Engineering Guide · Vendor-Neutral

MabSelect PrismA vs Amsphere A3 vs Toyopearl AF-rProtein A: Protein A Resin Comparison

Published June 9, 2026 · 14 min read
MabSelect PrismA, Amsphere A3, and Toyopearl AF-rProtein A HC-650F Protein A resin DBC comparison Dynamic binding capacity at 4-6 min residence time CHO mAb capture · 10% breakthrough 80 60 40 20 0 DBC (mg IgG / mL resin) 77 PrismA Cytiva · agarose Z-domain mutein 65 Toyopearl Tosoh · HC-650F Methacrylate · 45 µm 60.5 SuRe LX Cytiva · legacy Pre-2018 reference 54 Amsphere A3 JSR · methacrylate Surface-modified Alkaline stability (150 cycles · 0.5 M NaOH between every cycle) ~90% DBC retained 80% @ 40 cyc ~80% DBC retained ~60% DBC retained
Figure 1: Manufacturer-published dynamic binding capacity (DBC) for the four leading Protein A resins on CHO mAb capture at 4-6 minute residence time and 10% breakthrough. MabSelect PrismA leads on raw DBC; Toyopearl AF-rProtein A HC-650F is the strongest methacrylate alternative. The alkaline-stability annotation summarises a 150-cycle aggressive-cleaning lifetime study where each cycle ended in a 0.5 M NaOH CIP.
Quick Verdict

MabSelect PrismA is the default modern Protein A resin for CHO mAb capture, delivering ~77-80 mg IgG/mL DBC at 6 minute residence time and roughly 90% DBC retention after 150 cycles of 0.5 M NaOH CIP. Pick Amsphere A3 when methacrylate matrix rigidity, single-supplier strategy, or chemical robustness outweighs the lower raw DBC. Toyopearl AF-rProtein A HC-650F is the strongest methacrylate alternative with 200+ cycles at 0.1 M NaOH. Keep MabSelect SuRe LX only on locked legacy commercial processes where a comparability exercise is harder to justify than the productivity gap.

Key differences at a glance

Side-by-side comparison

Factor MabSelect PrismA Toyopearl HC-650F MabSelect SuRe LX Amsphere A3
Vendor Cytiva Tosoh Bioscience Cytiva JSR Life Sciences
Base matrix High-flow agarose Methacrylate (HW-65F) High-flow agarose Methacrylate, surface-modified
Mean particle size ~60 µm 45 µm ~85 µm ~50 µm
Ligand Alkali-hardened Z-domain mutein Multipoint-attached rProtein A Alkali-stabilised Z-domain Engineered alkali-stable mutein
DBC, 6 min residence (mg IgG/mL) 77-80 >65 60.5 ~54
DBC, 2.4 min residence (mg IgG/mL) ~60 >50 (at 2 min) ~43 ~40
0.1 M NaOH lifetime (cycles) >200, <5% DBC loss 200+ CIP cycles ~150, <15% DBC loss 200 exposures, <10% loss
0.5 M NaOH lifetime (150 cycles, DBC retained) ~90% 80% at 40 cycles ~80% at 150 cycles ~60%
Pressure-flow ceiling Moderate (agarose softens) High (rigid methacrylate) Moderate High (rigid methacrylate)
Typical Protein A leakage <25 ppm Single-digit ppm <25 ppm Single-digit ppm
BLA precedent (CHO mAb) Default for post-2020 platforms Established Extensive legacy Growing post-2022

Values reflect published vendor datasheets and peer-reviewed lifetime studies (McCaw et al. 2014; Cytiva PrismA performance brief; JSR Amsphere A3 benchmarking). Your clone- and titer-specific numbers will differ.

MabSelect PrismA in detail

MabSelect PrismA is the default modern Protein A resin for CHO monoclonal antibody capture and has been the reference choice for almost every commercial-stage program initiated since 2020. It combines an optimised high-flow agarose base matrix (smaller bead than the legacy 85 µm MabSelect SuRe) with a genetically engineered alkali-hardened Z-domain Protein A mutein, delivering a dynamic binding capacity of 77-80 mg human IgG per mL of resin at a 6 minute residence time and roughly 60 mg/mL at a 2.4 minute residence time. That is a 27% step up over MabSelect SuRe LX and a 53% step up over the original MabSelect SuRe on the same column geometry, which is the productivity argument that broke the SuRe family's installed-base lock.

