Corrections

Every correction and editor’s note the Journal has published, in date order, linked to the piece it amends. We do not silently edit a published article: the original text stays, the correction is printed with it, and it is recorded here.

• 8 Jun 2026
  Editor’s note
  Isobaric substitutions: identical masses, different moleculesTwo paragraphs on resolving power were rewritten after a reader objected that the article used the phrase “high resolution” to cover requirements differing by a factor of twenty-five. The Journal now states the discrimination required rather than the specification claimed.
• 1 Jun 2026
  Editor’s note
  Steady state is not a metaphor — it is an arithmetic fact about exenatideThe accumulation table in this article was recalculated after a reader pointed out that the original assumed a fourteen-day rather than a seven-day dosing interval in one row.
• 17 May 2026
  Editor’s note
  Aseptic technique for people who were never taught itThe reconstitution table was re-indexed by both vial mass and diluent volume, replacing an earlier single-concentration version that the desk judged too easy to misread as a fixed answer.
• 14 May 2026
  Editor’s note
  How a claim degrades between the bench and the listingThis article was edited by the standards desk rather than by the analytics desk, because two of the four organisations covered advertise in this publication. The advertising relationships are disclosed in the text, in a sidebar, and on the funding page.
• 9 May 2026
  Correction
  A certificate is a snapshot, and peptides do not stand stillAn earlier version gave the median interval between manufacture and analysis in the Journal’s audit as nineteen days. The correct figure among the fifty certificates carrying both dates is eleven days; the higher figure included documents where one date was absent.
• 20 Apr 2026
  Correction
  Two documents, two disciplines: purity and sterility are not the same tradeAn earlier version of this article stated that the threshold pyrogenic dose for intrathecal administration is 0.5 endotoxin units per kilogram. The figure in the compendial guidance is 0.2 endotoxin units per kilogram per hour.
• 11 Apr 2026
  Correction
  LAL, kinetic chromogenic, recombinant factor C: three ways to the same figureAn earlier version of this article stated that the threshold pyrogenic dose for intrathecal administration is 0.5 endotoxin units per kilogram. The figure in the compendial guidance is 0.2 endotoxin units per kilogram per hour.
• 3 Apr 2026
  Correction
  The submitted sample: the weakest link in the whole systemAn earlier version of the price table converted invoiced amounts at a single annual exchange rate. Amounts are now converted at the rate applying on each invoice date, which changes two of the seven bands.
• 1 Apr 2026
  Correction
  What PeptideMeter is able to certify about a sealed vial, and what it is notAn earlier version described dye ingress testing as a deterministic leak test method. It is probabilistic, and the compendial guidance is explicit in preferring deterministic methods.
• 27 Mar 2026
  Correction
  Twenty-eight days is a number from somebody else’s productAn earlier version described mean kinetic temperature as the average of the highest and lowest recorded temperatures. It is the single temperature which, held constant, would produce the same degradation as the measured profile, computed through an Arrhenius weighting.
• 22 Mar 2026
  Editor’s note
  Low endotoxin recovery: the interference nobody looks forThe section on aseptic process simulation was rewritten to state that the interventions performed during a media fill matter as much as the number of units filled. The original text gave a unit count without that qualification and was quoted back to us, fairly, as reassurance.
• 16 Mar 2026
  Editor’s note
  The imaging substudy is an afterthought in the protocol and the centrepiece of…A paragraph comparing lean-mass outcomes between semaglutide and tirzepatide substudies was removed before publication. The Journal concluded that cross-trial comparison of substudies with different scanners, durations and populations could not support the comparison, and that publishing it would have licensed exactly the ranking we criticise elsewhere in the piece.
• 9 Mar 2026
  Editor’s note
  What the four services actually offer, test by testTwo paragraphs setting out what this article does not allege were added at the request of the standards desk after the first version was read by two correspondents as an accusation against named companies. No company named here has been shown to us to have misrepresented a test result.
• 6 Mar 2026
  Correction
  The same signature, the same peak, three different batchesAn earlier version reported that two of twenty programme suppliers print the peptide sequence on the certificate. The correct count at the last quarterly cycle was six; the figure of two refers to those reporting counter-ion identity and content.
