Non-invasive spent culture media cfDNA analysis for preimplantation aneuploidy screening
By day 5–7 of in vitro development, a human embryo reaches the blastocyst stage: a fluid-filled sphere of ~100–200 cells comprising an outer trophectoderm (future placenta) and an inner cell mass, ICM (future fetus). As with essentially all proliferating tissue, a fraction of these cells undergo programmed cell death — apoptosis — and in doing so release small fragments of cell-free DNA (cfDNA) into the blastocoel cavity and, ultimately, into the surrounding drop of culture medium.
Embryo culture is not a passive process — normal development involves continuous cell turnover. As the blastocyst expands and hatches, some trophectoderm and ICM cells are naturally shed or undergo apoptosis, fragmenting their nuclear DNA into short double-stranded pieces (typically ~100–300 base pairs, similar in size distribution to circulating cfDNA seen in maternal blood during pregnancy).
These fragments accumulate in two compartments: the blastocoel cavity (the fluid-filled interior of the blastocyst) and, by diffusion across the zona pellucida and trophectoderm, the surrounding spent culture medium in which the embryo has been incubating. Longer culture duration generally allows more cfDNA to accumulate in the medium, which is why day 6–7 spent media samples tend to be more informative than day 5 samples.
Because this DNA is released as a normal by-product of development — not extracted by a physical biopsy — spent media analysis is often described as "biopsy-free" or "non-invasive" PGT-A (niPGT-A).
Preimplantation genetic testing for aneuploidy (PGT-A) screens embryos for the correct number of chromosomes before transfer, since chromosomal aneuploidy is a leading cause of implantation failure and miscarriage. The current clinical standard, trophectoderm (TE) biopsy, removes 5–10 cells from the trophectoderm using a laser and biopsy pipette.
While TE biopsy is considered safe in experienced hands, it is an invasive micromanipulation procedure performed on a living embryo, requires specialized equipment and training, and — like any physical intervention on the embryo — some clinicians and patients would prefer to avoid it if an equally reliable non-invasive alternative existed. Spent media analysis is being investigated precisely because it could, in principle, provide similar genetic information without ever touching the embryo.
The central technical challenge of niPGT-A begins right here, at the source: the total quantity of embryonic cfDNA available in a spent media droplet is extremely small — typically in the picogram range, several orders of magnitude less DNA than is obtained from a trophectoderm biopsy of 5–10 whole cells.
Working with such minute quantities of degraded, fragmented DNA — mixed with an unknown amount of contaminating non-embryonic DNA (addressed in Stage 3) — is far more technically demanding than working with a intact cell biopsy, and is the primary reason niPGT-A protocols remain under active development and validation rather than routine clinical use.
Collecting a niPGT-A sample could not be procedurally simpler compared to biopsy: at the point the embryo is normally transferred, frozen, or moved to a fresh droplet, the embryologist simply aspirates a small volume (typically ~10–20 µL) of the used culture medium the embryo has been incubating in. No laser, no biopsy pipette, and no cells are ever removed from the embryo itself.
Trophectoderm biopsy is a subtractive procedure: cells are physically removed from the embryo, and that removal itself carries a small but non-zero theoretical risk of disrupting subsequent development or reducing implantation potential — a risk that has been studied extensively but never fully eliminated as a concern in the field.
Spent media collection, by contrast, is a purely passive collection: the embryologist aspirates only the surrounding fluid the embryo has already released material into, using the same pipetting steps already built into standard IVF culture workflows (e.g., media changes, pre-vitrification handling). Nothing is added to the procedure that touches the embryo differently than routine culture already does.
Because no additional manipulation of the embryo itself is required, niPGT-A is frequently promoted as removing the theoretical biopsy-related risk to embryo viability entirely — its single clearest advantage over trophectoderm biopsy.
In a typical protocol: the blastocyst is cultured individually in a microdrop (rather than in group culture, to avoid cross-contamination between embryos); at the planned assessment day, the full spent media droplet is collected into a PCR tube before the embryo is vitrified (frozen) or transferred; a fresh, unused aliquot of the same culture medium lot is collected in parallel as a negative control/background blank.
This individual-culture requirement is itself a workflow change from some labs' standard group-culture practices, and represents one of the practical adoption costs of niPGT-A even though the collection step itself is simple.
