Article Text
PDF
Abstract
Objective To determine if the serum levels of neutrophil extracellular trap (NET) remnants (Elastase-DNA and HMGB1-DNA complexes) at the time of a lupus nephritis (LN) flare predict renal outcomes in the following 24 months.
Methods This was a retrospective study performed in prospectively followed cohorts. The study included two cohorts: an exploratory cohort to assess the association between NET remnant levels and the presence of active LN, and a separate LN cohort to determine the utility of NET remnants to predict renal outcomes over the subsequent 24 months.
Results Ninety-two individuals were included in the exploratory cohort (49 active systemic lupus erythematosus (SLE), 23 inactive SLE and 20 healthy controls (HC)). NET remnants were significantly higher in patients with SLE patients compared with HC (p<0.0001 for both complexes) and those with active LN (36%) had significantly higher levels of NET remnants compared with active SLE without LN (Elastase-DNA: p=0.03; HMGB1-DNA: p=0.02). The LN cohort included 109 active LN patients. Patients with proliferative LN had significantly higher levels of NET remnants than non-proliferative LN (Elastase-DNA: p<0.0001; HMGB1-DNA: p=0.0003). Patients with higher baseline levels of NET remnants had higher odds of not achieving complete remission (Elastase-DNA: OR 2.34, p=0.007; HMGB1-DNA: OR 2.61, p=0.009) and of progressing to severe renal impairment (Elastase-DNA: OR 2.84, p=0.006; HMGB1-DNA: OR 2.04, p=0.02) at 24 months after the flare.
Conclusions Elastase-DNA and HMGB1-DNA complexes predict renal outcomes, suggesting they could be used to identify patients requiring more aggressive therapy at flare onset.
- lupus nephritis
- lupus erythematosus, systemic
- outcome assessment, health care
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Lupus nephritis (LN) is characterised by significant variability in the response to treatment, to date, there are no reliable predictors to differentiate responders from non-responders.
WHAT THIS STUDY ADDS
This study showed that baseline Elastase-DNA and HMGB1-DNA complexes may serve as predictors of adverse renal outcomes 24 months after the LN flare.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
The levels of neutrophil extracellular trap remnants measured at the time of a renal flare may help identify patients at high risk of worse renal outcomes.
Introduction
Lupus nephritis (LN) can affect up to 65% of patients with systemic lupus erythematosus (SLE)1 and is characterised by significant variability in the response to treatment.2 3 Given that delays in treatment response are associated with an increased risk of renal damage,3 4 the ability to predict treatment response at the onset of the LN flare could help tailor treatment, potentially improving renal outcomes.
Compared with healthy control (HC) neutrophils, neutrophils from SLE patients have an increased capacity to undergo NETosis5: a specific form of neutrophil cell death that culminates in the formation of neutrophil extracellular traps (NETs), which are comprised of fibrils of chromatin coated with cytosolic and nuclear proteins.6 Increases in NET remnants7 or surrogates of NET formation (free DNA and neutrophil-specific proteins)8 9 in serum have been correlated with the presence of LN. NETosing neutrophils have also been found in proliferative LN renal biopsies,10 linking NETs to LN.
SLE neutrophils characteristically release greater amounts of High Mobility Group Box 1 (HMGB1) protein in comparison to HC neutrophils.5 Once in the extracellular space, HMGB1 acts as an alarmin, activating multiple inflammatory cells that are associated with LN pathogenesis.11 Peripheral blood and urine levels of HMGB1 have been shown to correlate with active LN.12 13 Furthermore, we have shown that NETs are a source of extracellular HMGB1 and that the amount of HMGB1 in SLE NETs correlates with the severity of LN.14 Elastase, a neutrophil-specific protein responsible for most of the proteolytic activity in NETs,15 has also been demonstrated to damage renal endothelial cells leading to renal dysfunction and proteinuria.16
Taken together, these findings suggest that NET-specific components, including Elastase and HMGB1, could be responsible for renal inflammation and damage, which may in turn dictate the severity of the disease and the response to therapy. The aim of this study was to determine if the amount of NET remnants, specifically Elastase-DNA and HMGB1-DNA complexes, in serum at the time of an LN flare predicts renal outcomes over the following 24 months.
