Table 2

Correlation of IGHV mutation status with FCRL, CD38, and ZAP-70 expression

FCRL1FCRL2FCRL3FCRL5CD38ZAP-70 (T/B)ZAP-70 (%)
UM 
    Median (range) 9.53 (1.01-30.7) 1.79 (1.00-3.90) 2.04 (1.02-12.1) 1.43 (1.00-2.72) 49.8 (0.68-99.4) 2.79 (0.85-6.28) 26.9 (0.58-94.7) 
    Mean ± SD 10.0 ± 7.05 1.93 ± 0.89 2.89 ± 2.36 1.52 ± 0.40 48.5 ± 36.6 2.82 ± 1.11 29.5 ± 19.1 
MT 
    Median (range) 26.0 (1.58-45.5) 8.89 (1.05-19.4) 8.15 (1.22-26.9) 3.45 (1.05-8.35) 1.63 (0.71-99.9) 4.80 (1.40-13.0) 10.1 (1.00-59.9) 
    Mean ± SD 24.2 ± 9.86 9.07 ± 3.73 9.49 ± 6.03 3.57 ± 1.31 11.3 ± 26.0 5.24 ± 2.17 14.6 ± 13.1 
Cutoff value ≥ 15 < 15 ≥ 4.2 < 4.2 ≥ 4.0 < 4.0 ≥ 2.3 < 2.3 < 30% ≥ 30% ≥ 4.0 < 4.0 < 20% ≥ 20% 
    UM (n = 52) 45 52 45 50 19 33 47 14 38 
    MT (n = 55) 45 10 49 46 49 49 39 16 42 13 
Sensitivity, % 81.8 89.1 83.6 89.1 89.1 70.9 76.4 
Specificity, % 86.5 100 86.5 96.2 63.5 90.4 73.1 
PPV, % 86.5 100 86.8 96.1 72.1 88.6 75.0 
NPV, % 81.8 89.7 83.3 89.3 84.6 74.6 74.5 
Concordance, % 84.1 94.4 85.0 92.5 76.6 80.4 74.8 
C-statistic 0.861 0.964 0.877 0.946 0.803 0.864 0.760 
FCRL1FCRL2FCRL3FCRL5CD38ZAP-70 (T/B)ZAP-70 (%)
UM 
    Median (range) 9.53 (1.01-30.7) 1.79 (1.00-3.90) 2.04 (1.02-12.1) 1.43 (1.00-2.72) 49.8 (0.68-99.4) 2.79 (0.85-6.28) 26.9 (0.58-94.7) 
    Mean ± SD 10.0 ± 7.05 1.93 ± 0.89 2.89 ± 2.36 1.52 ± 0.40 48.5 ± 36.6 2.82 ± 1.11 29.5 ± 19.1 
MT 
    Median (range) 26.0 (1.58-45.5) 8.89 (1.05-19.4) 8.15 (1.22-26.9) 3.45 (1.05-8.35) 1.63 (0.71-99.9) 4.80 (1.40-13.0) 10.1 (1.00-59.9) 
    Mean ± SD 24.2 ± 9.86 9.07 ± 3.73 9.49 ± 6.03 3.57 ± 1.31 11.3 ± 26.0 5.24 ± 2.17 14.6 ± 13.1 
Cutoff value ≥ 15 < 15 ≥ 4.2 < 4.2 ≥ 4.0 < 4.0 ≥ 2.3 < 2.3 < 30% ≥ 30% ≥ 4.0 < 4.0 < 20% ≥ 20% 
    UM (n = 52) 45 52 45 50 19 33 47 14 38 
    MT (n = 55) 45 10 49 46 49 49 39 16 42 13 
Sensitivity, % 81.8 89.1 83.6 89.1 89.1 70.9 76.4 
Specificity, % 86.5 100 86.5 96.2 63.5 90.4 73.1 
PPV, % 86.5 100 86.8 96.1 72.1 88.6 75.0 
NPV, % 81.8 89.7 83.3 89.3 84.6 74.6 74.5 
Concordance, % 84.1 94.4 85.0 92.5 76.6 80.4 74.8 
C-statistic 0.861 0.964 0.877 0.946 0.803 0.864 0.760 

FCRL cutoff values were determined according to the ROC and CART methods. The indexes including sensitivity, specificity, PPV, NPV, concordance (concordance with IGHV mutation status), or C-statistic (predictive of model accuracy) were evaluated. A C-statistic of 1.0 indicates perfect predictive discrimination, while values greater than 0.9 indicate outstanding, 0.8 to 0.9 indicate excellent, and 0.7 to 0.8 indicate fair discrimination.44 

or Create an Account

Close Modal
Close Modal