The Apheresis Market in 2026: Size, Growth, and Where the Money Sits

Apheresis is the process used to remove specific elements from blood and return the remainder to the body. A donor or patient is connected to a machine through a sterile intravenous line, blood is drawn continuously and separated by size and density, the wanted fraction is retained, and everything else goes back. That single sentence describes a clinical procedure. It also describes the operating core of an industry that research firms size in the billions and that most people writing business plans get wrong, because they treat "apheresis" as one market when it is at least three.

Start with the headline number. Reports and Data, 2025 puts the global apheresis equipment market at $4.43 billion with a 7.3% CAGR forecast through 2035. That is the figure we use in the hero stats above, and it is the one we would anchor a lender pack to. It is not the only figure available, and the spread between the estimates is itself worth understanding before you cite one in a plan.

Healthcare Foresights, 2025 sizes the therapeutic apheresis segment alone at $3.84 billion, projecting $8.94 billion by 2035 at a 7.9% CAGR. Straits Research, 2026 takes a narrower cut, plasmapheresis machines only, and gets $1.73 billion in 2026 rising to $3.16 billion by 2034 at 7.8%. Other houses have published 2025 apheresis figures ranging from $3.30 billion to $4.76 billion with CAGRs from 5.67% to 11.2%.

Those numbers are not contradicting each other. They are measuring different boundaries: devices only, devices plus disposables, therapeutic procedures, or the whole category. A lender does not need you to pick the biggest one. A lender needs you to say which boundary you chose and why. Every serious estimate agrees on the direction and roughly on the slope, somewhere between 7% and 8% annually, and that consistency across independent methodologies is a stronger argument than any single figure.

The growth is not evenly distributed, and this is where the commercial opportunity actually lives. Straits Research reports that US apheresis collections for cell and gene therapy starting material are projected to rise from 43,000 in 2021 to roughly 132,000 by 2025, a near-tripling driven by CAR-T and the wider regenerative medicine pipeline. Nothing else in the category is moving at that rate. Source plasma is a large, mature, contract-driven business. Therapeutic apheresis grows with clinical indications and payer policy. Cell collection is the segment where demand is outrunning capacity, and capacity is the thing a new entrant can build.

On the source plasma side, the installed base tells you what kind of market you are entering. There are more than 1,000 IQPP-certified collection centres in the United States. This is not a greenfield. It is a mature, consolidated, standards-governed supply chain with four names doing most of the volume: CSL Plasma, which operates over 300 US centres under a parent with roughly $10 billion in revenue; BioLife Plasma Services, owned by Takeda, at around $4 billion in plasma revenue; Octapharma Plasma at roughly $3 billion; and Grifols, trading in the US as Biomat USA.

Independents exist and they are the relevant comparable for anyone reading this page. ADMA BioCenters has built out ten centres at various stages with six FDA-approved, the sixth of which was approved at Myrtle Beach, South Carolina. ImmunoTek Bio Centers runs the independent collector model, building and operating centres that supply fractionators rather than a captive parent. Life Plasma achieved IQPP certification, which is the credential a smaller operator needs before a fractionator will sign a supply contract. Those three are the shape of the business a founder can realistically build.

On the therapeutic side the operators look completely different: Vitalant, Bloodworks Northwest, New York Blood Center Enterprises, Stanford Blood Center, and hospital-embedded programmes such as the one at UC San Diego Health. These are not-for-profits and academic centres, not roll-ups. If your plan proposes to compete with them on price, you have misread the market. If it proposes to take referral overflow, offer faster scheduling, or serve a catchment they do not cover, you have a business.

Staffing an Apheresis Operation: What the Wage Data Says

Labour is the second-largest line in a source plasma centre and the largest controllable one. Most plans get it wrong by staffing from a hunch. The Bureau of Labor Statistics publishes the actual numbers, and a lender who has financed clinical operations before will check them.

