A systematic review of peptide-based serological tests for the diagnosis of leishmaniasis

Serological methods should meet the needs of leishmaniasis diagnosis due to their high sensitivity and specificity, economical and adaptable rapid diagnostic test format, and ease of use. Currently, the performances of serological diagnostic tests, despite improvements with recombinant proteins, vary greatly depending on the clinical form of leishmaniasis and the endemic area. Peptide-based serological tests are promising as they could compensate for antigenic variability and improve performance, independently of Leishmania species and subspecies circulating in the endemic areas. The objective of this systematic review was to inventory all studies published from 2002 to 2022 that evaluate synthetic peptides for serological diagnosis of human leishmaniases and also to highlight the performance (e.g., sensitivity and specificity) of each peptide reported in these studies. All clinical forms of leishmaniasis, visceral and tegumentary, and all Leishmania species responsible for these diseases were considered. Following PRISMA statement recommendations, 1,405 studies were identified but only 22 articles met the selection criteria and were included in this systematic review. These original research articles described 77 different peptides, of which several have promising performance for visceral or tegumentary leishmaniasis diagnosis. This review highlights the importance of and growing interest in synthetic peptides used for serological diagnosis of leishmaniases, and their performances compared to some widely used tests with recombinant proteins.


Introduction
Neglected tropical diseases (NTDs) affect more than one billion people worldwide, largely in rural areas of low-income countries [80].Among NTDs, leishmaniases are considered a major global public health problem with over one billion people at risk of infection living in 98 endemic countries and territories on five continents, and almost 1.3 million new cases reported during the last five years (2015-2020) [77,81].
Leishmaniases are a group of vector-borne diseases caused by parasites of the genus Leishmania which present different clinical forms: visceral (VL) and tegumentary leishmaniasis (TL) that includes cutaneous (CL), and mucosal or mucocutaneous (ML) leishmaniasis.The most common forms are CL, which causes skin sores, and VL, which affects several internal organs, usually the spleen, liver, and bone marrow and which is fatal if not treated.To sustainably fight leishmaniases, control strategies should include early biological diagnosis, and active case detection, which will improve clinical diagnosis, treatment and follow-up, and reduce transmission [21,78,79].
To date, there is no gold standard available for the diagnosis of active leishmaniasis.A combination of tests (composite reference standards) is required to achieve the accurate diagnosis of leishmaniases.Three major methods are routinely used: parasitological examination (direct demonstration of parasites in Giemsa-stained smears by microscopic observation or parasite culture), molecular tests (polymerase chain reaction (PCR) technique targeting Leishmania DNA), and serology (indirect immunofluorescence (IFA), direct agglutination test (DAT), enzyme linked immunosorbent assay (ELISA), immunochromatographic test (ICT) also known as lateral flow test) [3,52].At the point of care, serological tests are widely used because they do not require well-equipped laboratories and experienced staff compared to parasitological or molecular techniques.So far, serological diagnostic tests were developed with soluble antigens or whole-cell lysates.These tests had variable sensitivity depending on the antigen used and low specificity due to cross-reactivity with other pathogens present in endemic areas, such as Trypanosoma cruzi, the parasite responsible for Chagas disease [25,39,71].To improve the sensitivity and specificity of the immunodiagnostic tests, several studies have attempted to replace soluble antigens with recombinant proteins.Different antigens such as KMP11, LiP2, K39, K26, A2, KE16 have been used in ELISA and ICT [33] with variable sensitivity and specificity depending on both the kind of recombinant antigens and the areas under study.ICTs based on the rK39 protein (fragment of Leishmania (L.) infantum kinesin-like protein) became the most commonly used tools for VL diagnosis because they are well adapted for field testing and have high specificity.However, WHO reported significant variations in sensitivity among the different manufacturers of rK39-based ICT [76].Moreover, several studies reported variations in sensitivity depending on the geographical region concerned.Sensitivity was lower in East Africa (36% to 87%), Brazil (61% to 92%), and the Mediterranean basin (52% to 91%), than in the Indian subcontinent (92% to 100%) [5,6,20,32,34,50,64].For CL and ML diagnosis, serological tests are not commonly used because of their low sensitivity and variable specificity [81].To increase antibody detection, synthetic peptides containing B-cell epitopes could be used in immunoassays.Their use could improve the accuracy and robustness of diagnostic tests because they are devoid of uninformative or less informative epitopes responsible for background reactions [38].Their chemical synthesis does not involve handling living organisms and can be standardized for a high level of purity and reproducibility, allowing the production of robust ELISA or ICT [7,42].Peptide phage display technology, overlapping peptide libraries covering the entire selected protein sequence or in silico B-cell epitope prediction can be used to identify specific peptides [28,61,67,73].Advances in computational techniques and bioinformatics have also enabled the development of algorithms using several physicochemical, structural, and geometrical aspects of amino acid sequences.Most algorithms are free of charge, can be used online, and provide a fast and scalable way to predict B-cell epitopes in silico [65].
To evaluate peptide relevance in the design of new serological diagnostic tests for leishmaniases, we reviewed all studies focusing on peptides for serodiagnosis of visceral and tegumentary leishmaniases published from 2002 to 2022.All clinical forms of human leishmaniasis and all Leishmania species responsible for these diseases were considered in this systematic review.We reported the diagnostic performance of peptidebased tests (index tests) against a reference standard.

