PGT testing: three tests, three different questions about your embryos

If your clinic has mentioned PGT, chances are they presented it as one thing: PGT-A, a chromosome count. What most families don't realize is that there are two other forms of preimplantation genetic testing, and they answer completely different questions about your embryos.

The genetic information is there from the moment of fertilization. PGT-A reads chromosome count. Preimplantation genetic testing for monogenic disorders (PGT-M) reads specific inherited variants. Preimplantation genetic testing for polygenic conditions (PGT-P) reads polygenic disease risk. Each one looks at a different layer of the same genome.

These tests are often complementary rather than alternatives. PGT-A has the longest clinical track record and roughly halves the number of transfers most patients need. PGT-M has prevented diseases like cystic fibrosis and sickle cell in families who knew they carried the variants. PGT-P is the newest layer, and it's been tested with a validation methodology (within-family comparison on sibling pairs) that most established genetic tests have never had to undergo. Which test or tests make sense for your situation depends on the question you're trying to answer, your cycle data, and your family history.

Counting chromosomes: what PGT-A can and can't tell you

PGT-A counts chromosomes. A healthy embryo needs exactly 46 (23 pairs). When cell division goes wrong during egg or sperm formation, an embryo can end up with too many or too few. That's called aneuploidy, and it's the most common cause of failed implantation and early miscarriage. Most aneuploid embryos can't develop into a pregnancy regardless of how they look under a microscope.

That last part matters. Without PGT-A, embryo screening relies on morphological grading: a trained embryologist evaluating the embryo's appearance. A top-graded embryo (3AA) has roughly a 45% live birth rate per transfer. A lower-graded one (1BB) sits around 25%. But morphological grading can't see chromosome count. A beautiful-looking embryo can be aneuploid. An average-looking one can be perfectly euploid.

That's why PGT-A was developed, and why clinics recommend it so often. On day 5 or 6, an embryologist biopsies a few cells from the embryo's outer layer and checks them for chromosome count. The logic is straightforward: find euploid embryos before transfer, skip the ones destined to fail.

PGT-A has the strongest evidence base in specific populations. In women 38 and older, it significantly shortened time to live birth (hazard ratio 1.46). For patients with recurrent pregnancy loss, it improved live-birth rates to 53% compared to 43% without testing. For most IVF patients, it also roughly halves the number of transfers needed, which typically covers the cost of the test itself.

For younger patients with no history of loss, the main gain is time-to-pregnancy rather than cumulative outcome. Across 263,521 cycles, cumulative live-birth rates were 74.1% with PGT-A and 74.0% without. The test shifts when a live birth happens (fewer failed transfers first) more than whether one happens at all, when patients have access to enough cycles.

PGT-A answers one specific question: does this embryo have the right number of chromosomes? It's the right question for most IVF patients because aneuploidy is the most common cause of embryo loss, and the benefit is largest when age or pregnancy history makes aneuploidy more likely. That's why clinics recommend it broadly. See our detailed guide on PGT-A for the full picture, plus what it costs to add to a cycle.

But chromosome count tells you nothing about single-gene conditions or polygenic disease risk. An embryo can be chromosomally normal and still carry the BRCA1 variant, or have a high genetic predisposition toward type 2 diabetes. PGT-A can't see those.

Reading single genes: where PGT-M fits

PGT-M asks a different question entirely: does this embryo carry a specific genetic variant that's already known to be in the family?

It screens for single-gene (monogenic) conditions: cystic fibrosis, sickle cell disease, Huntington's disease, BRCA1/2, Tay-Sachs. These are caused by variants in a single gene, and whether the variant is present in the embryo directly determines whether the future child will be affected or carry the risk. The variant is there from conception. PGT-M for BRCA1/2 has been offered since 2009, and no one calls that testing "unvalidated" because the children haven't reached breast cancer onset age yet. The variant is either there or it isn't.

For families who know they carry one of these variants, PGT-M is the most straightforward form of genetic testing available. The UK's Human Fertilisation and Embryology Authority currently approves PGT-M for over 2,000 genetic conditions. And families don't need a fertility problem to access it. If both partners are carriers of a recessive condition like cystic fibrosis, they can use IVF specifically to screen embryos before transfer, even if they'd conceive naturally without difficulty.

In countries where carrier screening and PGT-M have been widely adopted, the numbers are striking. In Israel, where screening and IVF were made free for cystic fibrosis carriers, the prevalence of the disease among newborns declined by over 90%. Families who were carriers could go to a fertility clinic, work with a doctor to create embryos through IVF, then transfer an embryo that didn't carry both copies of the disease-causing variant.

But PGT-M has a hard limit. It only works when you're looking for a known variant in a known gene. The condition has to be identified in the family first (typically through carrier screening), and the variant has to have high enough penetrance that its presence in the embryo predicts disease. Carrier frequency for any individual recessive condition is roughly 3%, so the odds of both partners carrying the same one are low. But if both partners are carriers, the stakes for that specific embryo are high, and PGT-M addresses them directly.

