Quality Management
GAMP 5 Principles. FDA Requirements. Imaging-Specific Evidence.
Clinical trial imaging is only as dependable as the technology systems behind it. Those systems receive images, protect blinded reviews, record measurements and deliver endpoint data. Software validation provides documented evidence that they perform as intended, every time. For sponsors, that means confidence in the imaging assessment itself. It also means confidence in the electronic record that supports it.
Two frameworks guide this work. GAMP 5 is the internationally used good-practice framework for controlling a computerized system across its lifecycle. FDA requirements set the legal obligations for regulated records and electronic signatures. Together, they connect intended use, risk-based testing, data integrity and ongoing control. Validation is not a single test before go-live. It is a discipline that lasts for the life of the system.
Intrinsic Imaging builds GAMP 5 principles and FDA-aligned validation activities into our imaging core lab quality framework. We connect system requirements to verification evidence, study-specific configuration and controlled change. The advantage for your trial is simple: validation grounded in the actual imaging workflow. Your endpoint data stays protected. Your sponsor qualification is supported. And the evidence behind system reliability is ready for audit review.
GAMP stands for Good Automated Manufacturing Practice. GAMP 5: A Risk-Based Approach to Compliant GxP Computerized Systems is published by the International Society for Pharmaceutical Engineering (ISPE). It is good-practice guidance for systems used in regulated life-science work. GxP refers to the relevant good-practice disciplines, including Good Clinical Practice. Despite its manufacturing origins, the framework covers computerized systems across all regulated activities. That includes clinical trial workflows.
The Second Edition emphasizes critical thinking. Understand what a system must do. Identify where a failure could harm participants or compromise data. Then use proportionate evidence to show control. Validation therefore covers the configured system, its users, its interfaces and its operating procedures — not just the supplier’s software product. GAMP 5 is guidance, not a regulation or a certification.
Define the imaging process, the regulated records and the functions that support it. Requirements provide the basis for acceptance criteria and verification.
Focus testing on functions whose failure could affect participant protection, blinding or endpoint integrity. Scale the evidence to the system’s complexity and risk rather than treating every function identically.
Assess supplier capability and relevant documentation, then address the additional verification needed for the study’s configuration, interfaces and workflow. Supplier evidence supports validation; it does not eliminate responsibility for intended use.
Maintain the validated state through approved release, trained users, change assessment, incident handling, periodic review and controlled record retention or retirement.
FDA’s requirements depend on the regulated activity and the records a system maintains. The underlying clinical investigation regulations — often called predicate rules — define which records must be created and retained. When those records are kept electronically, 21 CFR Part 11 may apply. It also applies when electronic signatures replace required handwritten signatures.
For closed systems within its scope, Part 11 §11.10(a) requires validation. The goal is accuracy, reliability, consistent intended performance and the ability to discern invalid or altered records. Other provisions cover record copies and retention, authorized access, audit trails, operational checks, authority checks, training and documentation control. Open systems and electronic signatures carry additional controls.
FDA’s October 2024 clinical electronic-systems guidance recommends a risk-based approach to validation. It weighs intended use and the potential impact on participant protection and trial results. FDA’s 2003 Part 11 guidance describes a narrower scope and enforcement discretion for certain provisions. It does not remove the underlying clinical-record obligations. Our approach considers both the applicable regulations and the clinical guidance. We do not treat every software system as subject to identical requirements.
Establish requirements and acceptance criteria, challenge critical functions and retain evidence showing that the configured system produces reliable results—including under invalid-input and failure conditions.
Control access, preserve accurate and complete records, and maintain applicable time-stamped audit trails without obscuring earlier information. Where electronic signatures are used, preserve identity, meaning and linkage to the signed record.
Use trained personnel, controlled procedures and documentation, and evaluate changes that could affect validated functions. Verification addresses the impact of a change before the revised system is released for use.
Retain validation documentation and protect clinical records for their applicable retention period. Support retrieval and accurate, complete copies so sponsors and inspectors can review the record and its relevant context.