How it works

The ligand is a recombinant protein derived from the B-domain of Staphylococcus aureus Protein A, further alkali-stabilised by site-directed mutagenesis that exchanges asparagine and other base-sensitive residues for more stable alternatives. Each ligand is multipoint-attached to the agarose bead surface, which reduces leaching under elution and CIP conditions and extends column life. The smaller, optimised bead size and tuned pore structure of the high-flow agarose deliver fast intraparticle mass transfer, which is why DBC holds up at the 2-3 minute residence times that intensified upstream demands. The standard operating window is bind/wash at pH 7.0-7.4, elute at pH 3.5-4.0, and 0.1 M NaOH CIP between cycles, escalating to 0.5 M NaOH for stubborn host-cell-protein deposits — typically tracked using the DBC decay model in our Protein A resin lifetime guide.

When MabSelect PrismA wins

PrismA wins whenever capture-step productivity matters — intensified fed-batch CHO mAb at 8-12 g/L titer, perfusion-with-batch-capture campaigns, capacity-constrained commercial plants, and almost any new platform program where the regulatory comparability cost is already absorbed by the new-product BLA. The PrismA X variant (DBC up to ~82 mg/mL at 6 min) extends this further for processes pushing the capture column to its DBC ceiling. PrismA also wins on alkaline-cleaning lifetime: roughly 90% DBC retained after 150 cycles of 0.5 M NaOH CIP, well above the 60% Amsphere A3 retains under the same protocol (Purolite benchmarking, 2021).

Amsphere A3 and the methacrylate alternatives

Amsphere A3 (JSR Life Sciences) and Toyopearl AF-rProtein A HC-650F (Tosoh Bioscience) are the two leading rigid-polymer Protein A resins competing for the second-source slot behind the PrismA Protein A resin. Amsphere A3 places an engineered alkali-resistant Protein A mutein on a surface-modified methacrylate base bead (~50 µm), and pairs a DBC of roughly 54 mg/mL polyclonal IgG with low non-specific binding through the hydrophilised surface. Toyopearl AF-rProtein A HC-650F immobilises an enhanced rProtein A ligand by multipoint attachment to the Toyopearl HW-65F methacrylate bead (45 µm), reaching a DBC of more than 65 g/L at 5 minute residence time and more than 50 g/L at a 2 minute residence time. Both resins solve a real engineering problem: methacrylate beads do not compress like agarose, hold tall beds and high linear velocities, and tolerate longer aggressive-cleaning programs without bed-structure damage.

How they work

The methacrylate base bead used in these two Protein A resins is a rigid synthetic polymer with a tightly controlled particle size distribution, which is the main reason both Protein A resins run reliably above 300 cm/h linear velocity without column compression. Amsphere A3's surface chemistry adds a hydrophilic modification on top of the methacrylate, which suppresses non-specific binding of host-cell proteins and lowers HCP carry-over into elution. Toyopearl AF-rProtein A HC-650F multipoint-attaches its rProtein A ligand to the bead via the same ligand-immobilisation principle as PrismA, which gives it 200+ cycles at 0.1 M NaOH and 80% DBC retention after 40 cycles of 0.5 M NaOH per the published Tosoh datasheet. Both resins use the same bind-wash-elute-CIP cycle as the agarose family, which keeps the buffer train and the harvest-to-elution timing unchanged when swapping in or out.

When the methacrylate resins win

The methacrylate Protein A resins win when pressure-flow at tall beds matters more than top-end DBC. That includes high-flow campaigns where the column is pushed at >300 cm/h linear velocity (PrismA agarose softens under these conditions), CMO multi-product suites that use methacrylate elsewhere in the polishing train and want one base-matrix family across the resin fleet, and chemistry-heavy cleaning programs that escalate to 0.5 M NaOH plus elevated temperature. Amsphere A3 also wins for sites already on JSR raw materials or those that want a Japanese supplier alongside Cytiva. Toyopearl AF-rProtein A HC-650F wins for sites on Tosoh process media elsewhere (ion exchange, mixed-mode) where a single-vendor strategy reduces qualification overhead.