Objective
To determine the diagnostic accuracy of peptide-based tests for the diagnosis of active human leishmaniasis.

Eligibility criteria
Original research articles meeting all the following inclusion criteria were eligible: -Population: any patient with clinical symptoms of leishmaniasis; -Intervention: diagnosis with index tests defined as any immunological test based on peptides derived from Leishmania parasite antigens allowing the detection of anti-Leishmania antibodies in human serological samples (serum or plasma).Peptide length must be less than 40 amino acids; -Comparison: diagnosis with one or several reference standards (parasitological methods, commercial serological tests and/or molecular tests); -Outcomes: Accurate human leishmaniasis diagnosis.Performance data were described by assessing sensitivity and specificity of the index test.Sensitivity is the probability that the index test result is positive given that leishmaniasis is present, reflecting the ability of the index test to correctly identify individuals with the disease through a positive response.Specificity is the probability that the index test result is negative, given that leishmaniasis is absent, reflecting the ability of the index test to correctly identify individuals without the disease through a negative response [82].

Information sources
A literature search of four electronic databases was performed in December 2022: PubMed, Web of Science, Worldwide Science and SciELO.

Search strategy
The search strategy used the following string: "((leishmaniasis*) AND (diagnosis*)) AND (peptide OR epitope)" and filters "(humans))".Language search was limited to English, Portuguese and Spanish and chronological research from January 2002 to December 2022.

Selection process
Study eligibility was assessed by two independent investigators (JP and OP) by review of article title and abstract and, when relevant, by text reading.Two other independent investigators (RBG and EP) analyzed pre-selected full texts and confirmed inclusion of studies.

Data collection process and data items
Data extraction from the full texts included was performed by two independent investigators (JP and OP).The following data were imported to a Microsoft Excel Ò worksheet: reference, peptide sequence, peptide name given by authors, antigen name, ID UniProt or NCBI accession number of antigen, reference standard, index test, clinical research phase, clinical form evaluated, case origin, and causative Leishmania species.The clinical research phases of diagnostic tests described in studies were classified according to Zhou et al. [82].Briefly, Phase I (exploratory) studies are those whose purpose is proof-ofprinciple of a new diagnostic test in a small number of patients (10-50 archived specimens, retrospective design) with a confirmed disease status versus healthy volunteers.Phase 2 (challenge) studies are those evaluating the diagnostic test with a case-control design in 10-100 individuals in each series.The sampling plan includes patients with a spectrum of targeted disease characteristics versus healthy people or patients with other diseases mimicking the targeted disease.Lastly, Phase 3 (clinical) studies are large-scale prospective studies validating the test in representative sample (hundreds of patients) from the target population (suspected cases).

Risk of bias assessment and effect measures
The Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool was used to analyze the quality of the included studies and their susceptibility to bias [74].QUADAS-2 consists of four domains: patient selection, index test, reference standard, and flow and timing.All four domains are judged in terms of the risk of bias and the first three in terms of applicability [74].
A descriptive analysis of quality assessment was performed on the data collected from eligible articles by two independent investigators (JP and OP).Data extraction was checked and any discrepancies were resolved by two others investigators (RBG and EP).

