Written and reviewed by the InVivoLabs team, drawing on 15+ years of hands-on work in preclinical surgical models, GLP toxicology and CNS drug delivery, on both the CRO and sponsor side of the table. Last reviewed October 1, 2026.
Why most biotechs outsource preclinical work
A virtual biotech owns the science, the strategy and the decisions, and contracts out nearly everything that needs a laboratory or a vivarium. Even well-funded emerging companies rarely build animal facilities, because the work is episodic, regulated and specialized. A single IND-enabling package can need rodent and non-rodent toxicology, safety pharmacology, genetic toxicology, bioanalysis and pathology, and few companies will run each of those often enough to justify doing it in-house.
Outsourcing works well when the sponsor stays in control of what is being asked and why. It goes badly when a company hands over a molecule and a budget and expects the CRO to decide what the program needs. CROs are very good at running studies. Deciding which studies your program needs, and what result would change your plans, is your job.
Map the program before you call a CRO
Start from the decision each study supports, not from a list of study types. Before the first CRO conversation, write down:
The target product profile. Modality, intended route and dosing frequency in patients, indication and patient population. These drive species selection, study duration and route of administration in every animal study that follows.
The questions in order. Does the molecule work in a disease model? At what exposure? Is that exposure tolerated? Which species is pharmacologically relevant? Each answer gates the next spend.
The regulatory target. Work back from the IND or CTA date. ICH M3(R2) sets how long the toxicology studies must be to support the planned clinical duration, and modality-specific guidance (ICH S6(R1) for biologics, FDA guidance for gene and cell therapies) changes what the package contains.
Our preclinical research overview and GLP toxicology guide cover the standard package in more detail, and the species selection guide covers how to choose and justify species.
What stays in-house, and who you need
Even a fully virtual company needs people who can design studies and read the data critically. Most build a small core and add fractional experts:
A nonclinical lead or consulting toxicologist who designs the program, reviews protocols and reports, and owns interpretation of findings. This is the role most often under-resourced, and the one that most often decides whether an IND package holds up.
Regulatory support to align the nonclinical plan with the intended clinical plan, and to prepare for a pre-IND meeting.
CMC and test article supply. GLP studies need a characterized test article with stability data, and the toxicology batch should be representative of the clinical material. Late or poorly characterized test article delays more studies than any CRO does.
A study monitor, often the nonclinical lead, who is the single point of contact with each CRO study director and can make decisions quickly when something unexpected happens in life.
One full-service CRO or several specialists?
There is no universally right answer, and most programs end up with a hybrid.
| Full-service CRO | Specialist labs | |
|---|---|---|
| Strengths | One contract and one quality system; in-life, bioanalysis and pathology under one roof; capacity for large GLP studies | Deeper experience in a model, species, surgical procedure or modality; often more senior attention for a small sponsor; more flexible scheduling |
| Watch for | Small programs can get less senior attention; less flexibility on non-standard designs | More contracts and handoffs; samples and data move between sites; you coordinate the interfaces |
| Common use | IND-enabling GLP toxicology and safety pharmacology | Disease models, CNS and other specialized delivery routes, device studies, specialty bioanalysis |
A common pattern is to run efficacy and early PK with specialists who know the disease model, then move IND-enabling toxicology to a GLP-capable CRO. If your program depends on a demanding procedure, such as intrathecal catheterization, intracerebroventricular dosing or a large-animal surgical model, weigh the surgical team's experience above almost everything else. A study that fails in life because of the procedure costs far more than the difference between two quotes.
Finding and qualifying CROs
Build a shortlist
Aim for three to five candidates per study type. Use the InVivoLabs directory to filter by species, service, GLP and AAALAC status and location, and ask colleagues which labs they have used for similar work. Put a confidentiality agreement in place before you share anything about the molecule.
Send the same request to everyone
A short, consistent study outline makes quotes comparable: objective, species and strain, route and frequency, group sizes, duration and recovery, endpoints, GLP or non-GLP, test article form and quantity available, sample analysis needs, report format and target start date. Without it, each CRO quotes a different study and you cannot compare them.
