Evaluation of the Urine POC-CCA Test Accuracy in the Detection of Schistosoma mansoni Infection: A Systematic Review and Meta-Analysis

Background Schistosomiasis is a common public health problem throughout the world and Schistosoma mansoni is the most prevalent species in Africa. Most endemic countries use the Kato–Katz (KK) stool smear examination for diagnosis, mapping, and monitoring of intervention programs. However, its poor sensitivity calls for an urgency to evaluate and use more accurate diagnostic tools, of which detection of circulating cathodic antigen (CCA) in urine seems promising. Methods Studies published until May 2022 were searched from PubMed, Google Scholar, and grey literature for systematic review and meta-analysis following the PRISMA guideline. Eligible studies were selected based on preset inclusion and exclusion criteria. Quality of included studies was assessed using the QUADAS-2 tool. Heterogeneity between studies was assessed using Cochrane Q test and I2 test statistics. Data were analyzed using Review Manager 5.4.1 and Meta-DiSc 1.4 software programs. Results Thirty-seven studies published in 29 papers and enrolling 21159 study participants were included for analysis. Overall analysis of Point-of-Care Circulating Cathodic Antigen (POC-CCA) test against KK reference standard revealed a pooled sensitivity and specificity of 0.86 (95% CI: 0.85–0.87) and 0.66 (95% CI: 0.65–0.67), respectively. Subgroup analysis among 24 studies comparing single POC-CCA with test single KK revealed a high sensitivity (0.88) but low specificity (0.66). Based on findings of 24 studies, the area under the curve (AUC) for the systematic receiver operating characteristic (SROC) curve was 0.7805, indicating that the POC-CCA test effectively separates those with the disease from those who do not have it. Higher sensitivity estimates of 0.93 and 0.90 were reported when comparisons were made between test results of 2 urine and 1 stool samples, and 3 urine and 3 stool samples, respectively. Single POC-CCA test resulted in a pooled sensitivity estimate of 0.81 (95% CI: 0.78–0.84) as evaluated by the polymerase chain reaction (PCR) reference test. Conclusions The POC-CCA test has higher sensitivity than KK and may serve as a routine diagnostic alternative for disease diagnosis, mapping, and monitoring of interventions. However, its accuracy should further be evaluated at different transmission settings and infection intensity.


Background
Schistosomiasis is a waterborne disease caused by blooddwelling fukes of the genus Schistosoma [1].Schistosoma mansoni, S. japonicum, and S. haematobium are the most common disease causing species [2].Te disease is endemic in 78 countries where 780 million people are at risk of infection.More than 250 million people are infected globally with more than 90% of the infections occurring in sub-Saharan Africa.In Africa, S. mansoni and S. haematobium are widespread causing intestinal and genitourinary schistosomiasis, respectively.Intestinal schistosomiasis poses most common public health problem throughout the continent.Schistosomiasis is a public health problem in Ethiopia where about 53.3 million people are at risk of infection and 4 million are already infected [3,4].Currently, two species are found in Ethiopia in the genus Schistosoma: Schistosoma mansoni is widespread throughout the country while S. haematobium has focal distribution in the low land borders of the country [5][6][7].
Intestinal schistosomiasis can be readily diagnosed using parasitological, immunological, and molecular techniques.Te World Health Organization (WHO) recommends the Kato-Katz (KK) technique as the "gold standard" technique for screening of intestinal schistosomiasis due to its feld applicability and possibility of quantitative reporting [8].However, the KK thick smear is poorly sensitive especially in low transmission areas [9,10].More importantly, the ongoing mass drug administration (MDA) program is thought to decrease infection intensity and reduce worm fecundity that the egg output will be too low to be detected by the KK method.Daily variations in egg excretion and unisex infection also afect the test performance [11][12][13].Tis calls for urgency in evaluation and use of new diagnostic methods which are more sensitive and afordable in resource-limited countries like Ethiopia.In recent years, antigen detection rapid diagnostic tests have sought great attention and both laboratory and feld-based evaluations have been done at diferent geographical settings.Studies show that the urine Point-of-Care Circulating Cathodic Antigen (POC-CCA) test has superior performance as compared to KK [14,15].Other studies, on the contrary, reported that the CCA test has low sensitivity, especially in low transmission settings [16].However, there is paucity of comprehensive data summarizing those fndings.Terefore, considering that systematic reviews provide the best evidence for decision makers, we conducted a systematic review and meta-analysis on urine POC-CCA test accuracy in the diagnosis of infection by S. mansoni.

