TomTeleRx said:

Steady state is faster than people expect: half-life about five days, so you are at plateau in roughly three to four weeks rather than five.

Dose-response modeling for tirzepatide: Emax model fitting to the STEP/SURMOUNT dose-finding data shows:

Semaglutide: ED50 ≈ 0.6mg, Emax ≈ -18%, Hill coefficient ≈ 1.3
Tirzepatide: ED50 ≈ 6mg, Emax ≈ -25%, Hill coefficient ≈ 1.5

Clinical implication: most patients achieve >80% of maximal response by the mid-range dose (1.7mg sema, 10mg tirz). Going to the maximum dose provides diminishing returns — possibly not worth the additional side effect burden for some patients. Individualize dosing based on response vs tolerability.

One thing that is still open after VendorMark’s answer:

How much of the tirzepatide advantage is the GIP component and how much is simply that the dose ladder goes higher in receptor-occupancy terms?

dan_philly said:

Dose-response modeling for tirzepatide: Emax model fitting to the STEP/SURMOUNT dose-finding data shows: Semaglutide: ED50 ≈ 0.6mg, Emax ≈ -18%, Hill…

There is a second half to this that has not been said yet. ApoB is the more informative number and it is cheap. LDL-C estimates cholesterol mass in atherogenic particles; ApoB counts the particles, and it is particle count that tracks risk — which is why the two can disagree and why the disagreement is the clinically interesting case. Triglycerides fall substantially with weight loss and improved insulin sensitivity, HDL moves modestly, and LDL-C often barely moves at all, which surprises people who expected everything to improve together.

Closing the loop on my own question.

Reporting back after another eight months at the same dose. Still losing slowly, no new side effects, and no reason I can find to climb further.

Dr.EM_Chicago said:

ApoB is the more informative number and it is cheap.

Agreed on the mechanism, with the caveat that the head-to-head used semaglutide 1mg, not 2.4mg. It is still the best direct evidence available, but it is not the comparison most people think they are citing.