☰

AndzonBio2

Turning academic science into IND-ready programs

Industry
Biotechnology
Headquarters
Paris, France
Public information as of
February 2026

A4BEE prepared this analysis from publicly available sources. It reflects our own reading of AndzonBio2's published strategy and is not endorsed by, or produced in cooperation with, AndzonBio2. Company website

Strategic priorities

AndzonBio2 emerged from stealth in October 2025 with a €3M seed round led by AdBio Partners and Inserm Transfert, structured as the second Pipeline Builder vehicle under the AndzonBio® alliance. The operating model is portfolio-based: identify high-potential European neuroinflammation projects, primarily through the Inserm Transfert network, and progress them as parallel programs rather than a single asset.

The stated target is IND-enabling studies by 2028, currently across two active programs — a Liver Fibrosis Small Molecule Entity and a Cardiac Fibrosis Antibody — with two further programs under evaluation. Achieving that target means moving data, processes and documentation out of the academic laboratory environment in which each project originates and into a regulatory-grade framework the FDA and EMA will accept.

Day-to-day R&D execution sits with an established network of CROs and scientific advisors, while the AndzonBio2 team in Paris operates the portfolio and the milestone calendar. The COO's framing of a "long-distance race" with a "final sprint" this year reflects a small team compressing an industrialisation step into a tight window.

The co-founder and acting CEO has described the company as "structured to efficiently turn breakthrough science into therapeutic solutions," and the founding investors are positioned to follow on. In 2026 the investor lens includes cyber resilience and data integrity, which makes the digital backbone that connects CROs and the central portfolio more than an internal IT question.

Challenges we see

  • Digital Integration

    Capturing academic data in an industrial format during tech transfer

    Projects sourced from diverse Inserm laboratories arrive with heterogeneous data formats, paper-based protocols and manual entry workflows. AndzonBio2's stated mandate is to "efficiently turn breakthrough science into therapeutic solutions," and the tech-transfer step is where that mandate either takes hold or doesn't.

    When each new program enters the portfolio with its own data conventions, the integration work scales with the portfolio count rather than staying fixed, and the time spent on each new asset becomes the bottleneck for adding the next one.

  • Operations Manufacturing

    Keeping portfolio-level visibility into work happening at CRO sites

    R&D execution is distributed across an established network of external CROs and expert consultants in multiple geographies. The COO has framed the year as a "final sprint" toward operational milestones, which makes the gap between portfolio oversight and laboratory execution directly visible.

    Where strategic oversight and laboratory data sit on different sides of a network boundary, portfolio decisions are made on retrospective reports rather than on what the experiment is producing today.

  • Compliance Regulatory

    Producing IND-grade documentation from academic-origin programs

    Academic laboratories typically do not operate under GAMP5 (Good Automated Manufacturing Practice) standards, yet the 2028 IND target requires complete audit trails and data integrity compliance for biological KPIs across all programs. AndzonBio2 has earmarked a substantial R&D budget per program for regulatory and clinical development.

    Where a program arrives without an audit-ready history, the documentation burden falls on the integration window rather than being collected by the running experiment, and the timeline to submission becomes a document-assembly problem.

  • Digital Integration

    Protecting IP that travels between CROs and the central team

    Core intellectual property — chemical structures of the Liver Fibrosis SME and sequences of the Cardiac Fibrosis Antibody — must transit between AndzonBio2 in Paris, the Inserm laboratory network and CRO partners. The COO has highlighted the priority of connecting with leading investors who, in 2026, weight cyber resilience and data integrity heavily in their diligence.

    Where high-value data moves between organisations without an agreed identity model and encryption baseline, the security posture of the chain is only as strong as the weakest link, which makes partner-by-partner security a portfolio-level concern.

  • Operations Manufacturing

    Handling high-dimensional bioprocess data from sequencing and antibody programs

    Programs such as the Cardiac Fibrosis Antibody generate complex datasets including single-nucleus RNA sequencing. The deep research record describes a "spaghetti code" of data integration where each new asset brings its own legacy systems and formats.

    Where analytical output from different programs lands in different structures, the platform work needed to make those structures comparable grows with each program added, and the portfolio's analytical capability moves more slowly than the science it is meant to support.

Opportunities, by urgency and business impact

Each bubble is one opportunity, numbered to match the list below. Further right means it bites sooner; higher means a bigger effect on the business. A bigger bubble means a bigger implementation effort.

Source: A4BEE analysis of public sources

  1. Standardising data capture during the tech transfer step

    Research projects arrive from Inserm laboratories with paper-based protocols, manual data entry and disconnected Excel files. The integration team must convert these into a format that supports both regulatory submission and ongoing program work.

    A digital lab integration framework that captures, validates and structures incoming academic data at the tech-transfer step — using OPC UA (Open Platform Communications Unified Architecture) instrument connectivity and protocol digitisation — establishes ALCOA+ (Attributable, Legible, Contemporaneous, Original, Accurate, plus Complete, Consistent, Enduring and Available) compliance from day one of the program.

