Getting Cytokine Profiling Right

Cytokine profiling has become a routine tool in many research workflows, but getting trustworthy results takes more than running the assay correctly. In this Q&A article, one theme comes up again and again: cytokine profiling results are only as reliable as the sample handling that precedes them. Pre-analytical variables such as the collection tube and technique used, and how quickly and consistently a sample is processed and stored, can shift measured cytokine levels before an assay ever begins.

Reproducibility introduces its own set of variables to control, from reagent and lot consistency, calibrator preparation, and instrument performance to operator technique and how the resulting data is analyzed. Left unchecked, these same factors are behind many of the most common mistakes researchers make.

To help researchers navigate these challenges, from sample collection through data interpretation, we asked experts at Luminex, Cytek Biosciences, Promega, and Bio-Rad Laboratories, representing both immunoassay-based and flow cytometry-based approaches to cytokine measurement, to share tips and insights.

What guidelines do you suggest for proper sample collection, storage, and preparation to maintain data integrity?

Vanitha Margan, Global Product Manager, Immunoassays, Bio-Rad Laboratories

"For cytokine profiling, consistency in sample handling is critical. While the most common samples for cytokine profiling are blood-derived samples like serum or plasma, cytokine production by cells such as cultured cell lines, PBMCs, tissue isolates (splenic cells), or even CSF, can also be done. For serum, plasma, CSF, tissue lysates, or culture supernatants, aliquot samples before freezing at −80°C to avoid repeated freeze–thaw cycles. For viable cell-based studies, cells should be cryopreserved using appropriate media and controlled-rate freezing so they retain functional cytokine responses after thawing."

Mike Blundell, Ph.D., Global Product Manager, Antibodies, Bio-Rad Laboratories

"For intracellular cytokine staining (ICS) by flow cytometry, cells must be viable and typically undergo ex vivo stimulation in the presence of a protein transport inhibitor (e.g., brefeldin A or monensin) to trap cytokines within the cell before fixation and permeabilization. Whole blood or freshly isolated PBMCs work best, though well-cryopreserved PBMCs with confirmed viability can also be used."

Sherry Dunbar, Ph.D., MBA, Senior Director, Scientific Affairs Programs, Luminex

"Pre-analytical variation can substantially affect results. A good set of guidelines should control the entire process from collection through assay preparation. Variables such as how a subject was prepared (fasting, time of day, etc.) can be mitigated by standardizing subject status. It’s also important to use a validated specimen type and collection tube specified for the assays, as well as a consistent collection technique. Finally, guidelines for time and temperature of processing should be established and documented based on a validated assay. There should be a maximum time from collection to processing, as well as a recommended storage temperature."

Diana L. Bonilla Escobar, Director, Scientific Commercialization, Cytek Biosciences

"The detection of intracellular cytokines is highly sensitive to pre-analytical variables in sample collection and preparation. Since the goal is to evaluate cytokine production, samples should be processed as soon as possible after collection to maintain cell viability and functionality. If cryopreserved cells are used, all steps in the sample processing protocol (freezing, thawing, and cell resting) should be validated to preserve cellular function.

"Because different cytokines exhibit distinct kinetics or are induced by different stimuli, no single stimulation protocol is optimal for every application, requiring optimization based on the study. Other factors to consider include the choice of anticoagulant (sodium heparin better preserves cellular functionality)."

Danny Mamott, Research Scientist, Promega

"Always store your samples at -80°C and aliquot them to avoid freeze-thaws. When performing induction experiments, determine the ideal time to collect the sample and be consistent. If you are collecting samples daily, collect them at the same time every day. There are diurnal fluctuations of cytokine levels that can confound results."

What factors might affect reproducibility between runs, and what can researchers do to achieve consistent data?

Sherry Dunbar, Luminex

"Reproducibility between runs is influenced by both pre-analytical factors and analytical assay variation. The key is to identify which variables can introduce systematic bias versus random variability and then standardize or monitor them. Examples include:

• Reagent lot: Different lots, or repeated warming and improper handling, can shift assay performance. Qualify new lots with bridging studies before use.
• Calibrator/standard preparation: Errors in standard concentration or dilution can affect the entire run. Use standardized procedures and calibrated pipettes.
• Control materials: Control drift may indicate assay instability. Include low, medium, and high QC controls in every run.
• Instrument performance: Calibration and temperature changes can shift results. Perform scheduled maintenance and performance verification as recommended by the manufacturer.
• Operator technique: Pipetting, mixing, and timing differences introduce variability. Use detailed SOPs, training, and automated pipetting where appropriate.
• Timing and incubation: Variable incubation time, temperature, or intervals between steps can change assay kinetics. Standardize the workflow with timers or controlled equipment.
• Sample freeze–thaw history: Freeze–thaw cycles can alter analyte concentrations and compromise results. Standardize aliquoting and minimize freeze–thaw cycles.
• Sample matrix effects: Serum/plasma differences or interfering substances can affect signal. Use a consistent, validated specimen type.
• Plate position: Edge effects and spatial variation can alter results. Where appropriate, randomize samples and use a consistent plate layout.
• Data analysis: Different curve-fitting models or exclusion criteria can produce different results. Lock analysis algorithms and acceptance criteria in advance."

