Creation of a Single Institutional Review Board for Collaborative Research in Nephrology: The APOLLO Experience

After January 25, 2018, multicenter research applications submitted to the National Institutes of Health (NIH) were required to use a single Institutional Review Board (IRB). This affected academic institutions by centralizing IRB oversight to reduce redundancy. Although single IRBs were common in industry-sponsored research and the United Kingdom,^1,2 academic IRBs in the United States were reluctant to cede responsibility for NIH studies. This article describes the development, operation, and onboarding timelines of the NIH “APOL1 Long-term Kidney Transplantation Outcomes” (APOLLO) single IRB assessing apolipoprotein L1 gene effects on transplantation.³ Lessons may be useful for streamlining the launch of multi-institutional studies.

In 1983, protection of human participants in research was codified under 45 Code of Federal Regulations 46, and IRBs were developed to evaluate single-center investigator-initiated projects.⁴ Over time, studies grew to thousands of participants from hundreds of sites.⁵ The early IRB model where each institution in multisite studies repeatedly reviewed identical protocols delayed start-up and caused difficulty controlling versions of documents.⁶ Sites operated with different versions of protocols/consent forms and froze enrollment until study-wide approvals were granted.⁷ Starting-up multiple sites was challenging due to differing IRB processes.^6,8

Industry sponsors had long used centralized, often unaffiliated commercial IRBs in multicenter studies.² Independent central IRBs often claimed rapid approval for site initiation, unencumbered by the time and burden of satisfying local institutional and state requirements. Although industry sponsors used independent IRBs, federal awards avoided them because of funding limitations. This changed in 2018 when the NIH introduced a policy requiring single IRBs in multicenter studies.⁹ Similar policies were extended to all federally funded studies through revisions to the Common Rule in 2020.⁴ Institutions and principal investigators now faced novel challenges and expectations with limited resources.

In September 2017, APOLLO was one of the first large-scale federally funded consortia required to apply the single IRB mandate.³ The Coordinating Center and single IRB are at Wake Forest University School of Medicine (WFU). The WFU single IRB oversees APOLLO and its ancillary studies, interacting with the Steering Committee comprised of the NIH, Coordinating Center, Study Chair, 13 Clinical Center principal investigators, Community Advisory Council leadership, United Network for Organ Sharing, and Association of Organ Procurement Organizations (OPOs). The single IRB also interacts with 100 engaged-in-research transplant programs, 46 nonengaged-in-research programs with patients recruited by APOLLO Clinical Centers, 57 OPOs, and 64 OPO-aligned HLA laboratories.

Before drafting study documents, Coordinating Center leadership met with the WFU IRB Director to plan for managing 146 recruitment sites. The Coordinating Center identified one lead study manager for single IRB processing, and it was decided that engaged transplant programs would complete the single IRB process in waves leading to “rolling-start-recruitment.” Similarly, the Coordinating Center worked with OPOs and HLA laboratories to permit deceased donor bio sample collection on a rolling basis.

Before the mandate, WFU limited its role as single IRB of record. First, negotiating and executing institutional reliance agreements required extensive revision and discussion regarding delegation of responsibilities, handling/communicating protocol deviations, and providing updates. The SMART IRB master reliance agreement (https://smartirb.org/agreement/) alleviated these issues by streamlining the process using a single master document, rather than institution-specific reliance agreements for each site. Second, our internal IRB system was not designed to support multiple external teams on one protocol. A technical solution was necessary to manage site communications and document reliance decisions and local considerations. Finally, our Human Research Protections Program (HRPP) was not staffed to support uploads of materials from large numbers of reliant sites. To alleviate these barriers, APOLLO used the IRB Reliance Exchange (IREx), a freely available web-based portal supporting single IRB review documentation and coordinating for multicenter clinical trials (https://urldefense.com/v3/__http://www.irbexchange.org__;!!GA8Xfdg!yUA7R-sU2W4jONG1vnBpRvTkKGVmLyGREj2Jw7R4snBTya0BgWw8rIlwwxnFTScGoidYO741RPSdFUsaNCk$).

The APOLLO single IRB required systematically capturing documentation from 100 reliant sites and efficiently disseminating single IRB approvals and documents. Challenges in implementing a single IRB are coordinating communications and capturing and managing bidirectional information exchange between study teams and their HRPPs/administrators.¹⁰ To satisfy “local considerations” (including institutional policies and state laws), the respective IRBs at each site were given the opportunity to acknowledge the study and sign off on its conduct, before the single IRB approval. The IREx platform has been used by >40 single IRBs.¹⁰ IREx maintains records of reliance decisions, collects local considerations from sites, controls versions of historical documents, and notifies researchers when new versions are approved. The IREx platform permits delegation of single IRB operational responsibilities to the Coordinating Center's “IREx Study Manager” who uses IREx to manage site communications, track progress for completing required agreements and documentation (local considerations and reliance decisions), store site approvals, and notify site HRPPs and study teams of new approvals (initial, continuing review, site amendments, and study modifications).

Figure 1 presents the process beginning when the lead site has been submitted to the single IRB. APOLLO sites were provided draft protocols and consent templates in July 2018; these were approved by the single IRB on January 17, 2019. Owing to revisions during the IRB review process, an updated protocol, manual of procedures, and consent form were redistributed, and some sites had to restart their local considerations process. Therefore, APOLLO strongly suggests that large studies approve their lead sites before onboarding other sites.