• 22 Feb 2026
  Editor’s note
  Why "muscle-sparing" is a marketing term and not a measurementThis article was revised after publication to state the imaging substudy sample size alongside every composition figure quoted, following correspondence pointing out that the original quoted parent-trial enrolments in two places.
• 19 Feb 2026
  Editor’s note
  What the survodutide schedule assumes about a person it has never metThis file was revised after publication to state the estimand behind every weight-change figure quoted from the STEP and SURMOUNT programmes. Two figures moved by less than a percentage point; no conclusion changed.
• 16 Feb 2026
  Editor’s note
  Sarcopenia is a diagnosis, not a synonym for losing lean massThe bone section was expanded after publication to give the soft-tissue attenuation artefact its own paragraph. Two readers, one a densitometry physicist, argued that treating it as a parenthetical understated its size relative to the effects being reported.
• 15 Feb 2026
  Editor’s note
  The convention nobody explains before handing over the vialThe reconstitution table was re-indexed by both vial mass and diluent volume, replacing an earlier single-concentration version that the desk judged too easy to misread as a fixed answer.
• 13 Feb 2026
  Editor’s note
  Vitamin D, adiposity, and a volume-of-distribution problemThe section on micronutrient monitoring was rewritten before publication. An earlier draft reproduced a surveillance schedule adapted from post-bariatric practice; the standards desk objected that printing a schedule while stating that no evidence supports it would function as a recommendation, and the section now describes the mechanisms and the evidence gap instead.
• 7 Feb 2026
  Correction
  Reading the survodutide titration schedule as a regulatory artefactAn earlier version of this file gave the semaglutide weight-management ladder as reaching 2.4 mg at week thirteen. On the standard four-week step the top rung is reached at week seventeen.
• 26 Jan 2026
  Correction
  What a low result means when intake has halvedAn earlier version reported that serum calcitonin monitoring is recommended for patients taking this drug class. It is not recommended for that purpose, and the sentence has been removed.
• 26 Jan 2026
  Correction
  The incidence tables, read line by lineAn earlier version stated that asymptomatic lipase elevation on treatment indicates subclinical pancreatitis. It does not; enzyme elevation is common on treatment and is not diagnostic.
• 22 Jan 2026
  Editor’s note
  The glycaemic panel before treatment: what is worth measuring and whenThe Journal declines requests to interpret readers’ own laboratory results and receives several each week. Our reasoning is set out on the standards page: a panel without a history, an examination and a reason for ordering cannot be interpreted by anybody, and a publication attempting it would be practising rather than reporting.
• 22 Jan 2026
  Correction
  Who signed this, and what did they undertake by signing itAn earlier version gave the median interval between manufacture and analysis in the Journal’s audit as nineteen days. The correct figure among the fifty certificates carrying both dates is eleven days; the higher figure included documents where one date was absent.
• 13 Jan 2026
  Editor’s note
  The supply gap as a clinical eventA paragraph on intermittent dosing schedules was rewritten before publication after the standards desk objected that describing a schedule in detail, even with an explicit statement that no evidence supports it, functions as a protocol. The section now states the schedules in general terms and the evidentiary position in specific ones.
• 11 Jan 2026
  Editor’s note
  The ceiling is a tolerability finding, not an efficacy oneThe residual-exposure table was recalculated and the steady-state assumption made explicit after a reader pointed out that interrupting mid-escalation clears faster than the original figures implied.
• 3 Jan 2026
  Editor’s note
  What the four services actually offer, test by testThe section on aseptic process simulation was rewritten to state that the interventions performed during a media fill matter as much as the number of units filled. The original text gave a unit count without that qualification and was quoted back to us, fairly, as reassurance.
• 1 Jan 2026
  Editor’s note
  What actually helps, ranked by evidenceThe management table was regraded after correspondence from a gastroenterologist and a clinical pharmacist, both of whom objected that the original version did not distinguish measures with transferred evidence from measures with none at all.
• 24 Dec 2025
  Correction
  The batch number is the only thing connecting the page to the vialAn earlier version of the model test table gave the residual acetonitrile limit as 41 ppm. The figure in the source certificate was 410 ppm, consistent with the relevant residual-solvent classification.
• 23 Dec 2025
  Editor’s note
  Oxidation, and the peroxide that came in with the surfactantA paragraph on headspace gas composition was added to the list of documentation items after a reader pointed out that the article discussed nitrogen backfill in the section on oxidation and then omitted it from the changes it asked for.
