ICH is expected to adopt the revised Q1 stability guideline this month. For the first time, small molecules, biologics and advanced therapies sit in one document, and Q5C — the short guideline most of us have used for biologics stability for almost thirty years, is folded into it.
At first glance, it reads as a document-management exercise: combining six guidelines into one. But for biologics, the new Q1 changes what a defensible stability strategy looks like.
The draft went out for public consultation in summer 2025. The ICH work plan targets adoption in November 2026. FDA and EMA will implement on their own timelines, so nobody is going to reject your IND next quarter for following Q5C. However, regulators read new submissions through the lens of the current thinking well before the official implementation date, and the draft guideline has been public for 18 months.
What is Changing for Biologics Stability
Stability is treated as a development activity, not a registration exercise. Q5C was written mostly about the formal studies that support a BLA. The new Q1 puts much more weight on what you learn earlier: stress and forced degradation, understanding degradation pathways, and showing that your methods are stability-indicating. What used to be good practice is is about to become an expectation you can be asked to demonstrate.
Modeling is legitimate for biologics. Q1E extrapolation was always awkward for proteins. The modeling annex opens the door to using kinetic and statistical models to support shelf life, with the knowledge and data to justify the model. Most protein degradation does not follow neatly Arrhenius' equation. Aggregation mechanism can change between 5 °C and 40 °C. The annex does not give you a free shortcut, but it rewards programs that understand their degradation mechanisms.
Reduced designs need a justification. Bracketing and matrixing have been used sparingly for biologics. The new guidance makes it explicit that it requires scientific justification.
In-use, short-term storage and excursions sit in the same framework. Bringing them into Q1 means the in-use study for your reconstituted lyophilized ADC and the excursion data supporting your clinical shipping lanes will be read as part of the same stability package.
Cell and gene therapies get their own annex. For products with a shelf life measured in hours or days, where potency is the attribute that matters most, a separate annex is welcome. If you work in CGT, read this annex first.
Where I Anticipate Programs Falling Short
The forced degradation study is a checkbox. It was done once, by a CRO, on early material, using the release panel. Nobody connected the degradants to CQAs, and nobody checked whether the release methods actually resolve them. Under the new Q1, that study is the foundation of the whole argument.
The stability-indicating claim rests on a single method. For a bispecific, a size-exclusion method that shows the monomer is stable can hide low-level clipping in the hinge or linker that shows up only by CE-SDS or intact mass. If your claim that "the product is stable" relies on SEC and a binding ELISA, you have not shown that the methods can detect the degradation that actually occurs.
There is no plan to use the data you already generate. Many programs collect accelerated and stressed data at every timepoint and never model it. If you have three years of 25 °C data on a platform mAb and never asked whether it predicts 5 °C behavior, you are leaving shelf life, and possibly a faster extension during the clinical program, on the table.
Clinical shelf-life extensions are ad hoc. Every extension is a new negotiation because nobody wrote down in the IND how shelf life would be extended as real-time data came in. A simple, pre-stated extension plan tied to real-time and supportive data saves time at every amendment.
Practical Takeaways
Rebuild forced degradation around CQAs. Map each stress condition to the degradation pathway it probes, the CQA it affects, and the method that detects it. If a CQA has no stability-indicating method, this is your first gap.
Start modelling the data you already have. Even if you never use a model to set shelf life, fitting your accelerated and long-term data tells you whether your degradation is predictable. Regulators will increasingly expect you to know.
For ADCs, bispecifics and cell and gene therapies, write the modality-specific stability rationale explicitly. Free payload and DAR drift, fragment and mispairing species, viability and potency loss after thaw. The generic mAb template does not cover these, and the new guideline gives you a framework to say so.
If you want a second opinion on whether your stability strategy will hold up under the new Q1, I am happy to take a look.