The map below summarizes key provisions, not every obligation in Part 11. Applicability and FDA’s enforcement approach must be assessed alongside predicate rules and clinical guidance. Technical controls and documented operating practices work together. A software feature list alone is not a compliance determination.
| Requirement and Reference | What It Means in Practice | Supporting Evidence |
|---|---|---|
| Validated Intended Performance§11.10(a) | Show accuracy, reliability and consistent intended performance. The system must be able to discern invalid or altered records. | Requirements, risk assessment, verification results and assessed deviations. |
| Complete Copies and Protected Records§11.10(b)–(c) | Generate accurate, complete copies in human-readable and electronic form. Protect records for accurate and ready retrieval throughout retention. | Export reconciliation, retrieval tests and record-retention arrangements. |
| Access, Sequencing and Authority§11.10(d), (f)–(h) | Limit access to authorized people. Apply appropriate operational, authority and device checks to the workflow. | Role permissions, blinded-visibility tests and challenges to invalid workflow sequences. |
| Time-Stamped Audit Trails§11.10(e) | Use secure, computer-generated histories for record creation, modification and deletion. Preserve earlier information and retain the history with the record. | Attribution, chronology, change-history retrieval and audit-trail review arrangements. |
| People and Controlled Documentation§11.10(i)–(k) | Ensure appropriate education, training and experience. Establish accountability for electronic signatures and control system documentation. | Training records, applicable procedures and documentation revision history. |
| Open Systems and Electronic Signatures§11.30; §§11.50, 11.70, 11.100, 11.200, 11.300 | For open systems, add measures needed to protect authenticity, integrity and confidentiality. Where signatures are used, address identity, signature display, record linkage and identification-code/password controls. | System classification, signature-identity verification, record-linkage tests and applicable security controls. |
Computerized-system validation (CSV) establishes fitness for intended use through objective evidence. Computer software assurance (CSA) emphasizes critical thinking and testing proportionate to risk. FDA’s February 2026 CSA guidance addresses medical-device production and quality management system software. It does not replace the October 2024 guidance for clinical investigation systems. Our approach applies the relevant framework to the regulated activity. We do not treat CSA as a blanket exemption from validation.
Choosing an imaging CRO means choosing the systems and controls behind your study evidence. Intrinsic Imaging connects software validation with deep imaging expertise, three FDA inspections with zero Form 483 observations and five ISO certifications. These strengths complement each other. Inspection history and certifications provide quality context. Intended-use validation addresses the reliability of the particular system and study workflow.
Verification follows the imaging record from receipt through assessment and delivery. Checks on identifiers, measurements, review sequencing and transfers help address errors that could compromise the evidence supporting your trial.
Requirements, risk decisions, traceability and release evidence give sponsors a structured basis for system qualification and audit discussions—not simply an assurance that a platform has been tested.
Study-specific configuration and change assessment keep validation connected to actual use. Security, privacy and continuity disciplines support reliable imaging operations throughout the study and record-retention lifecycle.
Validation is not a one-time test or a vendor checklist. Our approach connects each important requirement to its risk, its verification evidence and its operational controls. Documentation stays proportionate to the system’s complexity and the consequences of failure.
Identify the imaging workflow, users, regulated records and interfaces. User requirements define what the system must do. This includes study-specific configurations and acceptance criteria.
Intended-use statement · User requirements
Evaluate how a failure could affect participant protection, blinding, image interpretation or endpoint integrity. Assess supplier evidence. Focus verification on the functions that matter most.
Risk assessment · Supplier assessment
Test the actual workflow and configuration, not just the software name. Challenge permissions, calculations, data transfers and failure handling. Match test rigor to risk.
Test evidence · Requirements traceability
Review results against acceptance criteria. Assess deviations. Document the release decision. Approved procedures and user training support controlled use in the study.
Validation summary · Release authorization
Assess changes before implementation. Repeat relevant verification when risk warrants it. Periodic review, incident handling, backup recovery and controlled retirement protect the records throughout their lifecycle.
Change control · Review and retention records
Traceable System Evidence
Validation connects what a system must do with objective evidence that it performs as intended. Risk assessment directs the depth of verification; controlled change preserves that evidence over time.
Define the imaging process, critical functions and acceptance criteria.
Prioritize participant protection, blinding and data integrity.
Establish the study settings, access roles and approved baseline.
Link functional tests and end-to-end workflow checks to the requirements they verify.
Assess results and deviations against acceptance criteria before release.
Training, change assessment and periodic review keep evidence connected to use.