Synthesis methods
Sensitivity and specificity data were presented according to the clinical form of patient groups used.The reported sensitivity and specificity of peptide-based tests, and their confidence intervals, were plotted using the forest plot function generated by Review Manager version 5.3 (RevMan 5.3) [60].The predictive values were not specified in all articles, so these were estimated based on the percentage of sensitivity and specificity and the number of samples described by the authors in each group.
In some articles, several performances were calculated using different non-case groups (healthy individuals from endemic and non-endemic regions and patients infected by diseases other than leishmaniasis).When possible, the performance of peptide-based tests was reevaluated, including all non-cases in a single control group.

Included studies
A total of 1,405 published studies were identified using our search strategy from four databases (Fig. 1).After removing duplicates, 1,103 articles remained.Among them, 1,065 were excluded for several reasons: off-topic (n = 290), studies on animals (n = 182, including 168 on dogs), other diagnostic techniques (n = 135), medicine/report cases/treatment/pathology (n = 128), reviews (n = 124), recombinant protein or crude Leishmania antigens used for antibodies detection (n = 81), studies on other diseases (n = 51), epidemiology (n = 48), and Leishmania vaccine studies (n = 26).Finally, 38 articles were assessed for eligibility.After a thorough reading of the full text, 16 additional articles were excluded.The reasons for exclusion were the following: peptides were not derived from Leishmania antigens (n = 5), no performance of the index test was described (n = 3), the samples used were not serological samples (n = 2), no immunoassay was performed to validate the diagnostic potential of the peptides (n = 3), no reference standard was used (n = 2), the peptides were not used for antibody detection (n = 1).At the end of this screening, 22 studies were included in our systematic review (Supplementary data 1) and the extracted data were compiled in a data collection table (Supplementary data 2).

Study characteristics
The characteristics of the 22 studies included in this review have been compiled in Table 1.Six studies were classified in clinical research phase I and 16 studies in phase II according to Zhou et al. [82], in which the cases were selected from a health service or hospital.To assess the performance of peptide-based tests, three different formats were reported: one study used an ICT format [7], two studies used phage-ELISA [15,62] and 19 studies used an ELISA technique [10, 11, 16, 17, 26, 27, 29, 35, 37, 40-43, 47, 48, 58, 63, 69, 70].In these studies using an ELISA method, a marked difference was observed in the amount of peptides used, which ranged from 0.25 lg to 20 lg per well.

Performance of peptide-based serological tests
Performance of index tests based on different peptides was estimated by assessing sensitivity and specificity.The sensitivity and specificity of peptide-based tests for human VL diagnosis are presented as forest plots in Figure 2, for human TL diagnosis (CL and ML) in Figure 3, for only human CL in Figure 4 and for both VL and TL in Figure 5, based on data extracted from the 22 included articles (Supplementary data 2).
For VL diagnosis, the performance of 49 different synthetic peptides was evaluated from serum of VL patients.Their sensitivity ranges from 51% to 100% and their specificity from 60% to 100% (Fig. 2).Among them, the performance of 17 synthetic peptides was evaluated both in VL patients and VL patients co-infected with HIV.Their sensitivity ranges from 51% to 100% and their specificity from 60% to 100% [26,58].High performance with 100% sensitivity and specificity was reported for three peptides [58].Two studies also evaluated the performance of peptides in combination [11,17].The sensitivity of peptide combination ranged from 71% to 100% and specificity from 88% to 100%.Thus, for VL diagnostic, high performance with sensitivity 95% and specificity 98% was reported for 16 peptides [11,37,47,48,58,62,69] and four peptide mixtures [11,17].
For CL diagnosis only, the performance of ten different synthetic peptides was evaluated (Fig. 4).Their sensitivity ranged from 0% to 66%, and specificity from 82 to 100%.One study evaluated the performance of three synthetic peptides in combination [35].This peptide mixture obtained 77% sensitivity and 85% specificity.
The performance of seven synthetic peptides was evaluated using both VL and TL patients in same group (Fig. 5).Their sensitivity ranged from 28% to 57%, and specificity from 16% to 84% [70].

