Hepatocytes: the largest line item in a liver programme has no public price

Here is the buying problem, stated as an arithmetic gap.

A CN Bio PhysioMimix Liver-12 plate holds roughly 400,000 to 600,000 primary human hepatocytes per well across twelve wells. That is about 5 to 7 million primary hepatocytes consumed by a single plate, and primary human hepatocytes carry no public price at any supplier surveyed on 2026-09-01.

So the largest recurring consumable in a perfused liver programme is invisible while you are evaluating the platform. You can read the plate vendor’s page, you can read the instrument specification, and you cannot cost the cells until you are already inside a second sales conversation with a different company. Any comparison of a perfused route against a ready-made spheroid route that omits the cell input is not a comparison.

Three routes, one published price

The iPSC-derived route is roughly 67 per million viable cells, which makes it the cheapest hepatocyte on this page by a wide margin, an order of magnitude below what most buyers expect a human hepatocyte to cost. That figure deserves scepticism until you understand what it buys and what it does not, which is the subject of the next section.

The other two routes are quote-gated. We do not estimate. The honest statement is that a buyer cannot construct a budget for a primary hepatocyte programme from public information, and that is a structural feature of this market rather than a gap in this page.

The three routes are not substitutes

They answer different questions, and the price difference is mostly a proxy for that.

Primary human hepatocytes

The reference material. Freshly isolated or cryopreserved hepatocytes from a human donor liver, carrying the donor’s genotype, the donor’s drug-metabolising enzyme expression, and the donor’s history. Nothing else reproduces that.

The costs are lot-to-lot variability, limited and unpredictable supply, and a short useful window in 2D culture before dedifferentiation. Work on high-yield isolation from small liver samples (PMC12909712) and on two-step in vitro expansion of primary human hepatocytes with retained metabolic and regenerative capacity (PMC11373096) both exist because primary hepatocyte supply and expandability are unsolved problems, not solved ones.

Quality attribute definition for primary human hepatocytes is itself an active standards question, a 2025 paper in Cell Regeneration reports on the validation of quality attributes in a Primary Human Hepatocytes Standard (PMC12532551). That is worth knowing before a negotiation: the specification you should be demanding is a live topic in the field, not a settled catalogue field.

iPSC-derived hepatocytes

Deterministic manufacturing removes lot-to-lot variability, which is exactly what makes the price low and the supply predictable. What you give up is stated plainly: a single genetic background, and an expression profile that is not that of pooled primary hepatocytes.

For assay development, for high-throughput screening where relative response matters more than absolute human-equivalent metabolism, and for any experiment repeated over months where consistency between runs is the dominant concern, that trade is often correct. For a definitive drug metabolism prediction intended to support a regulatory position, it usually is not.

HepaRG and immortalised lines

The third route, and the underrated one. A terminally differentiated human hepatic cell line with a long publication record. Its value is that it makes method development, assay qualification and technician training affordable in a way primary hepatocytes never are.

The published evidence base is unusually practical. A CometChip assay in HepaRG has been assessed for transferability and reproducibility across laboratories (PMC12703573), which is precisely the kind of cross-site evidence that makes a model usable in a regulated environment. Epigenetic mapping of maturation versus retrodifferentiation in HepaRG (PMC13298337) documents the state transitions the line goes through, which is the thing you need to control to use it reproducibly.

We could not retrieve a public price. That is stated rather than estimated.

The invisible line item, worked through

The arithmetic that decides platform choice, using only published seeding figures.

Now consider what that means over a programme. A study running four plate conditions in triplicate consumes twelve plates, which is roughly 60 to 84 million primary human hepatocytes, for one study. Neither the plate quote nor the instrument quote contains that number. It arrives as a separate purchase order from a separate company at a price nobody would tell you in advance.

The procurement instruction that follows is simple and rarely followed: cost the cells before you choose the platform. Get a primary hepatocyte quote, in millions of viable plateable cells, at the volume your study design actually implies, before you compare a perfused route against a spheroid route or against a monolayer. Otherwise you are comparing two hardware prices and calling it a model comparison.

The ready-made alternatives, where the vendor absorbs the cell input into a finished unit, are covered in liver microtissues and spheroids.

Where the honest comparison between routes sits

A useful way to hold the three routes in mind is to ask what each one is protecting you from.

Primary hepatocytes protect you from being wrong about human metabolism. They cost lot variability, supply unpredictability and an unquotable price. Buy them when the claim you intend to make is about what a human liver does.

iPSC-derived hepatocytes protect you from lot variability. They cost genetic-background breadth and a metabolic profile that is not pooled primary material. Buy them when the claim is about relative response across many runs over many months, and when consistency between those runs is the thing that would otherwise sink the study.

HepaRG protects you from spending primary material on learning. It costs the argument that your model is donor-representative. Buy it to develop the method, qualify the assay, train the operator and generate the cross-site reproducibility evidence, then spend primary lots on the questions that require them.

Most well-run liver programmes use more than one of these, and the sequencing matters: method development on the cheap route, definitive work on the expensive one. Programmes that skip the first step tend to consume their most valuable material during the phase when their protocol is least settled, which is the same failure documented across the cell category generally.

