Raising the bar for reproducibility in stem-cell based research: Standards, best practices, and programmed cells as a potential solution

Reproducibility has become a pressing concern across the life sciences, including human stem-cell based research. Despite the tremendous promise of human induced pluripotent stem cell (hiPSC)-based disease models, scientists continue to encounter irreproducible results and variable data.

A recent perspective by Lucia Selfa Aspiroz et al.¹ highlights the scope of this reproducibility crisis. They discuss how, in many cases, studies cannot be replicated due to issues like misidentified hiPS cell lines, inaccurate protocols, hiPS cell line variability, and laboratory-specific quirks. Multi-site analyses, cited in the perspective, highlight that even when labs use the same hiPSC differentiation protocol and parental hiPS cell line, results can diverge significantly because of the way different labs and users interpret a protocol. The consequences are costly, irreproducible preclinical research is estimated to waste tens of billions of dollars annually and floods the literature with misleading data. These problems erode trust and slow down the translation of findings from human iPSC-derived cells. 

In response, the stem cell community is rallying around building a framework for standards and best practices as a solution. The 2025 perspective¹ (co-authored by experts from academia, biobanks, and industry, including bit.bio’s Founder Dr Mark Kotter) argues that embracing common standards is essential for ensuring reliable, high-quality stem cell-based research. For transparency, Dr Kotter’s involvement was in his academic capacity and does not represent bit.bio’s view.

Here, we discuss bit.bio’s take on the topic, examine the paper’s recommendations and consider how programmed human cells, specifically bit.bio’s human iPSC-derived ioCells, align with these efforts to address the challenge of reproducibility. Our goal is to spark a conversation on how academia and industry together can accelerate the adoption of reproducible human cells for research and drug discovery. 

The reproducibility crisis in hiPSC-derived models

The inability to reproduce results with hiPSC-derived cells originates from several well-documented issues. 

1. Cell line variability: hiPS cell lines from different donors (or even different clones from the same donor) can respond differently due to genetic background or epigenetic idiosyncrasies. 
2. Cell authentication and culture contamination issues: Misidentification of hiPS cell lines or undetected contamination with other cells or microbes remains surprisingly common. 
3. Cell handling and protocol complexities: Handling across labs and the complexity of some of the protocols introduce variability. Even when users ostensibly follow the same published differentiation protocol, subtle differences in reagents, operator technique, or cell passaging schedule can yield different outcomes. 
4. Protocol drift and evolving materials: The drift over time of methods and materials exacerbates these problems and threatens consistency. Standard operating procedures (SOPs) that are not rigorously maintained tend to evolve (“drift”) as they are handed off between staff or scaled up, causing results from earlier and later experiments to differ. 

The Selfa Aspiroz et al. (2025) perspective frames these issues as a call to action. Improving reproducibility is paramount if hiPSC technology is to reach its potential in redefining drug discovery and regenerative medicine.

Best practices and standards: a path forward

What concrete solutions are being proposed? Encouragingly, there is a growing consensus around establishing best practices and formal standards for stem-cell based models.

The  International Organization for Standardization (ISO) has begun publishing standardised protocols relevant to cell culture and specifically to pluripotent stem cells². The International Society for Stem Cell Research (ISSCR) recently released its Standards for Human Stem Cell Use in Research³. Other frameworks address the process of cell culture more generally. The concept of Good Cell (and Tissue) Culture Practice (GCCP) with the aim of instilling quality principles in day-to-day cell handling⁴. Similarly, the Organisation for Economic Co-operation and Development’s (OECD) Good In Vitro Method Practices (GIVIMP) guidance focuses on in vitro assays intended for regulatory use⁵.

Beyond lab practices, the perspective also highlights systemic measures to tackle reproducibility. For instance, it advocates establishing reference hiPS cell lines and public cell registries. The authors call for community consensus on core characteristics for each cell type and data exchange through databases to facilitate comparisons across studies.

Finally, the perspective outlines a multi-stakeholder strategy to drive these changes. They recommend that journals, funders, and institutions require or incentivise proper iPS cell line authentication, reporting of key metadata, and use of high-quality starting materials. If successful, the payoff would be huge: more reliable science, less waste, and faster progress translating lab discoveries into industrial and clinical applications.