Quality assessment of study reports
Although all publications included in this review were in clinical research phase I or II of diagnostic test development, we used the QUADAS-2 tool (more suitable for phase III) to assess the quality of the 22 articles in terms of risk of bias and applicability concerns.The results are summarized in Figure 6 and study details are provided in Supplementary data 3.
All the included studies followed a case-control design with archived or collected samples, selected for convenience and according to the outcome of the reference standard.Thus, they had a high risk of bias and high concern regarding applicability in the domain of patient selection.
No study specified whether the index test was conducted in a double-blind fashion or with random samples, so all studies were classified "unclear" for risk of bias concerning the index test.Three out of 22 studies were at high risk of applicability concerns in the index test domain, because two used prespecified cut-offs based on the control group value (mean with three or two standard deviations to determinate the threshold) instead of using the ROC curve [10,17], and the third study did not describe how the cut-off value was set [29].All included studies were evaluated at low risk of bias and applicability concerns regarding the reference standard domain.The reference standard or the composite reference standards used were made before the index test, so the results of reference tests were still interpreted without knowledge of the index tests results.
All included studies in this review were at high risk of bias regarding the domain of flow and timing.In most studies, there were no details on the period and conditions of sample storage or time interval between the performance of index test and the reference standard.However, except for two studies [7,17], an ELISA test was done to check the presence of antibodies directed against soluble Leishmania antigen (SLA) extracted from local Leishmania strain.Finally, in studies where the reference standards included a parasitological test, that is considered invasive (lesion biopsies, bone marrow aspirates), the control groups did not receive the same reference standard as patients.

Summary of main results
This literature search allowed a state-of-the-art review of the use of synthetic peptides in serodiagnostics for human leishmaniases from 2002 to 2022.Among included articles, most of them (19/22; 86%) were published in the last ten years, which shows the growing interest in the use of synthetic peptides in leishmaniasis diagnostic tests in the recent years.The development of several algorithms to predict in silico linear B-cell epitopes like Bepipred, BCpred, ABCPred and LBtope [24,59,65], and the increase in immune epitope data deposited in public databases such as the Immune Epitope Database (IEDB, http://ieddb.org/)[72], promote the surge in the use of synthetic peptides in health research [1].A combination of bioinformatics tools can be used in the absence of immunoproteomic data from previous research.For example, Carvalho et al. used bioinformatics tools in all processes to predict antigenic properties of proteins and to identify B-cell epitopes from this protein.These authors used available data (TritrypDB or SignalP) to identify secreted excreted proteins that are most exposed to the immune system [11].Furthermore, phage display is also a powerful strategy to rapidly identify diseasespecific B-cell epitopes, as has been done for other pathogens, such as those causing severe acute respiratory syndrome (SARS) [36] or dengue fever [68].
The present review reported promising performance with ELISA or phage-ELISA techniques with sensitivity 95% and specificity 98% for VL diagnosis, and with sensitivity 95% and specificity 95% for TL diagnosis for several peptide-based tests [11,15,17,37,43,47,48,58,62,69].Only one peptide was assessed in ICT cassette and dipstick formats for which 100% specificity and, 84% and 79% sensitivity were reported, respectively in Sudanese patients [7].While serological tests are known to have low sensitivity for the diagnosis of VL in HIV-infected compared non-HIV-infected patients [18,19], two studies reported promising performance (sensitivity from 51% to 100% and specificity from 60% to 100%) of peptide-based test evaluated on VL/HIV co-infected patients [26,58] of which three peptide-based tests with 100% sensitivity and specificity [58].
Most studies included in this review estimated the performance of peptide-based tests with sera of patients from single geographical areas (Sudan, Brazil or Peru) and infected by a same Leishmania species (molecularly identified or assumed), either by L. infantum for VL diagnosis or L. braziliensis for TL diagnosis.Only six studies, all performed in Brazil, evaluated the performance of the test using both sera of L. infantum VL patients and L. braziliensis TL patients [26,27,[41][42][43]70].

