Cell Viability: Definition, Assays, Acceptance Criteria and Limits of Interpretation
Cell viability is the fraction of live cells in a sample. It is usually measured by dye exclusion (whether the cell membrane is intact) or by metabolic assays such as tetrazolium, resazurin and ATP. This page covers how the main methods work, how viability is reported in cell therapy and fat grafting research, the 70% release benchmark commonly used for cell therapy products, and why results depend on the method used and the timing of the measurement.
What cell viability means
Cell viability describes how many cells in a population are alive at the moment of measurement. It is usually reported two ways: as a viable cell count (an absolute number or concentration) and as a viability percentage (viable cells divided by total counted cells, multiplied by 100). Viability is not measured directly. Each assay picks a marker linked to being alive, such as an intact plasma membrane, active enzymes, reducing capacity or intracellular ATP, and treats that marker as a stand-in for life. The NCBI Assay Guidance Manual notes that plate-reader viability assays work by measuring "a marker activity associated with viable cell number" [1].
Research laboratories use viability to measure the effects of cell proliferation, to screen compounds for cytotoxicity, and as an internal control in other cell-based assays [1]. For cell-based medical products, the viable cell count and viability percentage are quality-control attributes. They are checked before a product is released and reported in studies of cryopreservation and adipose-derived cell preparations [2][3].
Main types of cell viability assays
Researchers commonly sort the methods into two families. Membrane integrity assays (for example, trypan blue or propidium iodide staining) find cells whose membranes no longer keep dyes out. Functional assays (for example, MTT, MTS, XTT, BrdU and Alamar blue/resazurin) measure metabolic or proliferative activity [7].
Dye exclusion and membrane integrity
A live cell with an intact membrane keeps certain dyes out, while a damaged cell takes them up. In trypan blue exclusion, cells are viewed under a microscope or in an automated counter, and blue-stained cells are counted as non-viable. Fluorescent DNA-binding dyes such as propidium iodide (PI), 7-aminoactinomycin D (7-AAD) and DAPI work on the same principle. They cannot cross intact membranes and give little or no fluorescence until they bind DNA. They suit live, unfixed cells and cannot be used after fixation [5]. When samples must be fixed, amine-reactive viability dyes are used instead. These bind proteins irreversibly and stain dead cells more brightly because more of the dye gets inside [5]. Automated counters often pair acridine orange with propidium iodide (AO/PI), which stains nucleated live and dead cells separately [2].
Metabolic and ATP-based assays
Plate-based assays measure a population average instead of classifying individual cells. The NCBI Assay Guidance Manual chapter by Riss and colleagues (first published 2013, updated 2016) describes the main formats [1]:
- Tetrazolium reduction (MTT, MTS, XTT, WST-1): viable cells convert the reagent into a coloured formazan. This typically takes 1–4 hours at 37 °C, and MTT also needs a solubilisation step [1].
- Resazurin reduction: viable cells convert resazurin to fluorescent resorufin over a similar 1–4 hour incubation. It is slightly more sensitive than tetrazolium methods [1].
- Protease substrate (GF-AFC): a live-cell protease produces a fluorescent signal in about 30 minutes to 1 hour. The reagent is reported to be non-toxic, so it can be combined with other assays [1].
- ATP detection: a luciferase reaction produces light in proportion to ATP. The chapter reports detection of fewer than 10 cells per well, which makes it the most sensitive of the formats described [1].
- Real-time assays: a non-lytic luciferase format can track live cells in culture for up to 72 hours [1].
Flow cytometry and apoptosis staining
Flow cytometry combines viability dyes with surface markers, so viability can be reported for a defined cell subpopulation. Membrane-integrity dyes do not stain cells in early apoptosis, whose membranes are still intact. Combining Annexin V with PI or 7-AAD separates early apoptotic cells from late apoptotic or necrotic cells [5][8].
Comparison of common methods
| Method | What it detects | Output | Known limitations |
|---|---|---|---|
| Trypan blue exclusion | Membrane integrity (brightfield) | Viable count and % per cell | Tends to give higher viability than AO/PI; long dye exposure affects results; hard to separate cells from debris [2] |
| AO/PI fluorescence counting | Nucleated live vs dead cells | Viable count and % | Depends on instrument and gating; reported as robust for low-viability samples [2] |
| 7-AAD / PI flow cytometry | Membrane integrity plus phenotype | % viable within subsets | Misses early apoptosis; not compatible with fixation [5] |
| Tetrazolium / resazurin | Reducing (metabolic) activity | Relative signal per well | Reducing compounds or fluorescent test articles interfere; long exposure is toxic to cells [1] |
| ATP luminescence | Intracellular ATP | Relative signal per well | Lyses cells; reflects metabolic state as well as cell number [1] |
Viability as a quality-control measure in cell therapy
In cell therapy manufacturing, viability is a routine release test. A 2024 review in Cell Transplantation states that "the FDA advocates the establishment of minimum cell viability criteria, which are generally set to above 70%" [2]. This practice is associated with FDA's April 2008 final guidance on chemistry, manufacturing and control (CMC) information for human somatic cell therapy INDs [9]. A 2025 review of natural killer (NK) cell products found that phase 1 and 2 trials in the US and Europe used viability of ≥70% as the acceptance criterion across the board [4].