BLS, OEWS May 2025 puts the median annual wage for phlebotomists at $45,230, or $21.75 an hour, up from $43,660 in the prior release. There are 143,540 phlebotomists employed nationwide. The middle half of the profession earns between $38,190 and $48,930; the lowest tenth earns around $35,780 and the highest tenth around $58,780. Employment is projected to grow 6% from 2024 to 2034, faster than the average across all occupations.

Two things follow from that distribution. First, if your model prices donor-floor staff at the 10th percentile, you are planning to lose them, and turnover on a licensed clinical floor is not merely a recruitment cost — retraining and requalification are documented, audited activities. Second, the gap between the 25th and 75th percentile is only about $10,700, which means wage competition in this role is regional rather than dramatic. Site selection moves your labour cost more than negotiation does.

Above phlebotomy, the picture changes. BLS puts clinical laboratory technologists and technicians at a median of roughly $61,890 to $62,930, which is the pay grade for anyone running on-site testing. And the regulatory structure forces roles into your roster that a generic clinic model would not have. FDA requires that initial medical examinations be performed by a qualified doctor of medicine or osteopathy currently licensed to practise, though it permits trained physician substitutes working under that physician's supervision to perform examinations. That single sentence is worth several hundred thousand dollars a year of org-chart design.

Roster and Annual Base Cost — 45-Bed Source Plasma Centre

Roles marked "BLS median" use published national medians; roles marked "Avvale estimate" are planning assumptions from our clinical-operations models and should be replaced with local quotes before submission.

Note what the burden does. A $2.08M base becomes $2.66M loaded, and plans that forget the 28% are not off by a rounding error — they are off by more than half a million dollars, which on an 18% margin is most of the year's profit. Every business plan we rebuild for a clinical operator has this same defect somewhere.

Three Models, One Machine

This is the section most apheresis business plans are missing, and it is the reason so many get rejected. The word "apheresis" covers three businesses that share a device and share nothing else. Before you write a single financial assumption, you have to say which one you are building, because the payer, the capex, the sales cycle and the margin are all different.

Source plasma is a manufacturing business wearing a clinic's clothes. You are producing a raw material to a specification, on contract, at volume, and your entire economics reduce to cost per litre. Donor experience matters because retention matters, and retention matters because cost per litre falls as a centre matures. Historical SEC filings from plasma businesses show exactly this pattern: collection cost per litre declined as centres matured, and lower collection costs combined with higher volumes drove gross margin improvement. That sentence is the whole investment thesis, and it means your Year 1 margin is supposed to look bad.

Therapeutic apheresis is a reimbursement business. You are not selling anything; you are documenting medical necessity to a payer. Volume comes from referring nephrologists, neurologists and haematologists, which means your sales cycle is measured in relationships and your revenue is capped by what the payer mix will bear. The margin is thin and the working capital cycle is a claims cycle.

Cell collection for advanced therapies is a contract-manufacturing business. A sponsor needs apheresis-derived starting material collected to protocol, on a scheduled date, with chain-of-identity documentation that will survive an inspection. Price per collection is the highest of the three. Volume is entirely gated on winning the contract, which is why this is the model where a strong plan does the most work: nobody signs a starting-material agreement with a founder who cannot articulate their quality system.

Most operators stop at "the apheresis market is growing at 7.3%." The number that actually drives the business is revenue per chair-hour, because chair time is the constraint in all three models and the only asset you genuinely own. Two machines running nine procedures a day is a different company from two machines running four, and no market CAGR will tell you which one you are.

What It Costs to Open an Apheresis Business

There is no single answer, and any page that gives you one is guessing. The two viable entry models sit an order of magnitude apart.

A two-machine outpatient therapeutic apheresis service runs roughly $340K to $980K (£268K to £773K) on our planning model. You are buying clinical space, two systems, a small nursing roster, and the payer credentialling to bill. It is financeable by a specialist physician with a bank relationship, and it is the realistic starting point for most founders reading this.

