Definition
RUO vs GMP cells: what actually differs
RUO cells are supplied for research use only with no regulatory manufacturing standard behind them; GMP cells are manufactured under a quality system for clinical use. Comparison of documentation, testing, price, permitted use, and when RUO is sufficient.
RUO (research use only) cells are supplied for laboratory research with no regulatory manufacturing standard behind them and an explicit prohibition on human use. GMP cells are manufactured under a documented quality management system intended to support use in humans, with the traceability, testing and change control that implies.
The difference is not primarily about the cells. It is about the paperwork, the facility, the change control and the liability — and those are the things you are paying for.
The core distinction
An RUO product is characterised. A GMP product is manufactured under a system that establishes, in advance and in writing, how it will be made, how deviations are handled, who is qualified to make it, and how any lot can be traced back through every input to its origin.
You can have two vials of the same cell type, from the same donor, differentiated by the same protocol, where one is RUO and one is GMP. The cells may be indistinguishable at the bench. The GMP vial costs considerably more because it carries a manufacturing record, qualified personnel, a qualified facility, validated methods, controlled raw materials, and a person whose job is to release or reject the lot against predetermined criteria.
RUO is a use designation carrying restrictions. GMP is a manufacturing standard carrying obligations. They are not two points on one quality scale, and it is a mistake to read RUO as “lower quality” — a well-made RUO product can be more thoroughly characterised for a research purpose than a GMP product is, because characterisation and compliance are different axes.
Comparison
| Dimension | RUO | GMP |
|---|---|---|
| Primary document | Certificate of analysis for the lot | Batch manufacturing record, plus CoA, plus a quality agreement with the buyer |
| Manufacturing standard | Vendor’s internal process. Many suppliers hold ISO 9001, which certifies a quality management system, not pharmaceutical manufacture | Documented GMP quality system, qualified facility (cleanroom classification typically stated, for example ISO 7 background with ISO 5 critical zones), qualified equipment, trained and qualified operators |
| Traceability | Lot number, usually donor identifier | Full chain of custody from donor consent through every raw material and process step to the vial |
| Raw materials | Research-grade reagents. Media and supplements may be undisclosed proprietary formulations | Qualified, traceable, often animal-component-free, with vendor qualification files and specifications for each |
| Identity testing | STR profiling on the parent line where relevant; marker expression | STR plus a defined identity panel, with methods validated for the purpose |
| Sterility | Vendor release testing to internal criteria | Compendial sterility testing to pharmacopoeial method, with defined sample plans |
| Mycoplasma | Standard, method varies (PCR is common) | Required, by a validated method to a compendial standard |
| Adventitious agents / viral safety | Often not performed. Donor screening may or may not be documented | Required. Donor eligibility determination, viral marker testing, and adventitious agent testing on the bank |
| Endotoxin | Rarely specified | Specified with a numerical limit |
| Karyotype / genomic stability | Commonly provided for iPSC lines, less often for derivatives | Required with defined acceptance criteria and testing at defined points in the bank |
| Change control | None owed to you. The vendor can change a protocol between lots without telling you | Formal. Changes are assessed, documented, and typically notifiable to you under a quality agreement |
| Stability data | Rarely | Required, supporting the stated shelf life and storage condition |
| Auditability | No right of audit as standard | Right to audit is normal and is negotiated into the quality agreement |
| Permitted use | Research only. Explicitly not for administration to humans, clinical trials, or diagnostic purposes involving human subjects | Intended to support clinical use, subject to the applicable regulatory pathway and filings |
| Price basis | Published catalogue price is common | Quote only. No supplier in our survey publishes GMP cell pricing |
| Lead time | Days to a few weeks from catalogue stock | Long. Batch scheduling, campaign manufacture, release testing, and often a technology transfer phase first |
| Licensing | Research-use licence. Commercial derivative rights are separate — see cell line licensing | A separate licensing route entirely. Vendors typically maintain a distinct licensable GMP line programme apart from the research catalogue |
On the price multiple
We are not going to give you a number, and it is worth saying why.
Every supplier surveyed prices GMP cell material by quote. Not one publishes a GMP list price alongside a matched RUO price for the same cell type from the same donor. Any multiple we printed would be constructed from unmatched products or from hearsay, and a fabricated ratio in a purchasing decision is worse than no ratio at all.
What can be said honestly is structural. GMP cost is dominated by fixed costs that do not scale with your order — facility qualification, method validation, the quality function, release testing per lot — so the premium per vial is large at small volumes and compresses substantially at manufacturing scale. The corollary is that GMP material bought in research quantities is at its worst possible price point, which is one more reason not to buy it before you need it.
If you need a real figure, it comes from an RFQ against your specification, and the answer will depend on your volume and your phase.
When RUO is sufficient
Most of the time. RUO is the correct grade if all of the following hold:
- Nothing you make will be administered to a human, ever, on this programme.
- The work does not support a regulatory submission that requires GMP-grade starting material.