How a portfolio of consistent, defined human cells addresses unmet needs

bit.bio’s opti-ox powered ioCells deliver human cells that meet unmet needs and align with emerging standards.

“We built our manufacturing platform around the principle that every vial of cells should perform the same, regardless of who uses it or where it is used. That means designing processes with reproducibility in mind from the outset, not as a QC step at the end. Standards are important, but without industrialised production systems behind them, reproducibility remains aspirational.” 

Karl FirthSenior Director, Technical Operations, bit.bio

Product consistency and characterisation: Each ioCells product undergoes rigorous quality control to ensure consistent identity and function. Characterisation methods include immunocytochemistry, qPCR, and RNA sequencing, verifying marker expression and reproducibility across batches. This level of QC supports reliable, repeatable results, reducing variability and giving researchers confidence that the cells they use meet defined performance and identity criteria every time.

Scalability and lot uniformity: Traditional differentiation protocols often produce variable results between batches, even when following the same protocol. bit.bio’s ioCells were built to eliminate this problem by moving to a manufacturing paradigm. Through our opti-ox technology, we can convert hiPSCs into the desired cell type with extremely high efficiency and consistency in a single manufacturing step. The process has been optimised to be deterministic, meaning every starting pluripotent cell is driven to the target fate. Crucially, the process is highly scalable; we routinely produce billions of cells in a single manufacturing run without losing consistency, and the identity is programmed into every cell. From an end-user perspective, this uniformity means experiments become more reproducible and addresses the differences in cell handling and protocol complexities. 

See Dr Ania Wilczynska, bit.bio's Director of Bioinformatics and AI, presenting data on the unparalleled lot-to-lot consistency of opti-ox technology. 

Integrated quality controls: bit.bio’s production process incorporates quality control at multiple steps, aligning with principles from GCCP/GIVIMP about monitoring and documentation. For each ioCells type, we establish quality acceptance criteria (echoing the ISO standard’s emphasis on QC criteria). Every batch is tested to ensure it meets predefined benchmarks – for example, correct marker expression, viability, purity, and functional performance (where applicable). While these internal QC data may not all be visible to customers, the outcome is that each vial of ioCells contains a highly characterised, defined and functional population of human cells that are ready for experiments in a matter of days. It is also worth noting that each ioCells vial comes with documentation and recommended protocols, which help ensure users handle the cells in a reproducible way (reducing protocol drift). 

“In manufacturing, reproducibility is non-negotiable. We treat cell production like any other regulated bioprocess: controlled inputs, defined outputs, and rigorous testing at every stage. That mindset is what allows scientists to trust the cells they are working with.” 

Ben NewmanAssociate Director Manufacturing, bit.bio

Translational relevance: Perhaps the most important payoff of all these features is the potential for increased translational relevance of the data generated. A human cell model that is consistent and well-characterised is far more likely to yield predictive, reliable insights for drug discovery than an inconsistent, poorly defined model. For example, researchers at Charles River Laboratories noted that using more physiologically relevant, standardised human cells (like ioCells) in safety and toxicity testing allows them to identify unsafe drug candidates much earlier in the development process⁶. The combination of human origin and consistent performance makes ioCells powerful for bridging the gap between in vitro tests and clinical outcomes. Moreover, when every experiment is done on a defined cell population, it becomes easier to aggregate data and compare results across studies or sites, which is crucial for regulatory acceptance. The perspective paper itself argues that improving rigour and quality in stem cell research will increase the uptake of stem cell-based models in industrial and regulatory contexts¹. As a result, pharma and biotech companies are already more willing to adopt hiPSC-derived models if they come with assurance of consistency and quality. By tackling some of the root causes of variability, bit.bio’s ioCells aim to deliver that assurance. 

A shifting landscape: regulation and collaboration

It is worth noting that all these efforts are converging at an opportune time. Regulators and policy-makers are increasingly supportive of human-based, reproducible models. In the United States, the FDA Modernization Act 2.0 explicitly opened the door for drug developers to use non-animal methods – including human cell models – to satisfy preclinical testing requirements⁷. In April 2025, the FDA announced plans to phase out mandatory animal testing for monoclonal antibodies and other new drugs⁸. Most recently, in July 2025, the NIH confirmed it will no longer issue funding opportunities that rely solely on animal experiments, instead requiring the inclusion of non-animal methods⁹. This change reflects a recognition that human-relevant in vitro cell models can potentially be more predictive of human outcomes than traditional animal studies.