A 40 to 50 bed source plasma collection centre runs roughly $2.4M to $5.5M (£1.9M to £4.3M). You are building a licensed clinical facility of 10,000 to 12,000 square feet, buying 40 to 50 collection systems, funding an FDA licensure programme, and carrying working capital across a 90 to 120 day payment cycle before the first fractionator cheque lands.

The equipment line is the biggest swing factor and the one you can most easily get wrong. Straits Research puts basic plasmapheresis machines at $10,000 to $20,000, while high-end units with automated functions run $30,000 to $100,000. Multiply that spread by 45 units and your capex moves by more than $3.5 million on a single sourcing decision. The report is blunt about the consequence: these upfront costs are a significant barrier, particularly for smaller or newly established plasma collection centres.

Funding Routes

In the US, an apheresis collection business files under NAICS 621991, Blood and Organ Banks. The SBA small-business size standard for that code is $40 million in average annual receipts, which means essentially every independent apheresis operator in the country qualifies as a small business for SBA purposes. That matters more than it sounds: it puts the SBA 7(a) programme, capped at $5 million, within reach of a business whose entire capex sits below that ceiling.

The practical structure we see work is a stack rather than a single instrument. Senior SBA 7(a) debt covers the fit-out and the licensure programme. Equipment finance, secured on the collection systems themselves, covers the machines — lenders like this collateral because Spectra Optia and NexSys units hold resale value and there is a real secondary market. Founder equity covers the piece nobody will lend against: the months of collection you cannot sell. More on that below, because it is the single most common reason an apheresis plan gets declined.

In the UK the picture is different and smaller. Start Up Loans cap at £25,000 at 6% fixed, which does not touch a blood establishment build. UK apheresis ventures are financed through commercial lending, NHS or ICB service contracts, Innovate UK grant routes for the cell-therapy adjacent models, or equity. If you are UK-based and modelling a compensated-donor source plasma centre on US unit economics, stop — the donor compensation framework that underpins those numbers does not transfer.

Equipment: What You Buy, and What It Locks You Into

Platform choice in apheresis is not a procurement decision made after the plan is written. It sets your procedure time, your consumable cost per run, your staffing ratio and your maintenance contract for the next decade. Lenders who have financed clinical equipment know this and will ask. Here are the vendors that actually matter.

• Terumo BCT Spectra Optia — the reference platform for therapeutic apheresis. In a comparative study of five plateletpheresis machines at a tertiary care centre, the Spectra Optia recorded the shortest procedural time at 59.5 minutes, donor adverse effects of 6.3%, the highest collection efficacy at 52.8%, and the highest collection rate. Budget in the $30K–$100K high-end automated band.
• Terumo BCT Trima Accel — transfusion and multicomponent apheresis; the platform of choice where you need red cells, platelets and plasma from one donation.
• Haemonetics NexSys PCS and PCS2 — the donor plasma collection workhorses. If you are building a source plasma centre, this is the likely fleet. Haemonetics also supplies the blood-centre ecosystem around the device.
• Haemonetics MCS+ — multicomponent collection; still widely deployed and available on the secondary market at materially lower cost.
• Fresenius Kabi Amicus, COM.TEC and Aurora — the third major vendor family, with Fresenius Kabi supplying apheresis collection systems across both donor and therapeutic settings.
• Blast freezer (−30°C) and walk-in plasma storage — $60K–$180K (Avvale estimate). Non-negotiable and inspected. Cold-chain failure is the fastest way to write off a month of collection.
• Donor chairs, bariatric-rated, 45 units — $1,200–$2,600 each, so $54K–$117K (Avvale estimate). Cheap chairs cost you retention, and retention is cost per litre.
• Plasma protein and haematocrit analysers — $18K–$46K (Avvale estimate). FDA requires total plasma protein determination before each plasmapheresis procedure, so this is a throughput-critical device, not lab furniture.
• Donor management software and eQMS — Haemonetics NexLynk DMS for donor management; Veeva Vault QMS or MasterControl for the quality system. $120K–$310K including implementation (Avvale estimate).
• Backup generator and validated cold-chain monitoring — $35K–$90K (Avvale estimate).

