Details up front: a sterility test · retatrutide.
This is a planning question. I know what my options are; I do not know how to weigh them.
What I want is the minimum viable version, which I suspect is smaller than what I would design.
What is the minimum version of this that is still defensible?
Twelve months is 52 weeks, so the budget is set by lot turnover, not by the price of a sterility test. Take one lot a quarter as the low case: 4 lots a year, so a test-every-lot policy is 4 assays and a test-every-third-lot policy is 2 once you round up. Take one lot a month as the high case: 12 lots, and the same two policies are 12 assays and 4. The spread between the cheapest and the dearest defensible policy is therefore about a factor of six across the same 52 weeks. Choose the policy before the first result. One chosen after a disappointing figure is a reaction to that figure, and it will not survive the second one. Then spend it where it changes a decision: over a year, one content assay on each new lot tells you more than four purity figures on the same lot, because purity and content are independent and only one of them changes your arithmetic.
The practical consequence is that spot-testing one vial from a new supplier is better than assuming they are all the same.
Stratified sampling — testing one vial from the top, one from the middle, and one from the bottom of a shipment — is cheap insurance against segregation.
On the detail: the sample size determination requires choosing a confidence level and an acceptable error rate, and the smaller the error rate you want, the larger your sample must be.
Sampling plans for pharmaceutical manufacturing are defined in ISO 2859 and ANSI Z1.4, and they are based on statistical sampling theory.
I would treat a "complies with" statement without sampling details as a claim rather than as evidence.
The practical summary: a lot number without a sampling statement is a lot number without meaning.
Analytical standards and reagents with traceable certificates. Every quantitative result you read inherits the accuracy of the standard behind it.
Shop standardsOn the detail: most suppliers test one vial per lot and report the result as lot homogeneity, which is sampling one item from one lot and extrapolating wildly.
Acceptance Sampling by Attributes defines the number of samples you need from a lot to claim a specified quality level at a specified risk — it is in ANSI standard Z1.4.
Published segregation failures show that even modern automated processes sometimes produce lots with measurable vial-to-vial variation.
Lyophilised peptide homogeneity studies show that vial-to-vial variation is usually small but occasionally large, depending on the distribution in the freeze-dryer.
The limitation is that you cannot know for certain without testing every vial, and you almost never can afford to do that.
Assume segregation is possible, and design your sampling to catch it if it exists.
Batch testing establishes what can be claimed about the lot as a whole, and the sample size determines how much you can actually claim.
Under AQL sampling plans, testing two vials from a fifty-vial lot gives you an operating characteristic curve that tells you what risks you are accepting.
The part that matters: if the lot was manufactured in multiple batches, testing vials from each batch separately establishes whether batch-to-batch variation is acceptable.
Published data on lot homogeneity from manufacturers who sample multiple vials consistently find variation below the published specifications, suggesting the sampling plans work.
Worth noting that thermal excursions during shipping affect different vials differently, so the lot may not be homogeneous even if it left the factory that way.
If testing multiple vials, state how many you tested and why you chose those vials.
edited 19 Feb 2026 by Dr_Colm_Fitzhenry — corrected a unit error in the worked example
Ask PeptideStack is a static archive. Posting is closed, but the norms are worth stating: answer the question that was asked, show your working, cite the trial or the certificate, and say plainly where the evidence runs out.