Study Settings.
Studies conducted all over the world were included in the present review because we believe that geographical location has no direct efect in the performance of the POC-CCA test, rather level of transmission, test interpretation threshold, and number of samples examined mainly afect the accuracy.

Information Sources and Search of Literature.
Potential articles were searched in PubMed, Google Scholar, and grey literature following the PRISMA guideline and checklist updated in 2020 [17].Search in databases was done using the key terms: "Performance, accuracy, Circulating cathodic antigen, CCA, Kato Katz, Polymerase chain reaction (PCR), Schistosoma mansoni" combined with Boolean operators (AND, OR).A search in Google Scholar was made using the term "circulating cathodic antigen" in the title.Te search and study selection was done from June 03 to August 16, 2022, by two reviewers independently.

Study Selection.
Te study selection process is shown in Figure 1.Relevant studies were selected after sequential screening based on the title, abstract, and full text based on the following inclusion criteria: (1)

Statistical Analysis.
Pooled accuracy of urine POC-CCA test was analyzed against reference KK or PCR using Meta-DiSc 1.4 software.Te number of participants with TP, FP, FN, and TN POC-CCA test results was used to calculate sensitivity and specifcity of each study and an overall summary as well.Positive likelihood ratio (LR+), negative likelihood ratio (LR−), and diagnostic odds ratio (DOR) were also calculated.Subgroup analysis was done by number of stool and urine samples tested from each participant.In order to assess the ability of POC-CCA in discriminating participants with S. mansoni infection (TP rate) from noninfected (FP rate), summary receiver operating characteristic (SROC) curve was drawn and interpreted based on area under the curve (AUC) value as excellent (0.9-1.0), good (0.8-<0.9), fair (0.7-<0.8), poor (0.6-<0.7), and failed (0.5-<0.6) [18].Heterogeneity between studies was checked with forest plot, Cochrane's Q test, and I 2 test.Signifcant heterogeneity was declared at I 2 > 50% and Q-test p value < 0.10.Methodological quality of the included studies was assessed by the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool using Review Manager 5.4.1 software.
Figure 2 shows the risk of bias and applicability concern results of the 29 included papers as analyzed by QUADAS-2 tool in Review Manager 5.4.1.Te objective of doing this quality assessment is to assess the validity of estimates of POC-CCA test accuracy generated by this review and the applicability of included evidence to the review objective.Accordingly, except for the patient selection, nearly half or more of included studies got a score of low risk of bias.In majority of the studies, there was low applicability concern (Figure 2).

Single POC-CCA versus Single KK.
Twenty-four studies assessed the accuracy of a single POC-CCA test as compared to single KK reference test (duplicate 41.7 mg stool smear).Te pooled sensitivity was high (0.88, 95% CI: 0.87-0.90),but the specifcity was low (0.66, 95% CI: 0.65-0.67).Te SROC curve showed an AUC of 0.7805 with standard error of 0.0385.Tis reveals that single POC-CCA test fairly discriminates infected participants from those without Schistosoma infection (Figure 4).

POC-CCA versus KK with Diferent Number of Samples
Examined.Simultaneous increase in the number of urine and stool samples collected at diferent days results in a slight increase in the sensitivity (0.88 versus 0.90) but a signifcant decrease in the specifcity (0.66 versus 0.53) of POC-CCA test as compared to examination of single urine and stool samples.Te pooled DOR is 11.44 (95% CI: 6.00-21.84)and the AUC was 0.8525 (data not shown).A few other studies collected diferent number of urine and stool samples for evaluation of POC-CCA.Pooled performances of each combination are presented in Table 2.
In studies comparing single POC-CCA and KK, the pooled LR+ was 2.10 (95% CI: 1.79-2.46).Tis means that the probability of POC-CCA test being positive among S. mansoni patients is 2.1 times higher than the probability of getting positive POC-CCA result among noninfected     Journal of Tropical Medicine participants.Te LR− in the same pair of tests was 0.22 (95% CI: 0.10-0.50),i.e., in participants with negative POC-CCA result, the probability of being infected decreases by 22% as compared to those tested positive by POC-CCA.Te odds of getting a POC-CCA positive result among S. mansoni infected participants were 9.46 (95% CI: 6.03-14.85)(Table 3).