One constraint that catches new operators: Straits Research identifies the shortage of skilled professionals as a critical restraint on this market, because these systems require specialised expertise to operate and maintain. Your platform choice therefore determines your hiring pool. Choosing a system nobody within 100 miles has been trained on is a staffing decision disguised as a purchase order.

New versus refurbished is a genuine strategic fork. A refurbished fleet halves the capex and roughly doubles the maintenance line, and it complicates the validation package you will hand to an inspector. For a first centre with financing pressure, mixed fleets — new on the critical path, refurbished on the flex capacity — are usually where the model lands.

How Apheresis Businesses Make Money

Now the numbers that decide it. Both worked examples below are Avvale planning models built on cited unit prices. Treat them as structure, not as forecast.

Worked Example 1 — Source Plasma Collection Centre

The revenue side is almost brutally simple. Pharmaceutical companies pay $300 to $800 per litre for plasma, while donors receive $50 to $100 per donation. That gap is the business. Everything else is throughput and cost discipline.

We model 45 donor beds, 160 donations a day, six days a week, 50 weeks a year — 48,000 donations at 0.8 litres each, so 38,400 litres. We price at $300 per litre, the bottom of the cited band, deliberately. Long-term fractionator supply contracts do not price at spot, and a plan that models $600/L will not survive a lender's diligence call with anyone who has seen a real supply agreement.

Read the cost stack carefully, because it tells you what kind of company this is. Donor compensation is 27% of revenue and it is the largest single line — larger than your entire payroll. That means this is not a clinic that happens to pay donors. It is a procurement operation with a clinical licence, and the strategic question is always the same: what is my cost per litre, and is it falling?

It should be falling. That is the entire maturation thesis. Donor compensation per donation is roughly fixed, but a mature centre gets more litres from the same fixed cost base because repeat donors are faster to process, less likely to defer, and do not carry acquisition cost. Historical filings from plasma operators show collection cost per litre declining as centres mature, with lower costs plus higher volume driving gross margin. If your Year 3 cost per litre equals your Year 1 cost per litre, you have not modelled the business — you have modelled a spreadsheet.

Worked Example 2 — Therapeutic Apheresis Service

Different business, different failure mode. CPT 36514, therapeutic apheresis for plasma pheresis, carries a work RVU of 1.76 and an estimated Medicare payment of $714.43 at the 2026 CMS PFS conversion factor of $33.40, before locality adjustment. In the hospital outpatient setting, claims coded 36514 map to APC 5242, Blood Product Exchange, and the total cost per procedure includes both the professional component and supplies — particularly replacement fluids. Hold that last clause; it is the whole section.

We model two Spectra Optia systems, nine procedures a day, five days a week, 48 weeks: 2,160 procedures at a blended $1,150 (Medicare plus a commercial mix) = $2.48M. Then the costs: replacement fluid — albumin or fresh frozen plasma — at $480 a procedure is $1.04M. Disposable kits at $160 are $346K. Six apheresis RN FTE, loaded, are $741K. Medical director $80K. Space, equipment lease and G&A $260K. Total $2.46M. The service is roughly breakeven.

Look at what happened there. The replacement fluid line is larger than the nursing line. In a business everyone assumes is a staffing business, the biggest controllable cost is a bag of albumin. Move fluid cost by $100 a procedure and you move EBITDA by $216K — which is the difference between a 9% margin and a loss. Most therapeutic apheresis models we are asked to review do not have a replacement-fluid line at all. They have "supplies," at 8% of revenue, and they are wrong by a factor of five.

The ASFA Therapeutic Apheresis Reimbursement Guide, 2025 is the reference document here, and reimbursement varies by locality and by Medicare Administrative Contractor. Historical rates give a sense of the spread: outpatient payment for plasma and cellular apheresis has run around $951 on average, roughly $1,140 in Manhattan against $870 in Pittsburgh. Your geography is a revenue assumption, not a lifestyle choice.

