The Potential Role of Molecular Analysis in Hereditary Antithrombin (AT) Deficiency Diagnosis and Management.

AT deficiency is classified as type 1 (quantitative), or type 2 (qualitative), based on plasma AT activity and antigen levels. However, such levels can sometimes be only mildly and equivocally reduced. Type 2 deficiency can be subdivided into type 2a (reactive center loop [RCL]), type 2b (heparin binding domain), and type 2c (pleiotropic), usually by molecular analysis, but the role of AT molecular analysis in patient diagnosis and management is uncertain.

1) to estimate the frequency of mutation detection in possible hereditary AT deficiency, and 2) to correlate clinical manifestations with AT deficiency type as determined by plasma AT, and by AT genotype.

Mayo Clinic patients (n=16) were categorized as type 1 or 2 by plasma AT, or by molecular analysis for previously reported type 2a-c mutations. Patient characteristics were abstracted from medical records. The SERPINC1 putative promoter region, all exons (n=7) and splice junctions, and the 3'UTR were PCR amplified from leukocyte genomic DNA, and sequenced with both forward and reverse primers.

The mean patient age at diagnosis was 32 years (range 18-65); 11 (69%) were women. 12 (75%) patients were type 1 by plasma AT, while 4 were type 2 by molecular analysis (Table). The mean plasma AT (range) activity/antigen for types I and 2 patients were 52% (39-71%)/53% (39-68%) and 58% (43-67%)/85% (60-103%), respectively [normal range=80-130%]. 13 (82%) patients had VTE (8 [61%] idiopathic; 10 [77%] recurrent); 5 (32%) had stroke/TIA, 4 at a young age. Thrombosis “penetrance” appeared greater among type 1 families. Only 2 pregnancy losses occurred in 14 pregnancies among 6 type 1 women, and no losses occurred in 5 pregnancies among 2 type 2b women. Among the 14 (88%) patients with mutations identified, all were heterozygous and 8 patients had 6 novel mutations; 2 patients had no mutation detected. Plasma AT activity/antigen did not always correlate with genotype; 1 patient each with type 2b and 2c by molecular analysis had concordant decreases in plasma AT activity and antigen (i.e., a type 1 plasma phenotype), while amino acid changes (T85K and F239L) predicting reduced plasma AT activity only had a type 1 plasma phenotype. One type 2b (Q118P) patient had extensive recurrent VTE.

88% of patients had a detrimental mutation, suggesting that molecular analysis can be helpful for diagnosis. While our findings should be interpreted with caution, type 1 patients may have a higher prevalence of stroke/TIA at a young age and a higher family thrombosis penetrance compared to type 2. Plasma AT activity and antigen frequently did not correlate with genotype. However, whether molecular analysis is useful for clinical management remains uncertain. Finally, 50% of patients had a novel mutation, suggesting that reported mutations causing AT deficiency have not reached gene saturation.

No relevant conflicts of interest to declare.