Sequencing a fetus: everything depends on which fetus
Fifty-three percent in one group, two percent in another. Same test, same year. And in normal-looking pregnancies, most of what comes back is uncertainty.
Sequencing is the most powerful tool in this whole stage, and it is genuinely useful. Where it is useful is the part that gets left out of the sales pitch.
Before birth, in a fetus with something visible on the scan. Two studies were published on the same day in 2019. Both enrolled only fetuses that already had a structural anomaly on ultrasound, and only after the standard chromosome tests had come back normal.
| Study | Fetuses analysed | Diagnosis found | Findings that could not be interpreted |
|---|---|---|---|
| UK PAGE study | 610 | 8.5% (52) | 3.9% (24) uncertain but possibly useful |
| US study, same issue | 234 | 10% (24) | 20% (46) suggestive but not interpretable |
Read the last column. In the US study there were twice as many uninterpretable findings as diagnoses. Uncertainty is not a footnote in this test. It is often the main product.
The yield depends almost entirely on which fetus is being tested. In a pooled analysis of 4,350 fetuses with anomalies, sequencing added 53% in fetuses with isolated skeletal problems and 2% in fetuses whose only finding was increased nuchal translucency. Same test, same year, twenty-six-fold difference. Any advertised figure for prenatal sequencing that does not name the group it applies to is meaningless.
Before birth, in a fetus that looks completely normal. This is the situation being marketed to ordinary pregnancies, so it deserves its own numbers. Pooling four studies and 1,916 structurally normal fetuses with normal chromosome results, sequencing found something extra in 1.6% (confidence interval 1.0 to 2.6). For a severe condition, the figure was 0.5% (0.1 to 1.5).
The largest single series of this kind sequenced 1,020 low-risk pregnancies. It found something potentially significant in 28 of them (2.7%), and something with the potential for a severe outcome in 9 (0.9%). There were 13 terminations in the cohort. Four of those thirteen were for a variant of uncertain significance: a finding the laboratory itself could not classify.
ACOG's position has not changed and is still in force, reaffirmed in 2023: the routine use of whole-genome or whole-exome sequencing for prenatal diagnosis is not recommended outside of the context of clinical trials.
The single-gene blood test. A related product screens your blood for about 30 single-gene conditions. SMFM's view is short: it is not recommended for routine use
, and more peer-reviewed research is needed. In a series of 2,208 women, results varied enormously by why the test was ordered. Where a fetal long-bone abnormality had been seen, 33.7% were positive. Where the only reason was that the father was 40 or older, and the scan was normal, 2 of 912 were positive. Follow-up was available on only about half of all positives, so none of these are positive predictive values.
"In someone like me, with a normal scan, what is the chance this finds something serious?"
"If the result is normal, what does it still not rule out?"