Individual products can use higher limits. For tisagenlecleucel, a CAR T-cell therapy, the early clinical trials used a release specification of 70% or more viable cells, and the commercial specification is at least 80%. Viability was tested on a thawed sentinel vial by trypan blue exclusion or dual-fluorescence automated counting [3]. In a retrospective analysis published in Blood in 2019, 15 of 123 paediatric ALL products (12%) and 4 of 25 DLBCL products (16%) had viability below 80%. The authors found no statistically significant association between product viability and complete remission rate [3].
Cell viability in cryopreservation
Freezing and thawing damage cells, and some of that damage appears only after a delay. In a 2020 study of human bone marrow-derived mesenchymal stromal cells, fresh cells showed 98.4–99.2% viability. Viability of cryopreserved cells dropped most at 2–4 hours after thawing and recovered to 93.1–97.6% by 24 hours. However, metabolic activity and adhesion were still impaired at 24 hours [8]. The authors concluded that a 24-hour period is not enough for full recovery [8]. So a viability percentage measured right after thawing may not reflect later cell loss or function. For this reason, post-thaw studies often report viable cell recovery and functional assays alongside viability.
Cell viability in fat grafting and adipose-derived cells
In fat grafting research, adipocyte viability and the viability of the stromal vascular fraction (SVF) are used to compare harvesting and processing techniques. They are measurements taken in the laboratory, not measurements of graft retention in the patient. Mature adipocytes are large and fragile, which complicates standard methods. A 2023 study in Plastic and Reconstructive Surgery Global Open warned that with microscopy-based trypan blue, "dead cells without nuclei or mitochondrial activity may be interpreted as live cells," and named this as a possible cause of the wide range of reported results [7]. Using the functional Alamar blue assay, the same study reported median adipocyte viability of 85.46% for ultrasound-assisted liposuction and 73.16% for suction-assisted liposuction [7]. Such figures depend on the method and are not directly comparable with figures from other assays.
Limits of interpretation and standardisation
- Method dependence: different counting methods can give different cell counts and viability results for the same product. Trypan blue often reads higher than AO/PI [2].
- Marker is not function: membrane dyes can miss early apoptosis [5]. Different dyes can enter dying cells through different routes and at different times [6]. A viable cell is not necessarily a functional or potent one [8].
- Confidence in assays: a 2020 survey cited in the 2024 review found that only 18% of respondents had high confidence in cell viability assays [2].
- Standards: ISO 20391-1:2018 and ISO 20391-2:2019 cover cell counting methods and how to evaluate their performance. A separate ISO standard for cell viability methods (ISO/CD 8934) was reported to be in development [2].
Because of these limits, viability results are easiest to interpret when the report names the assay, the instrument, the time since thawing or processing, and whether debris and non-nucleated cells were excluded.
References
- journalCell Viability Assays (Assay Guidance Manual) — National Center for Biotechnology Information (NCBI Bookshelf), Riss TL et al. · Evidence page
- journalChallenges of Cell Counting in Cell Therapy Products — Cell Transplantation (Liu M et al.) · Evidence page
- journalCAR T cell viability release testing and clinical outcomes: is there a lower limit? — Blood (Chong EA et al.), 134(21):1873–1875 · Evidence page
- journalQuality standards for NK cell immunotherapies — Frontiers in Bioengineering and Biotechnology (von Werz et al.) · Evidence page
- academicCell Viability – Flow Cytometry Facility — University of Toronto, Temerty Faculty of Medicine Flow Cytometry Facility · Evidence page
- journalPlasma membrane permeabilization following cell death: many ways to dye! — Cell Death Discovery (De Schutter E et al.) · Evidence page
- journalAssessing Adipocyte Viability and Surgeons' Work Efficiency by Comparing Different Liposuction Methods — Plastic and Reconstructive Surgery Global Open (Urbonas T et al.) · Evidence page
- journalQuantitative assessment of the impact of cryopreservation on human bone marrow-derived mesenchymal stem cells: up to 24 h post-thaw and beyond — Stem Cell Research & Therapy (Bahsoun S, Coopman K, Akam EC) · Evidence page
- governmentContent and Review of Chemistry, Manufacturing, and Control (CMC) Information for Human Somatic Cell Therapy Investigational New Drug Applications (INDs) — U.S. Food and Drug Administration · Evidence page