The role of CXCL10 as a biomarker for immunological response among patients with leprosy: a systematic literature review

Study design

This systematic review is based on Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 and the protocol was registered in PROSPERO (ID: CRD42023460378). Five international databases including PubMed, Scopus, Science Direct, and Wiley Online Library were used to conduct the search in October 2023. Keywords based on medical subject headings (MeSH) were used in the search of relevant scientific studies. The keywords used are: (CXCL10 OR “CXC Chemokine Ligand 10” OR IP#10 OR “Interferon#Inducible Protein 10” OR “Small Inducible Cytokine B10” OR “IFN#gamma#Inducible Protein#10” OR gammaIP#10 OR “Chemokine (C-X-C Motif) Ligand 10” OR “Chemokine (CXC Motif) Ligand 10” OR “Interferongamma#Inducible Protein of 10”) AND (Leprosy OR “Hansen disease” OR “Hansen’s disease” OR “Morbus Hansen” OR Leprae) AND (Immunological Marker OR Marker$ OR Biomarker$ OR Chemokine OR Immunology OR “Immunologic Marker” OR “Immunological Marker$”). We did not place any limitations on publishing type or date.

Inclusion and exclusion criteria

Using rayyan.ai (https://www.rayyan.ai/), we performed an initial title and abstract screening. We selected studies limited in humans with data on CXCL10 presence in leprosy patients’ serum up to 30^th of September 2023. We only selected papers in English. This systematic review didn’t have any exclusion criteria for the studies used.

Study selection

Results from the initial search of studies were uploaded to rayyan.ai, which is a free website that could be accessed freely. Rayyan.ai helped organizing these manuscripts based on their tags. Then, we search if there are any duplicates. After the duplicates were deleted, we separate these manuscripts into included and not included based on its title, abstract, and full text. Two reviewers, F. R. S. P and N. H., evaluated the full texts, abstracts, and titles of the studies based on the inclusion and exclusion criteria. A third party (L. M. D.) would settle any differences amongst all reviewers.

Quality assessment

We utilized the Newcastle-Ottawa Scale (NOS) to assess the quality of the selected studies. The NOS evaluates the studies based on selection, comparability, and outcome, using seven questions for cross-sectional studies and eight questions for case-control and cohort studies with maximum score of nine points. First, we included all cross-sectional, case-control, and cohort studies we find as NOS is used for these types of studies. Then, we read the manuscript thoroughly to judge and score these studies based on questions provided from the scale. Studies were categorized as high-quality if they scored six or more points. The critical judgments were made by two reviewers (F.R.S.P. and N.H.), and in cases of disagreements throughout the process, a third reviewer (L. M. D.) was consulted.

Risk bias assessment

In this study, the Risk of Bias Assessment tool for Non-randomized Studies (RoBANS) questionnaire, consisting of six items, was employed to assess the risk of bias in the selected studies. This questionnaire evaluated various aspects, including participant’s selection, consideration of confounding variables, exposure measurement, blinding of outcome evaluations, handling of incomplete outcome data, and selective outcome reporting, each rated as low, unclear, or high risk. The study outcomes were visually presented using graphs, including a traffic-light plot and a summary plot. Because there are many aspects that influence the results, it will make the results not objective and unreliable if you do not consider these aspects. With unreliable results, this research couldn’t be used as reference. Similar to the previous process, two reviewers (F.R.S.P. and N.H.) conducted this evaluation, and in cases of disagreements, a third reviewer (L. M. D.) was involved.

Data extraction and analysis

The data analysis used in this study involved predefined sheets that included information on study design, date and geographic location of testing, aim of study, population description and setting, inclusion and exclusion criteria, diagnostic methods and findings, tools for measurements, classification of leprosy, treatments, and outcomes. This analysis was conducted by three reviewers (N. H., E. J. H., and M. F. I), and a qualitative method was employed to assess the collected data.

