First-Line Ovulation Induction in PCOS: Letrozole's Better Numbers vs. Clomiphene's Label
A PCOS patient trying to conceive is offered an off-label aromatase inhibitor with better trial numbers over the decades-old, FDA-labeled standard — a case about weighing real evidence against a regulatory default.
Meredith A., a 31-year-old woman, has worked as a veterinary technician for eight years and has spent the last fourteen months timing intercourse around ovulation- predictor kits that, for a woman with PCOS-related anovulation, keep coming back unhelpfully blank. She and her husband have been trying to conceive for just over a year, meet the standard definition of infertility for her age, and were referred for ovulation induction after a basic workup found nothing else contributing — normal semen analysis, normal hysterosalpingogram, and PCOS as her only identified factor.
The choice between the two first-line agents isn't close on the trial evidence, though it complicates on the regulatory picture. PPCOS II (Legro et al., NEJM 2014), a large randomized trial specifically in PCOS patients like Meredith, found a live- birth rate of 27.5% with letrozole against 19.1% with clomiphene — a meaningful absolute difference in the outcome that actually matters to her, and the finding behind ASRM's current preference for letrozole as first-line in this specific phenotype. Clomiphene, though, remains the only agent of the two actually FDA-labeled for ovulation induction; letrozole's use here is off-label, an aromatase inhibitor developed and approved for breast cancer whose reproductive-endocrinology use rests entirely on trial evidence built up outside its original indication. Meredith herself mentioned, unprompted, that she'd read something online about letrozole and birth defects and wanted that addressed before agreeing to either option.
Fourteen months of unsuccessful timed intercourse has already cost Meredith and her husband real ground on the calendar she'd hoped to conceive within, and she said plainly that she doesn't want to spend a full treatment cycle on a drug she's quietly worried about, only to switch afterward once the worry catches up with her. She brought a printed page of what she'd read, unprompted, and asked the team to look at it directly rather than simply reassure her in the abstract — a request the visit is being run around, since the actual source of that concern turns out to matter more to how it gets addressed than a general reassurance would. Her husband's semen analysis, drawn earlier this cycle and reviewed at today's visit, came back within normal parameters on every measured value, which the team noted directly removes one variable from the conversation and keeps the decision focused squarely on which ovulation-induction agent fits her.
Fertility consult, workup complete
My default is to start with the agent actually labeled for this indication. Clomiphene has decades of real-world use specifically for ovulation induction, an FDA label to match, and a well-characterized safety record. Letrozole's use here is off-label — approved for breast cancer, not fertility — and I'd want a clear, specific reason before reaching past the labeled option first.
If the trial evidence for the two drugs looked roughly equivalent, I'd stop right here — the labeled option would be the obvious default with nothing more to discuss.
The specific reason exists, and it's not a small one. PPCOS II randomized PCOS patients directly against each other on these two drugs and found a live-birth rate of 27.5% with letrozole versus 19.1% with clomiphene — an absolute difference large enough that ASRM has since named letrozole preferred first-line specifically for PCOS-related anovulation. Off-label doesn't mean under-studied here; it means the regulatory label hasn't caught up to a large, specific, randomized trial finding.
I'd agree the labeled option is the right default when the evidence is close or absent — but PPCOS II isn't a small or preliminary study, and treating "not FDA-labeled for this indication" as equivalent to "less evidence- based" gets the actual evidentiary picture backwards in this specific case.
Whichever drug is chosen, Meredith's own question deserves a direct answer, not a reassurance without substance behind it. The birth-defect concern traces back to an early, unreplicated conference abstract that was never substantiated in the controlled data that followed, including PPCOS II's own safety monitoring. It's worth naming that history specifically — not just telling her it's fine, but telling her why the concern she read didn't hold up, so she isn't quietly distrusting whichever drug she ultimately takes.
Agreed: letrozole started as first-line, with the PPCOS II data and ASRM's reasoning explained to Meredith directly, and the historical, unsubstantiated teratogenicity concern addressed by name rather than dismissed without explanation. Clomiphene named explicitly as the next step if letrozole doesn't produce ovulation within three cycles.
All three voices converged on the same regimen once the trial evidence and the regulatory-label question were separated from each other — the disagreement in this case resolved once it became clear the two positions were answering different questions (which drug has the better outcome data vs. which drug carries the formal label) rather than genuinely competing claims about the same one.