Despite its simplicity, the collection step is where a major source of downstream error is introduced: any cumulus cells, granulosa cells, or other maternal cells adhering to the outside of the zona pellucida (left over from oocyte retrieval and denudation) can be inadvertently aspirated along with the medium.
These maternal cells contribute their own DNA — genetically distinct from the embryo — directly into the sample before any laboratory processing even begins. Thorough embryo washing through several droplets before final culture is used to minimize this risk, but it cannot guarantee complete removal, which is why contamination remains a live concern carried into every subsequent stage.
The picogram quantities of cfDNA recovered from spent media are far below the input requirements of standard NGS library preparation, so the sample must first undergo whole-genome amplification (WGA) to generate enough material to sequence. This is also the stage where the field’s single biggest technical challenge becomes unavoidable: WGA amplifies whatever DNA is present — embryonic and maternal alike — without distinguishing between them.
Whole-genome amplification techniques (commonly multiple displacement amplification, MDA, or multiple annealing and looping-based amplification cycles, MALBAC) are used to exponentially copy the tiny amount of starting cfDNA — often a billion-fold or more — until enough DNA exists to build an NGS sequencing library.
This extreme amplification introduces its own artifacts: uneven ("biased") amplification across different regions of the genome, allele dropout, and chimeric amplification products. These technical artifacts must be computationally corrected for, and they add noise on top of the biological signal the test is trying to measure — chromosome copy number.
WGA has no way to tell embryonic DNA apart from any co-aspirated maternal cell DNA — both are amplified with equal efficiency. If cumulus or granulosa cells were inadvertently collected with the medium (Stage 2), their DNA can end up representing a substantial fraction of the final amplified sample, sometimes dominating the true embryonic signal entirely.
Because the mother's cells are chromosomally normal (euploid) in the vast majority of cases, contaminating maternal DNA tends to "dilute" or mask a true embryonic aneuploidy signal toward a false-normal result — a systematic bias that is especially concerning in a screening test meant to catch abnormal embryos.
Maternal cumulus/granulosa cell DNA contamination is widely cited in the literature as the single greatest technical barrier to reliable niPGT-A — several published protocols report measurable contamination in a meaningful subset of samples despite careful embryo washing.
Labs are pursuing several approaches to reduce contamination impact: extensive multi-step embryo washing before individual culture; SNP-based genotyping of the spent media sample against a maternal blood/buccal reference to detect and quantify contamination computationally; bioinformatic filtering algorithms that flag samples exceeding a contamination threshold as "no result" rather than reporting a potentially unreliable call; and stricter laboratory protocols (single-embryo culture, DNase treatment of plasticware, minimizing open-air handling).
None of these approaches fully eliminates the problem — they reduce its frequency and help flag when it has occurred, which is one reason niPGT-A results are still generally treated as adjunctive or investigational rather than a stand-alone replacement for biopsy-based testing.
The amplified library is sequenced using low-pass (shallow) next-generation sequencing, typically producing a few million reads per sample — far less depth than clinical whole-genome sequencing, but sufficient to estimate relative chromosome copy number by counting how many reads align to each chromosome and comparing that to the expected diploid baseline.
After sequencing, reads are computationally aligned to a reference human genome and counted in bins across each chromosome. In a chromosomally normal (euploid) sample, read density should be uniform and proportional across all 23 chromosome pairs. A whole extra copy of a chromosome (trisomy) shows up as a ~50% increase in relative read density for that chromosome; a missing copy (monosomy) shows up as a ~50% decrease.
Because the underlying DNA has already been through picogram-scale extraction and billion-fold WGA, the resulting copy-number profile is inherently noisier than one derived from a trophectoderm biopsy, where more, less-degraded starting material is available. Bioinformatic pipelines apply GC-content correction, bin normalization, and smoothing algorithms to reduce this technical noise before making a call.
Any maternal DNA mixed into the sample (Stage 3) is sequenced right alongside the embryonic DNA, and both contribute reads to every chromosome bin. If the embryo carries a trisomy but ~30% of the sequenced DNA is actually normal maternal DNA, the apparent read-density increase for that chromosome is diluted — potentially below the detection threshold — producing a false-negative (missed abnormal) result.
Conversely, uneven WGA amplification bias can, in some cases, create the appearance of copy-number variation where none truly exists, contributing to false-positive calls. Both failure modes are why result confidence is reported alongside the raw copy-number call, and why samples with high estimated contamination are often flagged as inconclusive rather than given a definitive normal/abnormal result.