Vanitha Margan, Bio-Rad

"Sample handling and storage as well as freeze-thaw cycles have a definite impact on reproducibility between runs. Other factors at play can include reagent and lot variation, the assay condition, instrument performance, and data-analysis approaches. Researchers can improve consistency by standardizing collection and processing workflows, using the same validated assay conditions, including appropriate controls and replicates, and minimizing freeze–thaw events. Using a reliable multiplex platform consistently across studies can also help reduce variability while measuring multiple cytokines from limited sample volume."

Mike Blundell, Bio-Rad

"For ICS, reproducibility depends heavily on standardizing stimulation conditions and timing, transport inhibitor exposure, and the fixation/permeabilization protocol, since small variations can change how much cytokine accumulates and how well antibodies access intracellular targets. Consistent gating strategies, careful fluorophore choice, and fluorescence-minus-one controls across runs further help keep multicolor panels comparable."

Diana L. Bonilla Escobar, Cytek

"Reproducibility depends on careful optimization of the protocols for sample collection, preparation, stimulation, staining, acquisition, and analysis. For example, it is critical to maintain consistency between runs when adding protein transport inhibitors, using the same reagent concentration, time of addition, incubation time, and temperature each time the assay is performed.

"Instrument performance is another critical factor. Use a flow cytometer that has successfully passed daily quality control, maintain consistent detector settings, and use the same acquisition template and gating strategy across runs."

Danny Mamott, Promega

"Perform a standard curve of your analyte on each plate that you assay. This acts as a positive control and allows you to properly compare results to any other run. Because absorbance, fluorescence, and luminescence do not translate from plate to plate, the standard curve can help you determine if there is something wrong with one of your reagents.

"Be painfully consistent with how you prepare and set up your assays, even down to your pipetting technique. Whatever you do is fine, but do it for every sample every time.

"For long-term experiments, try to predict how much of the assay reagents you will need and secure enough of it with the same lot number. If this is not possible, then add a few samples you analyzed with a previous batch and compare how the new batch of assay reagent performs relative to the previous batch."

What are the most common mistakes researchers make when conducting cytokine profiling studies, and how can they avoid them?

Diana L. Bonilla Escobar, Cytek

"One of the most common challenges is defining the optimal stimulation time. Short stimulation times might fail to capture the increase in cytokine production, whereas longer stimulation times might miss the peak of expression or lead to activation-induced cell death and altered cell function.

"Since intracellular cytokine detection depends on antibodies accessing intracellular targets, another common issue is inadequate cell fixation or permeabilization. An inappropriate permeabilization reagent can reduce signal intensity, increase background staining, or even compromise epitope preservation. Fixatives and permeabilization buffers must also be compatible with the surface markers in the panel, since some protocols can diminish marker expression."

Danny Mamott, Promega

"One of the biggest mistakes is not performing an initial interference test with each assay and with each matrix that is being studied. Prepare a standard dilution of your analyte in the matrix you are studying and confirm you can recover the spiked-in concentration after subtracting background. If there is too much interference, you may have to dilute your matrix more.

"Another common mistake is reporting values for samples that are below the lower limit of quantitation (LLOQ). Extrapolating your standard curve doesn’t make your concentration real. If this occurs, report the percentage of samples that had a detectable concentration of the cytokine instead, as the reported 'value' would likely just be noise."

Vanitha Margan, Bio-Rad

"One of the most common mistakes is treating sample handling as an afterthought. Small differences in how samples are collected, processed, stored, or thawed can have a big impact on cytokine levels and make results harder to reproduce. Another common mistake is interpreting single cytokines in isolation instead of looking at broader patterns across related immune pathways."

Mike Blundell, Bio-Rad

"A common ICS mistake is omitting a viability dye before fixation, since dead cells bind antibodies nonspecifically and can be mistaken for genuine cytokine-positive events. Another is mistiming stimulation and transport inhibitor exposure relative to a given cytokine’s expression kinetics, which can under- or over-estimate the true positive population."

