Started 2.5mg in February 2025, sitting at 7.5mg now. Down 38 lbs, but the number I actually care about is A1C: 5.9 to 5.2 in fourteen months. The thing nobody told me going in: fasting glucose on a GLP-1 isn’t stable. It moves with my cycle. Luteal phase, days 15-24, my morning CGM readings jump 8-12 mg/dL without any dietary change. Consistent rise that starts around confirmed ovulation and flattens near day 27. Before tirzepatide I didn’t have enough cycle regularity to see the pattern. Now I have four consecutive cycles and the signal is repeatable. First spontaneous ovulation showed up at month six. Confirmed by LH strips and temp shift, not just the app guessing.
That was the moment I understood this is treating the metabolic problem, not just moving the scale. Baseline fasting insulin was 22. Now sitting at 9. That number explains everything else. The cycle return, the A1C, the fact that I’m not a disaster in luteal anymore. What I haven’t seen anywhere: does the luteal glucose bump blunt at higher doses, or is it a fixed insulin-resistance pattern tied to progesterone that tirz can’t fully offset? My data suggests the latter but I only have four cycles to work with. Anyone else tracking cycle phase against CGM on a GLP-1? The research basically doesn’t exist and I’m curious if this pattern holds across more than one body.
the insulin drop from 22 to 9 is the actual story here, and yeah, four cycles is real data if the conditions are identical, which yours are. your hypothesis about progesterone being a fixed resistance pattern makes sense to me mechanistically, but that’s the limit of what i can say - your doc is the one who can actually tell you whether the bump blunts at higher doses or if you’re seeing the ceiling. i’d ask specifically about that next appt. the bigger win is that you went from not ovulating to regular cycles. tbh that recovery alone tells you the metabolic picture is actually changing. the glucose variability on top of it is just…
noise. meaningful noise, but noise. anyone else tracking this with their own data? fwiw.
“noise” is doing a lot of work there though. if i’m planning to come off tirz to try to conceive in the next 12-18 months, knowing whether the luteal spike is dose-dependent or fixed matters for timing. it’s not academic. but yeah, agree on the insulin drop being the primary signal. 22 to 9 is the thing that explains why everything else moved. the glucose variability is downstream of that, you’re right about the mechanism.
insulin drop explains the cascade, yeah. Four cycles is real data and it matters for timing if you’re planning conception. But the dose-response question needs your clinician. You’re tracking the right signal
i had a patient years ago, pre-glp1 era, pcos and prediabetic, who tracked fasting glucose against basal temp on graph paper for almost a year. same luteal bump, 6 to 10 points, held steady even after metformin brought her overall numbers down. small n of one, but progesterone-driven insulin resistance seemed structural in her case. four cycles is more than most endocrinologists ever see plotted. keep going.
fasting insulin 22 to 9 is the number that jumps out. that shift alone explains so much of what you’re seeing. my situation is T2D not PCOS, so different hormone picture, but I track CGM against cycle phase too bc I noticed the same directional thing. luteal bumps my 14-day average roughly 6-9 mg/dL. consistent across the cycles I’ve had data for. I’m on 7.5mg also, so same dose range as you. my read on your question: I think it’s probably a fixed progesterone-driven IR pattern that tirz attenuates but doesn’t eliminate. the mechanism (progesterone reducing glucose transporter sensitivity) isn’t something GIP/GLP-1 dual agonism fully overrides, at least not at doses we’re using. n=1 on my end, and I haven’t found literature that addresses this specifically either. what I’d want to know: does your luteal bump track with progesterone lab values, or just cycle day? day-based patterns could be masking variance if your cycles are still irregular post-ovulation restoration.
my luteal bump ran a consistent ~10 mg/dL at 7.5mg too, but it narrowed to roughly 5-6 after titrating up. still present, not flat, which partially supports your read. but “tirz can’t fully offset” from four cycles at one dose is a big conclusion - the offset question almost certainly varies with dose, and you haven’t tested that variable yet. the progesterone-IR mechanism is real; the ceiling on how much tirz can blunt it probably isn’t fixed.
progesterone upregulates gluconeogenesis and reduces peripheral insulin sensitivity - that’s not a tirz problem, that’s luteal phase physiology GLP-1s can only partially offset. I see the same 6-10 point fasting bump on roughly CD18-25, currently at 7.5mg, and I ran a small experiment four mos back where I shifted my injection day to CD13 (right before confirmed ovulation) to see if front-loading coverage helped. It didn’t move the luteal reading meaningfully, which supports ur read that this is progesterone-mediated IR rather than a simple coverage gap you can dose around. The one thing I’d add to your dataset: has your SHBG changed since cycles regularised? Mine went from 28 to 51 over twelve months on mounjaro and the luteal bump has modestly softened alongside it - not gone, but smaller. Which makes me think there’s an androgen component layered on the progesterone effect that complicates the clean narrative. Your fasting insulin trajectory (22 to 9) is nearly identical to mine so I’m curious whether your free testosterone has moved at all.
four consecutive cycles is a real dataset, and the luteal IR pattern you’re describing matches what i’d expect from progesterone’s direct effect on insulin receptor sensitivity. the tirz-can’t-fully-offset hypothesis tracks. what i’d push back on slightly: “fixed” might be too strong a word for four cycles. i’ve noticed my own luteal readings are more variable depending on sleep and stress in that phase specifically, which makes me wonder if the progesterone effect is less a ceiling and more an amplifier of other IR inputs. so the signal might be real and consistent, but the magnitude could still move if you control for confounders the CGM doesn’t see. the “eight to twelve point jump” could have a floor that looks different once you’re three or four months at a stable dose with better sleep hygiene in luteal. not trying to muddy your data, just saying the dose question and the lifestyle-in-luteal-phase question might be harder to separate than they look right now. good data collection though, seriously.