Methods
Patients
All patients met the revised 1997 American College of Rheumatology classification criteria for SLE17 or had three criteria and a supportive biopsy (skin or kidney) and were part of the University of Toronto Lupus Clinic cohort, where patients are followed at regular intervals (2–6 months). On each assessment clinical and laboratory information is collected according to a standard protocol.
The study included two cohorts. An exploratory cohort to assess the association between the NET remnants (Elastase-DNA and HMGB1-DNA) and the presence of active LN, and an independent LN cohort, to determine the utility of NET remnants to predict renal outcomes over the subsequent 24 months. The exploratory cohort contained (1) active SLE patients with a recent flare (≤1 month) defined as a change in clinical SLEDAI-2K ≥1 requiring modification of therapy; (2) quiescent SLE patients defined as a clinical SLEDAI-2K of 0 for at least a year, with a prednisone dose of ≤10 mg and (3) HC. Patients in this cohort were recruited from December 2018 onwards. The LN cohort consisted of patients with active LN. The serum samples from the subjects included in this cohort came from two biobanks, the University of Toronto Lupus biobank and the LuNNET biobank.18 19 Inclusion criteria included: (1) LN flare defined as a urinary protein excretion of ≥500 mg/day in a 24-hour urine collection requiring a modification in therapy for LN; (2) baseline estimated glomerular filtration rate (eGFR) ≥30 mL/min (3 months prior to the flare); (3) follow-up of at least 24 months and (4) serum stored ±3 months from the beginning of the LN flare. Patients were recruited from April 2006 to May 2016. In both cohorts, renal biopsies were performed when clinically indicated.
Outcomes
The primary outcome was renal response to therapy at 12 and 24 months after the LN flare, with patients being classified into one of three groups, as defined in the literature.20 Complete response (CR) was defined as a reduction in urinary protein excretion to <500 mg/day based on a 24-hour urine collection with serum creatinine within 15% of previous baseline. Partial response (PR) was defined as a >50% reduction in proteinuria to non-nephrotic levels with serum creatinine within 25% of previous baseline, and no response (NR) as a failure to achieve CR or PR.21 Secondary outcomes included decline ≥30% in the eGFR; progression to severe renal impairment defined as an eGFR <30 mL/min22 and percentage decline in the eGFR over the 24 months after the flare.
NET remnant ELISAs
Circulating NET remnants were measured by ELISAs developed in-house for Elastase-DNA and HMGB1-DNA complexes.7 23–25 To establish these assays, neutrophils from a HC were isolated from heparinised peripheral blood and stimulated with PMA to undergo NETosis, as previously described.5 The NETs were then dissociated from the cells by vortexing, cellular debris removed by centrifugation and the NET-enriched supernatant collected, aliquoted and stored at −80o C. Generation of NETs was confirmed by immunofluorescence microscopy.26 Using serial dilutions of the NET-rich supernatant, the assays for NET remnants were optimised by testing different concentrations of each antibody. Following further optimisation to minimise background, the final assay was as follows: ELISA plates (3455; Thermo Scientific) were coated overnight at 4°C with 3 µg/mL rabbit anti-human neutrophil elastase antibody (481001; Sigma Aldrich) or 3 µg/mL anti-human HMGB1 antibody (SAB4501401; Sigma Aldrich), in coating buffer (0.2 M sodium carbonate/bicarbonate, pH 9.4). After blocking with 2% Bovine Serum Albumin in Phosphate Buffered Saline (PBS) at room temperature (RT) for 6 hours, the plates were incubated with 100 µl of patient serum diluted 1:10 in dilution buffer (blocking buffer plus 0.05% Tween) overnight at 4°C. The plates were then washed and incubated at RT for 1.5 hours with 1 µg/mL mouse IgG anti-human DNA antibody (MAB030; EMD Millipore) in dilution buffer. After further washing, the plates were incubated with HRP-conjugated polyclonal goat anti-mouse IgG (405306; Biolegend) diluted 1:5000 in a 1:5 dilution of dilution buffer in PBS for 1 hour. The plates were then washed and TMB (3,3',5,5'-Tetramethylbenzidine) substrate (DY999; R&D Systems) was added. Following addition of stop reagent (DY994; R&D Systems), the absorbance was read at 450 nm.
