What Medutest can tell you about a vial by weighing its water

The Journal has taken to reading cake appearance as a proxy for cycle quality, because it is the only evidence that arrives in the box. A well-made plug occupies close to the volume of the solution that was frozen, presents a matte and uniform surface, holds a clean edge at the glass, and dissolves in seconds. A shrunken, translucent or glassy plug indicates that the amorphous matrix softened and flowed while ice was still subliming, which is called collapse, and collapse is associated with higher residual moisture, slower reconstitution and a shorter useful life. It is a weak proxy. It is not no proxy.

A shelf life is a claim about a condition

Every stability figure is a conditional statement, and the condition is the part that gets dropped. A twenty-four-month shelf life means twenty-four months at a specified temperature, in a specified container closure system, with a specified formulation, assessed against a specified set of acceptance criteria by methods capable of detecting the changes that matter. Remove any one of those qualifiers and the number stops being checkable.

The trade routinely reports the number and none of the qualifiers. A certificate stating a two-year shelf life without a storage condition is asserting nothing in particular, and the same document frequently carries a storage instruction that has been copied from another product. The Journal’s habit is to treat an unqualified shelf life the same way we treat an unqualified purity figure: as a decoration until the procedure behind it is disclosed.

There is also a vocabulary problem worth clearing up. An expiry date states that material should not be used beyond it. A retest date states that material should be re-examined against specification before use beyond it, and is the appropriate concept for a stable chemical entity held in a controlled environment. Research suppliers overwhelmingly print the first word while meaning something closer to the second, and readers are entitled to know which is intended.¹

What the drying cycle actually does

A lyophilisation cycle has three stages and the differences between them explain most of what can go wrong. Freezing solidifies the solution, converting bulk water into ice crystals and concentrating everything else into an interstitial amorphous phase. Primary drying holds the product below the temperature at which that amorphous phase would soften, reduces the chamber pressure, and sublimes the ice directly to vapour. Secondary drying raises the shelf temperature to desorb water that remains bound to the solid matrix.

The rate-limiting stage is primary drying, and it is the stage under commercial pressure, because sublimation is slow and freeze-dryer time is expensive. Raising the shelf temperature accelerates it and risks carrying the product above its collapse temperature; shortening it leaves ice in the cake, which then melts during secondary drying and produces a partially collapsed plug with elevated moisture.

Secondary drying is the stage most often truncated, and truncation is invisible in the finished appearance. A cake can look entirely correct and carry three or four per cent residual water because the final desorption step was cut by six hours. The only way to detect it is to measure the water, which is why residual moisture is a release test in regulated manufacture and why its absence from a certificate is a substantive omission rather than a formatting one.²

Collapse temperature and the glass transition

An amorphous solid does not melt at a defined temperature; it softens over a range, and the midpoint of that range is the glass transition. For a frozen solution the relevant quantity is the glass transition of the maximally freeze-concentrated phase, and for the dried cake it is the glass transition of the residual solid. Both matter, at different stages, and both depend on composition and on water content.

During primary drying the product temperature must stay below the collapse temperature, which sits a little above the glass transition of the freeze-concentrated phase. Above it the amorphous matrix has enough mobility to flow, the pore structure that permits vapour escape closes, and the cake collapses. Sucrose-containing formulations have a glass transition of the freeze-concentrated phase in the region of minus thirty-two degrees, which imposes a genuinely cold and therefore slow primary drying stage. Mannitol behaves differently because it crystallises, giving a mechanically robust cake at the cost of losing the protective amorphous phase.

Water is a plasticiser: adding it lowers the glass transition of the dried solid substantially. This is the mechanism connecting residual moisture to storage stability. A cake with high residual water has a lower glass transition, and if storage temperature approaches it the matrix acquires mobility and every degradation pathway speeds up. A vial stored above its own glass transition is, chemically, a slow solution.

Residual moisture, and how it is measured

Two methods dominate. Karl Fischer titration determines water specifically, by a stoichiometric reaction with iodine, and is the reference method; the coulometric variant works on the small sample masses a single vial provides. Loss on drying is simpler and less specific, measuring total volatile mass lost under defined heating, which for a formulation containing residual organic solvent overstates the water.

Typical release specifications for lyophilised peptides sit in the range of one to three per cent water by mass, with tighter limits where the molecule is particularly moisture-sensitive. The relationship to stability is not linear. Below roughly one per cent, further drying sometimes destabilises rather than helps, because a monolayer of water contributes to conformational stability in some solid-state systems. Above three per cent, deamidation and hydrolysis rates rise steeply and the glass transition falls towards ambient.

None of the twenty companies the Journal tracks reports residual moisture as a standard release test. Two will provide a figure on request. This is the omission we would most like to see closed, ahead of endotoxin and well ahead of anything else, for a straightforwardly practical reason: it is a cheap determination on a small sample, it is performed in any pharmaceutical analytical laboratory, and it predicts what the vial will be like in eighteen months better than the purity figure that is printed instead.

Aggregation, and the method that dissolves the evidence

Aggregation covers a range of species from soluble dimers to visible particles, formed by covalent routes such as disulphide scrambling or by non-covalent association of partially unfolded monomers. For peptides the process is often nucleated at an interface — the air-water interface of a shaken vial, the silicone oil layer on a siliconised stopper, the ice-water interface formed during freezing — which is why mechanical handling and freeze-thaw cycling matter as much as temperature.

