Systematic Review & Meta-Analysis of Positron Emission Tomography/Computed Tomography and Bone Scan in the Diagnosis of Prostate Lesions

Jiafu Wang; Yue Han; Lin Lin; Linhan Zhang; Jin Li; Huiqi Gao; Peng Fu

Disclosures

Transl Androl Urol. 2021;10(11):4231-4240. 

In This Article

Results

Search Results and Basic Information of Literature

In this meta-analysis, 392 documents were retrieved from the electronic database, 125 documents were retrieved from the register, 10 repeatedly published documents were eliminated, 73 unqualified documents were eliminated, and 8 were excluded for other reasons, leaving 426. After reading the titles and abstracts of articles, 32 articles were deleted, leaving 394 articles. After preliminary screening, the full text of each article was read, and 365 articles were excluded (for example, if 18F-FDG PET/CT or BS is not used; Comments, meetings, etc.). Another 26 articles were excluded in the final stage of screening, including 12 articles with incomplete data and 14 articles involving patients other than prostate lesions. Finally, 3 articles were confirmed to be included in the meta-analysis. Figure 1 presents a flow chart for the literature retrieval and screening process used.

Figure 1.

Literature screening process. 68Ga-PSMA, 68Ga-prostate specific membrane antibody; PET, positron emission tomography; CT, computed tomography; BS, bone scan.

A total of 215 patients were considered in the 3 articles that met the inclusion criteria. All of the included articles were small sample studies, with sample sizes ranging from 28 to 113 cases. They described in detail the TPs, TNs, FPs, FNs, sensitivity, and specificity of the diagnostic modalities considered. Table 2 shows the basic characteristics of the included literature.

Results of Risk Bias Evaluation of Literature

Figures 2,3 are the results of the risk bias assessments of the references, plotted using the RevMan 5.3 software according to the QUADAS2 quality assessment tool. In this study, from the 3 randomized controlled trials, only 2 (66.67%) randomized controls described the correct randomized allocation method, and only 1 (33.33%) described the hidden allocation scheme in detail. The measurement index in this study was the laboratory index determined by the computer, so it can be considered that the blind method was correctly used in all the papers.

Figure 2.

Risk bias evaluation results for the included articles.

Figure 3.

Risk bias assessment diagram for the included articles.

Meta-analysis of 68Ga-PSMA PET/CT Diagnosis of Bone Metastases

A total of 3 articles analyzed the diagnosis of bone metastases by 68Ga-PSMA PET/CT in randomized controlled trials. Figure 4 is a meta-analysis of the sensitivity and specificity of 68Ga-PSMA PET/CT in the diagnosis of bone metastases. The highest sensitivity for 68Ga-PSMA PET/CT was 0.96, with 95% CI: 0.87, 1.00, and the highest specificity was 1.00, with 95% CI: 0.96, 1.00.

Figure 4.

Meta-analysis of sensitivity and specificity of 68Ga-PSMA PET/CT in the diagnosis of bone metastases. 68Ga-PSMA, 68Ga-prostate specific membrane antibody; PET, positron emission tomography; CT, computed tomography; CI, confidence interval; TP, true positive; FP, false positive; FN, false negative; TN, true negative.

Meta-analysis of Bone Scan Diagnosis of Bone Metastases

A total of 3 articles analyzed the diagnosis of bone metastases by BS in randomized controlled trials. Figure 5 is a meta-analysis of the sensitivity and specificity of BS for the diagnosis of bone metastases. The highest sensitivity and specificity of BS were 0.92 with 95% CI: 0.81, 0.98 and 0.96 with 95% CI: 0.78, 1.00, respectively.

Figure 5.

Meta-analysis of sensitivity and specificity of bone scan in diagnosis of bone metastases. CI, confidence interval; TP, true positive; FP, false positive; FN, false negative; TN, true negative.

SROC Curve of 68Ga-PSMA PET/CT and BS

Figure 6 shows the SROC curve for diagnosis of bone metastases from 68Ga-PSMA PET/CT imaging. The results of meta-analysis of 68Ga-PSMA PET/CT diagnosis with confirmation by surgical and histopathological examination showed that the area under the SROC curve (AUC) =0.826 and standard error (SE) (AUC) =0.0425. Figure 7 shows the SROC curve for diagnosis of bone metastases by BS. The results of meta-analysis of BS diagnosis with confirmation by surgical and histopathological examination showed that the area under the SROC curve (AUC) =0.714 and SE (AUC) =0.0034. These data suggest that 68Ga-PSMA PET/CT is a better imaging modality than BS in the diagnosis of bone metastases from prostate tumors.

Figure 6.

SROC curve of 68Ga-PSMA PET/CT diagnosis of bone metastases. 68Ga-PSMA, 68Ga-prostate specific membrane antibody; PET, positron emission tomography; CT, computed tomography; SROC, summary receiver operating characteristics.

Figure 7.

SROC curve of bone scan diagnosis of bone metastases. SROC, summary receiver operating characteristics.

processing....