To allow standardisation of measurement between plates, serial dilutions of a serum from a patient with active SLE with high concentrations of NET remnants were included on each plate and the data expressed as units relative to this standard control. Representative results for the standard curves are shown in online supplemental figure S1. All samples were run in duplicate, averaged, and their concentration was computed from a ln–ln plot of the NET standard curve, with adjustment for the dilution factor. Any samples with raw absorbance values that were over the upper limit of the standard curve using the optimal dilution (1:10) were rerun at higher dilutions (1:30) and those that were below the standard curve were given the lowest standard curve value for ensuing calculations. The intra-assay coefficient of variation (CV) was 3.95% and 4% for Elastase-DNA and HMGB1-DNA, respectively, and the interassay (same sample measured on different days) CV was 6.1% and 12.2% for Elastase-DNA and HMGB1-DNA, respectively.
Supplemental material
Statistical analysis
Descriptive statistics were generated for patients’ baseline characteristics for the two cohorts, with baseline categorical variables being presented as counts and percentages. Continuous biomarker variables are presented as median and IQR. Kruskal-Wallis or Mann-Whitney tests were used to assess differences in the amount of NET remnants between groups, and depending on the Gaussian distribution of the variables, either Spearman or Pearson tests were used to determine correlations between the levels of NET remnants and the clinical and laboratory variables.
For the binary outcomes, including response to treatment, at least 30% decline in eGFR, and progression to severe kidney impairment (eGFR <30 mL/min), multivariable logistic regression analysis was used. For the analysis of response to treatment, non-CR (PR and NR) was pooled together and results reported as OR with 95% CIs. The models were adjusted for possible confounding variables.2 27–29 Age (p=0.7), sex (p=0.6), cumulative prednisone dose (in the prior 3 months to the sample collection) (p=0.84) and immunosuppressive medication used for induction of remission (p=0.98) were not included based on their high p values in the Likelihood ratio test suggesting no impact on our outcomes.
Multivariable linear regression analysis was used to assess the relationship between NET remnant levels at the time of the flare and percentage decline in eGFR after 24 months from the renal flare. Restrictive cubic splines were used to assess linearity.
To assess overfitting of the models, the bootstrap validation method with 100 iterations was used. DFBETAS (difference in beta values) and DFFITS (difference in fit(s)) were used to identify influential observations. Collinearity was assessed using variance inflation factor.30
All statistical analyses were performed using R software V.4.1.2 and all p values were two sided, with values <0.05 considered to be statistically significant. GraphPad Prism V.9 was used to generate graphs.
Patient and public involvement
Patients or the public were not involved in the design, or conduct or reporting or dissemination plans of our research.
Results
NET remnants are preferentially elevated in patients with proliferative LN and correlate with global disease activity
Ninety-two subjects were included in the exploratory cohort, 49 SLE patients with a recent flare (≤1 month), 23 inactive SLE patients and 20 HC. Table 1 shows the baseline characteristics for this cohort.
