CASRAI Dictionary

Tag: vivli

  • Clinical Trial Data Sharing Platforms Compared

    Clinical trial data sharing platforms now fall into two structural models: sponsor-branded request processes with their own eligibility rules and review committees, and independent multi-sponsor infrastructure that hosts data from dozens of contributors under one governance framework. Pfizer and GSK each set their own scope, timelines and review criteria, but both now direct external researchers through Vivli, the non-profit platform that has become the shared backbone for pharmaceutical-sponsored data access.

    Vivli is a non-profit, independently governed platform that lets qualified researchers search, request and analyse de-identified patient-level data contributed by pharmaceutical companies, academic institutions and funders under a single, standardised data use agreement.

    What are the main clinical trial data sharing platforms?

    Three governance types dominate pharmaceutical clinical trial data sharing: sponsor-specific review processes layered on shared infrastructure (Pfizer, GSK), standalone multi-sponsor consortia (ClinicalStudyDataRequest.com, CSDR), and independent non-profit platforms (Vivli). Each imposes different eligibility criteria, review bodies and access windows on top of a broadly similar workflow: proposal submission, feasibility and scientific review, a signed data use agreement, then time-limited access to de-identified data in a secure environment.

    The table below summarises how the three named models compare on governance, review body and typical timeline, based on each organisation’s published data sharing commitments.

    Platform Governance model Review body Typical data availability
    Pfizer Sponsor-defined criteria, requests routed through Vivli Internal Pfizer committee (statisticians, data-sharing staff, programme leads) 18 months after primary study completion, for approved/terminated programmes
    GSK Sponsor-defined criteria, requests routed through Vivli since 2023 Independent external expert review panel Within 6 months of regulatory approval (US/EU) and publication of primary results
    Vivli Independent non-profit, multi-sponsor infrastructure Vivli administrative check, then sponsor feasibility review, then independent scientific review Varies by contributing sponsor; access typically granted for 12 months, extendable
    ClinicalStudyDataRequest.com (CSDR) Consortium of sponsors, separate from Vivli Independent review panel Varies by sponsor; 3,046 studies listed and 757 proposals submitted to date

    How does Pfizer’s data sharing process work?

    Pfizer submits its studies to the Vivli platform rather than operating a fully separate request portal. Researchers search Vivli’s catalogue, then submit a research proposal naming a qualified statistician and a defined analysis plan.

    Pfizer makes de-identified patient-level data available from global interventional studies of medicines, vaccines and devices that have been approved in the US and/or EU, or from programmes that were terminated during development. Data becomes available 18 months after the primary study completion date, and Pfizer’s own internal review committee — comprising the relevant programme lead, statisticians and data-sharing staff — assesses scientific validity before Vivli releases the dataset.

    • Scope: approved or terminated global interventional trials only
    • Review: internal Pfizer committee, feeding into Vivli’s independent scientific review
    • Anonymisation: identifiers replaced, dates shifted, sensitive fields redacted before release

    How does GSK’s data sharing process compare?

    GSK historically ran its own clinical study register and, for many years, used the separate ClinicalStudyDataRequest.com consortium. As of 2023, GSK directs all new patient-level data requests to Vivli, but its review criteria and timeline remain distinct from Pfizer’s.

    GSK makes data available once a medicine or vaccine has been approved by regulators and the primary study results have been published — a release window of within six months of both events. Requests are assessed by an independent panel of external experts rather than an internal sponsor committee, and GSK excludes some rare-disease studies where anonymisation would be technically difficult, along with data that predates electronic recordkeeping.

    • Scope: approved, published interventional trials; some rare-disease and pre-electronic-record exclusions
    • Review: independent external expert panel (not an internal GSK committee)
    • Output: raw and analysis-ready datasets, plus protocol and amendments, for approved requests

    What makes Vivli’s model structurally different?

    Vivli is not a sponsor — it is the shared, independently governed infrastructure that Pfizer, GSK and dozens of other contributors now use to process requests. Per the European Federation of Pharmaceutical Industries and Associations (EFPIA), Vivli’s catalogue spans more than 6,000 clinical studies from pharmaceutical companies, academic institutions and funders including Harvard University and the NIH HEAL Initiative.

    Every request passes through the same three-stage sequence regardless of which sponsor contributed the trial: a Vivli administrative completeness check, a sponsor feasibility review, and an independent scientific review panel. A single, standardised data use agreement then governs access, rather than each sponsor negotiating its own contract. This is the structural difference that matters for researchers comparing platforms: sponsor-run processes (Pfizer, GSK) still set the eligibility and review criteria, but Vivli supplies the common technical and legal infrastructure — search catalogue, secure analysis environment and template agreements — that makes cross-sponsor research feasible without a separate negotiation for every trial.