• 17 Dec 2025
  Correction
  The upper rungs, ranked by evidenceAn earlier version described the tirzepatide label as requiring escalation to 15 mg. The label permits escalation in 2.5 mg increments to a maximum of 15 mg and identifies 5 mg as a maintenance option.
• 9 Dec 2025
  Correction
  The reviewer, the approver, and why regulated documents carry two namesAn earlier version of the model test table gave the residual acetonitrile limit as 41 ppm. The figure in the source certificate was 410 ppm, consistent with the relevant residual-solvent classification.
• 3 Dec 2025
  Correction
  Custody, condition and the temperature nobody loggedAn earlier version stated that all four services record the condition of samples on arrival. Two do so as standard; the others record some elements of it. The table and text have been corrected.
• 7 Nov 2025
  Correction
  Why people actually stop, in the order they actually stopAn earlier version reported that lifestyle support was withdrawn in the STEP 4 placebo arm. Lifestyle support continued in both arms of STEP 4; it was the STEP 1 extension in which support was withdrawn alongside the drug.
• 5 Nov 2025
  Correction
  Asparagine, glycine, and the two residues that decide a shelf lifeAn earlier version of the shipment table gave the mean kinetic temperature of parcel 9. That figure has been withdrawn: the logger memory was exhausted on day five of an eleven-day journey and no mean kinetic temperature can be computed from a truncated record.
• 26 Oct 2025
  Correction
  How to check a certificate in ten minutesAn earlier version stated that accreditation to ISO/IEC 17025 covers a laboratory as an organisation. Accreditation is granted for a defined scope of activities listed in a published schedule, and a test may fall outside it.
• 23 Oct 2025
  Correction
  What a batch number is supposed to meanAn earlier version of the model test table gave the residual acetonitrile limit as 41 ppm. The figure in the source certificate was 410 ppm, consistent with the relevant residual-solvent classification.
• 17 Oct 2025
  Correction
  Eating enough protein on a suppressed appetite is an arithmetic problem before…An earlier version described a fall in appendicular lean mass as meeting the definition of sarcopenia. Published definitions require low measured muscle strength as the entry criterion; reduced mass alone does not qualify.
• 14 Oct 2025
  Correction
  What SELECT tells us about maintenance, and what it does notAn earlier version reported that lifestyle support was withdrawn in the STEP 4 placebo arm. Lifestyle support continued in both arms of STEP 4; it was the STEP 1 extension in which support was withdrawn alongside the drug.
• 11 Oct 2025
  Editor’s note
  Target dose, effective dose, and the distance between themThis file was revised after publication to state the estimand behind every weight-change figure quoted from the STEP and SURMOUNT programmes. Two figures moved by less than a percentage point; no conclusion changed.
• 8 Oct 2025
  Editor’s note
  A vial can be 99.4 per cent pure and not sterile in the same afternoonTwo paragraphs setting out what this article does not allege were added at the request of the standards desk after the first version was read by two correspondents as an accusation against named companies. No company named here has been shown to us to have misrepresented a test result.
• 6 Oct 2025
  Editor’s note
  When a template becomes a misrepresentationThe section on cloned and templated documents was rewritten twice. The current text reports documentary observations and declines to infer a cause, on the standards desk’s judgement that honest transcription, forwarded documents for the wrong lot, and fabrication cannot be distinguished from the page alone.
• 3 Oct 2025
  Correction
  The badge economy, and what it is actually certifyingAn earlier version described the selection-effect calculation as a model of vendor behaviour. It is illustrative arithmetic demonstrating the sensitivity of a published mean to a publication threshold, and the caption now says so.
• 29 Sep 2025
  Editor’s note
  Retest date, expiry date, and the two conventions being confusedThe audit figures in this article are refreshed each quarter as the dossier programme cycle completes. Counts in earlier versions referred to earlier cycles and are preserved in the standards desk’s published register rather than in the article.
• 27 Sep 2025
  Editor’s note
  The label says four weeks. The clinic says whatever holds.This file was revised after publication to state the estimand behind every weight-change figure quoted from the STEP and SURMOUNT programmes. Two figures moved by less than a percentage point; no conclusion changed.