The interaction between contamination level and call reliability is direct and measurable: as the fraction of non-embryonic DNA in a sample rises, the statistical confidence of any copy-number call falls, and the risk of a mis-classified embryo rises correspondingly.
Because of these compounding sources of noise, niPGT-A laboratory reports typically include: a per-chromosome copy-number call (normal / gain / loss / mosaic-range), an estimated contamination fraction where SNP-based maternal-reference comparison was performed, and a qualitative or quantitative confidence indicator for the overall result.
Samples that fail internal quality-control thresholds (insufficient DNA amplified, contamination above a defined cutoff, or excessive technical noise) are generally reported as "no result" rather than forced into a normal/abnormal call — a deliberately conservative design choice given the test's investigational status.
The ultimate question for any new diagnostic is: does it agree with the established reference standard? For niPGT-A, that reference standard is trophectoderm biopsy. Published concordance studies comparing spent-media results to matched TE biopsy results from the same embryos report a wide range of agreement — commonly cited figures span roughly 70% to 90% — reflecting real differences in protocols, contamination-control rigor, and study populations across the field.
Different research groups report meaningfully different concordance rates between niPGT-A and trophectoderm biopsy — some studies report agreement in the high 80s–90% range for whole-chromosome aneuploidies, while others, especially those including mosaic or segmental calls, report figures closer to 70% or even lower.
This spread is not simply "noise" — it reflects genuine methodological differences: how rigorously embryos were washed before culture, whether individual vs. group culture was used, which WGA chemistry and NGS pipeline were applied, how contamination was measured and filtered, and how discordant calls were adjudicated (biopsy is itself an imperfect proxy for whole-embryo ploidy, due to mosaicism). Both tests are indirect samples of a complex, sometimes mosaic embryo, so "ground truth" itself is genuinely uncertain.
Despite its invasiveness, trophectoderm biopsy has over a decade of large-scale clinical validation, standardized laboratory protocols, and outcome data linking biopsy-based PGT-A results to live birth rates. It samples more, less-degraded DNA from more cells, generally yielding a more robust and reproducible signal than spent-media analysis currently achieves.
Major professional bodies in reproductive medicine continue to describe niPGT-A as a promising but not-yet-validated alternative: encouraging as a research and adjunctive tool, but not currently recommended as a stand-alone replacement for biopsy-based PGT-A in routine clinical decision-making about which embryo to transfer.
As of the most recent published guidance and literature, niPGT-A remains largely investigational and is not yet considered standard of care in most fertility clinics — larger, prospective, multi-center validation studies with long-term outcome data are still needed before it could replace trophectoderm biopsy.
For niPGT-A to move from investigational to routine use, the field generally agrees it needs: standardized, reproducible laboratory protocols across centers (culture conditions, washing steps, WGA chemistry, bioinformatic pipelines); robust, validated methods to detect and quantify maternal contamination in every sample, with clear rules for when a result should be withheld; larger prospective concordance studies powered to detect mosaicism and segmental abnormalities, not just whole-chromosome aneuploidy; and ultimately, outcome studies linking niPGT-A-based embryo selection to implantation and live-birth rates, the way biopsy-based PGT-A has already been studied.
Until that evidence base matures, niPGT-A is best understood as a genuinely promising, biopsy-free research direction — one that could someday reduce embryo manipulation risk — rather than a ready clinical substitute for trophectoderm biopsy today.
| Product | Indication | Trial Design | Key Result |
|---|---|---|---|
| Trophectoderm (TE) Biopsy | 5–10 cells physically removed from trophectoderm | Established reference standard; validated across large clinical cohorts; requires laser micromanipulation | High DNA input, most reproducible signal, longest track record |
| Spent Media (niPGT-A) | cfDNA aspirated from used culture droplet | Zero direct embryo contact; picogram DNA input requires heavy WGA amplification | No biopsy-related manipulation risk to the embryo |
| Contamination Risk | Maternal cumulus / granulosa cell DNA | Can co-aspirate with spent media; amplified indistinguishably from embryonic DNA | Mitigated (not eliminated) by washing + SNP-based QC filtering |
| Clinical Status | Concordance ~70–90% vs. TE biopsy across studies | Investigational; used mainly in research settings and select adjunctive protocols | Active area of validation research, not yet standard of care |