Discussion
Tis systematic review assessed the accuracy of POC-CCA test for the diagnosis of S. mansoni infection using stool KK and PCR as reference standards.A similar review was published previously [47]; however, the authors did not consider the more sensitive PCR as a reference test.Furthermore, the quality of included studies was not assessed in the previous review, making the strength of evidence uncertain.Moreover, more studies are published since the previous review was completed and we were interested to produce a comprehensive review by including recently published studies.
All the included studies were cross-sectional where urine and stool samples were simultaneously collected and processed by POC-CCA (urine), KK (stool), and PCR (urine or stool).Tis might minimize the strength of evidence because participants were screened without controlling factors responsible to afect test performances of both the index and reference tests.All the included studies have compared accuracy of POC-CCA with KK while only 3 studies compared the index test with PCR.As can be observed from

6
Journal of Tropical Medicine  Journal of Tropical Medicine the risk of bias and applicability concern graph (Figure 2), majority of the included studies are with good quality.Terefore, the evidence generated in this review is strong enough to infer.
Comparison of POC-CCA with KK regardless of the number of urine and stool samples examined revealed good accuracy (sensitivity of 0.86 and specifcity of 0.66).However, CCA excretion in urine shows day to day fuctuation [22] that pooled accuracy from studies examining diferent number of urine does not provide strong evidence.Similarly, the number of stool samples collected and the number of smears examined from each sample determine the performance of the reference KK test [11,12,22,25,31].Terefore, we have done a subgroup analysis based on the frequency of urine and stool samples collected and examined.Q 2 and p values presented in Figures 3, 4, and 5 imply that there is signifcant heterogeneity among studies.As a result, we used the random efect model during estimation of summary measures.POC-CCA and single KK.Tis is interpreted as use of POC-CCA decreases FN results by 22%.Due to its low detection threshold (20-50 eggs per gram of stool), KK is expected to miss signifcant number of patients who might be tested positive by CCA.Index tests with LR− > 0.1 are considered to be signifcant in decreasing the FN rate [49].Diagnostic odds ratio is the ratio of LR+ and LR− and indicates the impact of an index test in decreasing both the FP and FN rates.Te POC-CCA was found good in this respect as compared to KK (DOR ≥ 10) [49].

Limitations
We have excluded studies which assessed the accuracy of POC-CCA but reported sensitivity and specifcity without providing data on TP, TN, FP, and FN.We did not conduct any modeling to create a good reference method.Te other important limitation in the present review is that we did not stratify studies based on the endemicity level or intensity of infection, as we were unable to get complete data in most studies.

Conclusion
Te commercially available POC-CCA test has higher sensitivity than KK and may serve as routine diagnostic alternative for disease diagnosis, mapping, and monitoring of interventions.However, the results should be interpreted with caution as we did not consider variations in disease endemicity and intensity of infection.Te KK test itself is with poor sensitivity that evaluation of POC-CCA using more sensitive molecular reference standard tests is recommended.Terefore, we recommend well-designed multicenter accuracy studies involving diverse endemicity settings and geographic locations.

Figure 1 :
Figure 1: Flowchart showing selection process of eligible studies.

Figure 2 :
Figure 2: Risk of bias and applicability concerns graph.

Figure 3 :
Figure 3: Sensitivity (a) and specifcity (b) of POC-CCA compared to KK regardless of the number of samples examined (n � 37).

Table 1 :
Characteristics of included studies for analysis of POC-CCA accuracy versus KK (n � 37).

Table 2 :
Subgroup analysis of diferent combinations of urine and stool samples.