• 13 Sep 2025
  Correction
  What Medutest would need in order to certify sterility, and why it does notAn earlier version stated that sterilising filtration removes bacterial endotoxin. It does not; lipopolysaccharide aggregates pass a 0.22 micron membrane without hindrance, and this is the reason endotoxin control is a separate discipline.
• 12 Sep 2025
  Editor’s note
  What column temperature does to selectivityThis article was revised after publication to state that charged aerosol and mass-based detection respond more uniformly than ultraviolet absorbance and therefore mitigate the response-factor problem, following correspondence pointing out that the original text left the limitation looking fundamental rather than instrumental.
• 8 Sep 2025
  Editor’s note
  Aggregation: the impurity your chromatogram dissolved before it lookedThis article was revised to add that a coolant pack removed from a frequently opened freezer may begin its journey partially thawed, and therefore with less than its nominal latent-heat budget, following correspondence from a reader who ships temperature-sensitive material commercially.
• 29 Aug 2025
  Editor’s note
  Asparagine, glycine, and the two residues that decide a shelf lifeThis article was revised to add that a coolant pack removed from a frequently opened freezer may begin its journey partially thawed, and therefore with less than its nominal latent-heat budget, following correspondence from a reader who ships temperature-sensitive material commercially.
• 11 Aug 2025
  Correction
  Six words a verification badge would need to mean anythingAn earlier version described the selection-effect calculation as a model of vendor behaviour. It is illustrative arithmetic demonstrating the sensitivity of a published mean to a publication threshold, and the caption now says so.
• 11 Aug 2025
  Editor’s note
  Asparagine, glycine, and the two residues that decide a shelf lifeThis article was revised to add that a coolant pack removed from a frequently opened freezer may begin its journey partially thawed, and therefore with less than its nominal latent-heat budget, following correspondence from a reader who ships temperature-sensitive material commercially.
• 10 Aug 2025
  Correction
  A supply gap is a pharmacological eventAn earlier version reported the SURMOUNT-4 withdrawal arm as regaining 9.9 per cent of body weight. The published figure for weight regain in the placebo arm over the 52-week randomised period is approximately 14 per cent.
• 8 Aug 2025
  Correction
  The badge economy, and what it is actually certifyingAn earlier version stated that all four services record the condition of samples on arrival. Two do so as standard; the others record some elements of it. The table and text have been corrected.
• 26 Jul 2025
  Correction
  What happens to a stopper after the fourth needleAn earlier version reported that all four independent testing services offer bacterial endotoxin determination. None of the four offered it as a standard catalogue item at the time of writing; the endotoxin results reported here were obtained from a separate contract laboratory.
• 24 Jul 2025
  Correction
  The supply gap as a clinical eventAn earlier version reported that lifestyle support was withdrawn in the STEP 4 placebo arm. Lifestyle support continued in both arms of STEP 4; it was the STEP 1 extension in which support was withdrawn alongside the drug.
• 21 Jul 2025
  Correction
  The eighteenth month, and the conversation that should have happened in the…An earlier version stated that the step from 1.7 mg to 2.4 mg is a 1.7-fold increase. It is approximately 1.41-fold; the 1.7-fold step is from 1.0 mg to 1.7 mg.
• 6 Jul 2025
  Correction
  How to read the silence on a certificate you already holdAn earlier version described the compendial sterility test as requiring seven days of incubation. The requirement is fourteen days, with periodic examination for visible growth.
• 5 Jul 2025
  Correction
  The dose that got you here and the dose that keeps you hereAn earlier version of this article described the STEP 1 extension as a randomised withdrawal trial. It was an off-treatment observational extension in a subset of participants, in which both the drug and the lifestyle support were withdrawn.
• 3 Jul 2025
  Editor’s note
  Percentage of loss, absolute kilograms, and the sleight of hand between themA paragraph comparing lean-mass outcomes between semaglutide and tirzepatide substudies was removed before publication. The Journal concluded that cross-trial comparison of substudies with different scanners, durations and populations could not support the comparison, and that publishing it would have licensed exactly the ranking we criticise elsewhere in the piece.
• 2 Jul 2025
  Correction
  The mechanism behind the mechanismAn earlier version placed the area postrema in the hypothalamus. It lies in the floor of the fourth ventricle, in the brainstem.