Article Text
PDF
Abstract
Objective To estimate real-life European Alliance of Associations for Rheumatology (EULAR)/European Renal Association (ERA)-European Dialysis and Transplantation Association (EDTA) response rates and predictors for no response in patients with lupus nephritis (LN) managed with conventional immunosuppressive therapies.
Methods Ambidirectional cohort study of patients with new-onset LN (period 2014–to date). Response rates in the first year were calculated, and all treatment modifications were recorded. Univariate and multivariate regression analyses were performed to assess determinants of failure to respond at 12 months.
Results 140 patients were included (81.4% women, median (IQR) age at LN diagnosis 38 (22) years). Among them, 32.1% presented with nephrotic range proteinuria, 28.6% with glomerular filtration rate <60 mL/min, 76.6% had proliferative and 19.7% class V LN. Initial treatment consisted of cyclophosphamide in 51.4% of patients (84.7% high-dose, 15.3% low-dose) and mycophenolate in 32.1%. 120 patients had available data at 12 months. EULAR/ERA-EDTA renal response rates at 3, 6 and 12 months were achieved by 72.6%, 78.5% % and 69.2% of patients, respectively. In multivariate analysis, increased Chronicity Index at baseline was associated with failure to achieve either complete or partial response at 12 months (OR 2.26, 95% CI 1.35 to 3.77). Notably, 20% of patients required treatment modifications due to suboptimal response during the first 12 months, with the addition of or switch to a different immunosuppressive drug in seven and nine patients, respectively.
Conclusions More than two-thirds of patients with LN attain EULAR/ERA-EDTA response rates by 12 months, but 20% require therapy modifications within this time period. Patients with increased chronicity in baseline biopsy, when combined with histological activity, are at higher risk for a lack of clinical response.
- Lupus Erythematosus, Systemic
- Lupus Nephritis
- Cyclophosphamide
Data availability statement
Data are available upon reasonable request.
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
Recent advances in the treatment of lupus nephritis have steered discussions on the initial use of combination therapies to achieve response.
The efficacy of conventional treatments in real-world settings is variable, often in contrast to the results of randomized trials.
WHAT THIS STUDY ADDS
In a contemporary multicentre cohort of patients with lupus nephritis, renal response rates at 12 months are satisfactory, however, changes in initial immunosuppressive regimens are not uncommon during this period.
An increased Chronicity Index at baseline biopsy confers a higher risk for failure to respond at 12 months.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
A repeat kidney biopsy may be considered in non-responding patients, especially with increased baseline chronicity, to justify intensification versus tapering of immunosuppressive treatment.
Introduction
Lupus nephritis (LN) represents a milestone in the natural course of systemic lupus erythematosus (SLE), affecting up to 50% of patients depending on race and ethnicity.1 Notwithstanding significant improvements in LN prognosis over the past decades, cumulative 5-year and 10-year end-stage kidney disease incidence reaches 10% and 20%, respectively,2 3 signifying its impact on morbidity and mortality.4 5 To this end, the 2019 joint European Alliance of Associations for Rheumatology (EULAR)/European Renal Association (ERA)-European Dialysis and Transplantation Association (EDTA) recommendations, as well as the 2024 Kidney Disease: Improving Global Outcomes LN Work Group, focus on the achievement of complete renal response, ideally within the first year and prevention of flares, so that long-term kidney function is preserved.6 7
Therapy of LN has witnessed remarkable advances over the past 5 years, with the approval of belimumab and voclosporin for any patient with active LN, following large phase 3 trials.8 9 These approvals have steered vivid discussions as to whether initial combination therapies should be used indiscriminately in all patients presenting with LN.10 11 The issue was addressed in the recent update of the EULAR recommendations for the management of SLE, where the phrase ‘should be considered’ was selected for combination therapies.12 Ultimately, the choice of initial therapy depends on several considerations, since differences in demographic,13 histological14 15 and clinical features16 have been associated with long-term kidney outcomes.
