COPDGene is an ongoing observational study of current and former smokers with and without COPD, including a smaller number (n = 454) of non-smoking controls; the total number of enrolled participants across 21 clinical centers in the United States was 10,71813. By design, two-thirds of COPDGene study participants were non-Hispanic White, and one-third were non-Hispanic Black. RNA-seq has been obtained on peripheral blood samples from the five-year follow-up visit, and microbial transcripts were assessed on 3,984 COPDGene participants.

COPDGene study participants provided written informed consent for the study protocol, which was approved at 21 clinical centers. Return of the HCV results and the HCV Follow-Up Study were approved by the 17 clinical center IRBs with HCV subjects identified (Baylor College of Medicine, Brigham and Women’s Hospital, Columbia University Medical Center, Reliant Medical Group, Lundquist Institute for Biomedical Innovation at Harbor-UCLA Medical Center, Minnesota Health Partners, Johns Hopkins University Medical Center, Morehouse School of Medicine, VA Ann Arbor Healthcare System, National Jewish Health, University of Pittsburgh, Temple University, University of Texas Health Sciences Center at San Antonio, University of Alabama at Birmingham, University of Iowa, University of Michigan, and University of California San Diego). Since there were no subjects with research evidence of HCV at the other four COPDGene clinical centers, IRB approval to return HCV results was not sought from those clinical centers.

Prior return of results in COPDGene has included alpha-1 antitrypsin deficiency (a well-documented autosomal recessive genetic risk factor for COPD and liver disease), spirometry results, chest CT findings, complete blood count abnormalities, and abnormal HADS (Hospital Anxiety and Depression Scale questionnaire) scores; these items were specifically included in the COPDGene informed consent form. For alpha-1 antitrypsin deficiency, the COPDGene consent form indicated that alpha-1 antitrypsin analysis would be performed in a research rather than a clinical laboratory, and that any reported research results would need to be confirmed through clinical testing. Participants were then asked: “Would you like to be informed about any abnormal alpha-1 antitrypsin results?” The COPDGene enrollment visit (2008-2012) template consent form also included the following text regarding other genetic results: “The test results from this study are not known to have any clinical significance at this time, and we will not tell you or any other individual about your specific genetic results.” No specific language was included in the COPDGene consent form regarding return of gene expression results. However, the 5-year follow-up visit COPDGene consent form (when the blood samples for RNA-seq that generated the viral infection results were collected) specified broad use of COPDGene blood samples: “Research using your samples and whole genome information is important for the study of virtually all diseases and conditions. Therefore, the data repositories will provide study data for researchers working on lung and other diseases.” However, the consent form did not include any language regarding return of these results to participants.

COPDGene has maintained contact with as many study participants as possible through semi-annual Longitudinal Follow-Up Program contacts (by e-mail, automated phone call, or coordinator phone call)25.

In response to the incidental identification of HCV results in COPDGene, a committee of COPDGene investigators and experts in bioethics and infectious diseases was established to determine whether to return HCV results and, if so, to develop a process for returning these results to COPDGene participants (committee members were co-authors EKS, AYK, BJM, EAR, JDM, JO, and PO). A group of nine COPDGene study participants was enlisted to join a COPDGene Participant Advisory Panel, and the procedures for returning HCV results were reviewed with these volunteers. These study participants provided their perspectives regarding whether HCV results should be provided to COPDGene subjects.

As recently reported14, 228 COPDGene participants with suggestive evidence of HCV (defined as “HCV results”) were identified by applying PathSeq26 to blood RNA-seq reads that were not mapped to the human genome. As shown in Table 1, we developed a procedure for returning HCV results, which separated the clinically important notification about potential HCV infection from a follow-up research study regarding the downstream effects of the return of results. This return of results procedure involved an initial contact letter (sent by Certified Mail) that specifically notified participants about their HCV research testing. This was in contrast to the commonly used procedure of providing a general contact letter that indicated an abnormal result had been found in their research samples and then giving them the option to learn about the specific finding. This decision was based on the individual and public health importance of HCV infection, as well as guidance from the COPDGene Participant Advisory Panel. We also included a phone contact by the participant’s COPDGene clinical center after the initial mailing (with a phone script) to explain the HCV results and assess their interest in receiving information regarding the HCV Follow-up Study. Participants were referred to their healthcare providers to determine if clinical HCV testing was appropriate. If participants needed a healthcare provider, the COPDGene clinical centers assisted in finding one. Although the phone script for this first call did not explicitly ask about previous diagnosis of HCV, many of these participants volunteered that they had previously been diagnosed with HCV, and this information was recorded.

A second mailing and phone call included the consent for the HCV Follow-up Study that involved notification of their primary care physicians, participation in a phone questionnaire approximately three months after their initial notification about HCV, and access to medical records relevant to HCV. The medical record review included prior HCV testing, HCV treatment, liver imaging, liver function testing, and evidence of liver disease.

The COPDGene Participant Advisory Panel endorsed this approach for returning HCV results to COPDGene participants. In addition, the COPDGene OSMB and IRBs for the 17 COPDGene clinical centers with at least one HCV-positive participant approved the return of results procedures. We elected to notify only participants with evidence for HCV infection (since the significance of a negative result is uncertain). We also decided to notify only surviving participants with evidence of HCV, and not to contact relatives of deceased participants.

The HCV Follow-Up Study Questionnaire (Supplemental Materials) was developed to capture information regarding previous HCV diagnosis and treatment as well as the reactions of COPDGene participants to receiving HCV results. The HCV Follow-Up Study Questionnaire was IRB-approved at all relevant clinical centers, and written informed consent was obtained for participation in the HCV Follow-Up Study. Interested participants were asked to respond to this questionnaire by phone interview three months after receiving their initial contact letter regarding HCV results. Differences in reactions to receiving HCV results based on whether or not there was a previous HCV diagnosis were assessed with two-sided Fisher’s Exact Tests using SAS version 9.4.