the main study. Eligible participants will be identified through one of two mechanisms. Most will be identified through the electronic health record (EHR) system, which at Penn Medicine is Epic, and will be referred directly to the study team via an Epic registry. Participants will be screened by a study research coordinator to determine study eligibility, making sure they do not meet any of the study's exclusion criteria based on information contained in the EHR. Alternately, some eligible participants will be identified at the time that they see a genetic counselor at one of the Penn Medicine cancer genetics practices and recruited by their local genetic counselor with immediate randomization. These patients will include, for example, patients who need to be seen more quickly than would be picked up through the Epic registry (e.g., breast cancer participants pre-surgery). Patients who have an appointment already scheduled with cancer genetics at the time they are picked up by the Epic registry will be recruited through their genetic counselors. Participants who do not meet NCCN/ACMG guidelines for referral to a genetic counselor will have routine clinical care, which does not include a genetic counselor visit or genetic testing for cancer predisposition. Participants who do meet NCCN/ACMG guidelines will be referred to a genetic counselor at their treating institution. Genetic testing will be ordered based on the genetic counselor's clinical recommendation and patient preference. Participants on this arm who receive genetic testing will be tracked and results will be manually entered into the study database, although if they are tested at Ambry Genetics, positive results can be directly imported into PennChart (our local instance of Epic) via a secure interface. Participants randomized to the universal-testing arm will have genetic testing using a broad cancer risk gene panel through Ambry Genetics. Participants will have saliva samples collected. For low-risk patients, as well as for high-risk patients who decline Genetic Counselor (GC) visits, Research Coordinators (RCs) will mail a saliva kit, along with a prepaid mailer addressed to Ambry Genetics and a completed testing requisition form, to the participant. Orders will be placed in PennChart and will link directly with the Ambry Genetics application programming interface (API). This functionality already has been built and implemented in PennChart. For both arms, clinical reports from Ambry including variant interpretations are directly imported into PennChart. Participants seen by a GC will have their results returned by that care team. Participants consenting remotely will be told in advance that they will be notified by their preferred method (patient portal, mail, email) that their genetic results are available. Negative results will be provided by that method, with the option of a GC visit upon request. Participants with positive results pathogenic/likely pathogenic or variant of uncertain significance (P/LP or VUS) will be told that results are available and be returned by phone by the study genetic counselor or their clinical genetic counselor. VUS results will be returned to tested patients with appropriate caveats. Ascertainment of genetic risk among 1st-degree adult relatives: This portion of the study will explore differences in ascertainment of genetic risk among 1st-degree adult relatives of patients assigned to the universal-testing arm as compared with the phenotype-directed arm after a combination of referral by the proband and direct clinical and/or research team outreach. The investigators hypothesize that, by testing all young adult patients with solid tumors with a comprehensive germline cancer risk panel rather than following the standard guideline-directed approach to germline genetic testing of patients with cancer, the investigators will identify a higher proportion of their 1st-degree relatives as being at increased genetic cancer risk.