Apart from the higher cost and potential safety issues, a counterargument for the universal use of combination therapies at the initiation of treatment in LN is the fact that a substantial proportion of patients will respond to standard-of-care (SoC) treatments (i.e., mycophenolate mofetil (MMF) and cyclophosphamide (CYC)). In this regard, rates of complete response (CR) over time are variable in the literature. Indeed, while SoC arms in randomized trials typically reach CR rates of approximately 30–40% (including the Belimumab International Study in Lupus Nephritis (BLISS-LN) and AURORA trials), clearly suggesting an unmet need, real-life data from recent observational cohorts report considerably higher response rates.17–21 To this end, to decide for which subsets of patients additional treatments are needed, it is desirable to quantify the unmet need with SoC treatments in LN in real-world settings by assessing response rates to initial treatment and, more importantly, to identify predictors of non-response that lead to treatment modifications.22 23
In this study, we sought to evaluate in a contemporary multicentre cohort of patients with LN (i) 1-year and 2-year renal response rates and longitudinal treatment modifications to achieve these responses, and () predictors for failure to achieve response, which could identify potential candidate patients for initial combination treatments.
Methods
Study design and patient population
This is a multicentre, combined retrospective and prospective cohort study, which included patients with SLE (classified by 2019 American College of Rheumatology/EULAR classification criteria24) who had a first diagnosis of biopsy-proven LN (class III, IV, V or III/IV+V, according to the International Society of Nephrology/Renal Pathology Society 2003 LN classification system,25 followed-up in 10 academic rheumatology centres in Greece. For the retrospective part, in order to maximise the adequacy of data in a multicentre setting and also ensure that our results would reflect current clinical practice, we chose to include only patients who were diagnosed with LN from 2014 onwards. The prospective part of the study is ongoing, including all incident cases of LN who will be followed longitudinally until the completion of a minimum of 3 years of follow-up.26 Participants provided informed consent allowing collection of their data for inclusion in the study.
Recorded parameters
Collected data included demographic characteristics (sex, ethnicity, age at SLE and LN diagnosis), disease-related parameters (disease activity assessed by Systemic Lupus Erythematosus Disease Activity Index 2000 SLEDAI-2K score,27 irreversible organ damage assessed by Systemic Lupus International Collaborating Clinics (SLICC) Damage Index-SDI,28 comorbidities and medications received for SLE prior to LN diagnosis), laboratory parameters (autoantibody profile at SLE diagnosis, C3 and C4 levels, serum creatinine (SCr) and estimated glomerular filtration rate (eGFR, based on Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formula), serum albumin, 24 hours proteinuria (UPr) and urine sediment) and histological characteristics of kidney biopsy (LN class, National Institutes of Health (NIH) Activity and Chronicity Index, number of crescents, any presence of interstitial inflammation, interstitial fibrosis, tubular atrophy and features of thrombotic microangiopathy). Notably, the categorisation of interstitial inflammation/tubular atrophy as mild, moderate and severe was not available in most kidney biopsies, therefore we were not able to use it in our analyses.
We also recorded parameters related to administered treatments for LN (methylprednisolone pulses and subsequent oral prednisone dose, initial (‘induction’) and subsequent (‘maintenance’) treatment regimen). Importantly, treatment was decided according to physician’s judgement in each participating centre and was not directed by any prespecified study protocol.
The aforementioned parameters were recorded at the time of LN diagnosis and subsequently at 3, 6, 9, 12, 18, 24, 36 months, as well as at the last follow-up visit (for patients completing more than 36 months of follow-up). We also recorded any treatment modifications, driven by either inadequate response or renal flare, on a yearly basis: from LN diagnosis to 12-month follow-up visit, from 12-month to 24-month follow-up visit, from 24-month to 36-month follow-up visit and finally from 36 months to last follow-up. The results of the present report focus on the outcomes of the first 24 months.
Definitions
The primaryendpoint, that is, response to therapy over the first year following kidney biopsy, was assessed according to the proposed EULAR/ERA-EDTA recommendations’ treatment targets: (i) 25% reduction in UPr at 3 months, (ii) 50% reduction in UPr (and to subnephrotic levels, if initially nephrotic) at 6 months (hereafter referred to as partial response, PR) and (iii) UPr<0.5 g/day at 12 months (hereafter referred to as CR), all with stabilized or improved SCr (within 10% of baseline value). Similar definitions of PR and CR were used for time points beyond the first year. Rates of modified primary efficacy renal response (PERR: UPr≤0.7 g/day and eGFR ≥60 mL/min/1.73 m2 or ≤20% below baseline value) were also calculated at 1 and 2 years.8 29
Renal flares were defined as follows: (i) proteinuric, when UPr exceeded 1 g/24 hours if CR had previously been achieved, or UPr>2 g/24 hours if PR had been achieved, (ii) nephritic, as an increase in hematuria >10 red blood cells (RBCs) per high-power field and concurrent increase in SCr by >30% (or by >10% eGFR decrease), irrespective of UPr changes.30 31 Inadequate response was defined as (i) failure to achieve PR by 6 months and (ii) CR by 12 months and all subsequent time points.
