In CAR-T cell release testing, CAR expression positivity is the core QC readout. But here's something every flow cytometry lab has run into: why does the same CAR-T sample yield different positivity rates with different kits? The problem may not lie in your hands — it may be built into how the antigen-fluorophore reagent itself is manufactured.
Most conventional CAR-T detection kits on the market rely on chemical conjugation, which carries inherent shortcomings: the fluorophore may attach directly at the antigen's binding epitope, or too many fluorophores may crowd the antigen and create steric hindrance — interfering with antigen-CAR interaction and ultimately underestimating CAR positivity, adding data variability, and hurting reproducibility.
● Principle: Small-molecule fluorophores are randomly attached to lysine, cysteine, and other residues on the antigen protein via NHS ester, maleimide, and similar chemistries.
● Limitation: This is the industry's classic, low-cost labeling approach — but where the dye attaches and how many copies attach are decided purely by reaction probability. The resulting product is highly heterogeneous: some antigen molecules carry no fluorophore at all, while others carry 2 or even 5. Labeling states vary widely within a single batch, and performance differs even more between manufacturing batches.
● Principle: Labeling still happens through in-vitro chemistry, but the fluorophore's attachment site is pre-defined — for example, the protein terminus, Fc-region glycans, or engineered cysteine residues. Directional conjugation can also be achieved enzymatically (transglutaminase, Sortase A, glycosyltransferase) or via click chemistry — the technologies the industry knows as "terminal labeling" and "site-specific protein labeling."
● Limitation: Compared with Generation 1, this approach controls both the labeling site and the labeling copy number, largely preserves the antigen's native conformation, and meaningfully improves lot-to-lot consistency. Yet it never escapes the underlying logic of in-vitro chemical modification, and its technical drawbacks are unavoidable: reaction conditions (pH, organic solvents, reducing agents) can disrupt protein conformation; free dye and reaction by-products must be removed; labeling efficiency fluctuates during scale-up; and every batch requires full performance re-verification.
So how do we eliminate CAR positivity bias and signal distortion? The third generation offers a new answer — fusion at the genetic level.
To address these industry pain points, AlpVHHs leverages its mature nanobody discovery and protein conjugation platform to break through existing detection bottlenecks, launching the world's leading third-generation CAR-T Test kit.
● Principle: No in-vitro chemical conjugation at all. Using genetic engineering, the target antigen gene is fused directly to the PE/APC fluorescent protein gene, and the fusion protein is produced through a recombinant expression system.
What this delivers:
● Fixed labeling site, fully exposed epitopes. Labeling is positioned at the protein terminus and never randomly blocks the antigen's binding region. The antigen retains its native conformation and full biological activity, binding CAR molecules stably and efficiently — no more PE obstructing CAR recognition.
● A strict 1:1 molar ratio — truly quantitative. Antigen and PE exist in fixed stoichiometry, eliminating the over- or under-conjugation of chemical approaches. Signal intensity is stable and supports precise quantification, whereas randomly labeled conventional products are inherently non-quantitative.
● Ultra-high affinity, higher dilution tolerance, clean background. The recombinant fusion product achieves nM–pM affinity, far beyond conventional chemically conjugated products. The reagent can be diluted further, shows low non-specific binding, delivers a clean fluorescent background, and separates positive and negative populations more distinctly.
● Superior lot-to-lot consistency, high stability, long shelf life. Free from batch-to-batch fluctuations of in-vitro chemistry, the molecules are highly uniform with minimal lot variation. The fusion protein is stable and storage-tolerant, reducing experimental reproducibility risk — ideal for long-term cell therapy R&D and GMP quality control.
Take the APK151 Human CD19-PE CAR TEST Kit from AlpVHHs — publicly available product data:
Metric | Result |
Conjugation format | Recombinant CD19–PE fusion protein, antigen: PE = 1:1 |
Detection sensitivity | 99.7% positivity on CAR-Jurkat cells |
Non-specific background staining | 0.094% |
PBMC non-specific staining | 0.027% |
High-dilution performance | Bright, reliable signal at 1:800 dilution |
| Multi-cell-line validation | CAR-Jurkat 99.7% / HEK293T 97.1% / HEK293F 99.2% |
Among these numbers, the 99.7% positivity isn't the part worth pondering — many reagents can approach 100%. What deserves attention is the background and the dilution tolerance: 0.094% non-specific background staining, 0.027% PBMC non-specific staining, and a signal that stays bright at 1:800 dilution.
Low background means the conjugation process leaves the protein untouched. High dilution tolerance means the fluorescence on every molecule is defined and efficient. Together, they point to one conclusion: the third-generation CAR TEST Kit eliminates the signal bias and distortion inherent in traditional detection — at the root.
First, can it quantify? "Telling positive from negative" and "quantifying" are two different things. 1:1 fusion > site-specific conjugation > random conjugation — that ranking reflects how much you can trust the linearity between fluorescence signal and CAR molecule number.
Second, look at background — not positivity. A 99% positivity rate is a passing grade, not a competitive edge. The real dividing line lies in the negative population: whether background is 0.09% or 0.9% determines whether you can hold the gate in low-expression samples.
Third, look at dilution tolerance. If a reagent still delivers a bright, clean signal above 1:400 dilution, its labeling density and activity are both on target. Conversely, a signal that collapses as dilution increases usually signals labeling-efficiency or activity problems — and it also means a higher cost per test.
Fourth, look at lot-to-lot consistency. Release-testing data must be comparable across lots, months, and years. Doing it beautifully once isn't hard; the hard part is matching the same data three years later. On this point, what is defined by genetics holds a natural advantage.
