Rethinking ATMP Assurance When Sterility Timelines, Reality Misalign

A conversation with Jose Caraballo-Oramas, biopharma executive

For autologous cell therapies with shelf lives measured in days, waiting the standard two weeks for compendial release sterility testing results is often not an option.

In this environment, organizations must make critical release decisions before conventional results are available. Teams are balancing the push for speed against the need for scientifically sound, operationally controlled testing

At the March 2026 ISPE Aseptic Manufacturing Conference, Jose Caraballo-Oramas, a biopharma quality executive with extensive experience in cell and gene therapy manufacturing, shared insights into how manufacturers are evolving contamination control strategies to meet these demands. The solution, he explains, lies in a more robust, integrated aseptic process design that meets the rapid turnaround times required for products with the shortest of shelf lives.

Looking back, Caraballo offered to discuss the technical hurdles of rapid microbial methods, the pitfalls of late-stage comparability studies, and the potential for a dedicated compendial chapter for short-life products. From the limitations of growth-based systems to the regulatory flexibility provided by 21 CFR 610.12(h)(2), Caraballo explores how the industry can achieve greater assurance of safety without sacrificing precious patient material.

Can we accept that releasing product at-risk while waiting for the 14-day USP result has become the de facto operational baseline? In your view, what is the most significant clinical or logistical hurdle companies face when a post-infusion result eventually comes back as a positive or even a false positive?

Caraballo: For many autologous advanced therapy medicinal products (ATMPs) with very short shelf lives, release before completion of the traditional 14-day sterility test has, in practical terms, become part of the operating reality. I would not describe it as an ideal baseline, but rather as a clinical necessity in cases where waiting is simply not compatible with patient need. In that setting, risk is inherent. The question is how well it is controlled through the full manufacturing and release strategy. That includes process design, contamination control, and the quality system’s ability to support decisions under pressure.

The hardest situation is not only a true positive after infusion, although that is obviously serious. In many cases, the broader challenge is what happens next when a late positive or suspect result appears. You are suddenly dealing with a medical assessment, a quality investigation, a lot disposition question, physician communication, pharmacovigilance implications, and often a difficult discussion with a treatment center that has already infused the product. If the signal later proves to be false positive, the organization still absorbs the operational disruption, the anxiety, and the loss of confidence that comes with it.

That is why the issue is bigger than test turnaround time. The burden falls on the entire system: aseptic process design, operator practice, environmental control, contamination prevention, sample handling, investigation discipline, and the ability to distinguish a meaningful contamination event from an artifact.

Beyond speed, how do rapid microbial methods improve the robustness of the result for complex cell matrices that might otherwise cause turbidity issues in traditional culture media?

Caraballo: Speed is the most visible advantage of rapid microbial methods, but it is not the only one. For advanced therapies, one of the real technical limitations of traditional compendial sterility testing is that the product itself can interfere with the readout. Cell-based matrices may be opaque, metabolically active, or otherwise difficult to interpret in a growth-based system that depends on visual or indirect signs of contamination.

Rapid methods can improve robustness because many are built on different detection principles. Instead of relying only on visible turbidity or long incubation, they may use metabolic markers, pressure changes, fluorescence, or nucleic acid-based detection. These approaches are generally less affected by matrix interference and can provide a clearer signal in complex samples.

They still need to be demonstrated as fit for purpose with the specific product and expected contaminants. But when properly developed, they improve interpretability and reduce ambiguity, which is critical when decisions must be made quickly and with limited material.

How can we implement rapid testing protocols that meet statistical sensitivity requirements without sacrificing product? Are you seeing a shift toward nondestructive testing?

Caraballo: For ATMPs, sampling is always a constraint. The product is often patient-specific, limited in volume, and high value. Taking large samples for testing is not always feasible, so the strategy has to be designed around those limitations.

In practice, this means defining a risk-based sampling approach tied to the process, the container closure system, and the points most susceptible to contamination. It also requires developing methods that can achieve meaningful sensitivity with minimal sample volumes. Rapid testing should be part of a broader contamination control strategy that includes closed processing where possible, strong environmental monitoring, and well-trained operators.

There is a clear movement toward less-destructive approaches. In some cases, that includes nondestructive or minimally invasive testing concepts. The driver is not convenience but necessity. That said, any approach still has to demonstrate equivalent or greater assurance of safety. The direction is toward extracting more reliable information from smaller samples while preserving product for the patient.

When moving a rapid sterility method from a Phase 1/2 clinical setting to a Phase 3/BLA, what is the most common pitfall you see in demonstrating comparability?

Caraballo: The most common pitfall is that the method works operationally in early phases, but the supporting data package is not built to withstand late-stage scrutiny. Early on, the focus is often on feasibility and turnaround time. By the time the program reaches Phase 3 or BLA, expectations around validation and comparability are much higher.

Comparability is not just showing that two methods can detect microorganisms. It requires demonstrating that the alternative method is suitable for the product matrix, covers relevant organisms, performs consistently, and remains controlled as the process evolves. If the product or process has changed over time, that can complicate the bridge unless it has been managed carefully.

Another issue is assuming early acceptance translates directly to licensure readiness. Late-stage review places greater emphasis on validation rigor, life cycle management, and data integrity. The strongest programs plan for this early and treat rapid methods as part of the development strategy, not a late justification.

The (h)(2) provision in 21 CFR 610.12 allows for alternative procedures if they provide equal or greater assurance of safety. Rather than a step-around, do you see this flexibility as a permanent home for ATMPs, or do we need a dedicated stand-alone compendial chapter specifically for ultra-short-life products?

Caraballo: The current flexibility is both necessary and appropriate for ATMPs. For products with very short shelf lives, it allows approaches that are aligned with clinical reality. In that sense, it is not a workaround but a recognition that traditional frameworks do not always fit these therapies.

Over time, however, the field would benefit from more explicit and harmonized guidance. A dedicated compendial chapter or equivalent standard for ultra-short-life products could help define expectations around validation, sampling strategies, and method performance. That would provide greater consistency across industry and regulators.

The objective should not be to lower standards but to ensure they are relevant to the product. Until more formal frameworks are established, the existing flexibility remains essential.

In January, FDA announced it would take a more flexible approach to CMC requirements for CGTs. Does that signal to you that regulators could be open to a parametric release model? Do you think relying on validated processes, especially for products with so much inherent variability, is a valid goal?

Caraballo: The increased flexibility reflects an understanding that traditional models do not always fit cell and gene therapies. It does not reduce expectations, but it does allow for approaches better aligned with the science and clinical context.

A direct application of classical parametric release is challenging given the variability inherent in these products. However, the broader direction is clear: greater reliance on process understanding, validated controls, and integrated evidence rather than a single end-point test.

For ATMPs, the release decision should be supported by a robust control strategy that includes process performance, in-process monitoring, and contamination control. That is a valid and necessary goal. The model will likely continue to evolve, but it will need to be supported by strong data and consistent performance to maintain confidence.

About The Expert:

Jose Caraballo-Oramas is a biopharma quality executive with more than 30 years of experience across biotechnology and advanced therapy medicinal products. He has held leadership roles across development, technology transfer, quality, engineering, and manufacturing, supporting global operations and regulatory compliance. He serves as Chair of ISPE’s Guidance Documents Committee, is a director on the ISPE International Board of Directors, and is a member of the PDA ATMP Advisory Board. He holds a Master of Engineering in chemical engineering from the University of Puerto Rico, Mayagüez.