Treatment modification was defined as any of the following (provided it was undertaken for the treatment of LN and not for an extrarenal manifestation, as specified to physicians of participating centres): (i) increase in prednisone dose or use of methylprednisolone pulses, (ii) increase in dose of existing immunosuppressive drug, (iii) addition of a new immunosuppressive drug to existing regimen and (iv) switch to a different immunosuppressive drug.
Statistical analysis
The distribution of variables was examined by D’Agostino-Pearson and Shapiro-Wilk tests. Continuous variables are presented as mean (SD) and median (IQR) for normal and non-normal distribution of data, respectively, and categorical variables as frequencies (percentages). T-test and Mann-Whitney U test were applied for comparisons of continuous variables, and χ2 and Fisher’s exact test for categorical variables, respectively, to examine for differences in recorded parameters between patients following different therapeutic regimens as initial therapy. Kaplan-Meier analysis was applied to examine the association between initial treatment regimens or different LN classes and time to response.
Univariate logistic regression analysis was performed to examine for associations between lack of response (dependent variable) and all recorded parameters (described above in Methods). To test for multivariate associations after controlling for potential confounders, multiple logistic regression models with the enter method were applied; only parameters statistically significant at p<0.05 in univariate analysis (provided they did not have significant missing data), were entered in the multivariate models. All levels of statistical significance were set at p<0.05. Statistical analysis was performed using Stata software (V.13.0, College Station, Texas, USA) and SPSS V.26 software (IBM, USA).
Results
Cohort characteristics
A total of 140 patients were included in the study (81.4% women, median (IQR) age at LN diagnosis 38 (22) years); 26 patients (18.6%) were diagnosed after 50 years of age (late-onset LN). Detailed patient characteristics are presented in table 1. LN was a presenting SLE manifestation in 39 (27.8%) patients, whereas 50 (35.7%) patients were diagnosed with LN more than 5 years since SLE diagnosis. Regarding histological classes, 20 (14.6%) had class III, 48 (35%) had class IV, 37 (27%) had mixed class (III/IV+V) and 27 (19.7%) had class V (membranous) LN. In 3 patients (2.1%) we did not have available data from renal biopsies (ie, although a renal biopsy was performed, it was not diagnostic due to lack of glomeruli. Mean (SD) proteinuria at baseline was 3.2 (2.9) g/day, and 45 patients (32.1%) manifested nephrotic range proteinuria. Mean (SD) baseline eGFR was 85.3 (40.5) mL/min/1.73 m2, while 16 patients (11.4%) presented with eGFR<30 mL/min/1.73 m2 at LN diagnosis. The median (IQR) follow-up time was 47 months (26); however, the present study focuses on the first 2 years.
Baseline demographic, clinical, laboratory and histological characteristics of patients with lupus nephritis (n=140)
Initial (induction) and subsequent (maintenance) treatment regimens
Regarding initial (‘induction’) treatment, 72 patients (51.4%) were treated with CYC and 45 patients (32.1%) with MMF. Of CYC-treated patients, a high-dose (modified NIH) regimen was used in 84.7% of patients, while only 15.3% received low-dose CYC (Euro-Lupus regimen). The remaining patients received various other immunosuppressive regimens (online supplemental table S1), among which a combination treatment of CYC or MMF with belimumab was administered in four patients (after approval of the latter for LN). The majority (n=104, 75.9%) received intravenous methylprednisolone pulses at treatment initiation, followed by a median (IQR) starting prednisone dose of 40 (20) mg. At 12 months of follow-up, 111/120 (92.5%) patients (for whom all data were available) were receiving a median (IQR) prednisone dose of 5 (5) mg and 30.8% were receiving a daily prednisone dose ≥7.5 mg. Patients treated initially with CYC differed significantly in baseline clinical and histological characteristics compared with patients treated with MMF; the latter presented with higher eGFR and a more benign kidney biopsy, with less frequent presence of factors associated with poor prognosis. Additionally, regarding the different CYC regimens, high-dose CYC was administered in older patients, confounded by the lack of fear for gonadal toxicity (median (IQR) age: 42 (40) years vs 29 (9), p=0.02) with increased activity in kidney biopsy (median (IQR) NIH Activity Index: 10 (7) vs 3 (1), p<0.001) (online supplemental table S2).