The Third Model

Cell collection for CAR-T and gene therapy has the best per-procedure economics of the three and the fewest procedures. With US collections for cell and gene therapy starting material tracking from 43,000 in 2021 toward roughly 132,000 in 2025, demand is real and growing faster than anything else in the category. But revenue is gated entirely on a sponsor agreement, so the plan's job is different: it must prove your quality system, your chain-of-identity documentation and your scheduling reliability, because that is what a sponsor is actually buying. Financially, this model works best bolted onto one of the other two — the marginal cost of adding it to an existing licensed operation is small, and the marginal revenue is not.

Regulatory Approval: FDA, MHRA, and Beyond

Apheresis is one of the few small-business categories where the regulator, not the market, sets your launch date. Treat this section as a cash-flow input.

United States

• Biologics License Application (BLA), FDA CBER. Plasma donation centres must apply for a BLA and pass a pre-licence inspection. This is a programme, not a form.
• Donor eligibility, 21 CFR 630.15. Requirements specific to Whole Blood, Red Blood Cells and Plasma collected by apheresis — a separate rule from whole-blood donation, and one your SOPs must reflect line by line.
• Total plasma protein, 6.0–9.0 g/dL. Establishments must determine the donor's total plasma protein level before each plasmapheresis procedure; no less than 6.0 and no more than 9.0 grams per decilitre. This is a per-donation throughput cost.
• Informed consent, 21 CFR 640.61. Consent must be obtained on the first day of donation or no more than one week before, and re-obtained at intervals of no longer than one year.
• Qualified physician or trained physician substitute. Initial medical examinations require a currently licensed doctor of medicine or osteopathy; FDA permits trained physician substitutes under that physician's supervision. This shapes your org chart and your medical-director contract.
• IQPP certification, PPTA. The International Quality Plasma Program has run since 1991 and provides independent third-party evaluation against global source plasma standards. Voluntary in law; in practice, fractionators contract with certified centres, so treat it as mandatory and budget for it.
• State licensure. States may impose requirements that exceed federal standards, affecting staffing and procedure. Two centres with identical models in different states are not identical businesses.

United Kingdom

• Blood Establishment Authorisation, MHRA. MHRA holds the controls and authorisations for blood establishments. A BE is authorised for collection, testing, processing, storage and distribution of blood and blood components intended for transfusion, and can also be authorised to collect and test blood intended for further manufacture of medicinal products.
• Blood Safety and Quality Regulations 2005 (SI 2005/50). The governing statutory instrument for blood establishments and hospital blood banks, enforced by MHRA.
• ATMP starting material route, HTA and MHRA. The two regulators agreed to extend their dendritic-cell policy so that collection of blood as a starting material for an ATMP may be performed under either a Tissues and Cells licence or a Blood Establishment Licence. If you are building the CAR-T collection model in the UK, that choice is your first strategic decision.
• SABRE haemovigilance reporting. Serious adverse blood reactions and events must be reported. Apheresis donations split into multiple doses must be individually identified for traceability — a data-architecture requirement, not just a paperwork one.
• Insurance. Public liability (£2M minimum) and employers' liability (£5M minimum), plus clinical indemnity appropriate to the procedure.

Other Jurisdictions

• European Union: blood establishments require competent-authority authorisation in each member state, with EU-level rules on substances of human origin governing collection, traceability and vigilance. Donor compensation rules differ sharply between member states, which is precisely why US source plasma economics do not port to the EU.
• Canada: Health Canada regulates blood establishments and licences collection. Paid plasma collection is restricted or prohibited in several provinces, which removes the compensated-donor mechanism that the US model depends on. A Canadian plan built on Worked Example 1 above is not conservative — it is invalid.

The through-line: in this category the licence is the business model. Change jurisdiction and you have not changed your address, you have changed your revenue equation. Our bespoke business plan service builds the compliance timeline directly into the cash-flow model rather than parking it in an appendix.