Study selection

The study selection process of this study followed the PRISMA guidelines. Four databases (PubMed = 33; Scopus = 54; Wiley Online Library = 21; Science Direct = 7) collected a total of 115 studies retrieved from these databases. Reviewers removed seven duplicated records before the screening process. Based on the screening of titles and abstracts, we excluded 82 records. We assessed the eligibility of the remaining 26 reports in full text and excluded four studies due to inappropriate study design and two studies with incomplete outcome data. The final analysis included twenty articles in the qualitative synthesis of this review shown in Fig. 1.

Quality and risk of bias

We utilized The Newcastle-Ottawa Scale (NOS) and The RoBANS tool in the assessment process to evaluate the quality and risk of bias in the selected studies. The results of the quality assessment can be found in Table 1. The included studies scored between six and eight points in the quality assessment, and as a result, they were considered high-quality studies. The results of the risk of bias assessment using the RoBANS tool are depicted in Figs. 2 and 3. While the majority of the items received “low” scores, certain items related to selection of participants, confounding variables, measurements of exposures, and incomplete results data were rated as “unclear”.

Study characteristics

There are 20 articles included in this study. Out of twenty articles, fifteen were cross-sectional studies, two were case control studies, and three were cohort studies. Samples from these included articles ranged from 8 to 715, with total of 2.525 samples. Research was conducted across multiple nations. Table 1 provides a detailed overview of the study’s characteristics.

Study results

CXCL10 as diagnostic properties

Constant spread of M. leprae has led leprosy researchers to concentrate more on finding biomarkers to aid in diagnosis. As a result, there is an increasing demand for novel, sensitive diagnostic instruments based on particular biomarkers that, ideally, translate easily into tests that are suitable for the field (Van Hooij et al., 2016). Expression of CXCL10 were found to be higher in leprosy patients than control or household contact (Ferreira et al., 2021; Queiroz et al., 2021; Cunha et al., 2023). Queiroz et al. (2021) stated greater levels of CXCL10 levels in multibacillary (MB) group than the household contact (HHC). It was found that in comparison to paucibacillary contacts, M. leprae-induced IFN-γ production is greater in multibacillary contacts. People living in endemic environments are more vulnerable to leprosy because they are in close proximity to others who have the disease, particularly those who have the multibacillary (MB) form (Prakoeswa et al., 2021). Studies have shown that leprosy tends to affect those living in close proximity more often than those living apart, corroborating the higher risk of contracting the disease for individuals who have spent more than four years near leprosy patients (Prakoeswa et al., 2020, 2021). Additionally, M. leprae-induced IFN-γ production is stronger in multibacillary contacts than in paucibacillary contacts (Cunha et al., 2023). The more severe the leprosy, the higher the CXCL10 levels (Freitas et al., 2015). CXCL10 levels increased in all types of leprosy, but they were higher in patients with a reaction where usually happened in multibacillary leprosy (Sharma et al., 2015).

Cunha et al. (2023) found high levels of CXCXL10 in household contact (HHC) of PB and MB, but higher HHC of PB patients than HHC of MB. Paucibacillary (PB) patients show antigen-specific T cell responses (Th1 and Th17), where Th1 secrete IFN-γ (Geluk, 2013; Ferreira et al., 2021). High level of CXCL10 was as response to IFN-γ, where leukocytes, monocytes, activated neutrophils, epithelial cells, endothelial cells, stromal cells, and keratinocytes, release CXCL10 (Geluk, 2013; Corstjens et al., 2019; Bobosha et al., 2014). According to histological results, pure neural leprosy (PNL) nerves showed CXCL10 expression. These nerves showed moderate to strong CXCL10-immunoreactivity, which may indicate that this chemokine is involved in initiating the intraneural response. But, CXCL10 were not linked to neural fibrosis (Medeiros et al., 2015). Higher levels of CXCL10 were also found to be higher in PB patients than EC and could differentiate PB from EC in samples containing recombinant protein (ML0276 + LID-1 WBA). But, CXCL10 alone did not distinguish PB from EC as well as IFN-γ (Hungria et al., 2017). Study by van Hooij et al. (2018) in low and high endemic population showed that CXCL10 was the most significant marker to identify BT/TT and LL/BL patients. In low and high endemic countries, CXCL10 could be used to differentiate between LL/BL patients and HHC and EC. Meanwhile in BT/TT patients, CXCL10 could be used to differentiate from endemic controls (van Hooij et al., 2018).