Supplemental material
At 6 months from treatment initiation, treatment consisted of MMF in 66 patients (51.2%), CYC in 32 (24.8%) and azathioprine in 10 (7.8%) patients. Of note, 9 patients (28.1%) continued to receive CYC between 6 and 12 months of follow-up, due to flare, inadequate response, or as a continuation of the original treatment after meeting CR (1, 6 and 2 patients, respectively).
Finally, regarding adjunct therapies for LN, 121 patients (86.4%) received hydroxychloroquine at baseline, 64 (45.7%) ACE inhibitors (ACEi), 7 (5%) angiotensin receptor blockers (ARBs) and 2 (1.42%) sodium-glucose cotransporter 2 (SGLT2) inhibitors.
Renal response rates at 12 months
A total of 120 patients (85.7%) had all data available at 12 months of follow-up, with 80 (57.1%) and 62 (44.3%) completing 24 and 36 months, respectively (study is ongoing). Rates of attainment of EULAR/ERA-EDTA treatment targets within the first 12 months were 72.6% (90/124) for 25% reduction in UPr at 3 months, 78.5% (95/121) for 50% reduction in UPr at 6 months and 69.2% (83/120) for UPr<500 g/day at 12 months (all with eGFR±10% of baseline) (figure 1A). At 12 months, an additional 18 patients (15%) had a 50% reduction in UPr, reaching a total of 84.2% of patients attaining CR or PR at the end of the first year. Collectively, combined CR and PR rates at 6, 12 and 24 months were 78.5%, 84.2% and 93.7%, respectively (figure 1B). Additionally, modified PERR rates in years 1 and 2 were 72.5% (n=87/120) and 83.7% (67/80), respectively. The detailed baseline characteristics between patients who achieved CR or PR and no response are presented in online supplemental table S3. Of note, among patients who did not respond, only one patient had a repeat kidney biopsy, which showed increased renal activity over chronic lesions, justifying the modification of treatment.
(A) More than two-thirds of patients with lupus nephritis achieve the EULAR/ ERA-EDTA treatment target at 12 months. Percentages of attainment of EULAR/ ERA-EDTA treatment targets within the first year after kidney biopsy in patients with lupus nephritis. (B) Rates of partial, complete and no response at 6, 12 and 24 months. Percentages of patients who achieved complete, partial or no response at 6, 12 and 24 months, respectively. EDTA, European Dialysis and Transplantation Association; ERA, European Renal Association; EULAR, European Alliance of Associations for Rheumatology.
No significant difference in achieving CR at 12 months was noted between the two main initial treatment regimens (CYC vs MMF) (p for log-rank test=0.252) (figure 2A). On the contrary, a marginally significant difference in time to CR between class III, class IV, mixed class LN and class V was noted (p for Wilcoxon test=0.048); the latter was associated with longer time to CR (figure 2B).
Probability for complete response is similar, irrespective of the initial standard-of-care immunosuppressive regimen and lupus nephritis class. Kaplan-Meier survival estimates of probability for complete renal response according to (A) initial treatment regimen (IV CYC vs MMF), and (B) lupus nephritis class. CRR, complete renal response; CYC, cyclophosphamide; IV, intravenous; LN, lupus nephritis; MMF, mycophenolate mofetil.
Predictors for lack of response at 12 months
Next, we performed logistic regression to identify baseline predictors for failure to achieve treatment responses at 12 months, as this could assist in the identification of patients who could be candidates for early combination therapies to increase their possibility of response. In univariate analysis, failure to achieve the EULAR/ERA-EDTA target at 12 months (UPr<500 mg/day, ie, CR) was associated with baseline proteinuria >3 g/24 hours (OR 2.50, p=0.03) and increased NIH Chronicity Index in baseline kidney biopsy (OR 1.52, p=0.004); by contrast, low C3 at baseline showed a negative association with failure to respond (ie, was associated with increased possibility for response) (table 2). In multivariate analysis, every unit increase in the baseline Chronicity Index was independently associated with a 57% increased probability of lack of CR, after adjusting for other covariates (adjusted OR 1.57, 95% CI 1.15 to 2.14, table 2). Similar results were found for predictors of failure to achieve any response (ie, CR or PR) at 12 months, where Chronicity Index showed an even stronger association (adjusted OR 2.26, 95% CI 1.35 to 3.77) (table 2). Male sex was also strongly associated with a lack of CR/PR at 12 months; however, the significance of this result is limited by the small number of male patients in our cohort.
