Yaron Goikhman

Dr. Yaron | Reproductive Medicine & Genetics Expert | Innovating fertility care with AI | Speaker | Family Health Advocate | Making parenthood dreams real.

[Yaron Goikhman @YaronMD]

@dissproportion The embryo data backs you up. In 15,169 biopsied embryos (Franasiak, F&S 2014), aneuploidy was ~21-27% at ages 26-30, 58% at 40 and 88% at 44.

Most of those errors start in the egg. Sperm health matters, but "it's mostly the man" isn't what the chromosomes show.

[Yaron Goikhman @YaronMD]

@NightShiftMD The line I'd underline from that CBC piece: 20 to 25 eggs frozen in your early 30s gives about a 75% chance of a baby.

Fewer eggs or freezing later, and the odds fall. Only ~11% have come back for their eggs so far. Worth it for some, but it buys odds, not a guarantee.

[Yaron Goikhman @YaronMD]

Canada's 2025 IVF registry: PGT-A went from 20.5% of cycles in 2019 to 40.7%. ERA fell to 1.7%.

One add-on shrank as evidence came in. The other doubled.

PGT-A has real uses, but in younger patients with good embryos, big trials haven't shown more babies. So who is it for?

[Yaron Goikhman @YaronMD]

@obioraokeke1 Kind idea, and it works for one generation: AA + SS means every child is AS, never SS. AA + AS, half are AS.

But the gene doesn't disappear, it spreads. Those AS kids will need to check their partner's genotype too.

Testing every generation is what prevents SS.

[Yaron Goikhman @YaronMD]

Exome carrier screening in Mexico's Jewish community (preprint, Hadassah + Mexico City): 787 couples, 49 (6.2%) both carriers for the same recessive condition.

Standard panels would have missed 12 of the 49.

But ~two-thirds were mild conditions. More genes, more counselling.

[Yaron Goikhman @YaronMD]

@FT The pitch for editing embryos skips a step. For most at-risk couples, PGT-M already exists: with a recessive condition about 3 in 4 embryos are unaffected, with a dominant one about half.

Editing really matters only when no unaffected embryo is possible. That's rare.

[Yaron Goikhman @YaronMD]

@Current_Urology The counselling section is the part couples need. With an AZFc deletion, surgery still finds sperm about 67-74% of the time, but every son conceived inherits the deletion.

The consensus says PGT to transfer female embryos may be considered. Best discussed before retrieval.

[Yaron Goikhman @YaronMD]

Orchid's own team checked 10 embryo biopsies against the genomes of the kids later born.

All 42 de novo variants in 3 of those kids were in the biopsy data. Impressive.

But its 1,200-gene panel missed ~1 in 60 variants, and n=10. I'd want an outside lab and hundreds of pairs.

[Yaron Goikhman @YaronMD]

@FertStert Good case to pick. PGT-A spread well before the big trials. In one of the largest, Yan et al. (NEJM 2021, 1,212 women with 3+ good blastocysts), cumulative live birth was 77.2% with PGT-A vs 81.8% without.

Worth asking at journal club: benefit per transfer, or per patient?

[Yaron Goikhman @YaronMD]

@m_goes_distance Worth keeping these apart. Embryo screening edits nothing. It ranks embryos already made, so the gain is capped by those few.

ASRM's ethics and practice committees said last December that polygenic embryo screening isn't ready for clinical use. Fast adoption isn't proof.

[Yaron Goikhman @YaronMD]

ASRM 2026 opens tomorrow. One genetic counselling prize paper's title says a lot: "Increasing patient-initiated raw data requests for PGT."

Your embryo data is yours. But re-running it outside the lab's validated pipeline isn't a second opinion. Talk it over with a counsellor.

[Yaron Goikhman @YaronMD]

@iam_preethi Agreed, his health counts. One nuance from that paper (Andrology, 4,520 couples): high sperm DNA damage tracked with worse semen numbers, and lower fertilization in heavier men. But no link with pregnancy or live birth.

So test him too, just don't oversell a DFI result.

[Yaron Goikhman @YaronMD]

@dantawfik For families, the mutation type matters: m.3243A>G passes down the maternal line at unpredictable levels, while single large deletions are mostly sporadic (~1 in 24 recurrence from affected mothers, Lancet 2004).

Newcastle (NEJM 2025): PGT gave 18 births, mito donation 8.

[Yaron Goikhman @YaronMD]

Abu Dhabi: 11,748 couples had premarital genetic screening by July 2026, and 565 were flagged for "genetic incompatibility." About 1 in 20.

I'd drop that word. These couples are at risk, not incompatible. Counselling, prenatal testing or IVF with PGT-M give them real options.

[Yaron Goikhman @YaronMD]

@my_helix Good picks: BRCA1/2, Lynch and FH all have prevention that works, and ~1 in 75 positive is a lot of families.

One add: all three are dominant, so a positive parent has a 50% chance of passing it on. Counselling should cover testing relatives, and PGT-M for those who want it.

[Yaron Goikhman @YaronMD]

@TurkishMinuteTM 811 trips for 41 women: about 20 each on average. That's the scandal under the "blond babies" headline.

Repeated stimulation carries real risks like OHSS. And the network catalogued hair colour. Did anyone check carrier status? Regulated donor programs screen for that.

[Yaron Goikhman @YaronMD]

Overture's robotic ICSI is now on sale to US IVF labs, as a Class II device exempt from FDA premarket notification.

Their first-babies paper (RBMO 2023): 13 of 14 robot-injected eggs fertilized vs 16 of 18 by hand.

Steadier hands, great. Now show a proper trial.

[Yaron Goikhman @YaronMD]

@genomics_policy Great topic. Canada is an interesting case: the AHR Act bans identifying embryo sex except for sex-linked disease, yet PGT-A reads X and Y anyway. Did you look at how clinics handle that, or how they explain mosaic results?

[Yaron Goikhman @YaronMD]

@noor_siddiqui_ Lydia's family deserves every option. Worth adding: after an apparently de novo variant, sibling recurrence is usually around 1%. Deep sequencing of the parents' blood finds low-level mosaicism in a few percent, and then targeted PGT-M for that exact variant is the sharper tool.

[Yaron Goikhman @YaronMD]

Premarital thalassaemia screening often starts with one number: MCV of 80 or below.

Taiwan, 2000-2024, 9,949 confirmed carriers: that cutoff missed about 5.3%. MCV 82 or MCH 27 or below caught over 99%, with more false alarms.

Exploratory, but a cheap fix worth testing.