Combination of UCP-LFA were useful to detect various analytes, such as proteins, polysaccharides, antigens, drugs, nucleic acid, and antibodies. This instrument was useful and didn’t require expensive equipment to operate, making it appropriate for use in leprosy-endemic areas where access to advanced laboratory instruments is frequently limited (Bobosha et al., 2014). Multi-biomarker test (MBT) using UCP-LFA strip could accurately differentiate leprosy patients from endemic control in endemic and non-endemic settings (van Hooij et al., 2021). Similar test findings were achieved with the six-marker MBT strip as they were with the individual UCP-LFA strips for every biomarker individually in the past, which suggest this tool feasibility as a diagnostic tool (van Hooij et al., 2019). In both leprosy-endemic and non-endemic settings, MBT can be helpful as an additional diagnosis for individuals exhibiting signs that could indicate leprosy. The assay time can be greatly reduced and a sample-to-result can be obtained on the same day with the usage of UCP-LFA (Bobosha et al., 2014). In addition, the strips can be kept in patients’ file and revisited in another time because of the assay components’ chemical stability, which offered easier way to monitor patient’s profile. With this potential, leprosy detection can be done easily and at low cost, yet it gives reliable results. Thus, the scope of early leprosy diagnosis can be broader, enabling earlier initiation of leprosy treatment (Corstjens et al., 2019; Bobosha et al., 2014).

Using UCP-LFA, CXCL10 was higher in response to M. leprae antigen and higher CXCL10/IL-10 ratio in leprosy patient compared to endemic control (EC) (Bobosha et al., 2014). It was also found, CXCL10 could differentiate MB patients from EC (Van Hooij et al., 2016; van Hooij et al., 2019) by UCP-LFA sample. van Hooij et al. (2019) validated the use of CXCL10 as a biomarker for MB leprosy, which is consistent with earlier research. However, PB patients and HC could not be differentiated given that the indicators displayed comparable responses for CXCL10, particularly in highly endemic areas (Van Hooij et al., 2016; van Hooij et al., 2018). Patients with PB and HC also exhibit comparable immune responses and frequently have undetectable M. leprae bacilli counts (Van Hooij et al., 2016). The use of CXCL10 as a biomarker to diagnose leprosy is more effective and sensitive with the combination of other biomarkers like PGL-1 antibodies, IL-10, CRP, and CCL4 (van Hooij et al., 2018, 2021, 2019; Van Hooij et al., 2016). Combining anti-PGL-I IgM, IL-10, and CXCL10 was suggestive for leprosy classification, allowing MB and PB patients to be distinguished from one another. A plasma biomarker profile that included αPGL-I IgM, CXCL10, S100A12, ApoA1, and CRP accurately diagnosed leprosy patients of any kind with high sensitivity (86%) and specificity (90%) in the UCP-LFAs. This shows that this signature can be used to diagnose leprosy because it can identify patients with both high and low bacillary loads (Van Hooij et al., 2016).

To accommodate field-friendly diagnostic tools for leprosy, a study discovered that CXCL10 was detected in fingerstick blood (FSB) sample and showed to be higher in leprosy patients than in control. This suggest the use of FSB sample offered a simpler way of collecting samples and also could be beneficial in diagnosing leprosy (Corstjens et al., 2019). This was supported by van Hooij et al. (2019) where CXCL10 is an important marker in plasma for MB patients even in FSB samples. Since HHCs of patients with MB leprosy have the highest chance of contracting an M. leprae infection, they are good candidates for preventative medication delivery in multiple studies (Barth-Jaeggi et al., 2016; Mieras et al., 2018; Ortuno-Gutierrez et al., 2019; Richardus et al., 2021; Tiwari et al., 2020). Aimed at slowing the spread of the illness and preventing its progression, the MBT may help identify M. leprae-infected people who qualify for post-exposure prophylaxis (PEP), enabling a more effective and focused drug delivery strategy as diagnostic tools in early detection of leprosy (van Hooij et al., 2021). To minimize errors in the initial diagnosis of leprosy, one of important steps is the utilization of appropriate tools and parameters. Additionally, confounding variables that may influence the results, such as age and exposure to contacts, should be taken into consideration. It is advisable to do the test not only once but several times to get more reliable results. Furthermore, the results should be evaluated based on the patient’s clinical condition.

