38) (Fig. S1). Fifteen eligible studies[4, 13, 14, 16-18, 23, 25, 29, 30, 39, 44, 51, 56, 58] were selected in this part, 12 of which provided data about combination of both markers.[13, 16-18, 23, 25, 29, 30, 39, 44, 51, 56] Sensitivity estimates for DCP, AFP and combination of both markers ranged from 0.19 to 0.92, 0.08 to 0.63, and 0.48 to 0.92 and the specificities estimates for DCP, AFP selleck compound and combination of both markers

were from 0.70 to 0.99, 0.63 to 1.00, and 0.62 to 0.99, respectively (Fig. 5). The summary sensitivity and specificity were 45% (95% CI, 35%–57%) and 95% (95% CI, 91%–97%) for DCP, 48% (95% CI, 39%–57%) and 89% (95% CI, 79%–95%) for AFP, 70% (95% CI, 61%–78%) and 83% (95% CI, 79%–86%) for the combination of both markers. The AUROC of DCP 0.84 (95% CI, 0.81–0.87) was better than Ivacaftor price AFP 0.68 (95% CI, 0.64–0.72) and combination markers 0.83 (95% CI, 0.79–0.86) for detecting early stage HCC (Fig. 4b).

An analysis for funnel plot asymmetry suggested that there was no evidence of publication bias for DCP (P = 0.99), AFP (P = 0.11) and combination of both markers (P = 0.56) (Fig. S2). Based on the forest plot (Fig. 3) and the SROC plot (Fig. 4a), it was clear that the Marrero study[19] and sterling study[33] were important outliers for DCP, the Sassa’s study[14] and Morroto’s study[32] were the primary heterogeneity for AFP. After excluding the outlier studies, the summary sensitivity, specificity and AUROC of DCP were 62%, 91% and 0.83. The estimated values of AFP were 61%, 85% and 0.74. Therefore, the findings that serum DCP was superior to AFP for detecting HCC were robust. According to forest plot (Fig. 5) and the SROC plot (Fig. 4b), the main heterogeneity originated from Sassa’s study[14] for AFP and Volk’s study[25] for DCP. After excluding

the above studies, the summary sensitivity of AFP (54%) was better than DCP (42%), while the AUROC of DCP (0.71) was higher than AFP (0.55). Hence, it was robust that the diagnostic accuracy of Interleukin-3 receptor serum DCP was better than AFP for early stage HCC. Results of the present study show that the sensitivity and specificity of DCP are superior to AFP in diagnosing HCC. Furthermore the diagnosis accuracy of DCP is better than AFP for detection of early stage HCC, and the combination of both markers does not improve results when comparing with DCP alone in diagnosing early stage HCC. These results indicate that DCP might be better than AFP as a single marker for detection of HCC. α-fetoprotein is the most commonly used surveillance marker for HCC. But AFP has a suboptimal performance,[59] as a serological test, its levels may also elevate in patients with cirrhosis or chronic hepatitis in absence of HCC.[60] DCP is a useful marker not only in diagnosing HCC, but also in evaluating the effect of treatment, prognosis, and the risk of recurrence of HCC.[61] DCP has been widely adopted in surveillance and diagnosis of HCC in Japan since the late 1990s, and appeared to have promising results.