Strengths and weaknesses of the review
The review process followed PRISMA guidelines and all included studies were evaluated in accordance with the QUADAS-2 tool.However, the performance data for peptidebased tests should be interpreted with caution because all the studies included in this review are in clinical research phase I or II of diagnostic test development.Thus, the risk of bias for most of these studies was assessed as "unclear" and/or "high" for three of the four domains included in the QUADAS-2 tool, i.e., patient selection, index test, and flow and timing.The main risk of bias in the reviewed studies was related to patient selection.In the first step of diagnostic test development, the inclusion of patients is based on the availability of stored serum samples and on diagnostic information provided by health services or depended on consecutively recruited cases with confirmation using the reference standard.Suspected cases were excluded, leading to overestimation of the diagnostic accuracy of the index test.
The heterogeneity of reference standard and sampling used in the included studies makes it difficult to compare performance results.Accuracy of diagnosis was not the same based on the reference standards used.The parasitological method was used in 59% (13/22) of the studies, but among them, 23% (3/13) used only parasitology as the reference standard [10,29,35].Most other authors have incorporated another technique to confirm the diagnosis such as PCR to detect the kinetoplastid DNA (kDNA) of Leishmania parasites (9/13) [15, 17, 27, 40-43, 62, 63] and/or a commercial serological test using the rK39 protein (3/13) [7,62,63].Furthermore, most studies integrated in the control group, in addition to healthy individuals, patients with other diseases such as Chagas disease (18/22) [7,17,37].Also, further heterogeneity was found between studies, and even within some studies, regarding the length and concentration of used peptides, making it difficult to compare the performance results.In studies using the ELISA technique (19/22), peptide concentration was expressed in lg/well or lg/mL, without considering the molecular weight of the peptide, which varies according to the amino acid composition of each peptide.For example, Carvalho et al., used 10 lg/well of a 12-amino-acid peptide (peptide 1) that has a theoretical molecular weight of 1274.27g/mole, whereas for another peptide (peptide 2), they also used 10 lg/well for this 28-amino-acid peptide that has a theoretical molecular weight of 2514.60 g/mole [11].
Therefore, some of the peptides may not have been studied at a saturating molar concentration, which can impact density and immunoreactivity of the immobilized peptides in ELISA microplate wells, and therefore have an influence on test performance [22,44].
Standardization of the evaluation methodology (in particular the use of a reference standard and the appropriate composition of control group), as well as better knowledge of the QUADAS-2 tool and the STARD statement (Standards for Reporting of Diagnostic Accuracy Studies) [14] by investigators, would be beneficial for the qualities of assessment and the quality report of diagnostic accuracy studies.

Applicability of findings and implication for practice and research
Given that there are more than 70 endemic countries described with multiple circulating species responsible for different clinical forms of leishmaniasis [81], different treatments depending on the clinical form and the parasite species involved [8], and other diseases that have similar clinical symptoms in same leishmaniasis endemic areas, [31,57,66], accurate diagnosis of leishmaniases is an important need.The development of diagnostic tests based on peptides should meet this need.Moreover, synthetic peptides have several other advantages, such as their low cost, simple chemical production, reproducibility, ease of storage, stability and safety [13,46,55].The use of bioinformatics tools helps to easily predict immunodominant epitopes from protein sequences and reduce potentially costly and time-consuming laboratory work.The peptides can be identified from one or several antigens and used in combination to increase the number of reactive epitopes and improve diagnosis performance [9,23,45,56].
This systematic review provides evidence for recommending the use of synthetic peptides for biological serodiagnosis of both TL and VL.Despite promising performances with 100% sensitivity and specificity for several peptide-based tests for VL or TL diagnosis, evaluation on serum samples from patient groups infected by different Leishmania species present in a same endemic area is important for accurate diagnosis, such as in Brazil where eight different pathogenic species are present [2].Importantly, sensitivity can be variable and depend on geographical area, as has already been observed with the rK39-ICT.In VL due to L. donovani, rK39-ICT sensitivity was lower in East Africa (85.3%; 95% CI 74.5 to 93.2) than in the Indian subcontinent (97.0%; 95% CI 90.0 to 99.5) [6].
The performance results can be influenced by the accuracy of the reference standards chosen, the composition of control group used, the disease-causing Leishmania species of patients' groups, leishmaniasis clinical form of patient groups, and geographical area of study.The parasitological method, although routinely used in health services for the diagnosis of leishmaniasis, shows low sensitivity [30] and is therefore likely to lead to false-negative results.The results of the parasitological examination may depend on several factors, such as parasitemia or parasite load, the sample, and the skills of the operator for sample collection and microscopic examination.The lack of an accurate gold standard leads to diagnostic errors and can be resolved by using multiple diagnostic methods, such as visualization of amastigote by microscopy, parasite isolation by culture, molecular detection of parasite DNA, and for VL, serological rK39 test [3,12].In the evaluation of a new diagnostic test, the use of such a composite reference standard would allow better classification of samples and thus decrease the proportion of false negatives, especially in individuals with low parasitemia.
The specificity assessment of serological diagnostic test for leishmaniasis depends on the differential diagnosis because other diseases with similar clinical manifestation to leishmaniasis are common in endemic areas, but also diseases known to cross-react with anti-Leishmania antibodies.The differential diagnosis depends on clinical form and endemic area.For example, it is important to include patients with leprosy and lupus vulgaris for diagnosis of CL, paracoccidioidomycosis and tuberculosis for diagnosis of ML [31,53], or malaria, typhoid fever, arbovirus diseases (chikungunya, Zika, and dengue fever), acute Chagas disease, acute schistosomiasis, amoebic liver abscess, mononucleosis and hepatitis for diagnosis of acute forms of VL [53].For example, in the Americas region, many patients with leishmaniasis are co-infected with other tropical diseases [54].In Brazil, O'Neal et al. showed 88% co-infection with helminths in the state of Bahia [49], while Azeredo-Countinho et al. diagnosed 15% of patients in the state of Rio de Janeiro [4].In Argentina, several studies have shown that co-infection with Trypanosoma cruzi can be common and the antigenic cross-reactivity between these two parasites makes it difficult to discriminate Leishmaniasis and Chagas disease [25].As there are a large number of diseases with clinical symptoms similar to those of Leishmania infection and cross-reactivities with other parasites, it is pertinent to select the diseases according to the relevant area of co-endemicity to enable proper differential diagnosis.
Therefore, further investigations using large cohorts of cases and non-cases from endemic areas are still required to determine the promising performance of these peptides (alone or in combination with other peptides), and to estimate the accuracy of these peptide-based diagnostic tests in clinical practice.Therefore, peptide-based tests could be very helpful for the development of more efficient point-of-care diagnostic tests, and this regardless of the clinical form of leishmaniasis, the circulating Leishmania species and geographical area.Moreover, the synthetic peptides composed of species-specific and conserved epitopes could overcome the problems of specificity and sensitivity found with some antigens and according to geographical regions.