For those 2,000+ conditions, PGT-M works. But most diseases don't work this way. Type 2 diabetes, heart disease, Alzheimer's, schizophrenia, breast cancer (outside of BRCA). These aren't caused by one variant. They're influenced by many genetic variants, each with a small effect, combining differently in every person. PGT-M can't screen for those. And those are the diseases that fill most hospital beds.

(A note on PGT-SR: preimplantation genetic testing for structural rearrangements screens for chromosomal problems like translocations, deletions, and inversions. It's relevant for families with a known structural rearrangement and clinically established, but it's a narrower application than the other three types.)

Scoring polygenic risk: why PGT-P exists

PGT-P answers the question that PGT-A and PGT-M can't: what's this embryo's genetic predisposition toward common complex diseases?

It works differently from the other two. Instead of counting chromosomes or reading a single gene, PGT-P calculates a polygenic score for each embryo. Genetic studies over the past two decades have identified specific DNA variants associated with common diseases. For type 2 diabetes, there are hundreds of variants across the genome, each nudging risk up or down by a small amount. No single variant is decisive the way a BRCA1 mutation is for breast cancer. But taken together, they produce real differences in risk.

A polygenic score combines all of these effects into a single number for each embryo: an estimate of that embryo's genetic predisposition toward a given disease. The same biopsy taken for PGT-A can be used for polygenic scoring. Each variant contributes a small effect on its own. Together, they produce real differences between embryos from the same parents.

How much difference? Consider a family where one or both parents have type 2 diabetes and they produce 10 embryos through IVF. The absolute risk difference between the highest-risk and lowest-risk embryos can reach up to 23.5%. That's a concrete gap in predicted disease probability between siblings who share the same parents and the same environment. One child could face a much higher lifetime risk of diabetes than another, based purely on which combination of variants they inherited.

If you've read anything about PGT-P at all, chances are you've encountered skepticism. Professional societies including ASRM have expressed concerns about whether PGT-P is ready for clinical use, citing predictive uncertainties. The core question is fair: do polygenic scores actually predict disease risk within families, where embryos share parents and environment? Or do they only look accurate because they pick up population-level patterns that vanish when you compare siblings?

This has been tested directly. Moore et al. (2025) validated 17 disease polygenic scores on sibling pairs. 16 of 17 showed no decrease in predictive performance within families compared to unrelated individuals. The signal the scores capture is real genetic signal, not a statistical artifact of comparing strangers.

The first PGT-P baby was born in 2019. Compared to PGT-A and PGT-M, the field is young. But "young" and "unvalidated" aren't the same thing. PGT-P has been tested with a validation methodology (within-family comparison on sibling pairs) that most established genetic tests have never had to undergo, because embryos are the first place where within-family comparison actually matters for clinical decisions. Independent research has confirmed the within-family approach works outside of any single company's data.

PGT-P doesn't add a separate biopsy. It reads the same genetic material that PGT-A already collects, but analyzes it for a different kind of information. For families considering both PGT-A and PGT-P, the practical cost of adding polygenic scoring is incremental, not a second round of procedures. (We cover how long PGT testing takes and what it costs in separate guides.)

Matching the test to your question

None of these tests changes an embryo. An embryo's genetic makeup is fixed from the moment of fertilization. PGT reads what's already there. The question is which layers you read, and that depends entirely on what you're worried about.

Think about three different families walking into the same IVF clinic. A 40-year-old with no family history of monogenic disease and repeated failed transfers. Her primary obstacle is probably aneuploidy, and PGT-A directly addresses it. A couple in their early 30s, both cystic fibrosis carriers. They don't have a fertility problem; they're using IVF specifically to screen embryos for CF. PGT-M is why they're here, and most clinics add PGT-A alongside it because it roughly halves the number of transfers needed, which typically covers the cost. A family with a strong history of type 2 diabetes and coronary artery disease who've produced 10 embryos. Their concern is polygenic risk, and only PGT-P speaks to it, though PGT-A still has a role in prioritizing transfer order.

Same clinic, same biopsy procedure, same lab. Three different primary questions, often with more than one test in the answer.

Most clinics lead with PGT-A because it's the most widely applicable: aneuploidy is the most common cause of failed implantation across IVF patients, and PGT-A has the longest clinical track record. PGT-M gets added when there's a known family variant. PGT-P is still the newest layer, though it reads from the same biopsy PGT-A uses. The right starting point is your situation and the question you're trying to answer, and for many families that means more than one test.

Three questions sit underneath most PGT conversations: Does this embryo have the right chromosome count? Does it carry a known single-gene variant? What's its polygenic risk profile for common diseases? Each question has a test. Each test has evidence behind it. The families who make the best decisions are the ones who understand which questions their situation actually raises, and talk them through with a clinician who can weigh the answers alongside cycle data, family history, and cost.

Our counselors can help you figure out which question matters most for your family. For PGT-P specifically, you want someone who understands polygenic scoring and within-family validation, because the interpretation is different from traditional genetic testing. If you want to go deeper on a specific test, we've written guides on PGT-A and PGT-P individually, plus practical details on timing and cost.

If you're ready to talk about which questions matter for your family, reach out to us.