Association between fetal fraction and positive predictive value of noninvasive prenatal testing for sex chromosome aneuploidies: a retrospective study of 68,290 cases

Noninvasive prenatal testing (NIPT) for sex chromosome aneuploidies (SCAs) has consistently demonstrated lower positive predictive values than for common autosomal trisomies, and the extent to which fetal fraction and maternal characteristics are independently associated with true-positive sex chromosome aneuploidy results remains incompletely characterized. This retrospective cohort study included 68,290 consecutive pregnancies undergoing NIPT between January 2020 and December 2024, from which 399 cases with high-risk NIPT results for fetal sex chromosome aneuploidies (SCAs) were identified. Of these, 276 cases accepted prenatal diagnosis via one or more platforms. Of 399 SCA screen-positive cases identified among 68,290 NIPT examinations, 276 (69.2%) underwent confirmatory invasive prenatal diagnosis. The overall PPV was 38.4% (106/276). PPV varied markedly by SCA subtype: 47, XYY 80.0% (8/10), 47, XXY 62.9% (22/35), 47, XXX 57.7% (15/26), and 45, X 18.9% (25/132). The mean FF was significantly higher in the SCA cohort compared with the non-SCA population ( P < 0.001) and in true-positive cases compared with false-positive cases ( P = 0.001). In multivariable analysis, FF remained independently associated with true-positive status after adjustment for maternal age, gestational age, and BMI (OR 1.059 per 1% increase, 95% CI 1.024–1.094, P < 0.001). PPV exhibited a significant dose-response relationship with FF: 14.3% at FF 4–8%, 25.0% at FF 8–12%, and 42.3% at FF ≥ 12% ( P = 0.016). Stratification by four maternal age groups showed no significant overall trend in PPV (Cochran-Armitage trend test, P = 0.673), although subtype-specific analysis revealed age-dependent patterns for 45, X and 47, XXX. Subtype-specific PPVs for NIPT-based SCA screening are reported. Fetal fraction showed an independent association with true-positive status after adjustment for maternal age, gestational age, and BMI, with a dose-response relationship across FF categories, whereas maternal age was not associated with PPV, and twin pregnancies could not be reliably assessed because of the small number of confirmed cases. These findings suggest a potential role for FF in post-test SCA risk stratification, though invasive confirmation remains essential.