Pros and cons

MabSelect PrismA (Cytiva)

Advantages

  • Highest DBC in the segment at 4-6 min residence time (77-80 mg/mL) and 50% higher than SuRe at 2 min
  • ~90% DBC retained at 150 cycles of 0.5 M NaOH CIP — the strongest aggressive-cleaning lifetime in the comparison
  • Default reference for post-2020 CHO mAb BLA filings; lowest regulatory risk for new programs
  • PrismA X variant available (DBC ~82 mg/mL at 6 min) for capacity-constrained capture steps

Disadvantages

  • Agarose base matrix compresses under sustained high pressure or tall beds — limits linear velocity ceiling
  • Cytiva single-supplier exposure — pandemic-era allocation showed the risk concentration
  • Premium pricing per litre of packed resin (Cytiva has not commoditised the segment)
  • For legacy SuRe-based commercial processes, the comparability cost of switching is often higher than the productivity gain

Amsphere A3 (JSR Life Sciences)

Advantages

  • Rigid methacrylate base bead — no column compression at high linear velocity, holds tall beds
  • Surface-hydrophilised matrix lowers non-specific binding and HCP carry-over into elution
  • Engineered alkali-stable Protein A ligand; <10% DBC loss over 200 exposures of 0.1 M NaOH
  • Provides a credible second source to Cytiva for supply-chain resilience

Disadvantages

  • Lower DBC than PrismA (~54 vs 77 mg/mL at 6 min) — needs more cycles or a larger column at fixed harvest
  • ~60% DBC retention after 150 cycles of 0.5 M NaOH — half as durable as PrismA on the aggressive CIP profile
  • Smaller installed-base of BLA precedents than PrismA or the SuRe family
  • Long contact time with 0.5 M NaOH measurably reduces DBC — keep CIP holds short

Which Protein A resin should you choose?

Pick the Protein A resin based on the dominant constraint at the capture step: titer, regulatory stage, second-source strategy, or chromatography-skid pressure-flow envelope.

New mAb platform, intensified upstream

You are setting up a platform for 8-12 g/L intensified fed-batch or perfusion-with-batch-capture. The capture column is the binding constraint. Take the highest DBC and the longest 0.5 M NaOH lifetime.

Choose MabSelect PrismA

Legacy commercial CHO mAb on MabSelect SuRe

The reference process is on SuRe or SuRe LX and the BLA is approved. A resin swap forces full comparability. Stay on the locked resin unless capacity-constrained.

Stay on MabSelect SuRe / LX

Tall-bed, high-flow, methacrylate fleet

Skid runs >300 cm/h linear velocity, columns above 25 cm bed height, and the rest of the chromatography train (CEX, AEX, mixed-mode) is on Tosoh process media. Match the matrix family.

Choose Toyopearl AF-rProtein A HC-650F

Second-source strategy, supply-chain resilience

Cytiva allocation risk is the binding constraint. You need a qualified alternative that delivers similar leakage and CIP behaviour without rebuilding the buffer train.

Choose Amsphere A3 (JSR)

Real-world use cases

Typical setups where bioprocess teams have converged on one Protein A resin or another.

CHO mAb, 2,000 L intensified fed-batch
MabSelect PrismA on a 1.5 m diameter axial column

A 9 g/L harvest delivers ~18 kg per batch. PrismA at 75 mg/mL DBC and a 4 min residence time clears the harvest in 4-5 capture cycles on a 350 L packed bed. CIP at 0.1 M NaOH between cycles and 0.5 M NaOH every tenth cycle; column lifetime >150 cycles.

Legacy commercial CHO mAb
MabSelect SuRe LX, locked reference process

2,000 L N-stage at 5 g/L titer, 60.5 mg/mL DBC at 6 min, ~6 capture cycles per batch on the existing 500 L column. Resin is replaced every ~150 cycles. No move to PrismA until the next major facility refit or capacity expansion.

