the size 18 to 12 / XXL to L thing is the part I’d hang onto, honestly, because clothing fit is a better multi-channel readout than the scale on any given week. glycogen and intracellular water resettle for weeks after a caloric drop, so the morning number lies in both directions while a sleeve that suddenly closes does not. anyway, since people in these threads keep asking when to step up, the dosing side: the 4-week-per-step cadence everyone treats as gospel is a registration-trial artifact. SURMOUNT picked clean 4-week intervals because a trial needs uniform titration arms, not because tirz’s PK demands it. half-life is ~5 days, so steady state on a given dose lands around day 25-30 regardless of how you feel at week 2. that’s the actual clock for “is this dose done working on me yet.” the trap is conflating two clocks. GI tolerance (mostly the GIP arm, peripheral, gut motility) settles early. central appetite suppression (mostly GLP-1) shows up later. “i feel fine” answers the first question. “am i still losing” answers the second. titrating to side effects instead of steady state is how people end up at 12.5 wondering why 5 “stopped working” three weeks in. third clock nobody names: the vial. if appetite creep tracks the age of your compounded vial rather than your injection cycle, that’s not tolerance, that’s potency drift, and excursion history integrates over the whole 4 weeks (MKT, not min/max). fwiw the weekly trend summary in the CareClinic app is the only reason i could tell that one of my “stalls” was actually a 6-day flat spot bracketed by losses, not a real plateau. single datapoints lie; the rolling line doesn’t. hold longer than you think. ymmv.
the two-clock framing is the part I’d keep, GI tolerance settling early on the GIP side vs central appetite suppression showing up later. that’s a real distinction and most people titrate against the wrong one. where I’d gently push back is treating “steady state plasma at day 25-30” as the same thing as “this dose is done working on me.” those are two different curves. plasma concentration plateauing around day 25 only tells you the dose is done climbing in your blood. whether the glucose and appetite response has plateaued is a separate question, and it only collapses into the plasma curve if the concentration-effect relationship is roughly linear across that range. for glucose specifically I’m not convinced it is. receptor occupancy can saturate well below peak plasma, which means the back half of your titration to steady state might be adding concentration that isn’t buying you much additional effect. so “am i still losing at day 28” isn’t a clean readout of “has plasma stabilized,” it’s a readout of where you sit on a dose-response curve nobody has mapped for n=1. the vial-age clock I’ll grant outright, and the MKT point is the part more people need to hear. excursion history integrates over the whole storage window, so a vial that spent four hours warm in transit three weeks ago is still a different vial than the spec sheet says, and min/max readings undersell that. I log injection date but I’ll admit I don’t log fridge placement, which came up in a shelf-life thread and made me realize my own tracking has a hole there. on the trend-line vs single datapoint thing, agreed, a 6-day flat spot bracketed by losses is not a plateau and people abandon doses over exactly that noise. seven months of injection-day vs day-6 deltas is what finally gave me a trough signal I trust, and even then I can’t separate dose-response position from plasma trajectory without changing dose, which I’m not in a hurry to do. so I’m a lousy case study for my own hypothesis. hold longer, yeah, but know which curve you’re actually waiting on. ymmv.
One readout missing from all of this is the lipid panel, and it misleads in a specific way during active fat mobilization. People pull a panel a couple months into titration, watch calculated LDL-C tick up, and either panic or blame the drug, when part of that bump is a particle-skew artifact while triglyceride-rich remnants are still clearing. ApoB or an NMR particle count tells a cleaner story than the Friedewald-calculated value in that window, because the estimate gets unreliable exactly when triglycerides are in motion. I’m not saying ignore a genuine rise, just that the timing of the draw matters as much as the number, same point you made about reading day-28 as a setpoint when it isn’t one. A lipid value pulled while you’re still dropping liver fat is reading a transient. The fasting glucose number rides the same channel, for what it’s worth. The morning value is mostly a hepatic story, and it keeps falling for a while as liver fat and DNL come down, independent of where your appetite response has landed. So two “metabolic” readouts can still be improving on their own clocks well after the scale flattens. I’d re-draw the lipids once weight has been flat for a stretch before reading anything into them. What that implies for dosing I’d leave to whoever’s writing the script.
The case for ApoB over Friedewald in that window is solid, and I’d grant it: when triglycerides are in motion the calculated LDL-C is just a bad estimate, and a direct particle count sidesteps the math. But that fixes the estimation artifact, not the timing problem you raised in the same breath. ApoB pulled during active GI disruption on a dose ramp is still measuring a moving target, just measuring it accurately. Those are two separate issues, and switching assays only closes the first one. My own February panel landed right at peak GI disruption, and in hindsight any lipid draw from that window was reading a different patient than six weeks later. The re-draw-once-flat advice is the real fix, not the assay swap.
The two-clocks frame is genuinely the useful part here, and “titrating to side effects instead of steady state” is exactly the error I watched a friend make at 12.5. Where I’d push is the clean handoff you’ve drawn, GI tolerance to the GIP arm, central appetite suppression to GLP-1. That assignment reads tidier than the receptor work actually supports. Gastric emptying delay is largely GLP-1 mediated, and the tachyphylaxis people feel as “GI settling” is itself a GLP-1 phenomenon, so the early clock isn’t a clean GIP readout the way the split implies. It doesn’t undo your point, the two timescales are real and worth tracking separately, it just means “i feel fine” and “am i still losing” aren’t cleanly one-receptor-each. The vial-age clock is the one I’d keep front of mind though, that’s the confound most people never log. I started tagging reconstitution date on the daily check-in for exactly that reason, since potency drift and tolerance look identical on the scale.