Aberrant plasma MMP and TIMP dynamics in Schistosoma - Immune reconstitution inflammatory syndrome (IRIS)

Abstract

Background: Among the different faces of immune reconstitution inflammatory syndrome (IRIS) developing in HIV-patients, no clinical definition has been reported for Schistosomiasis-IRIS (Schisto-IRIS). Although Schisto-IRIS remains largely uninvestigated, matrix metalloproteinases (MMP) and tissue inhibitors of metalloproteinases (TIMP) have previously been associated with S. mansoni infection and tuberculosis-IRIS. Here, we aimed to investigate the relevance of these markers in Schisto-IRIS.

Methodology: Patients were diagnosed with IRIS related to S. mansoni within a cohort of patients with Schistosomiasis-HIV co-infection, using a clinical working definition of Schisto-IRIS. We compared 9 patients who developed Schisto-IRIS to 9 Schisto+HIV+ controls who did not, and 9 Schisto-HIV+ controls. Plasma levels of MMP-1, MMP-7, MMP-10, TIMP-1, TIMP-2, sCD14, intestinal fatty-acid binding protein, C-reactive protein, and 8 anti-nuclear antibodies (ANA) were analyzed prior to and during 3 months of ART.

Principal findings: Although no differences were observed for MMP-1 and -7, MMP-10 levels decreased significantly in Schisto+HIV+ controls during 3 months of ART (p = 0.005) while persisting in Schisto-IRIS patients at significantly higher levels compared to Schisto-HIV+ controls (p≤0.030). In contrast TIMP-1 levels only decreased significantly in Schisto-IRIS patients (p = 0.012), while TIMP-2 levels were lower compared to Schisto+HIV+ controls at 2 weeks (p = 0.007), 1 month (p = 0.005) and 3 months (p = 0.031) of ART. Five out of 8 ANAs studied decreased significantly in Schisto-IRIS patients after 1 month of ART(p≤0.039), whereas only 1 ANA decreased for Schisto+HIV+ controls (p = 0.027).

Conclusions/significance: In this study, we propose a working definition for the diagnosis of Schisto-IRIS in resource limited settings. We report persistent plasma levels of MMP-10, along with a more pronounced decrease in TIMP-1 and ANA-levels, and low levels of TIMP-2 during 3 months of ART. Corresponding to the clinical symptoms, these data suggest that Schisto-IRIS is marked by unbalanced MMP/TIMP dynamics which favor inflammation.

Fig 1. Flow chart of study population.

Chart shows the selection of patients for this case-control study, nested within a previous prospective study on HIV-schistosoma co-infection. A number of Schisto-IRIS patients (17/26) were excluded due to sample or data issues, resulting in 9 patients selected with clinical data and non-haemolitic samples available at each time point. Additional Schisto⁺HIV⁺ (n = 9) and Schisto^-HIV⁺ (n = 9) patients were randomly selected as controls, except for allowing comparable gender distribution between groups.

Fig 2. Patient follow-up.

Chart shows a time line of patient follow-up during the study. Before enrolment, adult HIV+ patients who were ART naïve and receiving voluntary counseling and testing (VCT) were screened for TB, malaria and hepatitis B virus (HBV), and soil transmitted helminths (STH). Patients then underwent clinical and lab-investigations at baseline (i.e. prior to starting ART and PZQ treatment), and at 2 weeks, 1 month and 3 months after starting ART. Additional screening for signs of S. mansoni infection by ultrasound were performed at baseline and month 3. ^aPatients initiated ART and PZQ treatment in close proximity to each other, prior the week 2 visit. Since patients had to be referred to the ART-treatment center, some patients received PZQ before starting ART and some shortly after. ^bKato Katz smear was performed at each visit by performing 2 smears on 1 stool sample for quantitative evaluation of S. manoni, Ascaris lumbricoides, Trichuris trichuria, hookworm, while S. haematobium eggs were evaluated by a urine filtration method.

Fig 3. Time analysis of CD4 counts.

Graph shows CD4 counts before and during ART in Schisto-IRIS patients (red circles), Schisto⁺HIV⁺ controls (blue triangles) and Schisto^-HIV⁺ controls (green squares). The significant change over for each patient group was calculated using the Friedman test (p-values shown in graphs), with Dunn’s multiple comparison post-hoc tests to indicate differences between time points (indicated by horizontal bars with an asterisk). B = baseline, W2 = week 2, M1 = month 1, and M3 = month 3.

Fig 4. Correlation of treatment interval with onset of IRIS.

Graph shows a correlation between ART-PZQ interval and onset of IRIS symptoms (#days since ART initiation), negative values indicate PZQ treatment prior to ART and dotted line represents 0 days interval.

Fig 5. Time analysis of MMPs and TIMPs.

Graphs show plasma levels of (A) MMP-1, (B) MMP-7, (C) MMP-10, (D) TIMP-1, and (E) TIMP-2 before and during ART in Schisto-IRIS patients (red circles), Schisto⁺HIV⁺ controls (blue triangles) and Schisto^-HIV⁺ controls (green squares). The significant change over time for each patient group was calculated using the Friedman test (p-values shown in graphs), with Dunn’s multiple comparison post-hoc tests to indicate differences between time points (indicated by horizontal bars with an asterisk). B = baseline, W2 = week 2, M1 = month 1, and M3 = month 3.

Fig 6. Time analysis of CRP, I-FABP and sCD14.

Graphs show plasma levels of (A) CRP, (B) I-FABP and (C) sCD14 before and during ART in Schisto-IRIS patients (red circles), Schisto⁺HIV⁺ controls (blue triangles) and Schisto^-HIV⁺ controls (green squares). The significant change over time for each patient group was calculated using the Friedman test (p-values shown in graphs), with Dunn’s multiple comparison post-hoc tests to indicate differences between time points when applicable (indicated by horizontal bars with an asterisk). One Schisto^-HIV⁺ patient did not have sufficient sample volume for CRP determination. B = baseline, W2 = week 2, M1 = month 1, and M3 = month 3.

Fig 7. Time analysis of 8 anti-nuclear antibodies.

Graphs show semi-quantitative values of plasma levels of ANAs, calculated as a ratio of positive cut-off control (index value). Index values of (A) U1 snRNP, (B) snRNP/Sm complex, (C) Sm, (D) SS-A, (E) SS-B, (F) Scl 70, (G) CenpB and (H) Jo-1 before and during 1 month of ART are shown in Schisto-IRIS patients (red circles) and Schisto⁺HIV⁺ controls (blue triangles). The significant change over time for each patient group was calculated using the Wilcoxon-signed rank test (p-values shown in graphs). B = baseline, M1 = month 1.