    • AndzonBio2 launch announcement, October 2025
    • AndzonBio2 company website, About / Portfolio pages

  2. Reading portfolio progress against what CROs are producing today

    Strategic portfolio decisions are made on reports that arrive after the experiment is finished, while the running experiment is invisible to the central team. The COO's "final sprint" framing makes the operational visibility gap directly material.

    An ontology-based data platform that connects CRO instruments through OPC UA and surfaces KPIs (viable cell density, growth rates, antibody titres) on real-time dashboards gives the central team the same view of the program that the CRO scientists have.

    • AndzonBio2 leadership commentary on operational milestones, 2025-2026
    • AndzonBio2 portfolio and pipeline pages

  3. Reducing the integration cost of each new program

    Each new asset acquired from the European academic network triggers a manual integration step that scales linearly with portfolio growth. Moving from two programs to four or more without a repeatable pattern would compound the work.

    An MTP-enabled (Module Type Package, a NAMUR standard for vendor-agnostic equipment integration) integration procedure standardises the onboarding of new laboratory modules and protocols, so the same documented steps run once per program instead of being re-engineered.

    • AndzonBio2 strategy: 'searching, evaluating and integrating the most promising European projects'
    • A4BEE MTP library reference

  4. Building GAMP5-grade audit trails across the program set

    Academic-sourced programs do not arrive with the tamper-proof audit trails and validated data capture expected by FDA and EMA reviewers. The 2028 IND target sits on a fixed timeline, so compliance work that happens late in the cycle is the same work that risks the deadline.

    A validated lab system with audit trails, electronic signature and 21 CFR Part 11 (the US rule on electronic records and signatures) controls embedded into the data capture path produces the regulatory evidence as a by-product of the experiment.

    • AndzonBio2 stated 2028 IND target
    • AndzonBio2 budget allocation language for regulatory development

  5. Drafting and reviewing the IND-enabling document set with reviewer-checked AI

    IND-enabling submissions, periodic reviews and tech-transfer reports for two active and two evaluating programs draw on the same source records and the same templates. The document-assembly load scales linearly with program count, and most of the reviewer time goes on assembling and checking rather than on the technical judgement inside the document.

    Narrow AI agents can draft from source records, check a document against its template before the human review queue, and find every controlled document a standards change touches, with a named reviewer approving each output. The agents sit on top of the validated lab data layer rather than replacing it.

    • AndzonBio2 stated 2028 IND target and multi-program portfolio structure
    • AndzonBio2 R&D budget allocation for regulatory and clinical development

What we'd propose

  • Digital Lab

    Digital lab integration for academic-origin programs

    An end-to-end framework for capturing, validating and structuring research data during the tech-transfer step from academic laboratories to AndzonBio2's portfolio management, so each program enters with industrial-grade data and a regulator-readable history.

    • Instrument-level data capture

      Reading data off the equipment

      Connect laboratory instruments through OPC UA or MQTT so experimental data leaves the equipment in a documented, vendor-neutral form with method version and timestamp attached, instead of being read off a screen and re-typed.

    • Protocol digitisation

      Paper protocols as structured workflows

      Convert paper-based experimental protocols into structured digital workflows with built-in validation checkpoints and ALCOA+ markers, so a scientist's protocol becomes a queryable artefact rather than a notebook page.

    • Data validation engine

      Anomalies found before the report

      Run automated integrity checks on incoming research data — ranges, units, expected relationships — so anomalies surface at capture rather than at submission, and reproducibility questions are answered from the record.

    • Each program enters the portfolio with industrial-grade data rather than needing a retrofit.
    • The integration step runs once per program using a documented procedure instead of being re-engineered.
    • Reproducibility questions are answered from the record rather than reconstructed from memory.

    We've done this before

    See how we do Digital Lab
  • Enterprise AI

    Ontology-based data platform for portfolio intelligence

    A unified data architecture that aggregates research output from AndzonBio2's distributed programs and CRO partners into a single source of truth, with real-time visualisation and Golden Batch comparison across the portfolio.

    • Shared program ontology

      One agreed vocabulary across programs

      Define assay, specimen, batch, run, instrument and KPI as explicit entities with agreed relationships, so a query written once returns comparable answers across the Liver Fibrosis and Cardiac Fibrosis programs.

    • Real-time KPI dashboard

      Program progress as it happens

      Surface viable cell density, growth rates, antibody titres and other critical biological metrics against configurable thresholds, so the central team sees the same view of the program the CRO scientists see.

    • Golden Batch comparison

      Comparing the current run to the best one

      Overlay current run data against historical best-run profiles to identify drift and flag process parameters that are moving away from the proven path, with the comparison itself living in the model rather than in a spreadsheet.

    • Portfolio decisions are made on live program data rather than on retrospective reports.
    • Comparing two programs stops being a reconciliation exercise and becomes a model query.
    • The same platform absorbs new programs as they enter, without rebuilding the integration.

    We've done this before

    See how we do Enterprise AI
  • Digital CDMO

    MTP-enabled modular integration framework

    A standardised integration architecture based on the MTP (Module Type Package) standard that enables rapid onboarding of new laboratory modules and research protocols with minimal engineering overhead per program.