Protocol and registration
The review protocol was conducted following the guidance of the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement, PRISMA 2020 statement [51].The PRISMA 2020 Checklist of the review is provided as supplementary material (Supplementary data 4).The review was not registered and was different from any other review registered or published.

Figure 1 .
Figure 1.PRISMA Flowchart of the selection steps undertaken in the systematic review.

Figure 2 .
Figure 2. Forest plot representing percentage of sensitivity and specificity of different peptide-based tests for visceral leishmaniasis (VL) diagnosis.* studies where the counts of true positive (TP), false positive (FP), false negative (FN), and true negative (TN) results were estimated from the sensitivity and specificity values.# studies where sensitivity and specificity were recalculated considering all non-cases in a single control group.

Figure 3 .
Figure 3. Forest plot representing percentage of sensitivity and specificity of different peptide-based tests for tegumentary leishmaniasis (TL=CL+ML) diagnosis.* studies where the counts of true positive (TP), false positive (FP), false negative (FN), and true negative (TN) results were estimated from the sensitivity and specificity values.# studies where sensitivity and specificity were recalculated considering all non-cases in a single control group.

Figure 4 .
Figure 4. Forest plot representing percentage of sensitivity and specificity of different peptide-based tests for cutaneous leishmaniasis (CL only) diagnosis.* studies where the counts of true positive (TP), false positive (FP), false negative (FN), and true negative (TN) results were estimated from the sensitivity and specificity values.# studies where sensitivity and specificity were recalculated considering all non-cases in a single control group.

Figure 5 .
Figure 5. Forest plot representing percentage of sensitivity and specificity of different peptide-based tests for both visceral leishmaniasis (VL) and tegumentary leishmaniasis (TL) diagnosis.* studies where the counts of true positive (TP), false positive (FP), false negative (FN), and true negative (TN) results were estimated from the sensitivity and specificity values.# studies where sensitivity and specificity were recalculated considering all non-cases in a single control group.

Figure 6 .
Figure 6.Risk of bias and applicability concerns for the 22 articles included in the review.QUADAS-2 results summarize quality assessment for patient selection, index test, reference standard and flow and timing.This figure was generated using the QUADAS website.(https://www.bristol.ac.uk/population-health-sciences/projects/quadas/quadas-2/).

Table 1 .
Characteristics of included studies.

Table 2 .
List of peptides described in included studies.