CDMO multi-product suite
Toyopearl AF-rProtein A HC-650F in a Tosoh-aligned fleet

CDMO chromatography skid already runs Toyopearl GigaCap S, Toyopearl GigaCap Q, and Toyopearl PPG. Matching the capture step to the same methacrylate family removes a base-matrix variable from cleaning validation and lets the skid run at 400 cm/h capture loading.

Second-source qualification
Amsphere A3 as the JSR backup to PrismA

Commercial mAb facility with two qualified capture resins — PrismA primary, Amsphere A3 backup. Identical buffer recipes, slightly larger column for the backup to compensate for the DBC gap. Supply-chain team triggers the backup when Cytiva allocation tightens.

Sizing a capture column or modelling resin lifetime?

The Chromatography Calculator sizes capture and polishing columns from DBC, residence time, and harvest mass — and pairs with the Protein A resin lifetime model for cost per gram of mAb purified across the resin lifetime.

Open the Chromatography Calculator

Cost and lifecycle considerations

Protein A resin TCO is dominated by cycles delivered, not list price

Process-grade Protein A resins all sit in the $12,000-$18,000 per litre band at process-media list pricing, with negotiated rates often well below list at scale. What separates the per-gram economics is total grams purified per litre of resin over its useful life — a function of DBC, cycles delivered before DBC drops below 80% of new, and harvest titer. PrismA's higher DBC and longer 0.5 M NaOH lifetime translate directly into more grams per litre of resin and a lower contribution to downstream COGS.

Take a 2,000 L bioreactor at 8 g/L harvest (16 kg of mAb per batch) and a 350 L packed Protein A column. At 75 mg/mL DBC and 80% loading, PrismA delivers ~21 kg of capture capacity per cycle and clears the batch in one cycle. Across 150 cycles before retirement that is 2,400 kg per litre of resin — at a list of $15,000/L that puts the resin contribution at roughly $6 per gram of mAb captured, before factoring in buffer and labor.

At the same titer, Amsphere A3 at 54 mg/mL DBC needs ~1.4 cycles per harvest (or a 30% larger column) to clear the same mass, and the 0.5 M NaOH aggressive-cleaning lifetime is shorter — meaning more frequent replacement at fixed cleaning intensity. The per-gram resin cost ends up roughly 40-60% higher than PrismA on a like-for-like 0.5 M NaOH CIP program, though closing on parity if the CIP escalates only to 0.1 M NaOH between cycles with 0.5 M reserved for the regen-and-strip every 10th cycle.

Cost component MabSelect PrismA Toyopearl HC-650F MabSelect SuRe LX Amsphere A3
Indicative list price ($/L packed)$15,000-$18,000$12,000-$15,000$12,000-$14,000$12,000-$15,000
Capture cycles per 16 kg harvest~1.0~1.2~1.3~1.4
Useful cycles at 0.5 M NaOH CIP~150~120~120~100
Grams of mAb captured per litre of resin~2,400 kg/L~1,700 kg/L~1,500 kg/L~1,200 kg/L
Resin contribution to per-gram COGS (relative)Baseline (1.0x)1.3x1.4x1.6x

Vendor landscape

The four Protein A resins above own most of the commercial CHO mAb capture market, but several credible Protein A resin alternatives matter at the margins.

