Improved positive predictive value of non-invasive prenatal testing through integration with second-trimester ultrasound soft markers for fetal chromosomal abnormalities: a retrospective cohort study.

内容摘要

Non-invasive prenatal testing (NIPT) achieves high sensitivity for common fetal aneuploidies; however, its positive predictive value (PPV) varies substantially with population prevalence, generating a considerable false-positive burden. Current guidelines offer limited guidance on ultrasound-based risk refinement after a positive NIPT result. To determine whether integrating second-trimester ultrasound soft markers with NIPT improves PPV for fetal chromosomal abnormalities and to characterize aneuploidy-type-specific phenotypic profiles to inform risk stratification. This retrospective cohort study included 303 singleton pregnancies at a Chinese tertiary center (September 2022-September 2025). All patients underwent NIPT (12+0-14+6 weeks) and detailed second-trimester ultrasound (19-23 weeks). Eight soft marker categories were assessed. Chromosomal abnormalities were confirmed by karyotyping, and neonatal phenotypic assessment served as the reference for NIPT-negative pregnancies. Predefined combined strategies were compared using diagnostic accuracy metrics and receiver operating characteristic analysis. Twenty-four pregnancies (7.9%) harbored confirmed chromosomal abnormalities. NIPT achieved 100% sensitivity and 100% negative predictive value (upper-bound estimates given differential verification bias; see Methods), but only 55.8% PPV (24/43), with 19 false positives. Among the false-positive cases, 63.2% exhibited no soft markers, and none had &#x2265;2 markers. The high-risk combined strategy (NIPT-positive with &#x2265;1 high-risk marker or &#x2265;2 markers) improved PPV to 83.3% with 98.9% specificity, potentially reducing invasive procedures by 58% through individualized counseling. Strategy (a) (NIPT-positive with &#x2265; 1 marker) achieved 100% sensitivity for trisomies 18 and 13 (uniformly severe marker burdens) but 62.5% for trisomy 21 (37.5% marker-negative). The combined model achieved an AUC of 0.987, significantly exceeding that of NIPT alone (0.966; p = 0.012), although these estimates warrant external validation given the enriched cohort. Women aged <35 years derived the greatest incremental benefit (+28.2 percentage points). In this enriched tertiary referral cohort, integrating second-trimester ultrasound soft markers with NIPT was associated with improved PPV, with particular efficacy for trisomies 18 and 13; however, the observed PPV improvement was predominantly driven by trisomies 18 and 13, which uniformly present with multiple markers, whereas the clinical utility for trisomy 21 was more limited owing to phenotypic heterogeneity (37.5% marker-negative). If confirmed by prospective multicenter validation, this approach may enable individualized post-positive NIPT counseling and complement existing guideline frameworks.