We translate regulatory expectations into imaging-specific verification objectives. The scope reflects the system’s intended use and applicable records. Where relevant, it includes added protections for open systems and electronic signatures.
| Control Area | Validation Focus | Verification Objective |
|---|---|---|
| Access and Blinding | Unique user identities, role-based permissions and separation of reader, adjudicator and administrative activities. | Challenge unauthorized access and study-specific visibility restrictions. |
| Image and Record Integrity | Preserve the relationship between participant identifiers, visits, images, assessments and released results. | Reconcile receipt, processing and export; test incomplete, duplicate or incorrectly mapped records. |
| Audit Trails | Secure, computer-generated, time-stamped histories for applicable record creation, modification and deletion, without obscuring prior information. | Verify attribution, chronology, reviewability and retention of relevant changes. |
| Electronic Signatures | Where used, link the signature to its record and preserve the signer’s name, date/time and meaning of the signature. | Test identity controls, signature manifestation and record linkage. |
| Measurements and Workflow | Defined measurement rules, review sequencing, adjudication logic and study-specific configuration. | Challenge expected results, boundary conditions, invalid inputs and workflow exceptions. |
| Interfaces and Data Delivery | Controlled mappings and transfers that preserve record content and meaning between systems. | Check completeness, units, precision, identifiers and handling of interrupted or failed transfers. |
| Recovery and Retention | Protected records, appropriate retention, recoverable backups and accurate, complete copies for review and inspection. | Exercise restoration, retrieval and export, including associated metadata and audit trails where applicable. |
These example scenarios show how a risk-based approach addresses imaging-specific failure modes. The actual test scope and acceptance criteria are defined for each system and study.
Challenge: An image arrives with a missing, duplicate or inconsistent participant or visit identifier.
Verification: Challenge the mapping and reconciliation rules, confirm the exception is visible and verify that a correction preserves traceability.
Sponsor Value: Protects the relationship between the image, the assessment and the intended endpoint time point.
Challenge: A user attempts to view restricted treatment information, another reader’s assessment or an adjudication task outside their role.
Verification: Test each study role against permitted and prohibited actions, including visibility after reassignment or a role change.
Sponsor Value: Protects independent review and the integrity of the study’s read design.
Challenge: A configuration change affects a calculation, measurement unit, precision rule or data-transfer mapping.
Verification: Compare expected results with the configured output, test boundary conditions and reconcile the released dataset with the source assessment.
Sponsor Value: Helps prevent a technically successful transfer from delivering clinically misleading data.
Challenge: An upload, downstream transfer or restoration process stops before completion.
Verification: Challenge failure handling, identify incomplete or duplicate records and verify recovery preserves the required content and associated history.
Sponsor Value: Supports continuity without silently losing, duplicating or changing study records.
ALCOA+ is a widely used way to describe data-integrity principles: attributable, legible, contemporaneous, original and accurate, plus complete, consistent, enduring and available. It is not a separate certification. In imaging, these principles connect four things: the source image, its metadata, the reader’s assessment and the delivered endpoint record.
Connect an assessment or change to the responsible individual and its recorded time. Shared identities and retrospective reconstruction weaken accountability.
Keep images, assessments and essential metadata readable and preserve the original record or a verified true copy. Verify calculations, units and transformations against expected results.
Retain relevant changes and the context needed to interpret the record. Reconcile identifiers, visits, assessment versions and data transfers across the workflow.
Protect records for their applicable retention period and make them retrievable for authorized review. Recovery and export must preserve meaning, not just produce a readable file.
A hosting provider’s security controls can support qualification. So can a software supplier’s testing. But neither establishes that a study’s configured imaging workflow is fit for use. Assessment considers the service, the interfaces, the records and the responsibilities assigned to each organization.
For cloud-hosted or supplier-managed systems, our validation approach covers access administration, configuration control, supplier changes, incident handling, backup and recovery, record location and retention, and retrieval if the service changes or ends. Relevant supplier evidence is evaluated alongside the verification needed for the imaging process.
Assess supplier evidence, intended use, configuration and interfaces. Define acceptance criteria for the functions that protect study data.
Evaluate supplier updates and study-setting changes for impact. Use relevant verification and regression testing to support the release decision.
Plan for complete records, essential metadata and applicable audit trails to remain accessible throughout the required retention period.
A defensible validation package answers three questions: what was assessed, why the evidence is sufficient, and how the system remains controlled. Sponsor discussions can address the relevant system scope, risk decisions and documentation needed for qualification and audit review.
The software running a clinical trial workflow and the investigational imaging product answer different validation questions. A validated reading platform does not, by itself, establish an AI algorithm’s clinical performance. Nor does it authorize a medical device.
Intrinsic Imaging also supports independent AI/CAD clinical validation studies and medical device imaging trials. These activities evaluate the product’s intended clinical task. Meanwhile, controlled systems protect the evidence generated during the study.
Our FDA inspection history and five ISO certifications provide further context for our quality framework. Neither is a blanket FDA approval of software.
Discuss system qualification, study-specific controls and validation evidence with Intrinsic Imaging.