The date stamped on a compounded vial and the date stamped on a pharmacy box look like the same object and are not. One is an expiration date established by testing that specific formulation in that specific container through a registered stability program. The other is a beyond-use date, and in the federal guidance that describes it the definition is blunt: the BUD is the date or time after which the product is to be discarded.[1] Not the date it stops working. The date it goes in the sharps bin. If you are being shipped three months of vials at once, that distinction is the whole story, and so is the question of what a vial is versus a pen in the first place.
A BUD is derived from the process, not from the molecule
Here is the part that surprises people. The number on the label has almost nothing to do with semaglutide or tirzepatide. It is set by how the preparation was made and what happened to it before release.
FDA’s current good manufacturing practice guidance for outsourcing facilities publishes a table of default beyond-use dates — the dates the agency says it generally does not intend to take regulatory action over, in place of a full stability program. The table is keyed to exactly three facts. Was the finished product aseptically processed, or terminally sterilized through a validated cycle? Does it contain a preservative? And was a passing sterility test completed before the batch was released?[1] Nothing about the active ingredient enters the calculation at all.
Refrigerated, the answers produce these defaults. Aseptically processed, unpreserved, no completed sterility test: 9 days. Terminally sterilized, unpreserved, no completed sterility test: 28 days. A completed passing sterility test before release, unpreserved: 42 days. Preserved, with that test: 42 days.[1] Room-temperature figures are shorter across the board, and the frozen figure is 45 days regardless.
Forty-two days is the ceiling of that table. Six weeks. It is worth sitting with that number before reading a shipping page.
The defaults expire above a thousand units
There is a second condition that gets almost no attention and changes who the table even describes. The default BUDs apply to an aggregate batch of 1,000 sterile units or fewer over the six-month period covered by a facility’s drug product report. Above that, the guidance marks the default column not applicable and says a data-driven stability program is required instead.[1] If at any point in the reporting period a facility crosses the 1,000-unit line, the small-batch conditions stop applying.
A thousand vials in six months is a compounding pharmacy filling prescriptions. It is not a facility supplying a national telehealth roster. So for the suppliers most compounded GLP-1 actually comes from, the published default table is not the applicable rule — the applicable rule is a stability study whose results nobody outside the facility has seen. Which category of facility is which is set out in the comparison of the two compounding sections.
And a caution that has to travel with every figure above: this guidance is a draft, marked nonbinding on every page, and it binds only outsourcing facilities registered under section 503B. A state-licensed 503A pharmacy is exempt from CGMP entirely, so none of Table D reaches it. Its dating comes from the USP compounding chapters as adopted by its state board — a chapter this article could not retrieve, because USP’s compendium is subscription-only and its site refused every request made while researching this page. That gap is stated rather than papered over.
Two clocks, and the second one does not add to the first
A multiple-dose vial runs two timers. The BUD runs from compounding. The in-use period runs from the first puncture. Most people assume the second resets the first, and it does the opposite.
FDA’s guidance allows labeling up to a 28-day in-use period for multiple-dose products, and conditions it: the acceptance criteria of USP General Chapter <51> on antimicrobial effectiveness have to be met, whether the product contains a preservative or the formulation is inherently inhospitable.[1] Then comes the sentence that reverses the common reading. The labeled directions should tell the provider or the patient that the time in storage plus the administration phase must not exceed the BUD.[1]
In practice: a vial with ten days of beyond-use date remaining gives you ten days after you puncture it, not twenty-eight. The in-use allowance is a ceiling inside the BUD, never an extension of it. FDA separately recommends discarding a multiple-dose vial within 28 days of first use even when a compounder’s own instructions say longer[6] — a point the switching article takes up in detail — and the two rules stack rather than compete. Whichever is shorter is the one that applies.
There is a third clock for any preparation that has to be manipulated before it is injected. Where a compounded sterile product requires reconstitution or dilution, FDA reads the statutory directions-for-use requirement as including an in-use time, because whoever performs the manipulation needs to know how long the result holds — and stability data is supposed to support whatever that time says. Absent such data, the guidance says the agency generally does not intend to act where the directions state an in-use time of less than 4 hours at room temperature or less than 24 hours refrigerated.[1] Hours, not weeks. That is a very different window from the one a patient told to mix a vial themselves usually has in mind, and it is the subject of the reconstitution article.
Microbial dating and chemical dating are separate questions
Everything above is about contamination. It is not about whether the peptide is still the peptide, and a compounder is supposed to answer both and then take the shorter answer. A practitioner review of the criteria puts it directly: the risk-level beyond-use date and the chemical stability of the formulation are evaluated separately, the shorter governs, and a compounder wanting to exceed the risk-level date has to sterility test under USP General Chapter <71>, including method suitability for the strength and batch size actually being prepared.[2]
For these molecules the chemical half of that is not a formality. Semaglutide in solution degrades in a way that depends heavily on pH. A stress study using reverse-phase UPLC with high-resolution mass spectrometry ran samples at 25 °C, 40 °C and 60 °C for 28 days and at 80 °C for 7 days, identified thirteen known impurities, and found six of them forming across every condition tested; pH was the key factor in thermal degradation.[3] A companion solid-state study found the alpha-helical content of semaglutide falling from 49.07% to 43.75% at 60 °C and to 0.2% at 80 °C — a conformational collapse rather than a slow loss of potency.[4]
Those are stress conditions, well above anything a refrigerated vial should meet, and they are not a prediction about your product. What they establish is that the buffer, the pH and the temperature history of a given preparation are inputs to its real dating, which means two vials at the same concentration from two pharmacies can honestly carry different dates. Where that history goes wrong in transit is the cold-chain article’s subject.
What a three-month shipment is actually asking of the date
Multi-month fills are ordinary in this market because they lower the headline monthly price. The dating consequence is rarely stated anywhere on the checkout page.
Twelve weeks is 84 days. That is double the longest refrigerated default in FDA’s table and a fortnight longer than the 45-day frozen figure. More to the point, the clock does not start when the box arrives. A BUD is assigned at compounding, and between then and your first injection sit batch release, any sterility testing, order processing, shipping and however long the product waited in inventory. The vial you open in week eleven has been dated since before you ordered.
That reframes what a multi-month discount is buying. The saving is real and it is calculated per month, but the risk it shifts onto the buyer is inventory risk: if a dose is skipped, a titration stalls, a course pauses for an illness, or the prescription changes, the vials keep aging on the original schedule. Nothing about a shorter dated product makes it worse medicine, either — a 9-day beyond-use date can mean a pharmacy is being candid about aseptic processing without a completed sterility test, which is more informative than a long date with no stated basis. The date is a claim about a process, and a short claim that is supported tells you more than a long one that is not.
There are legitimate ways for a supplier to support a longer date: a real stability program, a preserved formulation, terminal sterilization, a passing sterility test before release. Every one of those is a fact about the pharmacy, and none of them is visible from a price. The questions worth putting in writing before a multi-month order are narrow and answerable — what BUD is on the vials, what date it runs from, was the product sterility tested before release, and is it preserved. A supplier that answers all four is describing its process. A supplier that answers with a shipping policy has not answered.
One vial-level detail closes the loop. Every puncture is an event, and a review of studies measuring microbial contamination of punctured vials found rates ranging from none to 27% across studies with widely differing methods, with low contamination frequency where puncture happened under ISO 5 conditions.[5] A kitchen counter is not ISO 5. That is not an argument against self-injection, which is what the whole category is built on; it is the reason the in-use ceiling exists at all, and the reason technique matters as much as the date does — which is where the technique article picks up.