Type to search every page, guide, role page, event and post.

Security & Compliance

Built for regulators, not retrofitted for them.

Trelice handles the documents that define how patients are treated. This page says what is in place today and what is not yet, so your quality and security reviewers do not have to guess.

Regulatory standards

21 CFR
Part 11
Architected to support Part 11: electronic signatures, audit trails, access control. Validation is in progress.
ICH
M11
Harmonized protocol template & terminology.
CPT
TransCelerate
Common protocol template v8.
CDISC
USDM
Unified Study Definitions Model: a machine-readable study definition produced from the approved protocol.

Certifications

SOC 2
On the roadmap
Not audited yet. There is no report today.
ISO
27001: on the roadmap
Not certified today.
Q&A
Security questionnaire
Send yours, or ask for our answers to the standard questions. Email security@trelice.com.
Practices

Security practices

Encryption

AES-256 at rest, TLS 1.3 in transit.

Access control

Role-based access across 14 roles, from platform administrator to read-only viewer.

Audit trail

Every change and approval is recorded in a permanent, append-only log that can be exported. Trelice is architected to support 21 CFR Part 11 requirements, including audit trails and electronic signatures. Validation is in progress.

Backup & continuity

Point-in-time recovery to any second in the last 35 days.

Vulnerability management

Continuous dependency scanning. 48-hour critical patch objective.

Data residency

US or EU hosting. No cross-region replication without written consent.

AI use

The core of Trelice is deterministic: protocol authoring, review, approval, amendments, consistency checks, and the documents computed from the structured data of the protocol run on the structure of the protocol, not on a language model. AI is optional and bounded. Trelice uses it to import an existing document, to learn a house format, and to draft consent form sections and their translations. Those drafts are labeled, and a person reviews each one before it counts. Every AI feature can be turned off. A team can also connect its own AI agents, whose work lands in Trelice as drafts and comments. A person approves every change, and an AI agent never approves.

Disclosure

Responsible disclosure

Found a vulnerability? Email security@trelice.com. We aim to acknowledge within one business day and resolve critical issues within 7 days.