The detection of CXCL10 in leprosy patient can also be done using rt-PCR, ELISA, whole blood assay, and histopathological evaluation using immunostaining. With rt-PCR, mRNA expression levels of CXCL10 could be detected. It was shown that mRNA expression levels of CXCL10 from skin biopsies of borderline leprosy cases with or without T1R was consistently high. There was significant difference CXCL10 expression between borderline leprosy (BB) and BB-T1R, where BB-T1R have higher expression than BB (Sharma et al., 2015). Khadge et al. (2015) used ELISA to identify CXCL10 expression in leprosy patients. Levels of CXCL10 was found to be increased in all types of leprosy, but higher in patient with reaction. This demonstrates the use of rt-PCR and ELISA in diagnosing leprosy reaction (Sharma et al., 2015; Khadge et al., 2015). Bobosha et al. (2014) assess combined cytokine profiles using ELISA in responses to M. leprae antigen and found that CXCL10 levels were higher in infected patients than those in EC. Van Hooij et al. (2016) also identified CXCL10 profiles in M. leprae infection and type of leprosy using ELISA and showed significant levels of CXCL10 in leprosy patients. CXCL10 could differentiate MB patients from endemic control or EC. Using whole blood assay, higher CXCL10 levels was also found in PB sample compared to samples from EC, but couldn’t distinguished between PB and HHC (Hungria et al., 2017). For immunostaining, biopsy samples showed CXCL10 had more pronounced staining in patient with T1R than in those without reaction (Khadge et al., 2015). Compared to non-reactional BB leprosy, T1R biopsies showed more pronounced CXCL10 staining (Sharma et al., 2015). In addition, mass spectrometry (MS)-based approaches could be considered to be used in CXCL10 levels detection as it also had the ability to discover disease biomarkers and revealed non-biased profile of protein/metabolites. It has been used in discovering COVID-19 biomarkers (Zhong, Zhu & Cai, 2021).

CXCL10 in leprosy reaction

Type 1 reaction (T1R) in leprosy patients are a delayed type hypersensitivity reaction, caused by exacerbations of immune system (Chaitanya et al., 2013). This reaction usually occurred in borderline-borderline leprosy (BB), borderline-tuberculoid leprosy (BT), and borderline-lepromatous leprosy (BL) (Khadge et al., 2015). These inflammatory reactions are treated with immunomodulatory medications at high doses, which frequently increase morbidity. They can happen before, during, and after multidrug therapy (MDT) (Das et al., 2023). Erythematous and/or oedematous skin lesions with or without ulceration are among the clinical characteristics of T1R, as is oedema of the hands, face, and feet. Even in rare cases, systemic symptoms might result from acute inflammation of the peripheral nerves, which can compromise nerve function (NFI) (Khadge et al., 2015; Chaitanya et al., 2013). Dermal oedema, oedema, and lymphocytes within the granuloma are the histological characteristics indicative of T1R (Sharma et al., 2015).

Patients with T1R have higher mRNA expressions of CXCL10 than patients without reaction (Chaitanya et al., 2013). Several studies showed mRNA levels CXCL10 were consistently higher in BL and BT patients during T1R than before T1R, BB patients with T1R compared to those without reaction, as well as in T1R compared to T2R (Ferreira et al., 2021; Scollard et al., 2011; Sharma et al., 2015; Das et al., 2023). Khadge et al. (2015) supported these findings by conducting a study in four different countries, with levels of CXCL10 peaking at the start of T1R as a result. This means that CXCL10 levels increased in all types of leprosy, but they were higher in patients with a reaction (Sharma et al., 2015). Biopsy samples also showed CXCL10 had more pronounced staining in patient with T1R than in those without reaction where immune-expression of CXCL10 was localized in cytoplasm of macrophages and epithelioid cells (Sharma et al., 2015; Chaitanya et al., 2013). The higher levels CXCL10 was caused by vital role of proinflammatory cytokines regulations in leprosy reaction, as CXCL10 is expressed in a variety of Th1-type inflammatory disorders due to its’ crucial role in drawing activated T cells to tissue inflammation sites (Scollard et al., 2011; Teles et al., 2019; Geluk et al., 2014). Expression of CXCL10 is also induced by IFN-γ, and IFN-γ expression was increased in reactional skin, along with macrophage activation (Scollard et al., 2011; Chaitanya et al., 2013; Geluk et al., 2010). However, there was no correlation of CXCL10 levels with the severity of T1R and CXCL10 couldn’t predict the occurence of T1R (Scollard et al., 2011).

