How many freeze-thaw cycles will liraglutide at 10 mg/mL tolerate?

For reference: liraglutide · 10 mg/mL.

I have read the obvious sources and they disagree with each other, so I would rather ask people who have actually done this.

I have a working setup and a notebook, and I am prepared to be told that my setup is inadequate if that is the answer.

Which parts of this are load-bearing and which parts are habit?

Nobody has published a cycle count, and at 10 mg/mL the concentration tells you what a wrong guess costs: a 0.1 mL aliquot holds 1 mg and every microlitre is 10 µg. Damage from freezing is not gradual attrition — it is concentrated at the phase transitions, where ice excludes solute and the unfrozen fraction climbs well above 10 mg/mL for as long as the transition lasts. Two slow cycles can therefore do more than four fast ones, which is why a cycle count is the wrong unit in the first place. The way to make the number one is to make it one: split at reconstitution into single-draw aliquots, label each with 10 mg/mL and the date, and never thaw a container you will refreeze.

Answer first: each freeze-thaw cycle costs something through aggregation and pH shift, so the mitigation is aliquoting rather than choosing a better freezer.

During freezing, solutes are excluded from the ice lattice and concentrate into the residual liquid. Local concentrations can rise many-fold, which promotes aggregation independently of temperature.

Let a frozen vial reach room temperature before opening it. Opening a cold vial in humid air condenses water into the cake, which raises residual moisture and undoes what lyophilisation achieved.

The caveat is that aggregation is invisible in a clear solution below the threshold where it becomes visible.

Thaw slowly and never refreeze an aliquot.

edited 28 Jul 2026 by deamidation_watch — corrected a unit error in the worked example