Baseline demographic and clinical characteristics of the patients in the exploratory and LN cohorts
Elastase-DNA and HMGB1-DNA complexes were significantly higher in SLE patients compared with HC (p<0.0001 for both complexes) and tended to be higher in active SLE patients compared with inactive patients (p=0.11 and p=0.34 for Elastase-DNA and HMGB1-DNA, respectively) (figure 1A). Eighteen (36.7%) of the patients had active LN, of whom 13 had a kidney biopsy at the time of the flare (seven proliferative and six non-proliferative classes). Patients with active LN had significantly higher levels of both complexes compared with active SLE patients without LN (p=0.03 and p=0.02 for Elastase-DNA and HMGB1-DNA, respectively). Furthermore, patients with proliferative LN had higher levels of NET remnants compared with non-proliferative LN (p=0.008 and p=0.001 for Elastase-DNA and HMGB1-DNA, respectively) (figure 1B,C). There was a strong positive correlation between the levels of Elastase-DNA and HMGB1-DNA complexes (r=0.84, p<0.0001, figure 1D) and both complexes had weak to moderate correlations with the clinical and laboratory features of the SLEDAI-2K (online supplemental table 1).
NET remnant levels in patients with SLE and HC. Comparison of NET remnant levels between (A) active SLE (n=49), inactive SLE (n=23) and HC (n=20); (B) active SLE patients stratified for the presence (n=18) or absence of active LN (n=31); (C) proliferative (n=7) and non-proliferative LN (n=6) in the exploratory cohort. (D) Correlation between the levels of Elastase-DNA and HMGB1-DNA in the exploratory cohort. (E) Comparison of NET remnant levels between proliferative (n=62) and non-proliferative LN (n=23) in the LN cohort. For all graphs, the symbols represent the determination from a single individual, columns the median and the bars IQR. The p values for all comparisons between groups are shown and were calculated as outlined in the Methods. HC, healthy control; LN, lupus nephritis; NET, neutrophil extracellular trap.
Active SLE patients had higher circulating neutrophil counts compared with inactive SLE patients (p=0.01), but there was no difference between patients with active LN and patients with active SLE without LN (p=0.12). Furthermore, there was no correlation between the levels of NET remnants and the neutrophil count, as seen in online supplemental table 1.
Replication of the results from the exploratory cohort in the LN cohort
A separate cohort of 109 SLE patients with active LN was used to validate the findings of the exploratory cohort and determine the utility of NET remnants to predict renal outcomes over the subsequent 24 months. The median (IQR) age of the patients was 29 (23–41) years and 92 (84.4%) were women. Forty-three (39.4%) were Caucasian, 25 (22.9%) Afro-Caribbean and 26 (23.9.%) Asian. The median (IQR) time between the renal flare and the serum sample collection was 1 (0–2) months (table 1). Forty-one (37.6%) and 56 (51.4%) of the patients achieved a CR at 12 and 24 months, respectively (all patients achieving a CR also had at least a 50% decline in proteinuria). Twenty-three (21.1%) patients had at least a 30% decline in eGFR, 12 (11%) patients had an eGFR <30 mL/min and the median (IQR) eGFR percentage decline was 4 (0–22) at 24 months after the renal flare (online supplemental table 2).
As observed in the exploratory cohort, the levels of Elastase-DNA and HMGB1-DNA complexes had weak to moderate correlations with clinical and laboratory features of the SLEDAI-2K and did not correlate with the circulating neutrophil counts. Eighty-five (78.0%) of the LN patients had a kidney biopsy at the time of the LN flare, and as seen in the exploratory cohort, patients with proliferative LN had higher serum levels of NET remnants compared with non-proliferative LN patients (figure 1E) at the time of flare. Elastase-DNA complexes positively correlated with the activity index in the kidney biopsy (r=0.35, p=0.001) but did not correlate with other features of active LN, including serum creatinine or proteinuria at the time of the renal flare (online supplemental table 1).
