Real-World Performance of the Afirma Genomic Sequencing Classifier (GSC)—A Meta-analysis
Authors' conclusion
RW data for the Afirma GSC demonstrates significantly better oSP and oPPV performance than the VS, indicating an increased yield of cancers for resected GSC suspicious nodules. The higher BCR likely increases the overall rate of clinical observation in lieu of surgery.
Study context
- Population
- Indeterminate thyroid nodule cytology tested with Afirma GSC; NIFTP considered malignant13 real-world (RW) studies
- Comparison
- Real-world Afirma GSC performance vs 2018 validation study
- Outcome
- GSC performance (sensitivity, specificity, PPV, NPV) in RW studies vs VS
Pooled result
As reported in the abstract
GSC diagnostic performance
| Group | Sensitivity | Observed specificity | Conservative specificity | Specificity | Observed PPV | Conservative PPV | PPV | NPV |
|---|---|---|---|---|---|---|---|---|
| RW | 97% | 88% | 80% | — not stated in the abstract | 65% | 49% | — not stated in the abstract | 99% |
| Validation study (VS) | 91% | — not stated in the abstract | — not stated in the abstract | 68% | — not stated in the abstract | — not stated in the abstract | 47% | 96% |
Sensitivity
RW97%
Validation study (VS)91%
Observed specificity
RW88%
Validation study (VS)— not stated in the abstract
Conservative specificity
RW80%
Validation study (VS)— not stated in the abstract
Specificity
RW— not stated in the abstract
Validation study (VS)68%
Observed PPV
RW65%
Validation study (VS)— not stated in the abstract
Conservative PPV
RW49%
Validation study (VS)— not stated in the abstract
PPV
RW— not stated in the abstract
Validation study (VS)47%
NPV
RW99%
Validation study (VS)96%
— not stated in the abstract
Further pooled results
| Outcome | RW | P value |
|---|---|---|
| Benign call rateRW vs VS · higher in RW | 67% | P < 0.05 |
- Benign call rate: VS 54%
Original publication
Real-World Performance of the Afirma Genomic Sequencing Classifier (GSC)—A Meta-analysis. J Clin Endocrinol Metab. Published 6 December 2022. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →