Can chemokines be used as biomarkers for endometriosis? A systematic review.

Hum. Reprod. Advance Access published November 27, 2013 Human Reproduction, Vol.0, No.0 pp. 1 –14, 2013 doi:10.1093/humrep/det401

REVIEW Gynaecology

Can chemokines be used as biomarkers for endometriosis? A systematic review G. M. Borrelli 1,2, M. S. Abraõ2, and S. Mechsner 1 1 Department of Gynecology, Charite´, Universita¨tsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germany 2School of Medicine, Department of Obstetrics and Gynecology, Sao Paulo University, Saõ Paulo, Brazil

*Correspondence address. E-mail: [email protected]

study question: Can we use chemokines as biomarkers to diagnose patients with endometriosis in clinical practice? summary answer: Some chemokines, especially CXCL8 (IL-8), CCL-2 (MCP-1) and CCL5 (RANTES), have the potential to work as biomarkers to identify patients with endometriosis but their accuracy could be improved by combination with other non-inflammatory markers in a panel of biomarkers.

what is already known: The need for a good marker to diagnose endometriosis has increased in recent years and research in this field has intensified. Chemokines have been reported to be associated with endometriosis in several studies over the last 20 years. Many of these studies measured one or more chemokines in peritoneal fluid (PF) and peripheral blood (PB) or through endometrial biopsies in patients with and without endometriosis.

study design, size, duration: A systematic review was done on all published studies that compared chemokine concentrations in patients with and without endometriosis to evaluate their potential as biomarkers for the disease.

participants/materials, setting, methods: Using MEDLINE database from December 1993 to August 2013 and the MeSH terms ‘Endometriosis’ and ‘Chemokines’, we identified relevant studies to include in the present review, which was based on the PRISMA statement. Studies that measured at least one chemokine in patients with endometriosis and matching controls in PB, PF or endometrial samples were included. We did not include samples from ectopic lesions. All review articles as well as studies with animals and those not written in English were excluded from this systematic review. The studies were assessed using a modified version of the Quality Assessment of Diagnostic Accuracy Studies criteria. Two authors independently assessed studies for inclusion and risk of bias, and extracted data.

main results and the role of chance: After inclusion and exclusion criteria, 62 studies were selected to be included in this systematic review. A total of 27 different chemokines or their receptors were evaluated in the reviewed studies. The most studied chemokines (including their receptors) were CXCL8 (51.6%), CCL2 (38.7%) and CCL5 (19.3%) (% of studies). CXCL8 (IL-8) appears to have the best results among all the other chemokines as a marker for endometriosis. limitations, reasons for caution: Some studies included have low power due to small sample size and study designs vary in the assessment criteria for the markers, the state of the patients (e.g. phase of the cycle and stage of disease) and the nature of the controls. wider implications of the findings: Our findings could guide future research in this field to select the chemokines with the best potential, and to stimulate better-designed studies to determine whether they can become a useful diagnostic tool in clinical practice. study funding/competing interest(s): There was no funding to support this systematic review. The authors have no competing interest to declare. Key words: chemokines / endometriosis / biomarkers

Introduction The clinical diagnosis of endometriosis has remained a challenge for physicians and scientists since the disease was first defined by John Sampson (1927). Despite the advances observed in this field during the past

10 years, especially in imaging technology (Abraõ et al., 2007; Bazot et al., 2009; Gonçalves et al., 2009; Saba et al., 2011), which have mainly improved the diagnosis of deep endometriosis lesions, we still lack a good biochemical marker capable of diagnosing all stages of the disease, especially the early stages. Women with subfertility and/or

& The Author 2013. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: [email protected]

Downloaded from http://humrep.oxfordjournals.org/ at University of Connecticut on December 3, 2013

Submitted on June 7, 2013; resubmitted on October 6, 2013; accepted on October 14, 2013

2

Chemokines Chemokines represent a family of small cytokines or proteins released by cells, especially lymphocytes, and are capable of inducing chemotaxis (directed movement through the chemicals of the microenvironment) in nearby responsive cells, directing the cellular migration. Some chemokines control cells of the immune system directing lymphocytes to lymph nodes, and they are called homeostatic chemokines. Inflammatory chemokines are released from a great variety of cells in response to bacterial infections, virus and other pathogenic agents. Therefore, they act under both physiological conditions and pathological processes (Acker et al., 1996; Lira and Furtado, 2012). Chemokines interact with specific receptors as we have explained in detail in our previous publication (Borrelli et al., 2013). Several studies have already assessed single or combined chemokines in patients with and without endometriosis, as possible markers for the disease (Laudanski et al., 2006; Agic et al., 2008; Othman et al., 2008; Socolov et al., 2011; Drosdzol-Cop et al., 2012; Margari et al., 2013).

Objectives The aim of this systematic review is to assess and review all studies that measured chemokines in patients with endometriosis and matching

controls, to try to answer whether they could or could not be used as a marker for the disease.

Methods The goal of this review was to verify the possibility of using chemokines as biomarkers for endometriosis, as a noninvasive test using PB samples, or even as a semi-invasive test using endometrial samples (eutopic endometrium), as well as a more invasive test using PF, which is currently obtained during laparoscopy—a surgical procedure. The review was based on the PRISMA statement (Moher et al., 2009). The quality of the individual studies was judged using a modified version of the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) criteria (Whiting et al., 2003; May et al., 2010) (Table I).

Search strategy We used the MEDLINE database to seek relevant studies and searched the literature published from December 1993 to August 2013. The words ‘endometriosis’ and ‘chemokines’ or ‘endometriosis’ and ‘biomarkers panel’ as well as ‘endometriosis’ and ‘chemokines’ and ‘biological markers’ were used as key words to recover all possible publications on this theme at PubMed database. The strategies for electronic search at the MeSH database were the following combined MeSH terms: ((‘endometriosis’[MeSH Terms] OR ‘endometriosis’[All Fields]) AND (‘chemokines’[MeSH Terms] OR ‘chemokines’ [All Fields])); ((‘endometriosis’[MeSH Terms] OR ‘endometriosis’

Table I Modified QUADAS criteria used for assessing studies. Criteria

Yes

No

Unclear

........................................................................................ 1. Were patients and controls recruited from women with symptoms of endometriosis? 2. Were selection criteria clearly described? Did the study describe time frame, consecutive recruitment and inclusion/ exclusion criteria? 3. Was the time period between the diagnosis and biomarker test short enough to avoid a change in disease status? 4. Were controls surgically verified (not to have endometriosis)? 5. Were the methods for testing sufficiently explained? 6. Were the biomarker test results interpreted in a blinded fashion? 7. Was the diagnosis of endometriosis made without knowledge of the biomarker test results? 8. Were uninterpretable/intermediate test results reported? 9. Were withdrawals from the study explained? 10. Were samples collected at a consistent phase of the cycle, or results corrected for cycle phase? 11. Were samples collected from women with a particular stage(s) of disease, or results corrected for stage? QUADAS, quality assessment of diagnostic accuracy studies.


pelvic pain, and a normal clinical examination as well as normal imaging results, would be the best beneficiaries from a noninvasive or semiinvasive test (Fassbender et al., 2013). A good biomarker may serve to identify risk patients for disease prevention, as a potential drug target or marker for a drug response, and finally, as a control after treatment, being capable of detecting recurrence and progression of the disease (Palmer and Barnhart, 2013). Currently, the most used marker with this goal is the serum level of the antigen CA 125, which was first related to endometriosis reported by Niloff et al., (1984). Since then, several studies have succeeded in showing variable sensibility and specificity rates, which are not good enough for precise diagnosis of all stages and clinical presentations of the disease (Abraõ et al., 1997; Somigliana et al., 2004; Xavier et al., 2005; Maiorana et al., 2007; Socolov et al., 2011). Although different markers have been tested and shown to be altered in patients with endometriosis compared with controls, as a noninvasive test (peripheral blood (PB) samples) and/or semi-invasive/invasive tests (eutopic endometrium samples/PF aspiration), to date none has been validated for clinical use as a diagnostic test in patients with endometriosis (May et al., 2010; Fassbender et al., 2013). Biomarkers were assessed alone (Khorram et al., 1993; Bourlev et al., 2006; Cho et al., 2007) or in combination with a panel of markers (Agic et al., 2008; Seeber et al., 2008; Mihalyi et al., 2010; Vodolazkaia et al., 2012). Khorram et al. (1993) were the first to demonstrate the relation between high levels of chemokines and endometriosis. During the past 20 years, many researchers have evaluated different types of chemokines in patients with endometriosis and matching controls in order to establish a possible role of chemokines in the pathogenesis of the disease (Borrelli et al., 2013). Although the exact role of chemokines in the pathogenesis of endometriosis remains unsolved, it is possible that they are related to the aggressiveness, invasiveness and progression of the disease (Li et al., 2012a,b).

Borrelli et al.

3

Chemokines as biomarkers for endometriosis

[All Fields]) AND (‘biological markers’[Mesh Terms] OR ‘biological’ [All Fields]) OR ‘markers’[All Fields] OR ‘biological markers’[All Fields] OR ‘biomarkers’[All Fields] AND ‘panel’[All Fields])); ((‘endometriosis’ [MeSH Terms] OR ‘endometriosis’[All Fields]) AND (‘chemokines’ [MeSH Terms] OR ‘chemokines’[All Fields]) AND (‘biological markers’ [MeSH Terms] OR ‘biological’[All Fields] AND ‘markers’[All Fields]) OR ‘biological markers’[All Fields])). Two authors independently assessed studies for inclusion and risk of bias, and extracted data.

Results

procedures. From 329 initial papers originated through the primary computerized search (238 + 22 + 69), 148 studies were excluded by filters (Reviews, animal studies, not English Language) and duplicates. The abstracts of the 181 remaining references were read and using the inclusion/exclusion criteria described above, 117 were excluded and 64 were selected to be read in full, as well as two additional papers selected from the reference lists. Of the remaining 66 articles, three were excluded because they did not meet the inclusion criteria and one because it contained duplicate data thus 62 papers were included in this systematic review. They all measured chemokines in patients with endometriosis and compared with matching controls (Fig. 1).

Eligibility criteria

Figure 1 Flow diagram showing selection of articles for systematic review.

Data items We extracted the following data from the selected studies—names of the chemokines measured, the type of samples taken, the phase of the menstrual cycle when the chemokines were measured, number of patients, stage of endometriosis (ASRM) of the patients in the study group, which patients composed the control group and, finally, their results in order to establish or not significance to the findings. We also draw, where available, the processing methods and storage conditions from used samples.


We included all studies that measured at least one chemokine in patients with endometriosis and matching controls in PB, PF and eutopic endometrial samples (EES), and we excluded the studies that measured expression of chemokines in ectopic tissue, as this requires the removal of a suspect lesion, which is already the most used diagnostic tool for endometriosis worldwide through histologic examination. Although it might be possible to obtain PF in the future by means of a less invasive (semi-invasive) test such as the transvaginal ultrasound-guided aspiration, depending on the identification and validation of a reliable test in the PF, such samples are currently mostly obtained during laparoscopic

4

Borrelli et al.

Table II List of all assessed chemokines divided into their families and the number (n) of the 62 studies that included them. Chemokine families

Chemokine biomarkers

n

% of studies

........................................................................................ CXC chemokines (a-chemokines)

CXCL-1 (GRO-a) CXCL-5 (ENA-78) CXCL-6 (GCP-2) CXCL-8 (IL-8) CXCR1 (CXCL-8/ CXCL-1 receptor) CXCR2 (CXCL-8 receptor) CXCL-9 (MIG) CXCL-10 (IP-10) CXCL-11 (IP-9) CXCL-12 (SDF-1)

2 2 1 28 2

3.2 3.2 1.6 45.1 3.2

2

3.2

1 3 1 1

1.6 4.8 1.6 1.6

CC chemokines (b-chemokines)

CCL-2 (MCP-1) CCR-2 (CCL2 receptor) CCL-3 (MIP-1 a) CCL-4 (MIP-1 b) CCL-5 (RANTES) CCR-1 (CCL5 receptor) CCL-7 (MCP-3) CCL11 (Eotaxin) CCL-15 (HCC-2) CCL-16 (Monotactin-1) CCL-17 (TARC) CCL-19 (MIP-3 b) CCL-20 (MIP-3 a) CCL-21 (6 Ckine) CCL-22 (MDC)

23 1 2 1 10 2 1 1 1 1 2 1 1 2 1

37 1.6 3.2 1.6 16.1 3.2 1.6 1.6 1.6 1.6 3.2 1.6 1.6 3.2 1.6

CX3C chemokines (d-chemokines)

CX3CL1 (Fractalkine)

2

3.2

XC chemokines (g-chemokines)

XCL1 (SCM-1a)

1

1.6

ENA-78, epithelial-derived neutrophil-activating peptide-78; GCP-2, granulocyte chemotactic protein-2; GRO-a, growth regulated oncogene a; HCC, hemofiltrate CC cytokines; IL-8, interleukin-8; IP-10, interferon gamma induced protein-10; MIG, monokine induced by interferon gamma; MIP, macrophage inflammatory protein; RANTES, relegated upon activation, normal T cell expressed and secreted; SCM, single C motif-1; SDF, stromal cell derivded factor; TARC, thymus and activation regulated cytokine.

and validation analyses, as well as Standard Operating Procedures for banking of biological samples from women with endometriosis and controls, with respect to collection, transport, processing and long-term storage. As this normative will be ready and published soon, in Supplementary Table SI we present the processing methods and storage conditions from samples—liquids and tissue—of the papers included in this systematic review, where they were available, to later compare them with the EPHect consensus. As we can see, the methods used to process and store the samples vary considerably among the studies reviewed. Table III shows the modified QUADAS criteria analysis, the chemokines that were assessed and the number of subjects and controls included in each study. None of the selected studies fulfilled all methodological criteria. The most common failures were the uncertainty about the interpretation of the biomarker result, if it was or not blindly interpreted, if the diagnosis of endometriosis was made without knowledge of the test results, the poor selection criteria to define patients and control subjects, in addition to lack of adjustment for menstrual cycle and stage of disease. Finally, Table IV summarizes the main data extracted from each study reviewed, including the chemokine(s) measured, the type of the sample, phase of the cycle of the collection (when mentioned), characteristics of the subjects and controls recruited, and, most importantly, their results showing or not statistical significance to the findings.

CXCL8 (IL-8) More than half of the studies reviewed had evaluated the a-chemokine CXCL8 (IL-8/Interleukin-8) as a marker for endometriosis, measuring it in the PF (50%), PB (40.6%) and tissue—EES (12.5%). When measured in the PF, CXCL8 was found to be significantly higher in the endometriosis patients compared with controls in almost all studies reviewed (Ryan et al., 1995; Arici et al., 1996; Rana et al., 1996; Gazvani et al., 1998; Iwabe et al., 1998; Barcz et al., 2002; Pizzo et al., 2002; Calhaz-Jorge et al., 2003; Skrzypczak et al., 2005; Bersinger et al., 2006; Kalu et al., 2007; Kuroda et al., 2010; Milewski et al., 2011; Barcz et al., 2012; Malhotra et al., 2012). Only Kim et al. (2009) found no statistically significant difference (P . 0.05) between patients with the disease and controls. However, their study included a small number of subjects and the controls had either dermoid cysts or cervical cancer, which could have influenced the results. Interestingly, Gazvani et al. (1998) observed better results in patients with endometriosis Stage I (ASRM) compared with controls, than all endometriosis patients together (Stages I –IV) versus controls. On the other hand, the studies that evaluated the serum levels of CXCL8 presented conflicting results. Carmona et al. (2012) and Ohata et al. (2008) obtained similar results, both reporting significantly higher serum levels of CXCL8 in patients with ovarian endometriomas versus controls (Ohata et al., (2008) used ovarian benign cysts as controls), but did not achieve the same results for Deep Infiltrating Endometriosis (DIE) patients. While another four studies (Braun et al., 1996; Pizzo et al., 2002; Mihalyi et al., 2010; Vodolazkaia et al., 2012) found higher and significant PB levels of CXCL8 chemokine in patients with endometriosis (Stages I –IV) compared with controls, seven others found no statistically significant difference (P . 0.05) of this potential marker among endometriosis subjects and controls (Gazvani et al., 1998; Fasciani et al., 2001; Barcz et al., 2002; Daraı¨ et al., 2003; Kalu et al., 2007; Othman et al., 2008; Socolov et al., 2011). Overall, from the 13 studies that


A total of 27 different chemokines were assessed from the 62 studies reviewed. The most studied chemokines, including their receptors, were CXCL8—51.6% (32/62), CCL2—38.7% (24/62) and CCL5—19.3% (12/62), respectively, as shown in Table II. The samples most often used to assess chemokines in the included papers were PF in 56.4% (35/62) of the studies, PB in 35.5% (22/62) and tissue—EES in 20.9% (13/62) of the studies. The way data and samples are differently collected and stored in many institutions, as well as the clinical data from studied subjects and the ideal controls are still a problem in endometriosis research. In order to reach a consensus on that, the World Endometriosis Research Foundation (WERF) created the Endometriosis Phenome and Biobanking Harmonisation Project (EPHect), whose purpose is to facilitate and enable large-scale, cross-centre, longitudinal, biomarker and treatment target discovery research, developing a detailed clinical phenotyping (phenome) data to be collected from women with endometriosis and controls to allow collaborative sub-phenotype discovery

5


Table III Modified QUADAS scoring for studies and main chemokines assessed. Paper

Modified QUADAS criteria

.................................................................................

1

2

3

4

5

6

7

8

9

10

11

Number

.........................

Endo

Chemokines assessed

Control

............................................................................................................................................................................................. Agic et al. (2007)

n

n

y

y

y

u

u

y

y

y

n

83

51

CCR1mRNA

Agic et al. (2008)

n

n

y

y

y

u

u

y

y

y

y

102

49

CCR1mRNA, CCL2

Akoum et al. (1995)

n

n

u

y

y

u

u

y

y

n

n

06

06

CCL2

Akoum et al. (1996a)

n

n

u

y

y

u

u

y

y

n

n

36

21

CCL2

Akoum et al. (1996b)

n

u

u

y

y

u

u

y

y

y

y

57

44

CCL2

n

u

y

y

y

u

u

y

y

y

y

11

12

CCL2

n

n

y

y

y

u

u

y

y

n

y

45

28

CXCL8

Arici et al. (1997)

n

n

y

y

y

u

u

y

y

n

y

60

18

CCL2

Barcz et al. (2002)

u

y

y

y

y

u

u

y

y

y

y

52

32

CXCL8

Barcz et al. (2012)

u

u

y

y

y

u

u

y

y

y

y

65

37

CXCL8

Bellelis et al. (2013)

n

y

y

y

y

u

u

y

y

y

u

32

32

CCL17, 21, CXCL9, 10, 11, 12, XCL1, CX3CL1

Bersinger et al. (2006)

u

y

y

y

y

u

u

y

y

y

y

77

55

CXCL8, CCL5

Braun et al. (1996)

u

u

u

u

y

u

u

y

y

n

n

20

10

CXCL8

Calhaz-Jorge et al. (2003)

n

u

y

y

y

u

u

y

y

y

y

68

24

CXCL8

Carmona et al. (2012)

n

y

u

y

y

u

u

y

y

u

u

33

24

CXCL8

Chand et al. (2007)

u

u

y

y

y

u

u

y

y

y

y

10

04

CCL16, CCL21

Darai et al. (2003)

u

u

y

y

y

u

u

y

y

u

u

34

43

CXCL8

Drosdzol-Cop et al. (2012)

y

y

y

y

y

u

u

y

y

y

y

33

17

CCL2

Fang et al. (2009)

n

u

y

n

y

u

u

y

y

y

y

11

05

CCL5

Fasciani et al. (2001)

n

y

y

n

y

u

u

y

y

n

y

38

110

CXCL8

Galleri et al. (2009)

y

y

y

y

y

u

u

y

y

y

y

77

70

CXCL10

Gazvani et al. (1998)

n

u

y

y

y

u

u

y

y

n

y

25

22

CXCL8

Gmyrek et al. (2005)

n

y

y

u

y

u

u

y

y

y

y

18

32

CCL2

Hornung et al. (2000)

n

y

y

y

y

u

u

y

y

y

y

15

07

CCL11


n

y

y

y

y

u

u

y

y

n

y

19

09

CCL5

Iwabe et al. (1998)

n

u

y

y

y

u

u

y

y

y

n

24

16

CXCL8

Jolicoeur et al. (1998)

n

y

y

y

y

u

u

y

y

y

y

47

22

CCL2 CXCL8, CCL5

Kalu et al. (2007)

y

u

y

y

y

u

u

y

y

y

y

26

31

Kharfi and Akoum (2001)

n

y

y

y

y

u

u

y

y

y

y

46

22

CCL2

Khorram et al. (1993)

n

u

y

y

y

u

u

y

y

n

y

24

12

CCL5

Kim et al. (2008)

y

n

y

y

y

u

u

y

y

y

y

105

101

CCL2

Kim et al. (2009)

n

y

y

y

y

u

u

y

y

y

y

12

05

CXCL8, CXCL10

Kuroda et al. (2010)

n

y

y

y

y

u

u

y

y

n

n

42

31

CXCL8

Kyama et al. (2008)

y

u

y

y

y

u

u

y

y

y

y

24

11

CCL2, CCL5

Laudanski et al. (2006)

y

u

y

y

y

u

u

y

y

y

y

34

18

CCL2, 3, 4, 5, 15, 17, 19, 20, 22

Li et al. (2012a)

u

u

y

u

y

u

u

y

y

y

y

10

10

CCL2, CCR2

Li et al. (2012b)

u

u

y

u

y

u

u

y

y

n

n

12

18

CXCL8, CXCR1/R2

Luk et al. (2010)

n

u

y

y

y

u

u

y

y

y

n

16

18

CCL2

Malhotra et al. (2012)

n

y

y

y

y

u

u

y

y

n

y

58

28

CXCL8

Margari et al. (2013)

y

u

y

y

y

u

u

y

y

u

y

54

42

CCL2, CCL5, CCL7

Mihalyi et al. (2010)

y

y

y

y

y

u

u

y

y

y

y

201

93

CXCL8

Milewski et al. (2011)

n

y

y

y

y

u

u

y

y

y

y

67

16

CXCL8

Mueller et al. (2003)

n

y

y

y

y

u

u

y

y

n

n

18

09

CXCL5

Na et al. (2011)

n

y

y

y

y

u

u

y

y

y

y

12

05

CCL2, CCL3, CCL5

Ohata et al. (2008)

y

n

u

y

y

u

u

y

y

y

y

70

21

CXCL8

Continued


Akoum et al. (2002) Arici et al. (1996)

6

Borrelli et al.

Table III Continued Paper

Modified QUADAS criteria

.................................................................................

1

2

3

4

5

6

7

8

9

10

11

Number

.........................

Endo

Chemokines assessed

Control

............................................................................................................................................................................................. Oral et al. (1996)

n

u

y

y

y

u

u

y

y

n

y

63

19

CXCL1

Othman et al. (2008)

y

n

y

y

y

u

u

y

y

y

y

68

70

CCL2, CXCL8

Pizzo et al. (2002)

y

n

y

y

y

u

u

y

y

y

y

26

05

CCL2, CXCL8

Rana et al. (1996)

n

u

y

y

y

u

u

y

y

y

n

17

08

CXCL8

Rocha et al. (2012)

y

u

y

y

y

u

u

y

y

y

y

06

06

CXCL8

Ryan et al. (1995)

n

u

y

y

y

u

u

y

y

n

n

18

09

CXCL8

Seeber et al. (2008)

n

n

u

y

y

u

u

y

y

y

y

63

78

CCL2

Shimoya et al. (2005)

y

y

u

y

y

u

u

y

y

u

y

14

25

CX3CL1

y

y

y

y

y

u

u

y

y

y

n

48

30

CXCL8

Socolov et al. (2011)

n

y

y

y

y

u

u

y

y

u

y

24

24

CXCL8

Suzumori et al. (2004)

y

u

y

y

y

u

u

y

y

u

y

35

24

CXCL5


y

u

y

y

y

u

u

y

y

y

y

38

26

CXCL6

Szamatowicz et al. (2002)

y

u

y

y

y

u

u

y

y

n

u

22

21

CXCL1

Tao et al. (2011)

n

y

y

y

y

u

u

y

y

n

y

28

47

CCL2

Ulukus et al. (2005)

n

u

y

y

y

u

u

y

y

y

y

25

27

CXCR1, CXCR2


n

u

y

y

y

u

u

y

y

y

y

24

27

CXCL8, CCL2

Vodolazkaia et al. (2012)

y

u

u

y

y

u

u

y

y

y

y

77

41

CXCL8, CCL2, CCL5

QUADAS, quality assessment of diagnostic accuracy studies; y, yes; n, no; u, unknown.

measured CXCL8 in the PB, 6 (46.1%) demonstrated its potential as a marker for endometriosis with diagnostic value. Few studies measured the expression of CXCL8 chemokine or its receptors CXCR1 and CXCR2 in EESs of women with and without endometriosis. All studies (Ulukus et al., 2009; Li et al., 2012a, 2012b; Rocha et al., 2012) found higher expression of CXCL8 in endometrial cells among patients with endometriosis compared with controls. When the CXCL8 receptors CXCR1 or CXCR2 were measured in place of the ligand (CXCL8), Ulukus et al. (2005) found higher expression from both receptors among endometriosis patients, while Li et al. (2012a,b) observed that only the expression of the receptor CXCR1 was higher.

CCL2 (MCP-1) The second most evaluated chemokine, as a potential marker for endometriosis was the monocyte chemotatic protein-1 or CCL2, a member of the b-chemokine family. It has been measured in the PF (45.8%), PB (41.6%) and tissue —EES (33.3%). While 54.5% of the authors found statistically significant higher concentration of CCL2 in the PF among endometriosis patients compared with controls (Akoum et al., 1996a, 2002; Arici et al., 1997; Pizzo et al., 2002; Kalu et al., 2007; Tao et al., 2011), 36.3% found no statistically significant difference (P . 0.05) between endometriosis patients and controls (Laudanski et al., 2006; Kim et al., 2008; Na et al., 2011; Drosdzol-Cop et al., 2012). Furthermore, one group of investigators found significant lower concentrations of CCL2 among endometriosis patients compared with controls (Margari et al., 2013). After assessing the studies that measured serum levels of CCL2, we found the results well divided, with 50% showing significantly higher CCL2 concentration in PB in patients with the disease compared with

that of controls (Akoum et al., 1996b; Pizzo et al., 2002; Gmyrek et al., 2005; Agic et al., 2008; Othman et al., 2008), and the other 50% showing no statistically significant difference (P . 0.05) between the groups with or without endometriosis (Kalu et al., 2007; Kim et al., 2008; Seeber et al., 2008; Drosdzol-Cop et al., 2012; Vodolazkaia et al., 2012). The majority of the studies that evaluated CCL2 in the tissue—EES— found significantly higher expression of CCL2 in the EESs of patients with endometriosis compared with controls (Akoum et al., 1995; Jolicoeur et al., 1998; Kharfi and Akoum, 2001; Ulukus et al., 2009; Li et al., 2012a), while two studies found no statistically significant difference (P . 0.05) (Kyama et al., 2008; Luk et al., 2010). Analyzing the results from Li et al. (2012a, b), we observe that they identified significant higher expression of CCL2 in endometrial samples of endometriosis patients compared with controls, but found no statistically significant difference (P . 0.05) of CCR2 (CCL2 receptor) expression in the same group of patients.

CCL5 (RANTES) The b-chemokine CCL5, also known as RANTES (regulated upon activation, normal T cell expressed and secreted), was the first chemokine evaluated in association with endometriosis back in 1993. In this literature review, we observed that this chemokine was the third most assessed during the past 20 years as a possible marker for endometriosis. Studies on the CCL5 chemokine, as we have seen for the first two chemokines, CXCL8 and CCL2, also measured CCL5 in the PF (58.3%), PB (33.3%) and tissue—EES (16.6%). Here, the analysis of the PF also shows controversial results, with three publications (42.8%) revealing statistically significantly higher concentrations of CCL5 among endometriosis


Skrzypczak et al. (2005)

7


Table IV Summary of chemokine markers measured in patients with endometriosis. Authors (year)

Biomarker

Samples

Phase of cycle (E/C)

Endometriosis/ stage (ASRM)

Controls

Significance (P value)

............................................................................................................................................................................................. CCR1 (CCL5 receptor)

PB

Follicular

n ¼ 83

n ¼ 51(myoma or tubal ligation)

Higher (,0.001)

Agic et al. (2008)

CCR1 (CCL5 receptor) CCL2 (MCP-1)

PB

Follicular

I –II (n ¼ 37) III–IV (n ¼ 65)

n ¼ 49 (myoma or tubal ligation)

Higher CCR1 (,0.001) CCL2 (,0.05)

Akoum et al. (1995)

CCL2 (MCP-1)

Tissue (endometrium)

Not mentioned

I –II (n ¼ 5) III–IV (n ¼ 1)

n ¼ 6 (tubal ligation)

Higher

Akoum et al. (1996a)

CCL2 (MCP-1)

PF

L (15/8) F (14/3)

I –II (n ¼ 32) II– IV (n ¼ 4)


Higher (,0.01)

Akoum et al. (1996b)

CCL2 (MCP-1)

PB

Luteal (L)

I –II (n ¼ 47) III–IV (n ¼ 10)


Higher (0.01)

Akoum et al. (2002)

CCL2 (MCP-1)

PF macrophage

L (4/5) F (7/7)

I –II (n ¼ 9) III–IV (n ¼ 2)


Higher—F (6/20/ 44 h incubation) (0.03/0.01/0.057)

Arici et al. (1996)

CXCL8 (IL-8)

PF

Not mentioned

I –II (n ¼ 24) III–IV (n ¼ 21)


Higher (0.021)

Arici et al. (1997)

CCL2 (MCP-1)

PF

Not mentioned

I –II (n ¼ 28) III–IV (n ¼ 32)


Higher S.III (0.01) S.IV (0.03)

Barcz et al., (2002)

CXCL8 (IL-8)

PF/PB

Not mentioned

I –III (n ¼ 46) – P.F. I –III (n ¼ 47) – P.B.

n ¼ 20—P.F. n ¼ 22—P.B.

PF-Higher (,0.05) PB—NS

Barcz et al. (2012)

CXCL8 (IL-8)

PF

Follicular

I –II (n ¼ 25) III–IV (n ¼ 40)

n ¼ 37

Higher (0.0002)

Bellelis et al. (2013)

CCL17 CCL21 CXCL9 CXCL10 CXCL11 CXCL12 CX3CL1 XCL1


Luteal and follicular

n ¼ 32 (n ¼ 12— bowel endometriosis/ n ¼ 10— retro-cervical)


Higher—Bowel EDT—CCL17 (P,0.05) Others—NS

Bersinger et al. (2006)

CXCL8 (IL-8) CCL5 (RANTES)

PF

L (34) F (98)

I –II (n ¼ 18) III–IV (n ¼ 59)

n ¼ 55 (no evidence of EDT)

Higher (CCL5 and CXCL8): (,0.001)

Braun et al. (1996)

CXCL8 (IL-8)

PB monocytes

Not mentioned

n ¼ 20

n ¼ 10

Higher

Calhaz-Jorge et al. (2003)

CXCL8 (IL-8)

PF

Luteal and Follicular

I –II (n ¼ 53) III–IV (n ¼ 15)


Higher (0.019)

Carmona et al. (2012)

CXCL8 (IL-8)

PB


Ovarian Endometrioma (OE)—(n ¼ 19) DIE (n ¼ 14)


OE: Higher (,0.01) DIE: NS

Chand et al. (2007)

CCL16 (LEC) CCL21 (6Ckine)


Mid-luteal

I –II (n ¼ 5) III–IV (n ¼ 5)


Higher (0.049)

Darai et al. (2003)

CXCL8 (IL-8)

PB


Endometrioma n ¼ 34

n ¼ 30 + 13 (Benign + malignant ovarian tumors)

Drosdzol-Cop et al. (2012)

CCL2 (MCP-1)

PF/PB

Follicular

I –II (n ¼ 19) III–IV (n ¼ 14)

n ¼ 17 (pelvic pain and no evidence of EDT)

Fang et al. (2009)

CCL5 (RANTES)


Mid-follicular

III–IV (n ¼ 11)

n ¼ 5 (IUD removal)

Fasciani et al. (2001)

CXCL8 (IL-8)

PB

Not mentioned

Endometrioma (III– IV) (n ¼ 38)

n ¼ 110 (53 + 9 + 10) Benign + malignant ovarian tumors + healthy controls)

Galleri et al. (2009)

CXCL10 (interferon-g-induced protein 10/IP-10)

PF/ PB

L (35/32) F (32/38)

I –II (n ¼ 40) III–IV (n ¼ 37)


Higher (,0.05)

PB—lower (,0.001) PF— Lower-S. III– IV: (,0.001)

Continued


Agic et al. (2007)

8

Borrelli et al.

Table IV Continued Authors (year)

Biomarker

Samples



Controls


............................................................................................................................................................................................. CXCL8 (IL-8)

PF/PB

Not mentioned

I –II (n ¼ 14) III–IV (n ¼ 11)


PF—higher all S. (0.02) S.I (0.001) PB—NS

Gmyrek et al. (2005)

CCL2 (MCP-1)

PB

L (0/20) F (18/12)

I –II (n ¼ 10) III–IV (n ¼ 8)

Myoma (M): n ¼ 16 Healthy (H): n ¼ 16

HigherM: (,0.05) H: (,0.005)


CCL11 (Eotaxin)

PF

Mid-follicular

I –II (n ¼ 8) III–IV (n ¼ 7)


S.I– II: NS S.III–IV: Higher (,0.05)


CCL5 (RANTES)

PF

Not mentioned

I –II (n ¼ 11) III–IV (n ¼ 8)


Higher (,0.05)

Iwabe et al. (1998)

CXCL8 (IL-8)

PF

L (14/6) F (10/10)

n ¼ 24


Higher (0.0005)

Jolicoeur et al. (1998)

CCL2 (MCP-1)


L (30/13) F (17/9)

I –II (n ¼ 35) III–IV (n ¼ 12)

n ¼ 22

Higher (0.0001)

Kalu et al. (2007)

CXCL8 (IL-8) CCL2 (MCP-1) CCL5 (RANTES)

PF/PB

Luteal

I –II (n ¼ 26)

n ¼ 31 (un-explained infertility)

PF: CXCL8-higher (0.003) CCL2-higher (,0.012)

Kharfi and Akoum (2001)

CCL2 (MCP-1)


L (28/10) F (18/12)

I –II (n ¼ 35) III–IV (n ¼ 11)


Higher (0.002)

Khorram et al. (1993)

CCL5 (RANTES)

PF

Not mentioned

I –II (n ¼ 12) III–IV (n ¼ 12)


Higher

Kim et al. (2008)

CCL2 (MCP-1)

PF/PB

Follicular

I –II (n ¼ 62) III–IV (n ¼ 43)


Kim et al. (2009)

CXCL8 (IL-8) CXCL10 (IP-10)

PF

Follicular

III–IV (n ¼ 12)

n ¼ 5 (dermoid cysts and cervical cancer)

Kuroda et al. (2010)

CXCL8 (IL-8)

PF

Not mentioned

n ¼ 27 (GnRH) n ¼ 15 (no GnRH)

n ¼ 18 (GnRH) n-13 (no GnRH)

Kyama et al. (2008)

CCL2 (MCP-1) CCL5 (RANTES)


Luteal (14/6) Menstrual (10/5)

I –II (n ¼ 12) III–IV (n ¼ 12)

n ¼ 11(no evidence of EDT)

Laudanski et al. (2006)

Mini-array CCL22 CCL17 CCL2 CCL5 CCL3 CCL4 CCL15 CCL20 CCL19 (MIP-3b)

PF

Follicular

I –II (n ¼ 14) III–IV (n ¼ 10)

n ¼ 18 (infertile with no evidence of EDT)

CCL19: Higher (0.0007)

Li et al. (2012a)

CCL2 (MCP-1) CCR2 (MCP-1 receptor)


Follicular

I –II (n ¼ 10)

n ¼ 10

CCL2 –Higher (,0.001)

Li et al. (2012b)

CXCL8 (IL-8) CXCR1 CXCR2 (IL-8 receptors)


Not mentioned

n ¼ 12

n ¼ 18

CXCL8: Higher (,0.05) CXCR1: Higher (,0.05)

Luk et al. (2010)

CCL2 (MCP-1)


L (8/10) F (8/8)

n ¼ 16

n ¼ 18

Malhotra et al. (2012)

CXCL8 (IL-8)

PF

Follicular

I –II (n ¼ 18) III–IV (n ¼ 40)

n ¼ 28

Higher (,0.002)

Margari et al. (2013)

CCL2 (MCP-1) CCL5 (RANTES) CCL7 (MCP-3)

PF


I –II (n ¼ 30) III–IV (n ¼ 24)

n ¼ 42

CCL2: Lower (0.024)

Higher (0.032)

Continued


Gazvani et al. (1998)

9


Table IV Continued Authors (year)

Biomarker

Samples



Controls


............................................................................................................................................................................................. CXCL8 (IL-8)

PB

L (78/38) F (83/36)

I –II (n ¼ 132) III–IV (n ¼ 69)

n ¼ 93

Higher (,0.0001)

Milewski et al. (2011)

CXCL8 (IL-8)

PF

Follicular

I –II (n ¼ 20) III–IV (n ¼ 47)

n ¼ 16 (infertile with no evidence of EDT)

Higher (,0.007)

Mueller et al. (2003)

CXCL5 (ENA-78)

PF

Not mentioned

I –IV (n ¼ 18)

n¼9

Higher (,0.05)

Na et al. (2011)

CCL2 (MCP-1) CCL3 (MIP-1a) CCL5 (RANTES)

PF

Follicular

Endometrioma (III– IV)—n ¼ 12

n ¼ 5 (dermoid cysts and cervical cancer)

Ohata et al. (2008)

CXCL8 (IL-8)

PB

L (38/9) F (32/12)

n ¼ 70 (ovarian endometrioma)

n ¼ 21 (benign ovarian cyst)

Higher (,0.005)

Oral et al. (1996)

CXCL1 (GRO-a)

PF

Not mentioned

I –II (n ¼ 32) III–IV (n ¼ 31)

n ¼ 19

S. III/IV: Higher (0.04/0.01)

Othman et al. (2008)

CCL2 (MCP-1) CXCL8 (IL-8)

PB

L (39/39) F (29/31)

I –II (n ¼ 32) III–IV (n ¼ 36)


CCL2 –Higher (,0.001)

Pizzo et al. (2002)

CCL2 (MCP-1) CXCL8 (IL-8)

PF/PB

Luteal

I –II (n ¼ 20) III–IV (n ¼ 6)


Higher PF/PB (CCL2 and CXCL8): (,0.001)

Rana et al. (1996)

CXCL8 (IL-8)

PF

Mid-luteal

n ¼ 17


Higher

Rocha et al. 2012)

CXCL8 (IL-8)


Follicular

III–IV (n ¼ 6) (Endometrioma)

n ¼ 6 (ovarian cyst or myoma)

Higher (,0.01)

Ryan et al. (1995)

CXCL8 (IL-8)

PF

Not mentioned

II– IV (n ¼ 18)


Higher

Seeber et al. (2008)

CCL2 (MCP-1)

PB

L (7/18) F (46/45)

II (n ¼ 22) III–IV (n ¼ 41)

n ¼ 78 (pain, infertility and tubal ligation)

Shimoya et al. (2005)

CX3CL1 Fractalkine

PF/PB

L (NM/11) F (NM/14)

I –II (n ¼ 9) III–IV (n ¼ 5)

n ¼ 25 (infertility)

PF—Lower (,0.005)

Skrzypczak et al. (2005)

CXCL8 (IL-8)

PF

Luteal

n ¼ 48 (Infertility ¼ 28 Fertility ¼ 20)

n ¼ 30 (Infertility ¼ 20 fertility ¼ 10)

Infertility: Higher (,0.01)

Socolov et al. (2011)

CXCL8 (IL-8)

PB


I –II (n ¼ 7) III–IV (n ¼ 17)

n ¼ 24 (benign ovarian cysts or tubal ligation)


CXCL5 (ENA-78)

PF


I –II (n ¼ 15) III–IV (n ¼ 20)

n ¼ 24 (ovarian cystoadenomas)

Higher (,0.001)


CXCL6 (GCP-2)

PF

L (16/12) F (22/14)

I –II (n ¼ 16) III–IV (n ¼ 22)

n ¼ 26 (ovarian cysts)

Higher (0.002)

Szamatowicz et al. (2002)

CXCL1 (GRO-a)

PF

Follicular

II (n ¼ 10) III (n ¼ 12)

n ¼ 21

Higher (0.05)

Tao et al. (2011)

CCL2 (MCP-1)

PF

Not mentioned

I –II (n ¼ 28)

n ¼ 47 (Fallopian obstruction ¼ 23 and myoma ¼ 24)

Higher (,0.01)


CXCR1 CXCR2 (IL-8 receptors)


L (13/18) F (12/9)

I –III (n ¼ 25)

n ¼ 27 (other gynecology pathologies than EDT)

Higher (,0.05)


CXCL8 (IL-8) CCL2 (MCP-1)


L (10/18) F (14/9)

III–IV (n ¼ 24)

n ¼ 27 (other gynecology pathologies than EDT)

Higher CXCL8: (0.04) CCL2: (0.003)

Vodolazkaia et al. (2012)

CXCL8 (IL-8) CCL5 (RANTES) CCL2 (MCP-1)

PB

L (27/16) F (32/17) M (18/8)

I –II (n ¼ 46)* III–IV (n ¼ 31)* *Patients from the Test set

n ¼ 41 (subfertility and no evidence of EDT)* *Controls from the Test set

CXCL8: Higher— (0.05) CCL5: Higher— (0.01)

NS, not significant; PF, peritoneal fluid; PB, peripheral blood; S, stage; L, luteal; F, follicular; M, menstrual; EDT, endometriosis; NM, not mentioned; E/C, endometriosis/controls; ENA-78, epithelial-derived neutrophil-activating protein-78; GCP, granulocyte chemotactic protein; GRO-a, growth regulated oncogene a; LEC, liver expressed chemokine; IL-8, interleukin-8; IP-10, interferon gamma induced protein-10; IUD, intrauterine device; MCP-1, monocyte chemotactic protein-1; MIP, macrophage inflammatory protein-1; RANTES, regulated upon activation, normal T cell expressed and secreted.


Mihalyi et al. (2010)

10

Others Here, we describe the main findings for the other chemokines, which were assessed only once or twice in the studies reviewed. Bellelis et al. (2013) assessed the following chemokines—CCL17, CCL21, CXCL9, CXCL10, CXCL11, CXCL12, CX3CL1 and XCL1 via endometrial biopsies. They found significantly increased levels of CCL17 among patients with bowel endometriosis compared with controls and to patients with retrocervical endometriosis; the other chemokines did not show statistically significant difference (P . 0.05) in EES in patients with endometriosis compared with controls. Chand et al. (2007) evaluated the b-chemokines CCL16 (HCC4) and CCL21 (6Ckine) in tissue samples (EES) and observed that their expression was significantly higher in endometriosis patients compared with controls. After assessing the a-chemokine CXCL10 (IP-10/interferon-g-induced protein-10) in the PF and PB, Galleri et al. (2009) found significantly lower concentrations of this chemokine among all stages endometriosis patients in the PB and also significantly lower concentrations among Stages III –IV in the PF, while Kim et al. (2009) found no statistically significant difference (P . 0.05) of CXCL10 levels in the PF of patients with the disease compared with controls. Hornung et al. (2000) studied the CCL11 chemokine (eotaxin) in the PF and found significantly higher concentrations among endometriosis patients with Stages III –IV, but no statistically significant difference (P . 0.05) for Stages I – II compared with controls. Laudanski et al. (2006) used a mini-array of nine different b-chemokines (CCL2, CCL3, CCL4, CCL5, CCL15, CCL17, CCL19, CCL20 and CCL22) and assessed their concentrations in the PF of women with and without endometriosis, finding significantly higher concentration of only the CCL19 (MIP-3b) among endometriosis patients, and this was not related to the stage of disease. Margari et al. (2013) evaluated another b-chemokine, the CCL7, in the PF and found no statistically significant difference (P . 0.05) between the group with the disease and controls. Two studies assessed the a-chemokine CXCL5 (ENA-78) and both found that its concentrations were significantly higher in the PF of women with endometriosis compared with controls (Mueller et al., 2003; Suzumori et al., 2004). The CX3CL1 chemokine (Fractalkine) was assessed in the PF and in PB samples by Shimoya et al., (2005), but no

statistically significant difference (P . 0.05) was found between patients and controls in serum samples; however, in the PF, its levels were significantly lower in patients with endometriosis compared with controls, suggesting a lack of peritoneal immunological protection in patients with the disease. Two studies (Oral et al., 1996; Szamatowicz et al., 2002) assessed the chemokine growth-regulated-alpha (GRO-a)—CXCL1— in the PF of patients with endometriosis and controls: the former found significantly higher levels only among patients with endometriosis Stages III –IV compared with controls, while the latter found significantly higher levels of this chemokine in all endometriosis patients, regardless of the stage of disease. The last different type of chemokine assessed among the reviewed studies was the CXCL6 (GCP-2), by Suzumori et al. (2005), and their results showed a higher concentration of this chemokine in the PF of women with endometriosis (any stage) compared with controls.

Discussion It is definitely not an easy task to find a good marker capable of diagnosing endometriosis at all stages, although it is a very important issue in endometriosis research (May et al., 2010). This might be more difficult to establish with chemokines due to their role in several unspecific inflammatory pathways and chronic diseases, which sometimes makes it hard to distinguish patients and controls when using them as biomarkers. Nevertheless, this review of the literature has shown a large number of studies that investigate many different subtypes of chemokines as possible biological markers for endometriosis. Despite the controversial results presented, there were a great number of significantly positive findings among the 62 studies included in this systematic review. Considering all the chemokines assessed and all types of samples across the studies reviewed, ca. 55% of the tests showed significantly increased chemokines levels among endometriosis patients, and if we consider only the three most studied chemokines (CXCL8, CCL2 and CCL5), this percentage rises up to 63.5%. Furthermore, if we look at the results of some specific chemokines alone, CXCL8 (IL-8) showed significantly higher levels in 72.7% of the tests, considering all samples, or even 93.75%, when assessed in the PF and 75% in the tissue—endometrial samples. The blood evaluation of CXCL8 showed significantly higher levels in 46.1% of the tests, while 53.9% of them did not find any difference. Nevertheless, we highlight the findings from Mihalyi et al. (2010) and Vodolazkaia et al. (2012), two well-designed studies with a great number of subjects that showed statistically higher levels of this chemokine in the PB, demonstrating its potential for diagnostic value. Generally, all the findings suggest that CXCL8 certainly has a stronger tendency to have diagnostic value in the PF and endometrial samples, regardless of the stage of disease, and it might also have this potential in the PB. Although the results from the chemokine CCL2 are controversial in all three different types of samples, overall they show a tenuous tendency to have diagnostic value, as 54.5% of the authors found statistically significantly higher concentration of CCL2 in the PF among endometriosis patients versus controls, while 71.4% presented such results in the endometrial samples, although, in both cases, the results generally did not vary according to the stage of disease. The results were well divided only when assessed in the PB, with 50% showing significantly higher levels, while the other 50% did not find any difference. We observed the same tenuous tendency for CCL5 (RANTES) to have diagnostic value as we did for CCL2. The best results were in the


patients versus controls (Khorram et al., 1993; Hornung et al., 2001; Bersinger et al., 2006), while another four (57.2%) found no statistically significant difference (P . 0.05) between the group with the disease and controls (Laudanski et al., 2006; Kalu et al., 2007; Na et al., 2011; Margari et al., 2013). In blood samples, Agic et al. (2007, 2008) found significantly higher levels of CCR1 (CCL5 receptor) mRNA in patients with endometriosis compared with controls without the disease, while Vodolazkaia et al. (2012) found increased levels of CCL5 (the ligand) in endometriosis patients compared with controls, which means that 75% of the studies had positive results assessing CCL5 in the blood. Only one paper (25%) from Kalu et al. (2007) found no statistically significant difference (P . 0.05) in the blood titles of CCL5 between the groups of patients and controls. Only two studies assessed CCL5 in tissue and they did not have the same findings: Fang et al. (2009) found significantly higher expression of CCL5 in EESs from endometriosis patients compared with controls, while Kyama et al. (2008) found no statistically significant difference (P . 0.05) between the groups.

Borrelli et al.

11


were appraised in a systematic matter and we could increase our knowledge with relevant information. Based on that, future research in the field should be conducted in better-designed, large-scale and multi-centre studies, as we have mentioned before. In our opinion, chemokines have the potential to be used as a marker and, perhaps, as a diagnostic test for endometriosis. Nevertheless, further investigations should be considered in this matter following the reasoning of the combination of markers as an ideal test for endometriosis, especially if associated with other non-inflammatory markers. We may then reach the validation stage of a good test.

Supplementary data Supplementary data are available at http://humrep.oxfordjournals.org/.

Authors’ roles All three authors conducted this systematic review. G.M.B. and M.S.A. assessed and selected the studies for eligibility. G.M.B. extracted data from the studies and prepared the manuscript. M.S.A. and S.M. critically reviewed and corrected the manuscript. S.M. supervised the project and, together with G.M.B., edited this final version of the manuscript.

Funding There was no funding to support this systematic review.

Conflict of interest None declared.

References Abraõ MS, Podgaec S, Filho BM, Ramos LO, Pinotti JA, de Oliveira RM. The use of biochemical markers in the diagnosis of pelvic endometriosis. Hum Reprod 1997;12:2523 – 2527. Abraõ MS, Gonçalves MO, Dias JA Jr, Podgaec S, Chamie LP, Blasbalg R. Comparison between clinical examination, transvaginal sonography and magnetic resonance imaging for the diagnosis of deep endometriosis. Hum Reprod 2007;22:3092 – 3097. Acker FA, Voss HP, Timmerman H. Chemokines: structure, receptors and functions. Mediators Inflamm 1996;5:393– 416. Agic A, Xu H, Rehbein M, Wolfler MM, Ebert AD, Hornung D. CCR1 mRNA expression in peripheral blood as a diagnostic test for endometriosis. Fertil Steril 2007;87:982 – 984. Agic A, Djalali S, Wolfler MM, Halis G, Diedrich K, Hornung D. Combination of CCR1 mRNA, MCP-1, and CA125 measurements in peripheral blood as a diagnostic test for endometriosis. Reprod Sci 2008;15:906– 911. Akoum A, Lemay A, Brunet C, He´bert J. Secretion of monocyte chemotactic protein-1 by cytokine-stimulated endometrial cells of women with endometriosis. Le groupe d’investigation en gynećologie. Fertil Steril 1995; 63:322– 328. Akoum A, Lemay A, McColl S, Turcot-Lemay L, Maheux R. Elevated concentration and biologic activity of monocyte chemotactic protein-1 in the peritoneal fluid of patients with endometriosis. Fertil Steril 1996a; 66:17– 23. Akoum A, Lemay A, Mccoll SR, Paradis I, Maheux R. Increased monocyte chemotactic protein-1 level and activity in the peripheral blood of


PB, where 75% of the studies found significantly higher levels of this chemokine among endometriosis patients compared with controls, regardless of the stage of disease. Many other chemokines were assessed alone, only once or twice, and they also showed significantly higher levels among endometriosis patients versus controls, such as CCL11 (Hornung et al., 2000), CCL16 (Chand et al., 2007), CCL17 (Bellelis et al., 2013), CCL19 (Laudanski et al., 2006), CCL21 (Chand et al., 2007), CXCL1 (Oral et al., 1996; Szamatowicz et al., 2002), CXCL5 (Mueller et al., 2003; Suzumori et al., 2004) and CXCL6 (Suzumori et al., 2005). We consider three major issues to explain the current controversial results in the evaluation of chemokines as possible markers for endometriosis patients: (i) first, the studies design, including here the sample size, the patients and controls selection criteria, as well as the criteria to assess the markers (phase of the menstrual cycle, stage of disease), as we could assess using the modified QUADAS criteria; (ii) second, we believe that evaluating one chemokine alone may not be the best way to find a good biological marker for endometriosis, but rather the association between some of them, in a panel of markers, as proposed before with other kinds of potential biological markers in a multivariate analysis (Vodolazkaia et al., 2012); and (iii) third, the real value of inflammatory markers, as chemokines, in the distinction of patients with endometriosis from women without endometriosis but with other pelvic pathologies. Therefore, future analyses should test the most relevant chemokines in a panel of biomarkers models, combined or not with other noninflammatory markers that have already been tested and shown to have good potential in previous studies, in order to reach better and concrete results. The three main chemokines assessed to date—CXCL8, CCL2 and CCL5—would certainly present better results in welldesigned, large-scale and multi-centre studies, as the new consensus from the project ‘EPHect’ will enable. Agic et al., (2008) have already done it using the combination of CCR1 (RANTES/CCL5-receptor) mRNA, MCP-1 (CCL2) and CA125 measurements in the PB as a diagnostic test for endometriosis. This combination showed good results, with 92.2% sensitivity, 81.6% specificity, negative predictive value of 83.3%, positive predictive value of 92.3% and, most importantly, a likelihood ratio of a positive test result of 5.017, and a likelihood ratio of a negative test result of 0.096 to predict the presence or absence of endometriosis. Vodolazkaia et al. (2012) also evaluated a panel of 28 biomarkers for the non-invasive diagnosis of endometriosis, which included three chemokines (IL-8/CXCL8, MCP-1/CCL2 and RANTES/CCL5). Although they have found significantly positive results for IL-8 and RANTES, as we have presented in the results of this present review, they did not include any of them in the two models of four biomarkers (anexin V, VEGF, CA125 and glycodelin or anexin V, VEGF, CA125 and sICAM-I) suggested as a potential blood test with good performance for endometriosis, with high sensitivity and acceptable specificity in patients with subfertility and/or pain without imaging findings. In the two publications discussed above, the association of markers improved the diagnostic performance better than any of the biomarkers assessed alone, and this is probably the best approach for future investigations, regardless of the type of markers involved in the study. With this systematic review, we can conclude that chemokines certainly play an important role in the pathogenesis of endometriosis. However, it remains unclear whether they can or cannot be used as markers to diagnose and predict the development of endometriosis. For the first time in the literature, all chemokines that were once tested for endometriosis

12

Fang CL, Han SP, Fu SL, Wang W, Kong N, Wang XL. Ectopic, autologous eutopic and normal endometrial stromal cells have altered expression and chemotactic activity of RANTES. Eur J Obstet Gynecol Reprod Biol 2009;143:55 – 60. Fasciani A, D’Ambrogio G, Bocci G, Luisi S, Artini PG, Genazzani AR. Vascular endothelial growth factor and interleukin-8 in ovarian cystic pathology. Fertil Steril 2001;75:1218– 1221. Fassbender A, Vodolazkaia A, Saunders P, Lebovic D, Waelkens E, De Moor B, D’Hooghe T. Biomarkers of endometriosis. Fertil Steril 2013; 99:1135 – 1145. Galleri L, Luisi S, Rotondi M, Romagnani P, Cobellis L, Serio M, Petraglia F. Low serum and peritoneal fluid concentration of interferon-gammainduced protein-10 (CXCL10) in women with endometriosis. Fertil Steril 2009;91:331 – 334. Gazvani MR, Christmas S, Quenby S, Kirwan J, Johnson PM, Kingsland CR. Peritoneal fluid concentrations of interleukin-8 in women with endometriosis: relationship to stage of disease. Hum Reprod 1998;13:1957– 1961. Gmyrek GB, Sozanski R, Jerzak M, Chrobak A, Wickiewicz D, Skupnik A, Sieradzka U, Fortuna W, Gabrys M, Chelmonska-Soyta A. Evaluation of monocyte chemotactic protein-1 levels in peripheral blood of infertile women with endometriosis. Eur J Obstet Gynecol Reprod Biol 2005; 122:199– 205. Gonçalves MO, Dias JA Jr, Podgaec S, Averbach M, Abraõ MS. Transvaginal ultrasound for diagnosis of deeply infiltrating endometriosis. Int J Gynaecol Obstet 2009;104:156– 160. Hornung D, Dohrn K, Sotlar K, Greb RR, Wallwiener D, Kiesel L, Taylor RN. Localization in tissues and secretion of eotaxin by cells from normal endometrium and endometriosis. J Clin Endocrinol Metab 2000;85: 2604 – 2608. Hornung D, Bentzien F, Wallwiener D, Kiesel L, Taylor RN. Chemokine bioactivity of RANTES in endometriotic and normal endometrial stromal cells and peritoneal fluid. Mol Hum Reprod 2001;7:163 – 168. Iwabe T, Harada T, Tsudo T, Tanikawa M, Onohara Y, Terakawa N. Pathogenetic significance of increased levels of interleukin-8 in the peritoneal fluid of patients with endometriosis. Fertil Steril 1998;69:924– 930. Jolicoeur C, Boutouil M, Drouin R, Paradis I, Lemay A, Akoum A. Increased expression of monocyte chemotactic protein-1 in the endometrium of women with endometriosis. Am J Pathol 1998;152:125– 133. Kalu E, Sumar N, Giannopoulos T, Patel P, Croucher C, Sherriff E, Bansal A. Cytokine profiles in serum and peritoneal fluid from infertile women with and without endometriosis. J Obstet Gynaecol Res 2007;33:490– 495. Kharfi A, Akoum A. Correlation between decreased type-II interleukin-1 receptor and increased monocyte chemotactic protein-1 expression in the endometrium of women with endometriosis. Am J Reprod Immunol 2001;45:193 – 199. Khorram O, Taylor RN, Ryan IP, Schall TJ, Landers DV. Peritoneal fluid concentrations of the cytokine RANTES correlate with the severity of endometriosis. Am J Obstet Gynecol 1993;169:1545 – 1549. Kim JY, Kim H, Suh CS, Kim SH, Choi YM, Kim JG. The G(-2518)A polymorphism of monocyte chemotactic protein-1 (MCP-1) and its serum and peritoneal fluid levels in Korean women with endometriosis. Eur J Obstet Gynecol Reprod Biol 2008;139:106 – 110. Kim JY, Lee DH, Joo JK, Jin JO, Wang JW, Hong YS, Kwak JY, Lee KS. Effects of peritoneal fluid from endometriosis patients on interferon-gammainduced protein-10 (CXCL10) and interleukin-8 (CXCL8) released by neutrophils and CD4+ T cells. Am J Reprod Immunol 2009;62:128– 138. Kuroda K, Kitade M, Kikuchi I, Kumakiri J, Matsuoka S, Kuroda M, Takeda S. Peritoneal vascular density assessment using narrow-band imaging and vascular analysis software, and cytokine analysis in women with and without endometriosis. J Minim Invasive Gynecol 2010;17:21– 25. Kyama CM, Overbergh L, Mihalyi A, Meuleman C, Mwenda JM, Mathieu C, D’Hooghe TM. Endometrial and peritoneal expression of aromatase,


women with endometriosis. Le Groupe D’investigation en Gynećologie. Am J Obstet Gynecol 1996b;175:1620 – 1625. Akoum A, Kong J, Metz C, Beaumont MC. Spontaneous and stimulated secretion of monocyte chemotactic protein-1 and macrophage migration inhibitory factor by peritoneal macrophages in women with and without endometriosis. Fertil Steril 2002;77:989 – 994. Arici A, Tazuke SI, Attar E, Kliman HJ, Olive DL. Interleukin-8 concentration in peritoneal fluid of patients with endometriosis and modulation of interleukin-8 expression in human mesothelial cells. Mol Hum Reprod 1996;2:40 – 45. Arici A, Oral E, Attar E, Tazuke SI, Olive DL. Monocyte chemotactic protein-1 concentration in peritoneal fluid of women with endometriosis and its modulation of expression in mesothelial cells. Fertil Steril 1997;67:1065–1072. Barcz E, Ro´zewska ES, Kaminski P, Demkow U, Bobrowska K, Marianowski L. Angiogenic activity and IL-8 concentrations in peritoneal fluid and sera in endometriosis. Int J Gynaecol Obstet 2002;79:229 – 235. Barcz E, Milewski L, Dziunycz P, Kaminski P, Ploski R, Malejczyk J. Peritoneal cytokines and adhesion formation in endometriosis: an inverse association with vascular endothelial growth factor concentration. Fertil Steril 2012; 97:1380 – 1386. Bazot M, Lafont C, Rouzier R, Roseau G, Thomassin-Naggara I, Daraı¨ E. Diagnostic accuracy of physical examination, transvaginal sonography, rectal endoscopic sonography, and magnetic resonance imaging to diagnose deep infiltrating endometriosis. Fertil Steril 2009;92:1825– 1833. Bellelis P, Barbeiro DF, Rizzo LV, Baracat EC, Abraõ MS, Podgaec S. Transcriptional changes in the expression of chemokines related to natural killer and T-regulatory cells in patients with deep infiltrative endometriosis. Fertil Steril 2013;99:1987–1993. Bersinger NA, von Roten S, Wunder DM, Raio L, Dreher E, Mueller MD. PAPP-A and osteoprotegerin, together with interleukin-8 and RANTES, are elevated in the peritoneal fluid of women with endometriosis. Am J Obstet Gynecol 2006;195:103 –108. Borrelli GM, Carvalho KI, Kallas EG, Mechsner S, Baracat EC, Abraõ MS. Chemokines in the pathogenesis of endometriosis and infertility. J Reprod Immunol 2013;98:1–9. Bourlev V, Larsson A, Olovsson M. Elevated levels of fibroblast growth factor-2 in serum from women with endometriosis. Reproduction 2006;132:501–509. Braun DP, Gebel H, House R, Rana N, Dmowski NP. Spontaneous and induced synthesis of cytokines by peripheral blood monocytes in patients with endometriosis. Fertil Steril 1996;65:1125 – 1129. Calhaz-Jorge C, Costa AP, Santos MC, Palma-Carlos ML. Peritoneal fluid concentrations of interleukin-8 in patients with endometriosis depend on the severity of the disorder and are higher in the luteal phase. Hum Reprod 2003;18:593 – 597. Carmona F, Chapron C, Martıńez-Zamora MA´, Santulli P, Rabanal A, Martıńez-Florensa M, Lozano F, Balasch J. Ovarian endometrioma but not deep infiltrating endometriosis is associated with increased serum levels of interleukin-8 and interleukin-6. J Reprod Immunol 2012;95:80 – 86. Chand AL, Murray AS, Jones RL, Hannan NJ, Salamonsen LA, Rombauts L. Laser capture microdissection and cDNA array analysis of endometrium identify CCL16 and CCL21 as epithelial-derived inflammatory mediators associated with endometriosis. Reprod Biol Endocrinol 2007;17:18. Cho SH, Oh YJ, Nam A, Kim HY, Park JH, Kim JH, Park KH, Cho DJ, Lee BS. Evaluation of serum and urinary angiogenic factors in patients with endometriosis. Am J Reprod Immunol 2007;58:497– 504. Daraı¨ E, Detchev R, Hugol D, Quang NT. Serum and cyst fluid levels of interleukin (IL) -6, IL-8 and tumour necrosis factor-alpha in women with endometriomas and benign and malignant cystic ovarian tumours. Hum Reprod 2003;18:1681 – 1685. Drosdzol-Cop A, Skrzypulec-Plinta V, Stojko R. Serum and peritoneal fluid immunological markers in adolescent girls with chronic pelvic pain. Obstet Gynecol Surv 2012;67:374– 381.

Borrelli et al.


Palmer SS, Barnhart KT. Biomarkers in reproductive medicine: the promise, and can it be fulfilled? Fertil Steril 2013;99:954– 962. Pizzo A, Salmeri FM, Ardita FV, Sofo V, Tripepi M, Marsico S. Behaviour of cytokine levels in serum and peritoneal fluid of women with endometriosis. Gynecol Obstet Invest 2002;54:82 –87. Rana N, Braun DP, House R, Gebel H, Rotman C, Dmowski WP. Basal and stimulated secretion of cytokines by peritoneal macrophages in women with endometriosis. Fertil Steril 1996;65:925 –930. Rocha AL, Carrarelli P, Novembri R, de Pascalis F, Luisi S, Reis FM, Petraglia F. Activin A stimulates interleukin 8 and vascular endothelial growth factor release from cultured human endometrial stromal cells: possible implications for the pathogenesis of endometriosis. Reprod Sci 2012;19:832–838. Ryan IP, Tseng JF, Schriock ED, Khorram O, Landers DV, Taylor RN. Interleukin-8 concentrations are elevated in peritoneal fluid of women with endometriosis. Fertil Steril 1995;63:929 –932. Saba L, Guerriero S, Sulis R, Pilloni M, Ajossa S, Melis G, Mallarini G. Learning curve in the detection of ovarian and deep endometriosis by using magnetic resonance: comparison with surgical results. Eur J Radiol 2011; 79:237– 244. Sampson JA. Peritoneal endometriosis due to the menstrual dissemination of endometrial tissue into the peritoneal cavity. Am J Obstet Gynecol 1927; 14:422– 469. Seeber B, Sammel MD, Fan X, Gerton GL, Shaunik A, Chittams J, Barnhart KT. Panel of markers can accurately predict endometriosis in a subset of patients. Fertil Steril 2008;89:1073 – 1081. Shimoya K, Zhang Q, Temma-Asano K, Hayashi S, Kimura T, Murata Y. Fractalkine in the peritoneal fluid of women with endometriosis. Int J Gynaecol Obstet 2005;91:36 – 41. Skrzypczak J, Szczepanska M, Puk E, Kamieniczna M, Kurpisz M. Peritonela fluid cytokines and sICAM-1 in minimal endometriosis: search for discriminating factors between infertility and/or endometriosis. Eur J Obstet Gynecol Reprod Biol 2005;122:95 –103. Somigliana E, Vigano` P, Tirelli AS, Felicetta I, Torresani E, Vignali M, Di Blasio AM. Use of the concomitant serum dosage of CA 125, CA 19 – 9 and interleukin-6 to detect the presence of endometriosis. Results from a series of reproductive age women undergoing laparoscopic surgery for benign gynaecological conditions. Hum Reprod 2004;19:1871 – 1876. Socolov R, Butureanu S, Angioni S, Sindilar A, Boiculese L, Cozma L, Socolov D. The value of serological markers in the diagnosis and prognosis of endometriosis: a prospective case-control study. Eur J Obstet Gynecol Reprod Biol 2011;154:215– 217. Suzumori N, Katano K, Suzumori K. Peritoneal fluid concentrations of epithelial neutrophil-activating peptide-78 correlate with the severity of endometriosis. Fertil Steril 2004;81:305 – 308. Suzumori N, Zhao XX, Suzumori K. Increased granulocyte chemotactic protein-2 concentrations in peritoneal fluid of women with endometriosis. Acta Obstet Gynecol Scand 2005;84:1141 – 1144. Szamatowicz J, Laudanski P, Tomaszewska I, Szamatowicz M. Chemokine growth-regulated-alpha: a possible role in the pathogenesis of endometriosis. Gynecol Endocrinol 2002;16:137 – 141. Tao Y, Zhang Q, Huang W, Zhu H, Zhang D, Luo W. The peritoneal leptin, MCP-1 and TNF-alpha in the pathogenesis of endometriosis-associated infertility. Am J Reprod Immunol 2011;65:403 – 406. Ulukus M, Ulukus EC, Seval Y, Zheng W, Arici A. Expression of interleukin-8 receptors in endometriosis. Hum Reprod 2005;20:794– 801. Ulukus M, Ulukus EC, Tavmergen Goker EN, Tavmergen E, Zheng W, Arici A. Expression of interleukin-8 and monocyte chemotactic protein 1 in women with endometriosis. Fertil Steril 2009;91:687 – 693. Vodolazkaia A, El-Aalamat Y, Popovic D, Mihalyi A, Bossuyt X, Kyama CM, Fassbender A, Bokor A, Schols D, Huskens D et al. Evaluation of a panel of 28 biomarkers for the non-invasive diagnosis of endometriosis. Hum Reprod 2012;27:2698 – 2711.


cytokines, and adhesion factors in women with endometriosis. Fertil Steril 2008;89:301 – 310. Laudan´ski P, Szamatowicz J, Oniszczuk M. Profiling of peritoneal fluid of women with endometriosis by chemokine protein array. Adv Med Sci 2006;51:148 – 152. Li MQ, Li HP, Meng YH, Wang XQ, Zhu XY, Mei J, Li DJ. Chemokine CCL2 enhances survival and invasiveness of endometrial stromal cells in an autocrine manner by activating Akt and MAPK/Erk1/2 signal pathway. Fertil Steril 2012a;97:919 – 929. Li MQ, Luo XZ, Meng YH, Mei J, Zhu XY, Jin LP, Li DJ. CXCL8 Enhances proliferation and growth and reduces apoptosis in endometrial stromal cells in an autocrine manner via a CXCR1-triggered PTEN/AKT signal pathway. Hum Reprod 2012b;27:2107 –2116. Lira SA, Furtado GC. The biology of chemokines and their receptors. Immunol Res 2012;54:111 – 120. Luk J, Seval Y, Ulukus M, Ulukus EC, Arici A, Kayisli UA. Regulation of monocyte chemotactic protein-1 expression in human endometrial endothelial cells by sex steroids: a potential mechanism for leukocyte recruitment in endometriosis. Reprod Sci 2010;17:278 – 287. Maiorana A, Cicerone C, Niceta M, Alio L. Evaluation of serum CA 125 levels in patients with pelvic pain related to endometriosis. Int J Biol Markers 2007; 22:200– 202. Malhotra N, Karmakar D, Tripathi V, Luthra K, Kumar S. Correlation of angiogenic cytokines-leptin and IL-8 in stage, type and presentation of endometriosis. Gynvolecol Endocrinol 2012;28:224 – 227. Margari KM, Zafiropoulos A, Hatzidaki E, Giannakopoulou C, Arici A, Matalliotakis I. Peritoneal fluid concentrations of b-chemokines in endometriosis. Eur J Obstet Gynecol Reprod Biol 2013;169:103 – 107. May KE, Conduit-Hulbert SA, Villar J, Kirtley S, Kennedy SH, Becker CM. Peripheral biomarkers of endometriosis: a systematic review. Hum Reprod Update 2010;16:651 – 674. Mihalyi A, Gevaert O, Kyama CM, Simsa P, Pochet N, De Smet F, De Moor B, Meuleman C, Billen J, Blanckaert N et al. Non-invasive diagnosis of endometriosis based on a combined analysis of six plasma biomarkers. Hum Reprod 2010;25:654 – 664. Milewski Ł, Dziunycz P, Barcz E, Radomski D, Roszkowski PI, Korczak-Kowalska G, Kamin´ski P, Malejczyk J. Increased levels of human neutrophil peptides 1, 2, and 3 in peritoneal fluid of patients with endometriosis: association with neutrophils, T cells and IL-8. J Reprod Immunol 2011;91:64 – 70. Moher D, Liberati A, Tetzlaff J, Altman DG. PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097. Mueller MD, Mazzucchelli L, Buri C, Lebovic DI, Dreher E, Taylor RN. Epithelial neutrophil-activating peptide 78 concentrations are elevated in the peritoneal fluid of women with endometriosis. Fertil Steril 2003; 79:815– 820. Na YJ, Lee DH, Kim SC, Joo JK, Wang JW, Jin JO, Kwak JY, Lee KS. Effects of peritoneal fluid from endometriosis patients on the release of monocyte-specific chemokines by leukocytes. Arch Gynecol Obstet 2011; 283:1333–1341. Niloff JM, Knapp RC, Schaetzl E, Reynolds C, Bast RC Jr. CA125 Antigen levels in obstetric and gynecologic patients. Obstet Gynecol 1984;64:703–707. Ohata Y, Harada T, Miyakoda H, Taniguchi F, Iwabe T, Terakawa N. Serum interleukin-8 levels are elevated in patients with ovarian endometrioma. Fertil Steril 2008;90:994 –999. Oral E, Seli E, Bahtiyar MO, Olive DL, Arici A. Growth-regulated alpha expression in the peritoneal environment with endometriosis. Obstet Gynecol 1996;88:1050 – 1056. Othman Eel-D, Hornung D, Salem HT, Khalifa EA, El-Metwally TH, Al-Hendy A. Serum cytokines as biomarkers for nonsurgical prediction of endometriosis. Eur J Obstet Gynecol Reprod Biol 2008;137:240 – 246.

13

14 Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J. The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC Med Res Methodol 2003; 3:25.

Borrelli et al.

Xavier P, Beires J, Belo L, Rebelo I, Martinez-de-Oliveira J, Lunet N, Barros H. Are we employing the most effective CA 125 and CA 19 – 9 cut-off values to detect endometriosis? Eur J Obstet Gynecol Reprod Biol 2005; 123:254– 255.


Can platelet indices be new biomarkers for severe endometriosis?

Can vascular endothelial growth factor and microvessel density be used as prognostic biomarkers for colorectal cancer? A systematic review and meta-analysis.

Can melatonin be used as a marker for neonatal sepsis?

Circulating microRNAs as potential biomarkers for endometriosis.

Can milk cell or skim milk miRNAs be used as biomarkers for early pregnancy detection in cattle?

Can aging in place be cost effective? A systematic review.

Cryopreservation can be used as an anti-aging strategy.

Can readmission rates be used as an outcome indicator?

Biomarkers for Wilms Tumor: A Systematic Review.

Can clinical features be used to differentiate type 1 from type 2 diabetes? A systematic review of the literature.

Can the Neutrophil to Lymphocyte Ratio Be Used to Determine Gastric Cancer Treatment Outcomes? A Systematic Review and Meta-Analysis.

Can Radiofrequency Ablation Be Used as a Treatment Modality for the Management of Pulmonary Metastatic Ameloblastoma?

Low-dose irradiation can be used as a phytosanitary treatment for fresh table grapes.

Can preoperative thyroglobulin antibody levels be used as a marker for well differentiated thyroid cancer?

Shape of glucose curve can be used as a predictor for screening prediabetes in obese children.

Can Dermatoglyphics Be Used as a Marker for Predicting Future Malocclusions?

Can live weight be used as a proxy for enteric methane emissions from pasture-fed sheep?

Endometriosis and physical exercises: a systematic review.

Can white blood cell count be used as a predictor of atrial fibrillation following cardiac surgery? A short literature review.

Cytokines as Biomarkers of Pancreatic Ductal Adenocarcinoma: A Systematic Review.

Biomarkers as outcome predictors in subarachnoid hemorrhage--a systematic review.

Hyperthermia: How Can It Be Used?

Can Control Banding be Useful for the Safe Handling of Nanomaterials? A Systematic Review.

Two simple questions may be used as the first screening step for alcohol problems in primary care: systematic review.