Plateable versus suspension is a specification, not a grade

Cryopreserved primary hepatocyte lots are qualified for different uses and the qualification is not a quality ranking.

Suspension-qualified lots are validated to survive thaw and perform in short suspension incubations, metabolic stability assays, clearance measurements over a few hours. They are not necessarily qualified to attach.

Plateable lots are validated to attach and form a monolayer, which is what any longer-duration culture, sandwich configuration, spheroid or perfused plate requires.

Ordering a suspension-qualified lot for a plate-based experiment is the most common expensive error in this category, and because the lot is otherwise a perfectly good lot, the failure looks like a technique problem for the first week. Put the word “plateable” on the purchase order in writing, and ask for the attachment specification and the confluence achieved at the stated seeding density.

The specifications that decide whether the lot is usable

Hepatocytes are bought for function, so a viability number alone is close to useless.

Metabolic activity. Cytochrome P450 activity for the specific isoforms your assay depends on, measured at release, with the substrate and the units stated. “CYP active” is not a specification. A lot with good viability and poor CYP3A4 activity will pass an incoming inspection and fail the experiment.

Albumin secretion and urea synthesis. Function markers that indicate the cells are behaving hepatically rather than merely surviving.

Attachment and confluence at a stated seeding density, for plateable lots.

The medium the specification was measured in. This matters more than it sounds. A 2026 study in Physiological Reports on physiologically relevant media in primary human hepatocytes (PMC13330593) reports overlapping metabolic responses across media formulations, which is a direct statement that the medium is part of the measurement. A release figure generated in one formulation does not transfer to another without evidence.

Donor demographics and medication history, for primary lots, because drug-metabolising enzyme expression varies with both.

Pooled or single donor, and for pooled lots the donor count. Pooling averages away the individual, which is right for a population estimate and wrong for a genotype-attributed claim.

Companion cells, and why they need the same donor conversation

A hepatocyte monoculture is a hepatocyte monoculture. Most current liver work is not.

Published 3D co-cultures of primary human hepatocytes with Kupffer-like cells addressing innate immune responses to rAAV (PMC12795831) illustrate the direction: the non-parenchymal compartment is the point when the question is immunological, inflammatory or fibrotic.

That imposes procurement constraints that a hepatocyte-only order does not capture:

• Which non-parenchymal cells, Kupffer, stellate, sinusoidal endothelial, and from which donor. Matched-donor material removes a confound that unmatched material introduces; the routes for obtaining it are covered in the matched donor material datasheet.
• The ratio, which comes from your protocol and is not a product specification anywhere.
• The delivery week, because every component has to thaw into the same culture.
• One medium that supports all populations, or a documented and justified transition.

QC checklist

• Lot-specific certificate of analysis, obtained and read before you commit to the lot, not after it ships.
• Plateable or suspension qualification, stated explicitly in writing.
• Post-thaw viability specification and the assay used to generate it.
• Viable yield per vial, stated as viable post-thaw rather than as a count at freeze.
• CYP isoform activity at release, per isoform, with substrate and units.
• Albumin secretion and urea synthesis where the supplier measures them.
• Attachment efficiency and confluence at a stated seeding density, for plateable lots.
• The medium and matrix configuration every functional specification was measured in.
• Donor demographics, disease state and medication history for primary lots.
• Pooled or single donor, with the donor count for pools.
• Sterility, mycoplasma and viral screening with agents named.
• Consent scope covering your intended use, including any genomic analysis.
• Shipping format, dry ice or dry shipper, and the receiving procedure it implies.

Specification checklist for a hepatocyte quote

• Route: primary, iPSC-derived, or immortalised HepaRG-class line. State which, and state why, because the supplier will otherwise quote whichever they hold.
• Total viable cells required, worked against the plate or construct that will consume them, not guessed. Do this arithmetic in the enquiry itself.
• Plateable or suspension, in writing.
• Single-donor or pooled, with the number of donors for a pool.
• The metabolic specification you require at release, named by isoform and assay.
• Non-parenchymal companion cells required, their donor provenance, and whether donor matching is a requirement.
• Species, where cross-species metabolic comparison is the question.
• Lot reservation: whether a single lot can cover the whole programme, and whether the supplier will hold it. Splitting a study across two primary lots introduces a variable no statistic will remove.
• Delivery schedule against your plate production or study calendar.
• Licence terms for anything downstream that will be sold rather than published.

Where to go next

The datasheet, unit, lead time and QC scope, is at human-hepatocytes. If the destination is a finished model rather than a vial, liver microtissues and spheroids covers what ships ready to use and liver model suppliers compared puts the vendors side by side. For the whole-programme view including instrument and consumable costs, see setting up a liver model programme.

How we can help

The specific value here is getting the invisible number out of two suppliers at once. Tell us the plate or construct, the study design and the number of conditions, and we will do the cell-count arithmetic, obtain primary hepatocyte quotes in viable plateable millions at that volume, put them next to the iPSC-derived catalogue figure on the same units, and ask for the CYP activity specification per isoform before you commit to a platform.

If the answer is that an iPSC-derived hepatocyte at a published catalogue price does your job, we will say so.