Figure 1.: The process for single IRB approval in the APOLLO Consortium. After lead site approval, study materials were distributed to participating sites to begin their HRPP review processes, starting with submission to their local IRB as required in the SMART IRB Agreement for institutions to communicate local considerations to the single IRB. Local considerations and “cede decisions” were documented in IRB Reliance Exchange. The IRB Reliance Exchange Study Manager prescreened information before single IRB submission, often resulting in changes by the local HRPP and reducing time to approval. The IRB Reliance Exchange Study Manager submitted sites to the APOLLO single IRB system in batches. Although participating sites may have different workflows for documenting reliance and local considerations, IRB Reliance Exchange milestones are consistent across all studies and sites. The standard time points used for descriptive analysis included dates when sites were contacted by the single IRB to begin their local processes, when local considerations assessment began and ended, and when submitted to and approved by the single IRB. APOLLO, APOL1 Long-term Kidney Transplantation Outcomes; HRPP, Human Research Protections Program; IRB, Institutional Review Board; IREx, IRB Reliance Exchange.

Owing to staggered start-up times for onboarding, APOLLO assessed differences in activation times for early sites versus later sites. The median time from protocol/consent form single IRB approval (January 17, 2019) or the date sites were sent materials (whichever was later) to completion and documentation of local considerations (steps 2–4 in the Figure 1) was 201 days (min=0/max=1020). Local investigators and coordinators were responsible for initiating single IRB processes at their institutions, including submission before the local HRPP could initiate and document reviews (steps 4–5 in Figure 1). Among the subset of 35 programs capturing these data, the median time for local review was 25 days (min=0/max=211). In all 100 APOLLO engaged sites, the median time from completion of local considerations documentation in IREx to single IRB submission for review (step 5 in Figure 1) was 7 days (min=0/max=144) and from IRB submission of a site to approval 5 days (min=0/max=17). Single IRB staff and reviewers only controlled the timeline from local consideration completion to IRB approval (step 5 in Figure 1), a median of 14 days (min=1/max=146). Most sites were submitted to the single IRB within 14 days of completing local considerations and received IRB approval within 7 days.

The first APOLLO site was approved on May 13, 2019, and the last was approximately 3 years later. The time required by the WFU Coordinating Center and single IRB staff/reviewers remained consistent. Time from completion of local considerations to single IRB review and approval averaged 12 days for the first 10 sites (in 2019) versus 8 days for the last 10 sites (in 2020).

The APOLLO single IRB made participant information consistent and promoted efficient study start-up. To ease implementation, the standardized consent form template had limited areas for editing. Although sites requested preferred institutional language, format changes outside areas designated for editing were prohibited. The single IRB collected and respected local considerations and differences in state laws, especially differences in Health Insurance Portability and Accountability Act authorizations and laws governing return of genetic results. Critical to APOLLO, approval of materials in one submission by the single IRB allowed 100 sites to simultaneously incorporate the latest version of the protocol and consent form.

Because study teams often did not know how to navigate the review process, a median of 201 days was required for sites to complete local review and be ready for single IRB review. This is consistent with 2019 IREx data showing the median time from Lead Site Approval to Reliant Site single IRB submission approximating 180 days, decreasing to <100 days in 2021. In comparison, APOLLO had longer time between local considerations completion and submission to the IRB, but shorter times to single IRB approval postsubmission. The Coordinating Center IREx Study Manager prereviewed single IRB materials to verify accuracy and completeness. Preliminary review allowed more efficient IRB approval processes postsubmission and contributed to the WFU single IRB having shorter approval times than other IREx studies. Time from completion of local considerations to IRB approval remained similar between early and late participating sites.

APOLLO includes the collection and biobanking of DNA, serum, and urine from deceased donors and participants at engaged transplant programs; however, the single IRB has had the greatest effect on APOL1 genotyping and planning for eventual return of APOL1 genotypes to participants and deceased donor next-of-kin. To that end, the single IRB mandated that genotyping must be performed in a Clinical Laboratory Improvement Amendments laboratory to permit return of results to participants in all 50 states (despite variable local laws). In addition, the NIH will hold a conference devoted to the return of genetic data to study participants in which APOLLO and J. Brian Moore will participate.

In conclusion, WFU successfully implemented a single IRB model for APOLLO, consistent and compliant with its determinations, and maintained reasonable timelines for approval. Communication and collaboration between multiple groups was pivotal. We recommend identifying an IREx Study Manager with sufficient effort dedicated to the study. Without the coordination, support, and resources available from researchers, staff, and collaborators, these processes would have been more burdensome, onerous, and time-consuming. Researchers developing multicenter studies should plan for budgeting, staffing, and coordinating closely with single IRB representatives.

Disclosures

A.A. Alexander reports an advisory or leadership role for Boston Children's Hospital. B.I. Freedman reports employment with Wake Forest University School of Medicine and Health Systems Management, Inc.; consultancy for AstraZeneca Pharmaceuticals, RenalytixAI, and Xinthera; research funding from AstraZeneca Pharmaceuticals and RenalytixAI; advisory or leadership roles on Editorial Boards of American Journal of Nephrology, JASN, and Kidney International; and other interests or relationships as Chief Medical Officer of Health Systems Management, Inc., Wake Forest University Health Sciences. B.I. Freedman has rights to a US patent related to APOL1 genetic testing. M.H. Spainhour reports employment with and research funding from Atrium Health Wake Forest Baptist. All remaining authors have nothing to disclose.