The levels of NET remnants at the time of the LN flare predicted the response to treatment
The levels of both complexes were higher in patients who failed to achieve a CR at 12 and 24 months after the LN flare (online supplemental table 1). Furthermore, on multivariable regression analysis, after adjusting for possible confounding variables, patients with higher levels of NET remnants had higher odds of not achieving a CR at 12 (Elastase-DNA: OR 1.97 (1.05–3.74), p=0.03; HMGB1-DNA: OR 2.51 (1.17–5.73), p=0.01) and 24 months (Elastase-DNA: OR 2.34 (1.24–4.38), p=0.007; HMGB1-DNA: OR 2.61 (1.19–5.96), p=0.009) after the flare (table 2).
Predictors of renal outcomes in the following 24 months from the LN flare (LN cohort, N=109)
The baseline levels of NET remnants continued to predict non-response to treatment when pooling together CR and PR and when directly comparing CR and NR (removing PR), as seen in online supplemental table 3.
Of the conventional biomarkers, only serum creatinine at the time of flare had a significant impact on the response to treatment, with higher values being associated with failure to achieve a CR (OR: 1.33 (1.02 to 1.73), p=0.03) at 24 months after the flare (table 2). However, when patients with an eGFR <60 mL/min at the time of the LN flare were excluded from this analysis (N=89), the effect of serum creatinine on response to treatment was lost (OR: 1.77 (0.79 to 3.99), p=0.16), whereas the ability of NET remnants to predict a CR to treatment at 24 months following flare was retained (Elastase-DNA: OR 1.81 (1.03 to 3.16), p=0.03; HMGB1-DNA: OR 2.37 (1.08 to 5.20), p=0.03) (table 3).
Predictors of renal outcomes in the following 24 months from the LN flare. Cohort excluding patients with an eGFR<60 at the time of the LN flare (N=89)
NET remnant levels at the time of the flare-predicted decline in eGFR at 24 months after the renal flare
The levels of both NET remnants also tended to be higher in patients who had ≥30% decline in the eGFR and progressed to severe kidney impairment (eGFR <30 mL/min) at 12 and 24 months from the flare (online supplemental table 2). Patients with higher levels of these complexes had higher odds of having a decline in eGFR ≥30% (Elastase-DNA: OR 1.55 (1.16 to 2.08), p=0.002; HMGB1-DNA: OR 1.54 (1.04–2.34), p=0.03) and developing severe renal impairment (eGFR <30 mL/min) (Elastase-DNA: OR 2.84 (1.34 to 6.04), p=0.006; HMGB1-DNA: OR 2.04 (1.12-3.72), p=0.02) at 24 months following the flare.
There was a linear relationship between the amount of Elastase-DNA and HMGB1-DNA complexes at the time of the flare and the decline in renal function in the subsequent 24 months. For every 100 U/mL increase in NET remnants, there was a decline in eGFR of 7.9% (95% CI 4.3 to 11.5, p<0.0001) and 6.8% (95% CI 2.5 to 11.0, p=0.001), for Elastase-DNA and HMGB1-DNA complexes, respectively (figure 2).
Linear relationship between the amount of NET remnants and decline in kidney function 24 months after the LN flare. Regression analyses were adjusted for ethnicity, adherence to treatment, serum creatinine at the time of the flare, history of a prior LN flare, immunosuppressive therapy received in the prior 3 months to the sample collection and the time in months from the sample collection to the NET remnants quantification. eGFR, estimated glomerular filtration rate; LN, lupus nephritis; NET, neutrophil extracellular trap.
Of the conventional biomarkers assessed, only serum creatinine had an effect on the decline in eGFR. When analysing the subgroup of patients with an eGFR ≥60 mL/min at the time of the LN flare (N=89), the effect of serum creatinine was lost, except for the outcome of decline of ≥30% in the eGFR 24 months after flare, where it remained significant, although the effect was attenuated. In contrast, the effect of the NET remnants was similar in patients with and without an eGFR ≥60 mL/min (table 3 and online supplemental table 4).
