SURMOUNT-4 and what withdrawal data does and does not tell an individual — one year on posts 31–60

I read post #27 twice before replying, because I had assumed the opposite.

I disagree with the framing of SURMOUNT-4 above, and I think it is a substantive disagreement rather than a terminological one. Setting out why, so it can be checked.

The reasoning depends on an assumption that is doing a lot of work and is never stated. If the assumption holds, the conclusion follows. I do not think it holds generally.

Source for the SURMOUNT-4 figure, since it was asked for. It is in the discussion rather than the abstract, which is why the version circulating is stronger than the paper is.

Reading the surrounding paragraph is worth the two minutes. The authors are more careful than their summarisers.

On storage: published stability data covers the licensed formulation at its licensed concentration. A research preparation reconstituted at home in a different diluent at a different concentration is outside every one of those conditions.

I have no interest in any supplier named above.

SURMOUNT-1 reported weight reduction of a magnitude that was the largest for a pharmacological intervention at that time of publication. It also reported a clear dose response across three doses. The categorical thresholds (people reaching 10%, 15%, 20% loss) got the most attention but read less informatively than the mean weight change.

The general answer and the answer for your case may diverge here.

Coming back to post #31, because the follow-up matters more than the original answer.

Comparisons between the tirzepatide and semaglutide programmes across trials rather than within one are weak. Different populations, different durations, different baseline characteristics; the only fair comparison is a head-to-head one.

The uncertainty is in the assumption, not in the calculation.

Marking my place. If it changes for me I will come back and say so.

I changed my mind about SURMOUNT-4 after someone here asked me for the source and I could not produce one. That is worth saying out loud because it is the ordinary way it happens.

SURMOUNT-4 used a randomised withdrawal design: everyone titrated, then those on maintenance were randomised to continue or placebo. Continuation maintained effect; withdrawal was followed by regain. The finding is robust but it says nothing about a lower effective maintenance dose, because that was not studied.

I have said this before in a thread nobody could find, so it is worth repeating.

Injection-site reactions were reported at a low but non-zero rate across the trials. The practical point is that a reaction at one site does not predict a reaction at the next, and rotating properly makes the question moot.

A qualification I should have led with rather than closed on.

The arithmetic in post #39 is right; the assumption feeding it is the part to check.

The 2.5 mg starting dose is a tolerance step and not a therapeutic one. Judging efficacy at that dose is the single commonest reasoning error in this subcategory.

The rule of thumb is fine; the edge cases are where it earns its keep.

Where I have landed on SURMOUNT-4, having got it wrong once in public: the direction is clear, the magnitude is not, and anyone quoting a precise magnitude has borrowed it from somewhere that did not measure it.

On storage: published stability data covers the licensed formulation at its licensed concentration. A research preparation reconstituted at home in a different diluent at a different concentration is outside every one of those conditions.

Adding the measurement that post #40 says would settle it.

Heart rate rises modestly across this class, tirzepatide included. It is consistent, small, and worth knowing about rather than worth alarm — and it is one of the reasons the trials monitored it explicitly.

I would rather say I do not know than round it up to an answer.

Categorical response thresholds — the proportion reaching ten, fifteen or twenty per cent reduction — are more persuasive and less informative than the mean. They depend entirely on where the threshold was drawn.

That is all I can say without guessing.