    Common questions about clinical trial data sharing platforms

    What is Vivli used for?

    Vivli is used to let qualified researchers request de-identified patient-level data from multiple pharmaceutical, academic and funder contributors through one standardised process. It replaces the need to negotiate separately with each trial sponsor, providing a shared catalogue, review workflow and secure analysis environment.

    How do I request Pfizer clinical trial data?

    Researchers submit a research proposal through the Vivli platform, naming a qualified statistician and a defined analysis plan. Pfizer’s internal review committee assesses scientific validity, and approved requests receive access to de-identified data 18 months after primary study completion, within Vivli’s secure environment.

    Does GSK still use ClinicalStudyDataRequest.com?

    No. As of 2023, GSK directs all new patient-level data requests to Vivli rather than ClinicalStudyDataRequest.com (CSDR), which it used for many years as a founding consortium member. Older GSK-sponsored studies may still appear in CSDR’s historical listings.

    What is a clinical trial data sharing agreement?

    A data use agreement (DUA) is the binding contract a researcher’s institution signs before receiving trial data. It sets permitted uses, publication obligations, data-security requirements and the access period — typically 12 months on Vivli, extendable on request.

    What this means for research administrators

    For institutional research offices, the practical implication is that due diligence now happens at two levels: the sponsor’s eligibility and review criteria, and the shared platform’s data use agreement and technical access terms. A proposal that meets Vivli’s administrative requirements can still be rejected at the sponsor feasibility stage if it falls outside Pfizer’s or GSK’s specific scope — so checking sponsor-level policy before drafting a proposal saves review cycles.

    The International Committee of Medical Journal Editors (ICMJE) has required a data sharing statement at trial registration since 1 January 2019, which has pushed sponsors toward standing infrastructure like Vivli rather than one-off, trial-by-trial arrangements. As more sponsors consolidate onto shared platforms, research administration teams negotiating data access agreements should expect fewer bespoke contracts but more variation in sponsor-specific eligibility rules layered on top of a common technical process. Institutions that track these sponsor-level differences — rather than assuming one data use agreement covers every contributor — will move proposals through review faster.

  • Vivli Explained: Clinical Trial Data Sharing

    Vivli is an independent, non-profit clinical trial data-sharing platform that acts as a neutral broker between organisations that hold participant-level trial data and researchers who want to reuse it. Rather than owning the data itself, Vivli operates the repository, search engine, and secure analysis environment, while two contractual instruments — the Data Use Agreement and the Data Contributor Agreement — govern who can access what, and under which conditions.

    Vivli is a global, cloud-based data-sharing and analytics platform for individual participant-level clinical trial data, operated as an independent non-profit since 2016. This article explains how the platform actually functions in practice: what the two governing agreements require, how a data request moves through review, and what it costs to analyse data once access is granted.

    What is Vivli?

    Vivli grew out of work at the Multi-Regional Clinical Trials (MRCT) Center of Brigham and Women’s Hospital and Harvard. In 2013, the MRCT Center convened a diverse group of global stakeholders to define and design a shared solution to the fragmentation of clinical trial data sharing, and the resulting non-profit platform launched in 2016.

    Vivli’s stated mission is to promote, coordinate, and facilitate the scientific reuse of clinical research data by building a sustainable global data-sharing enterprise, while explicitly positioning itself as a neutral broker rather than a data owner. That distinction matters: Vivli does not decide who gets access on commercial or institutional grounds — governance sits with the contributing organisation and independent review panels, not with Vivli’s own leadership.

    How does the Vivli platform work?

    The platform combines three technical components: an independent data repository, a search engine indexed to clinical trial metadata, and a secure cloud research environment. Researchers can run keyword or structured PICO (Population, Intervention, Comparator, Outcome) searches to identify relevant, listed studies before submitting a request.

    • Data repository — holds anonymised, individual participant-level datasets contributed by academic institutions, pharmaceutical companies, and funders.
    • Search engine — indexes study-level metadata so researchers can find eligible trials without needing prior access.
    • Secure research environment — a cloud workspace where approved researchers analyse data without downloading it to a local machine.

    The Vivli platform is built entirely within the Microsoft Azure cloud and has achieved SOC 2, Type 2 compliance following an independent audit, alongside a successful penetration test — infrastructure detail that matters to institutional data protection officers assessing whether to approve a data request through Vivli.

    What is the Vivli Data Use Agreement?

    The Data Use Agreement (DUA) is the contract every data requestor must execute before receiving access to a dataset on the Vivli platform. It is a mandatory, non-negotiable instrument: Vivli describes it as “the product of extensive negotiation” with the organisations that contribute data, meaning individual researchers cannot renegotiate its terms — they can only accept or decline.