Internationally, similar trends are visible. The OECD has been coordinating an in vitro Developmental Neurotoxicity (DNT) testing battery, which comprises a suite of stem cell-derived neural assays to predict neurotoxic effects without animal tests. 

On the industry front, we see growing uptake of hiPSC-derived models in pharma and biotech, especially as companies realise the cost of late-stage drug failures due to human-irrelevant models. Partnerships between cell providers and pharma (including some involving bit.bio) indicate a desire to bring standardised human cells into mainstream workflows. For example, contract research organisations (CROs) are now offering screening services using hiPSC-derived cells, banking on their improved predictive power. The earlier quote from Charles River scientists is telling; standardised human cells are viewed as a way to de-risk drug discovery. None of this is to say animal models will disappear overnight, but the momentum is clearly toward integrating more human-based testing earlier in R&D. Reproducibility is a key selling point in that integration. A drug company will understandably be wary of a new human cell assay if it seems finicky or lab-specific. But if the assay comes with a robust protocol and a consistent cell supply, those concerns diminish.

Conclusion: toward a collaborative future

In closing, the drive to improve reproducibility in stem cell-based research is both a scientific imperative and a gateway to broader adoption of human models. The perspective by Selfa Aspiroz et al. has crystallised a set of actionable recommendations – from embracing ISO/ISSCR standards and better cell authentication, to creating reference lines and incentivising best practices¹. At bit.bio, we find these guidelines deeply resonate with our mission to democratise access to consistent, defined human cells. The ioCells portfolio was built on the premise of consistency, a defined cell identity backed up by extensive characterisation data, and human relevance, which directly addresses many of the field’s longstanding pain points. By providing ready-to-use human cells with lot-to-lot consistency, we hope to exemplify how implementing rigorous standards can yield tangible improvements in research outcomes.

No single organisation can fix the reproducibility problem. It requires coordinated action across academia, industry, journals, and regulators. Researchers need to keep developing and sharing best practices. Companies must provide reliable tools. Journals should enforce reporting standards. Regulators need to support and recognise new approaches. Collaboration between academia and industry is essential. The tools exist. Now it is about adoption, alignment, and follow-through.

References

1. Selfa Aspiroz, L. et al. Promoting the adoption of best practices and standards to enhance quality and reproducibility of stem cell research. Stem Cell Reports, epub ahead of print. https://doi.org/10.1016/j.stemcr.2025.102531
2. ISO 24603:2022, Biotechnology, Biobanking; Requirements for human and mouse pluripotent stem cells; https://www.iso.org/standard/79046.html
3. International Society for Stem Cell Research, Standards for Human Stem Cell Use in Research; https://www.isscr.org/basic-research-standards
4. Bal-Price, A. and Coecke, S. Guidance on Good Cell Culture Practice (GCCP). Cell Culture Techniques. Neuromethods 2011; 56. https://doi.org/10.1007/978-1-61779-077-5_1
5. Organisation for Economic Co-operation and Development (OECD), Guidance Document on Good In Vitro Method Practices (GIVIMP) 2018. https://www.oecd.org/en/publications/guidance-document-on-good-in-vitro-method-practices-givimp_9789264304796-en.html
6. Dr Marijn Vlaming, Head of Biology, Beerse & Leiden, Charles River Laboratories. https://www.bit.bio/discover-iocells#:~:text=Dr%20Marijn%20Vlaming 
7. Food and Drug Administration - FDA Modernization Act 2.0. https://www.congress.gov/bill/117th-congress/senate-bill/5002
8. Food and Drug Administration - FDA Announces Plan to Phase Out Animal Testing Requirement for Monoclonal Antibodies and Other Drugs. https://www.fda.gov/news-events/press-announcements/fda-announces-plan-phase-out-animal-testing-requirement-monoclonal-antibodies-and-other-drugs
9. NIH Funding Announcements to Align with NIH Initiative to Prioritize Human-based Research. https://grants.nih.gov/news-events/nih-extramural-nexus-news/2025/07/nih-funding-announcements-to-align-with-nih-initiative-to-prioritize-human-based-research