The analytical difficulty is severe and specific to the trade’s chosen method. Reversed-phase chromatography runs in an acidic, partly organic mobile phase which dissociates most non-covalent aggregates before or during separation. The aggregate is loaded and the monomer is detected. Size-exclusion chromatography under non-denaturing conditions separates by hydrodynamic volume and reports high molecular weight species directly; analytical ultracentrifugation and light scattering methods characterise them further. None of these is offered as a routine service to this market.

The consequence for a reader is that the aggregate content of a research vial is, at present, an unmeasured quantity. It is not necessarily a large one — well-made lyophilised peptides are frequently very low in aggregate — but no certificate in circulation addresses it, and the purity figure that is printed instead is generated by the one method guaranteed not to see it.³

What a logger records, and what it cannot

A single-use electronic logger is a thermistor, a clock, a small memory and a battery, and its output is a series of readings of the air adjacent to its housing at a configured interval. Four configuration choices determine what it can detect. The sampling interval sets the shortest excursion it can see: a device sampling every fifteen minutes will record a ten-minute spike only by luck. The memory depth sets the maximum journey length before it stops or overwrites. The start delay determines whether the packing operation appears in the trace. And the calibration interval determines whether the numbers mean anything at all.

Placement matters more than any of them. A logger in the outer carton records the transport environment; a logger inside the insulated payload against the vials records something much closer to what the product experienced; and the two can differ by twenty degrees during the same afternoon. A trace without a stated placement is close to uninterpretable, and placement is almost never stated.

Our own nine shipments used devices calibrated within the preceding twelve months, sampling at five-minute intervals, placed inside the insulated payload adjacent to the vials, started at the point of packing where we controlled it and at the point of receipt of the outbound label where we did not. We state all of that because a trace is a measurement and a measurement without its method is decoration.⁴

Nine parcels, instrumented

We ordered nine parcels of lyophilised material at catalogue prices, as ordinary customers, from suppliers we did not identify ourselves to, and arranged for a calibrated logger to travel inside the insulated payload of each. Eight arrived and produced complete traces. The ninth was held at a border for eleven days, exhausted its memory on day five, and arrived with a partial record and a cake that had visibly shrunk.

The headline results are in the table below. Transit times ran from two to nine days against quoted estimates of three to five. Five of the eight complete traces exceeded twenty-five degrees at some point; three exceeded thirty; the maximum recorded was thirty-eight degrees for three hours and forty minutes on a road leg. Mean kinetic temperature exceeded the arithmetic mean in every case, by between one and six degrees, with the largest gap on the trace with the shortest and hottest excursion — which is precisely the pattern the calculation is designed to expose.

Two of the eight arrived with their coolant packs still partly frozen, both on two-day domestic journeys. On the six longer journeys the coolant had fully melted and equilibrated, in most cases before the halfway point of the trip. We regard that as the central finding: for anything beyond about forty-eight hours, the gel pack is a delay rather than a control, and the shipment is relying on the material tolerating ambient temperature.

The unfashionable finding: dried peptide is fairly tough

The alarmed version of this story would end with the excursions and leave the reader frightened. The evidence does not support that ending, and the Journal would rather publish the awkward finding than the satisfying one. A lyophilised peptide at low residual moisture, stored below its glass transition, has very little molecular mobility available for degradation. Short warm excursions in that state cost comparatively little, and the published solid-state stability literature is consistent on the point: dried peptides tolerate transient thermal insult far better than solutions do.

Two caveats keep this from being a licence. First, the protection depends on the cake being genuinely dry, which is the unmeasured variable this article keeps returning to. A cake at four per cent moisture has a much lower glass transition and much less margin. Second, repeated cycling is worse than a single excursion, particularly where a warm interval permits moisture redistribution within the cake or condensation inside the container on cooling.

The reordered risk list, on our reading, puts the reconstituted vial first, the cake with unknown residual moisture second, the multi-week domestic storage of an opened vial third, and the four hours at thirty-eight degrees in a courier van somewhere well below all of them. That ordering is not what the anxiety in this market reflects, and we think it is the more defensible one.⁵

A note on method and sourcing

The regulatory framework in this article is taken from the harmonised guidelines on stability testing and on biotechnological products, read in the original, and from the current compendial chapters on storage definitions, distribution of temperature-sensitive products and stability in dispensing practice. The degradation chemistry is drawn from the peptide and pharmaceutical sciences literature, and where a claim is a generalisation across sequences this piece says so, because sequence dependence is the rule rather than the exception.

The shipment data is ours. Nine parcels, ordered at catalogue prices as ordinary customers, with calibrated loggers placed inside the insulated payload and sampling at five-minute intervals. Eight complete traces and one truncated by a customs hold. We disclose that nine parcels is not a survey, that we did not control the packing operation, and that a single logger cannot characterise a payload with a thermal gradient across it.

Nothing in this department is a recommendation about storing, reconstituting or administering anything. The compounds discussed are sold for research use only and are not approved for human use in any jurisdiction. Corrections and disputes go to standards@compoundjournal.com; documents, traces and certificates readers would like examined go to letters@compoundjournal.com, and we do not identify the source of anything sent to us.

Residual moisture is the omission we would close first. It is cheap, it is fast, it is performed in any pharmaceutical analytical laboratory, and it predicts what a vial will be like in eighteen months better than the purity figure that appears in its place. That one line would tell a reader more about the future of a cake than everything currently printed on the certificate combined.