    • MTP library deployment

      Vendor-agnostic equipment integration

      Implement the MTP standard (a NAMUR working-group specification for vendor-neutral process module description) so laboratory modules from different vendors describe themselves in a common language, allowing them to be added to the platform without per-vendor integration work.

    • Plug-and-produce recipes

      Recipes composed by configuration

      Build manufacturing and experimental recipes through drag-and-drop composition rather than per-asset programming, so reconfiguring a workflow for a new program does not require custom code.

    • Repeatable integration procedure

      The same onboarding run, every time

      Document the asset-acquisition workflow as a step-by-step procedure with validated templates, so future European academic projects follow the same path and the integration cost per new program is known in advance.

    • Onboarding a new program becomes a known-cost exercise rather than a re-engineering one.
    • Equipment and protocols from different vendors coexist on the same platform without bespoke bridges.
    • The portfolio can scale beyond four programs without proportional growth in engineering overhead.

    We've done this before

    See how we do Digital CDMO
  • Digital Lab

    GAMP5 compliance infrastructure for the program set

    Regulatory-aligned lab infrastructure that establishes tamper-proof audit trails, validated data capture and documentation workflows required for IND submissions to FDA and EMA, embedded into the data capture path rather than retrofitted.

    • Audit trail system

      Complete data lineage

      Implement tamper-proof logging of every data modification with timestamp, user identification and change rationale, so the regulatory evidence chain is built by the running experiment rather than assembled at the end.

    • Validated workflows

      GxP-compliant data capture

      Design workflows that verify analyst training and instrument calibration status before allowing data collection, so a missing prerequisite fails the experiment rather than the audit.

    • 21 CFR Part 11 and EU Annex 11 alignment

      Electronic records that hold up

      Implement electronic signature, versioning and audit-trail handling to FDA 21 CFR Part 11 and EU Annex 11, the US and EU rules on electronic records and signatures, so the program record stands on its own during an inspection.

    • IND submissions carry a regulator-readable evidence chain from day one of the program.
    • Retroactive compliance work is replaced by evidence collected at capture.
    • Investor diligence on data integrity reads against the running platform rather than against a separate documentation exercise.

    We've done this before

    See how we do Digital Lab
  • Agents

    AI agents for IND-enabling and tech-transfer documents

    Narrow, reviewer-checked agents that take the repetitive part of document work for a portfolio of IND-bound programs: drafting IND sections, tech-transfer summaries and periodic reviews from source records, checking completeness against templates before human review, and finding every controlled document a standards change touches. A named scientist or regulatory reviewer approves each output.

    • Drafting from validated source records

      First drafts from program data

      Generate the first draft of an IND module section, a tech-transfer summary or a periodic review directly from the validated program records, so the author edits and judges rather than assembles.

    • Template and completeness checking

      Gaps found before review

      Check a submitted document against its regulatory template and the program's own checklist, returning missing or inconsistent sections before the document enters the human review queue.

    • Change impact search across the document set

      Which documents a change touches

      When a standard, method or specification changes, retrieve every controlled document across the program set that references it and rank them by how directly they are affected, so the update scope is known on day one.

    • Document review queues move faster because submissions arrive complete and against the right template.
    • The scope of a standards change is established by search across the program set rather than by recollection.
    • Every output is traceable to the validated records it came from and signed off by a named reviewer.

    We've done this before

    See how we do Agents

Digital maturity: today and target

Scored out of 100 across six dimensions. The target is what AndzonBio2's own published ambition implies — not a perfect score.

Source: A4BEE analysis of public sources
Data integration 25 → 80
Two active programs sit on academic-origin data structures with manual reconciliation between them. A portfolio-scale ontology would be ahead of where the company operates today, and the tech-transfer window is where it gets established.
Process automation 22 → 72
Lead and Drug Discovery stages today rely on expert-driven and project-managed coordination. The R&D budget allocated per program supports the automation layer once the integration procedure is in place.
Regulatory compliance 30 → 88
GAMP5 infrastructure is a 2028 IND prerequisite rather than an operational nicety. The audit-trail work is on the critical path of the timeline rather than running in parallel to it.
Cybersecurity 35 → 80
IP transit between Paris, the Inserm laboratory network and CRO partners is documented in the deep research as a porous digital perimeter. Investor diligence in 2026 reads cyber resilience as a portfolio-grade concern.
Real-time analytics 18 → 75
Single-nucleus RNA sequencing and bioprocess KPIs land in disconnected tools today. A shared dashboard model would change the analytical cadence from retrospective to in-program.
Modular architecture 22 → 78
Each new program is integrated ad hoc. An MTP-based pattern would let the same documented steps run once per program rather than per asset, which becomes material when the portfolio grows beyond four.

Check this yourself

Our Service Portal has free self-assessments and market comparisons. These are the ones that line up with what we've read above — no sales call required.

Think we've read this right?

Talk to us

Related reading

This is an independent analysis prepared by A4BEE from publicly available information as of February 2026. It reflects A4BEE's own interpretation and opinion, is not affiliated with, endorsed by, or verified with AndzonBio2, and may be incomplete or inaccurate. All company names and trademarks are the property of their respective owners. To request a correction or removal, contact [email protected].