Sherry Dunbar, Luminex

"Many of the same issues that affect reproducibility also apply here. Changing kit vendor mid-study and underpowered study design are additional mistakes that can cause problems in cytokine profiling studies. Defining a clear strategy and key parameters, such as lower limit of quantification, will help to ensure more reliable results."

Do you have any suggestions for interpreting cytokine profiling results?

Danny Mamott, Promega

"Always remember, ELISAs and immunoassays report epitope availability, which should not be confused with bioactivity. Cytokines binding to receptors, protein complexes, or isoforms can still have epitopes while reducing bioactivity. For reporting the results, I find that it is more meaningful to show the fold change against a matched baseline rather than seeing absolute concentration."

Sherry Dunbar, Luminex

"Start with the study question: what is the biological process or hypothesis being tested? Does analysis happen before or after an intervention? What is the objective — diagnosis, prognosis, response, disease characterization, or mechanism? Also, look at magnitude of change, not just statistical significance. Finally, look at patterns of changes across multiple biomarkers for the most comprehensive view of cytokine biology."

Diana L. Bonilla Escobar, Cytek

"For accurate data interpretation, it is important to include appropriate biological and technical controls. Every experiment should include unstimulated cells to establish baseline cytokine expression, as well as a positive stimulation control to confirm that the cells are functionally competent and that the stimulation conditions are adequate. In addition, fluorescence-minus-one (FMO) controls should be included to accurately define gating boundaries and identify true cytokine-producing cells.

"One of the strengths of evaluating intracellular cytokine expression by flow cytometry is the ability to simultaneously assess multiple cytokines across different cell subsets. I recommend including sufficient lineage markers to distinguish the various subpopulations present in the sample, then layer on markers of differentiation, activation, or memory status to understand which cell types and functional states are producing which cytokines."

Vanitha Margan, Bio-Rad Laboratories

"Researchers should interpret cytokine results in the context of the biological question, sample type, and pathway being studied. Focus on multiple cytokines for a broader view rather than focusing on individual analytes in isolation. Consider the magnitude and direction of changes, looking for coordinated patterns that may indicate specific immune or inflammatory pathways. Always compare results with appropriate controls and baseline samples. Statistical significance is important, but biological relevance should also be evaluated carefully."

Mike Blundell, Bio-Rad

"With ICS, results are best interpreted as the percentage of cytokine-positive cells within a clearly defined lineage-gated subset."

What are the most common reasons for unexpected cytokine results, and how should researchers troubleshoot them?

Sherry Dunbar, Luminex

"Sample collection and processing techniques may be the wrong fit for the specific study or assay, or they may have been applied inconsistently. Excessive freeze–thaw cycles can cause problems. If matrix effects crop up, it’s important to evaluate dilutional linearity and parallelism. Signs of assay interference can point to heterophile antibodies, rheumatoid factor, etc. If values hover near the lower limit of quantification, it’s time to check the raw signals and verify the assay limits."

Diana L. Bonilla Escobar, Cytek

"Successful intracellular cytokine profiling requires optimization at every stage of the workflow. A negative result might indeed indicate that the cells do not express the cytokine of interest; however, it could also reflect a technical issue rather than a true lack of biological expression, caused by poor cell viability, failure to identify the optimal stimulation time or concentration, omission of protein transport inhibitors, degraded or expired antibody conjugates, an incorrect permeabilization buffer, or suboptimal acquisition settings."

Vanitha Margan, Bio-Rad

"Unexpected cytokine results relate to sample degradation, inconsistent handling, repeated freeze–thaw cycles, hemolysis, or values that fall outside the assay’s optimal range. Researchers should evaluate assay controls, standard curves, replicate agreement, and dilution linearity to determine whether the issue is biological or technical.

"It may be important to check for assay-specific matrix effects, out-of-range values, and potential interference or cross-reactivity between analytes. When necessary, test additional sample dilutions or perform spike-and-recovery experiments to help identify whether the issue is related to the sample matrix or the assay itself."

Mike Blundell, Bio-Rad

"In ICS, unexpected results often stem from insufficient permeabilization that prevents antibody access to intracellular cytokine, high background from non-specific antibody binding, or stimulation conditions that didn’t sufficiently activate the cells. Troubleshooting should include confirmed positive and negative stimulation controls and verifying permeabilization efficiency before questioning the biology of the result."

Danny Mamott, Promega

"The most common explanation is human error. This can include incorrectly recording the concentration of an analyte, adding inappropriate amounts of assay reagents due to mathematical errors, or even reading a plate in the wrong orientation."