Five Mistakes That Sink Apheresis Ventures

These are drawn from plans we have been asked to rebuild after a decline. Each one is specific to this category.

1. Budgeting for FDA approval as a filing, not a programme

This is the one that kills deals. A source plasma centre needs a BLA and a pre-licence inspection. In the meantime the building is leased, 39 people are on payroll, machines are depreciating, and donors are being collected — and none of that plasma can be sold into fractionation. Founders model revenue from month one because the centre is operating from month one. It is operating and earning nothing. Price the gap explicitly, fund it with equity because no lender will advance against it, and show the lender you know it exists. Doing so converts your biggest weakness into your strongest credibility signal.

2. Pricing at the top of the band

The cited range is $300 to $800 per litre. The temptation to model at $600 is enormous — it roughly doubles your EBITDA and makes every ratio work. It is also the fastest way to lose a credit committee, because long-term supply contracts price near the bottom of that band and anyone who has financed this sector knows it. Model at $300, show the business works, and let the upside be upside.

3. Treating it as acquisition when it is retention

Plans arrive with a donor acquisition budget and a target donor count. Neither is the driver. The driver is cost per litre, and cost per litre falls because repeat donors process faster, defer less, and carry no acquisition cost. A centre with 4,000 donors donating twice a month beats a centre with 7,000 donors donating once a quarter, on every line. Build the retention curve, then derive the donor count from it — not the other way round.

4. Under-costing replacement fluid on the therapeutic side

Covered in the revenue section but it earns a second mention because it is that common. Albumin and fresh frozen plasma routinely cost more than your nursing roster. A model with a generic "supplies at 8% of revenue" line is not conservative or aggressive — it is simply not a model of this business. Break replacement fluid out, tie it to procedure volume, and sensitise it.

5. Siting on population instead of catchment

A map showing 400,000 people within 30 minutes proves nothing. What matters is the donor pool the IQPP community-based donor standard actually permits you to draw from, the state licensure regime you land in, and whether four competitors are already inside that radius — with 1,000-plus certified centres nationally and CSL alone running 300-plus, they very often are. Site selection in this business is a regulatory and competitive analysis wearing a demographics report's clothes.

Inside a Finished Apheresis Plan

Preview the structure and financial outputs a buyer receives. These mockups use a UK therapeutic and cell-collection model, distinct from the US source plasma economics above, to show how the same template flexes across the models.

Everything the Template Gives You

Every Avvale business plan template includes these sections, pre-structured for your industry:

• Executive Summary — Your business at a glance, written to hook investors in 60 seconds
• Company Overview — Legal structure, ownership, location, and founding story
• Industry Analysis — Market size, growth trends, and the regulatory picture
• Customer Analysis — For apheresis, that means the fractionator, the payer, or the sponsor — whichever model you chose
• Competitor Analysis — Catchment mapping against the certified centres already in your radius
• Marketing Plan — Donor retention, referral development, or sponsor business development by model
• Operations Plan — Chair-hour throughput, staffing structure, quality system, and licensure milestones
• Management Team — Founder bios, medical director, quality lead, and key hires planned

The optional Financial Forecast add-on (included in our $300/£250 and $1,000/£800 packages) provides a 5-year Excel model with income statement, cash flow, balance sheet, break-even analysis, and startup capital requirements — with the licensure gap, cost-per-litre maturation curve, and replacement-fluid sensitivity built in as live drivers rather than static cells.

If you are researching adjacent categories, our blood bank business plan template covers the transfusion-services side of the same regulatory framework, and the patient blood management business plan template addresses the hospital-programme model. Full worked client examples are in our case study library, and if you would rather hand the whole thing over, our market research and content service covers the research and narrative for $300 / £250.

Questions Founders Ask Before They Commit

Useful Links & Resources

These links were preserved from the live page so important references and partner links are not lost during the page refresh.

• Facebook
• Instagram
• LinkedIn
• YouTube