How large pharma qualifies CRO partners, scaled down
Large companies qualify vendors through formal audits by their quality assurance group, written quality agreements, and preferred-provider lists that are reviewed periodically. A small sponsor can borrow the essentials:
Audit or visit before the first GLP study. A consultant QA auditor can do this for you. Look at the facility, the animal care program, training records and how deviations are handled.
Ask about inspection history. When was the last FDA or OECD GLP inspection, and were there findings? Is the animal program AAALAC accredited?
Meet the people, not just business development. Who will be the study director, surgeon, pathologist and bioanalytical lead? How long have they been there, and how many similar studies have they run?
Ask for evidence. Historical control data, example (redacted) reports, and success rates for the procedures your study depends on, such as catheter patency or implant survival.
Our CRO selection guide has a full evaluation checklist and the red flags to watch for.
Comparing quotes
The lowest quote is rarely the cheapest study. Before comparing totals, make sure each quote covers the same scope, and check:
Pass-through costs. Animal purchase (especially nonhuman primates, whose cost and availability have been volatile), shipping, test article analysis and archiving are often quoted separately or as estimates.
What is included in the in-life price. Clinical pathology, toxicokinetic sample collection, bioanalysis, histopathology, pathology peer review and SEND datasets may each be separate lines.
Change orders and cancellation. How are protocol amendments priced? What happens to the fee if the study is postponed or cancelled because test article is late?
Payment milestones and report timing. Tie payments to deliverables such as study start, in-life completion, draft report and final report, and agree the draft report timeline in the contract.
Realistic timelines
Calendar time in preclinical programs is dominated by things other than the in-life phase. Plan for:
Slot availability. Popular CROs book GLP study starts months ahead, particularly for non-rodent species.
Animal sourcing. Nonhuman primates can need long lead times for sourcing, quarantine and acclimation, and supply has been constrained in recent years. Confirm availability before you commit to a start date.
Enabling work. Dose-range finding studies, a validated bioanalytical method and formulation stability data usually need to be in place before the pivotal study starts.
Reporting. Draft reports for GLP studies commonly arrive weeks to a few months after the last necropsy, followed by sponsor review, QA audit and finalization.
Managing outsourced studies
Once a study is contracted, the sponsor's job shifts from selection to oversight.
Review the protocol line by line. Dose levels, sampling times, sample volumes and necropsy tissue lists are where expensive errors hide. Anything not in the protocol will not be done.
Hold a kickoff meeting with the study director and key scientists, and agree how and when you will be told about unexpected findings, mortality or deviations, and who on your side can make decisions.
Be present for the critical days. For surgical and CNS delivery studies, attend (or watch remotely) the first surgeries and first dosing day. A short pilot cohort to confirm catheter placement, patency or dose delivery before the main study often saves the main study.
Ask for data during the study, not just at the end: body weights, clinical observations, and clinical pathology as it becomes available. Early signals give you time to adjust dose or design while it still matters.
Common mistakes we see
Going straight to a GLP study without a non-GLP dose-range finder, so the high dose is either untolerated or uninformative. Choosing a species because a CRO has capacity rather than because it is pharmacologically relevant. Starting in-life work before the bioanalytical method is ready, leaving samples in the freezer for months. Running out of characterized test article halfway through a program. And letting the CRO design the program without independent review: a good CRO will design a sound study, but it is not responsible for whether that study is the one your program needs.
Questions we hear from emerging biotechs
How early should we contact CROs?
Earlier than you think. Informal conversations while you are still planning help you learn what is feasible, what lead times look like and which designs a lab recommends. Contact GLP toxicology providers six months or more before you want the pivotal studies to start.
Does every study need to be GLP?
No. Pivotal safety studies that support a clinical trial application are run under GLP. Efficacy, early PK, dose-range finding and most discovery work are not, and running them non-GLP is faster and less expensive.
Can one CRO handle everything?
Large full-service CROs can run most of a standard package. Specialized models, delivery routes and some modalities are often better served by a specialist, so many sponsors use one main partner plus one or two specialists.
Where can we find a preclinical partner for our program?
Start with a filtered search of the InVivoLabs CRO directory, or browse by study area: toxicology, in vivo pharmacology, DMPK, bioanalysis and CNS. Sponsors use the directory free of charge and contact CROs directly.