Cytiva — agarose, Z-domain mutein platform

Methacrylate-base alternatives

Other alkali-stable contenders

Frequently asked questions

Which Protein A resin has the highest dynamic binding capacity?
MabSelect PrismA has the highest dynamic binding capacity among the commodity Protein A resins, at roughly 77-80 mg human IgG per mL of resin at a 6 minute residence time. That is about 27% above MabSelect SuRe LX (60.5 mg/mL) and 53% above the original MabSelect SuRe (50.3 mg/mL) on the same column geometry. Toyopearl AF-rProtein A HC-650F sits at >65 g/L at 5 min residence time, and Amsphere A3 is in the same band as MabSelect SuRe LX. The PrismA X variant pushes DBC to ~82 mg/mL.
How alkaline stable is MabSelect PrismA versus Amsphere A3?
Both resins are alkali-stable enough for 0.5 M NaOH cleaning, but PrismA holds capacity longer. In a 150-cycle lifetime study with 0.5 M NaOH CIP between every cycle, MabSelect PrismA retained about 90% of its initial mAb DBC, while Amsphere A3 retained about 60% over the same 150 cycles (per Purolite's published benchmark). At the milder 0.1 M NaOH set point, both resins show <10% DBC loss over hundreds of CIP cycles. PrismA's ligand uses additional site-directed alkali-hardening on top of the Z-domain that Amsphere A3 also engineers, which accounts for the gap at aggressive cleaning concentrations.
What is the difference between an agarose and a methacrylate Protein A resin?
Agarose (Cytiva MabSelect family, Purolite Praesto, Repligen) and methacrylate or synthetic polymer (JSR Amsphere, Tosoh Toyopearl, Merck Eshmuno) describe the base bead the Protein A ligand is immobilised on. Agarose beads compress under high column pressure and have softer flow characteristics, but they have decades of regulatory precedent and tend to deliver higher dynamic binding capacity per mL at short residence times in CHO mAb capture. Methacrylate beads are mechanically rigid, hold tall columns and high flow rates without compression, and run reliably above 300 cm/h linear velocity. Pore structure and ligand density matter more than the matrix family in raw DBC terms (McCaw et al. 2014).
Should I move my legacy MabSelect SuRe process to MabSelect PrismA?
If the program is in late clinical or commercial and the reference SuRe process is stable, the regulatory comparability cost usually outweighs the productivity benefit. If the program is in early development or you are platform-defining for a portfolio, switch to PrismA. The 1.5x DBC at 4-6 min residence time directly translates into a smaller column (or fewer cycles per harvest), lower buffer volumes, and longer resin lifetime under 0.5 M NaOH CIP. Most CHO mAb platforms set up post-2021 default to PrismA. The economics also tilt sharply toward PrismA when an intensified fed-batch upstream is pushing harvest titer above 8 g/L.
What residence time do MabSelect PrismA and Amsphere A3 need?
Both modern resins are engineered to work at sub-4 minute residence times, which is the central reason they exist. MabSelect PrismA reaches roughly 60 mg IgG/mL DBC at 2.4 min — comparable to MabSelect SuRe LX at 6 min — which means you can load the column 2-3x faster without losing capacity. Amsphere A3 and Toyopearl AF-rProtein A HC-650F are similarly tolerant of short residence times. The legacy MabSelect SuRe loses capacity rapidly below 4 min and is the limiting factor in many 2015-era harvest schedules.
How much Protein A ligand leakage should I expect?
MabSelect PrismA is specified at <25 ppm Protein A leakage relative to the eluted mAb under standard mAb capture conditions. Amsphere A3 and Purolite Praesto Jetted A50 publish leakage data in the same single-digit-ppm band. Toyopearl AF-rProtein A HC-650F leakage is similarly low. All four resins clear the typical 10 ng Protein A per mg mAb final drug substance specification with the standard polishing-step ion exchange/SEC train. Leakage spikes are almost always a sign of column overloading, low elution pH, or end-of-life resin rather than the ligand itself — track them against the DBC-decay model in the Protein A resin lifetime guide.
What is MabSelect PrismA X and when do I need it?
MabSelect PrismA X is the same PrismA ligand on a smaller-pore-tuned agarose bead, with published DBC up to about 82 mg human IgG per mL of resin at 6 min residence time. It is targeted at capacity-constrained capture steps where a few more grams per cycle change the harvest schedule — typically intensified fed-batch CHO mAb at 8-12 g/L titer where the upstream gain has pushed the capture column to its DBC ceiling. For 4-6 g/L standard fed-batch titers, plain PrismA still has headroom and the upgrade is hard to justify.
Which Protein A resin is best for high-titer intensified fed-batch?
For intensified fed-batch CHO mAb at 8-12 g/L titer, MabSelect PrismA (or PrismA X) is the default capture choice — the high DBC and 0.5 M NaOH alkaline stability let you keep the capture column compact and recycle it through enough cycles per harvest to clear the upstream output. Amsphere A3 is the credible methacrylate alternative when chemical robustness, pressure-flow at tall beds, or single-supplier strategy matters more than the top-end DBC. Both pair well with the N-1 perfusion seed and inline Raman feed strategies typical of intensified fed-batch.

Resources and references