Skip to main content

Table 2 Sensitivity and specificity of the WID™-qCIN test

From: DNA methylation-based detection and prediction of cervical intraepithelial neoplasia grade 3 and invasive cervical cancer with the WID™-qCIN test

 

Diagnostic validation

Predictive validation

LBC-CIN Diagnostic

LBC-CIN Predictive

All, % (95% CI)

 Specificity (≤ CIN1)

90 (87–93)

93 (87–97)

 Sensitivity (CIN2)

55 (42–67)

 Sensitivity (CIN3)

69 (56–80)

35 (27–44)

 Sensitivity (CIN3+)

78 (67–86)

35 (27–44)

 Sensitivity (CC)

100 (85–100)

 PPV (CIN3+)

59 (52–68)

57 (43–72)

 NPV (≤ CIN1)

93 (91–95)

84 (82–86)

 NPV (≤ CIN2)

92 (90–95)

84 (82–86)

Age < 30 years, % (95% CI)

 Specificity (≤ CIN1)

95 (89–98)

92 (82–97)

 Sensitivity (CIN2)

46 (28–66)

 Sensitivity (CIN3)

59 (36–79)

15 (7–27)

 Sensitivity (CIN3+)

65 (44–83)

15 (7–27)

 Sensitivity (CC)

100 (40–100)

 PPV (CIN3+)

76 (55–91)

33 (17–58)

 NPV (≤ CIN1)

91 (86–94)

80 (78–82)

 NPV (≤ CIN2)

89 (84–94)

80 (78–82)

Age ≥ 30 years, % (95% CI)

 Specificity (≤ CIN1)

88 (83–92)

94 (85–98)

 Sensitivity (CIN2)

61 (43–76)

 Sensitivity (CIN3)

75 (59–87)

52 (40–64)

 Sensitivity (CIN3+)

83 (71–92)

52 (40–64)

 Sensitivity (CC)

100 (82–100)

 PPV (CIN3+)

57 (49–65)

70 (50–86)

 NPV (≤ CIN1)

95 (92–97)

88 (85–90)

 NPV (≤ CIN2)

94 (91–97)

88 (85–90)

  1. We assessed the WID™-qCIN test in controls (control and CIN1) for different stages of CIN (CIN2, CIN3, CC) in different age groups and datasets as well as in relation to other gynaecological diseases (endometrial and ovarian cancer). For assessment of positive and negative predictive values (PPV, NPV), we included only HPV+ individuals and assumed a prevalence of 21.5% in line with results from a previous systematic review reporting CIN3+ prevalence in triaged women [21]. NPV is reported separately when including samples ≤ CIN1 or ≤ CIN2 as controls
  2. CIN cervical intraepithelial neoplasia, CC cervical cancer (includes adenocarcinoma in situ and invasive cervical cancer), PPV positive predictive value, NPV negative predictive value