- You are doing discovery, target validation, assay development, toxicology screening, device development, or any engineering application where the cells never enter a body.
- You can tolerate a process change between lots without a formal notification, and you have a bridging plan if it happens.
Almost all research procurement, all biohybrid robotics and biological computing work, all organ-on-chip development, and the great majority of preclinical discovery sits here comfortably. Buying GMP for a research application is a common and expensive error — it buys compliance infrastructure that your project has no use for and no way to exploit.
When you genuinely need GMP
- The material or something derived from it will be administered to a human.
- You are preparing a regulatory filing whose requirements specify the manufacturing standard of the starting material.
- You are a contract manufacturer whose clients will inherit your material into their clinical supply chain.
- Your investors or partners require it as a condition, which is a commercial reason rather than a technical one, but a real one.
Note the phrase phase-appropriate GMP, which you will meet in contract manufacturing discussions. Requirements tighten as a programme progresses toward and through clinical phases, and competent manufacturers will scope to your current phase rather than selling you full commercial-scale compliance in year one. Ask for that scoping explicitly; it is a significant cost lever.
The trap: switching grades late
The expensive mistake in this area is not choosing the wrong grade at the start. It is choosing RUO, building an entire programme on a specific RUO line, and then discovering that the clinical path requires GMP starting material — and that the line you have standardised on has no GMP equivalent, or has one under a different licence, or has donor consent that does not extend to clinical use.
That is a restart, not a swap. The differentiation protocol has to be re-established on new starting material, the characterisation has to be redone, and any comparability argument between your old and new data has to be constructed and defended.
If a clinical path is plausible within your programme’s life, the questions to ask at first purchase are:
- Does a GMP-grade version of this exact line exist, and from whom?
- Does the donor consent extend to clinical use, or does it stop at research?
- What is the licensing route to clinical use, and is it with you or with an upstream depositor?
- What would a technology transfer from this RUO line to a GMP equivalent involve, in months and in cost?
Answering those four at the beginning costs an afternoon. Answering them in year three costs a programme.
Boundary cases
“GMP-compatible” and “GMP-grade” are not GMP. They are marketing formulations meaning, roughly, that the material was made with qualified raw materials or in a suitable facility but without the full quality system, batch record and release process. It may be a perfectly good and sensible intermediate product. It is not GMP, and it will not satisfy a requirement for GMP. Ask what specifically is and is not in place.
ISO 9001 is not GMP. ISO 9001 certifies that a quality management system exists and is followed. It says nothing about pharmaceutical manufacturing controls. Vendors state it because it is true and meaningful; buyers sometimes read it as more than it is.
Cleanroom classification is not GMP either. An ISO 7 or ISO 5 cleanroom is a facility attribute. It is a component of GMP manufacture, not a substitute for the quality system.
A GMP cell does not make your product GMP. If you buy GMP starting material and then process it in a research laboratory, the output is not GMP. The standard applies to the whole chain.
Research use restrictions bind regardless of grade. RUO terms typically state explicitly that the material must not be administered to human subjects, used in clinical trials, or used for diagnostic purposes involving human subjects. Repository MTAs go further and treat use “in a clinical trial or other testing regulated by a government agency … or in any human” as a commercial use requiring a separate licence. Buying GMP material does not by itself resolve the licensing question — it is a manufacturing standard, not a permission.
What to ask a supplier
- Which grade is this SKU, in your own terminology, and what exactly does that terminology mean here?
- May I see a redacted example certificate of analysis and, for GMP, a batch record index?
- What release tests are performed on every lot, as opposed to periodically or on the master bank only?
- Will you notify me of a process change, and is that a contractual commitment or a courtesy?
- Is there a GMP equivalent of this line, and what is the path from one to the other?
- Do you offer a quality agreement, and will you accept an audit?
Ask the change-notification question even for RUO. It costs nothing and it is the single most common source of unexplained variance in a long-running programme.
Sources
Every figure above traces to one of these. Accessed on or before 2026-09-01.
- bit.bio — Ordering support FAQs: ioCells are Research Use Only, must not be administered to human subjects or used in clinical trials https://www.bit.bio/support/ordering-faqs
- ATCC — Material Transfer Agreement: materials are not for use in humans; clinical trial and regulated testing use is Commercial Use requiring a licence https://www.atcc.org/policies/product-use-policies/material-transfer-agreement
- FUJIFILM Cellular Dynamics — GMP iPSC lines, licensable for clinical and commercial manufacturing (separate from the RUO iCell catalogue) https://www.fujifilmcdi.com/cdmo-gmp-ipsc/
- Cook MyoSite — contract services: phase-appropriate GMP cell manufacturing, ISO 7 / ISO 5 cleanrooms https://www.cookmyosite.com/contract-services
- Wetware World supplier survey, 2026-09-01 https://wetwareworld.com/sourcing-methodology
rev 2026-09-01 · research use only · list prices are supplier-published and change without notice · not a quotation