Clinical implication

The WHO recommended two standardized multidrug therapy (MDT) regimens for leprosy in 1982. Regardless of the patient’s categorization, the uniform therapy for leprosy (U-MDT) is taking clofazimine and dapsone daily for six months, along with monthly rifampicin (Penna et al., 2017). A decrease in CXCL10 production was found in PB, MB, and HHC after treatment with MDT. After 12 doses of MDT, production of CXCL10 in skin lesion was reduced, along with serum levels of CXCL10 (Ferreira et al., 2021; Queiroz et al., 2021). Khadge et al. (2015) also discovered similar findings, where serum levels of CXCL10 decreased in the beginning of anti-reactional treatment with corticosteroid. Consequently, the study’s findings imply that CXCL10 may be utilized to assess the effectiveness of MDT in patients (Ferreira et al., 2021; Khadge et al., 2015). It is commonly known that phagocytosis, growth factor and cytokine secretion, and matrix remodeling are just a few of the overlapping processes that make up tissue repair. In this case, tissue repair may be linked to the decrease in proinflammatory mediators in skin cells, such as IFN-γ and CXCL10 (Ferreira et al., 2021; Novak & Koh, 2013). With the evidence that CXCL10 can assess the effectiveness of MDT, it is also possible to simultaneously monitor drug compliance of leprosy patients. Regular intake of 12 doses of MDT shows a decrease in CXCL10 levels, indicating the potential for CXCL10 to be used in monitoring patient compliance. Further research is needed explore this potential (Ferreira et al., 2021; Queiroz et al., 2021).

Besides assessing MDT effectiveness’ in leprosy patients, CXCL10 could be used as one of diagnostic biomarker for leprosy as described before. There are several studies focusing on using CXCL10 as one of biomarkers use in diagnosing leprosy. These studies also investigated field-friendly diagnostic tool, such as UCP-LFA, to help bigger scale of leprosy diagnosis. All of included studies was conducted in humans. Although there are some studies about CXCL10 as diagnostic biomarkers of leprosy and as MDT effectiveness’ parameter, further studies to investigate its’ properties in larger populations (endemic and non-endemic settings) with consideration should be conducted. Future research focusing on the development of these tools and CXCL10 as a biomarker in diagnosing different types of leprosy is needed due to the varying outcomes of the included studies, making it difficult to draw certain conclusions.

Limitation of study

Due to the heterogeneity of data from the included studies and differences in assessed parameters, this systematic review couldn’t proceed to a meta-analysis study. Most of the included studies did not address confounding factors in their investigations and did not explain if there was incomplete outcome data. Furthermore, there was no ascertainment of exposure stated in the included case-control and cohort studies. Additionally, some of the included studies did not specify any inclusion or exclusion criteria. Each of the studies included has its own limitations. A relatively modest sample size, primarily resulting from insufficient data or a shortage of blood samples, as well as the loss to follow-up of certain patients, represented one of the study’s constraints. The use of non-probabilistic sampling methods, the absence of corrections for confounding variables, and the lack of pre and post-test data were identified as additional limitations in these studies. The involvement of various examiners in data collection also introduced potential bias, contributing to the studies’ limitations. To establish the role of CXCL10 as a biomarker for diagnosing leprosy in infected individuals and controls, it is imperative to conduct larger-scale studies that carefully account for confounding variables.