Given that proliferative LN classes have been associated to worse renal outcomes,31 and that in our cohorts NET remnants were predominantly elevated in proliferative classes, we analysed the subgroup of patients who had biopsy-proven proliferative or mixed LN class at the time of the flare or with a previous LN flare (N=82). As shown in online supplemental table 5, NET remnants continued to predict worse renal outcomes at 24 months after the flare in this cohort of patients.
NET remnant levels 12 months after the LN flare do not predict treatment responses at 24 months after the LN flare
Seventy-six patients (69.7%) had frozen serum stored 12±3 months after of the LN flare, of whom 31 (40.8%) were Caucasian, 19 (25.0%) Afro-Caribbean and 16 (21.1%) Asian. Thirty-Five (46%) of the patients achieved a CR at 24 months after the flare. The NET remnant levels at 12 months after flare were significantly lower than those seen at baseline (median (IQR), Elastase-DNA: 43.9 (20.7 to 91) vs 30 (15 to 55.2), p=0.02; HMGB1-DNA: 34 (11 to 80) vs 20 (7.35 to 48), p=0.007; baseline vs 12 months). Even though patients who failed to achieve a CR at 24 months following flare tended to have higher levels of both NET remnants compared with those with a CR (online supplemental table 2), the levels of NET remnants at 12 months following flare did not predict treatment responses at 24 months (table 2).
Discussion
The pathogenic role of neutrophils in SLE has been acknowledged since 1948 when the ‘LE cell’ phenomenon was first discovered.32 More recently, genomic studies have shown that the neutrophil signature is the second most robust signature in SLE patients and that this signature is enriched in patients with LN.33 34 Furthermore, NETs are an important source of autoantigens and have the potential to amplify the immunologic response causing tissue damage,11 26 35 positioning them as key effectors in SLE and LN pathogenesis.
Although multiple studies have found that patients with LN have increased levels of circulating NET remnants,7 9 36 the majority of these studies were cross-sectional and did not address the impact of NETs on clinical outcomes. One exception is the study by Moore et al,36 who found that the levels of serum NET complexes correlated with the SLEDAI scores 3 months after the NET remnant measurement. The results from the current study suggest that the serum levels of NET remnants at the time of a LN flare predict renal outcomes 24 months later. After adjusting for other confounding variables, patients who had higher levels of serum NET remnants at the time of the flare were less likely to achieve a complete renal remission and had a greater decline in eGFR. Notably, NET remnant levels outperformed proteinuria, anti-dsDNA antibodies, and complement as predictors of renal outcome. While serum creatinine and NET remnants were similarly predictive when all LN patients were included, the effect of serum creatinine on the renal outcomes was mostly lost when analysing the subgroup of patients with an eGFR ≥60 mL/min at the time of the flare. In contrast, the effect of the NET remnants remained significant in this clinical setting, suggesting that they are more predictive of renal outcomes when compared with serum creatinine in the setting of preserved renal function at the time of the flare.
Consistent with prior studies,7–9 37 we found that NET remnants were higher in patients with SLE, as compared with HC, and higher in patients with active LN, as compared with patients with active disease without LN. Although NETosing neutrophils have been found preferentially in proliferative LN kidney biopsies,10 no prior studies measuring circulating NET remnants have been able to differentiate between proliferative and non-proliferative classes.7 36 In our study, patients with proliferative LN had significantly higher serum levels of NET remnants compared with non-proliferative LN patients, and this finding was consistent in both SLE cohorts examined. Furthermore, the levels of Elastase-DNA complexes correlated with the activity index on renal biopsies. These results further corroborate the potential pathogenic role of NETosis in LN, particularly in proliferative classes.