    In practice, the DUA process runs alongside a three-step data request review:

    1. Administrative check — Vivli confirms the request form is complete and that a qualified statistician is named on the research team.
    2. Data contributor feasibility review — the organisation that holds the trial data assesses whether the request is feasible to fulfil.
    3. Independent Review Panel or Scientific Committee review — an external body assesses the scientific merit of the research proposal.

    Once a request clears review, Vivli issues the appropriate DUA version to the requestor via DocuSign, appends the approved Final Research Plan as “Exhibit A” once negotiations conclude, and countersigns on behalf of the platform. Researchers from a different institution than the lead investigator must each execute their own DUA before touching the data — a detail that trips up multi-site teams who assume one signature covers the whole consortium.

    What is the Vivli Data Contributor Agreement?

    Where the DUA governs the researcher requesting data, the Data Contributor Agreement (DCA) governs the organisation supplying it. Institutions and companies that want to make trial data available through Vivli sign a master DCA that sets the standing terms under which their studies can be listed, requested, and reused — independent of any single research request.

    This upstream/downstream split is easy to miss in generic clinical-research-governance writing, but it explains why the same university can appear on both sides of the platform: as an institution with a signed master DCA covering data it contributes, and separately as the employer of principal investigators who must still sign individual DUAs to request other organisations’ data. The University of California, San Francisco’s own library guidance to researchers illustrates this exact pattern — a signed institutional DCA does not substitute for a PI’s individual acknowledgement when requesting access.

    Agreement Who signs it What it governs
    Data Contributor Agreement (DCA) The organisation supplying trial data Standing terms for listing and releasing studies on the platform
    Data Use Agreement (DUA) Each individual data requestor/researcher Terms for accessing, analysing, and citing a specific dataset

    What does it cost to analyse data on Vivli?

    Requesting a dataset through Vivli carries no fee, but the secure research environment used to analyse it is metered by compute tier, with published 2025 pricing that scales from free introductory periods to specialist high-performance options.

    Research environment Free period Charge after free period Specification
    Standard 365 days $12/day 2 vCPU, 7GB RAM
    Premium (adds SAS) 90 days $25/day 4 vCPU, 14GB RAM
    Large — $65/day 16 vCPU, 64GB RAM
    Compute Optimised — $125/day 32 vCPU, 64GB RAM
    Memory Optimised — $275/day 64 vCPU, 128GB RAM
    GPU-enabled (small) — $50/day 8 vCPU, 56GB RAM, 1 Nvidia T4 GPU

    All environments support two concurrent logins, and there is no cap on the total number of researchers who can access an environment across the life of a project. Standard statistical software — R, Python, STATA, and Jupyter Notebook — is available in every tier; SAS is reserved for Premium and above.

    Common questions about Vivli

    Is Vivli free to use?

    Requesting and being granted access to a clinical trial dataset through Vivli carries no charge. The secure research environment used to analyse that data is free for an introductory period — 365 days on the Standard tier, 90 days on Premium — after which daily compute charges apply based on the tier selected.

    Who owns Vivli?

    No single company, government, or funder owns Vivli. It operates as an independent, non-profit organisation governed as a neutral broker between data contributors and data requestors, a structure that grew out of Harvard’s Multi-Regional Clinical Trials Center and was designed deliberately to avoid commercial or institutional gatekeeping.

    How do I request data on Vivli?

    Researchers search listed studies on the Vivli platform, submit a data request naming a qualified statistician, and pass a three-step review — administrative check, data contributor feasibility review, and independent scientific review — before executing a Data Use Agreement to receive the dataset.

    What is Vivli AMR?

    Vivli AMR is a dedicated instance of the platform, hosted at amr.vivli.org, built specifically for antimicrobial resistance surveillance data. It applies the same contributor/requestor agreement model as the main Vivli platform but to AMR-specific studies, reflecting demand for a distinct governance track for that data domain.

    Why this matters for research administrators

    For institutional research offices, the DUA/DCA split is the practical detail that most generic “data sharing” explainers skip. Signing a master Data Contributor Agreement is an institutional act, typically negotiated once by a research office or general counsel. Executing a Data Use Agreement is a per-request act tied to a named principal investigator and research plan, and it does not transfer automatically to co-investigators at other institutions.

    Research administrators supporting multi-site clinical research should treat these as two separate compliance tracks: one covering what the institution has agreed to contribute, and one covering what each individual investigator has been approved to receive. Budgeting for a data request should also account for research-environment compute charges beyond any free introductory period, particularly for large-scale reanalyses that outgrow the Standard or Premium tiers.

    As data-sharing mandates from funders and journals continue to tighten, platforms structured around explicit, separable contributor and user agreements — rather than a single blanket licence — are likely to remain the reference model for how third-party clinical trial data sharing is operationalised at scale.