Recently, the protein cargo of NET complexes has gained attention, as variations in the protein content of NETs have been found not only between SLE patients and HC5 but also between SLE patients with different clinical manifestations, including the presence of LN.38 These differences in protein composition have been shown to affect the immunologic amplification and tissue damage related to NETosis, and in this context, both Elastase-DNA and HMGB1-DNA complexes have been implicated as important mediators of renal damage in SLE. HMGB1 is a known inducer of IFN-I production by pDCs via Toll-like receptor (TLR) 9 activation,5 14 which in turn has been postulated to play an important role in LN pathogenesis.39 40 Recently, the levels of IFN-induced genes in renal tubular cells were shown to be an important determinant of subsequent disease course.41–43 HMGB1 also activates multiple innate and adaptive immune effectors5 44–46 and appears to promote renal disease, not only through induction of IFN-I in the kidney but also by direct effects on mesangial cells, leading to their activation and proliferation in a TLR2-dependent manner.47 48 Elastase-DNA complexes could also contribute to renal damage by causing direct renal endothelial damage.16 In addition, both Elastase and HMGB1 have been shown to induce renal endothelial-to-mesenchymal transition,16 49 resulting in a matrix producing mesenchymal cell that has the capacity to produce fibrosis, a hallmark of the progression to end stage kidney disease.50 51
The current study has multiple strengths. A rigorous approach was used to ensure the validity and reproducibility of the measurement of NET remnants, and both NET assays had good interassay and intra-assay CVs. In addition, our study included two independent cohorts, allowing us to validate our findings. Furthermore, subjects included in the LN cohort came from two multiethnic prospectively followed cohorts of SLE patients, who were actively followed for at least 2 years at 3–6 months intervals allowing the reliable determination of outcomes.
Our study also has several limitations. Even though the NET complex ELISA assays were developed using purified NETs and employed methods similar to those that have been widely used in prior studies to quantify NET remnants,7 23–25 we cannot state unequivocally that the protein–DNA complexes that are detected in the serum arise solely from NETs. It is possible that other forms of neutrophil death may lead to generation of these complexes or alternatively, that extracellular DNA (as a product of death from other cells) binds to circulating neutrophil-derived proteins, or in the case of HMGB1, protein released from other cellular populations. However, the strong correlation between the levels of Elastase-DNA and HMGB1-DNA in both our cohorts suggests that they were generated by a similar process. Finally, the findings from the LN cohort suggesting the role of NET remnants as predictors of renal outcomes, will need to be validated in an independent cohort.
In conclusion, circulating Elastase-DNA and HMGB1-DNA complexes are not only predominantly elevated in patients with LN but may also serve as predictors of adverse renal outcomes, including response to therapy and decline in kidney function 24 months following the LN flare. Our findings, in conjunction with previous work, suggest that NETosis plays a role in LN and that the measurement of NET remnants could serve as a biomarker to identify patients at a high risk for poor outcomes.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by Research Ethics Board of the University Health Network (REB#11-0397 and REB#05-0869). Participants gave informed consent to participate in the study before taking part.
Acknowledgments
This work has been previously presented in conference meetings.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors All authors were involved in drafting the article or revising it critically for important intellectual content and all authors approved the final version to be published. LPW-G, MU and JW were responsible for study conception and design. LPW-G and FN were responsible for the acquisition of data. LPW-G, MU, DDG, ZT, AK and JW were responsible for the analysis and interpretation of data. JW is the guarantor for the research.
Funding JW is supported by The Arthritis Centre of Excellence of the University of Toronto, a Pfizer Chair Career Award, and the Schroeder Arthritis Institute. LPW-G is a recipient of the Lupus Foundation of America, Gary S. Gilkeson MD career award, and FN was a recipient of the Gina M. Finzi Student Fellowship award. ZT is supported by a University of Toronto Temerty Faculty of Medicine Merit Award. Support for this study also came from funds donated to the Lupus Program, Centre for Prognosis Studies in the Rheumatic Diseases, including donations from Lupus Ontario, the Marissa and Lou Rocca Family, and the Diana and Mark Bozzo Family. The funders were not involved in the study design, collection, analysis, interpretation of data, the writing of this article or the decision to submit it for publication.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer-reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.