An increased NIH Chronicity Index predicts failure to respond to immunosuppressive therapy at 12 months
Since LN class V is different from proliferative LN classes, a logistic regression analysis was performed after we had excluded patients with class V LN. We found similar results in multivariate analysis, as failure to achieve complete renal response at 12 months was associated with an increased Chronicity Index (online supplemental table S4).
Treatment modifications within the first 12 and 24 months
Among 120 patients who reached the 12-month follow-up time point, 24 patients (20%) had at least one treatment modification within this period (9 due to a flare and 15 because of inadequate renal response). A schematic representation of these changes in initial treatments is shown in figure 3. As initial treatment, these 24 patients had received CYC (n=15, high-dose and low-dose in 6 and 9, respectively), MMF (n=5), Rituximab (RTX) (n=1), RTX plus MMF (n=1), Cyclosporin (CsA) (n=1) and MMF plus CsA (n=1). Treatment modifications consisted of an increase in prednisone dose (58.3% of cases); an increase in the dose of the immunosuppressive drug (46%); a change of immunosuppressive drug (37%, most frequently to CYC or MMF); or an addition of a second drug (29.1%, belimumab in 3, rituximab and a calcineurin inhibitor, in 2 patients each). Ultimately, 45% of these patients (n=11/24) achieved EULAR-ERA/EDTA treatment target at 12 months.
Immunosuppressive treatment modifications within the first year are not infrequent in lupus nephritis. Flow diagram depicting immunosuppressive treatment modifications within the first year after kidney biopsy in patients with lupus nephritis. Treatment modifications were decided as per the treating physician judgement, either due to inadequate response or due to flare after the initial response.
80 patients completed 24 months of follow-up. Of these, an LN flare was documented in 8 patients (10%, proteinuric in 6/8), which led to a change in therapy. A repeat renal biopsy was performed in 4/8 (50%) patients. In all, a worsening in the NIH Chronicity Index was evident in repeat biopsies. An additional 8 patients underwent treatment modification due to inadequate response. In most patients, a prednisone dose was increased (11/16, 69%) and/or a second immunosuppressive drug was added (9/16; rituximab, belimumab and cyclosporine in 6, 2 and 1, respectively) or switched to (6/16; MMF, rituximab or CYC in 3, 2 and 1 patient, respectively).
Discussion
In this study, we collected real-world data from multiple rheumatology departments in Greece to map the current state of LN in terms of treatment and short-term outcomes, in an attempt to quantify the unmet need that has to be addressed by treatment strategy modifications. We found that almost 70% of patients reach an EULAR/ERA-EDTA response at 12 months, a milestone that has been linked to favourable long-term prognosis. Adding the percentage of patients achieving PR, more than 80% of patients (84.2%) with new-onset LN have a satisfactory response to SoC treatment regimens.
Our findings are in agreement with a large LN cohort from Italy, which reported 84% combined complete and PR rate at 12 months (58% and 26%, respectively).32 Other European cohorts have reported similarly high response rates (88–91%) at the same time point or later.17 18 33 In striking contrast, a recent report from the Accelerating Medicines Partnership (AMP) Lupus Network in the US found a combined response rate of hardly 44% at 1 year, in close resemblance to the results of randomised clinical trials. This discrepancy could possibly be attributed to differences in the population (less than 30% white patients in the AMP cohort, compared with an almost exclusive white population in ours), but also in treatment regimens (64% used MMF as induction in AMP).34 In our cohort, the majority of patients (52%) were treated with a modified high-dose intravenous CYC regimen, despite the fact that the ‘oldest’ case was diagnosed in 2015. Importantly, the multicentre design of our study precludes bias of one particular centre towards high-dose CYC therapy. Of note, CYC was the predominant induction regimen in three recent cohort studies from European populations (percentages ranging from 50% to 79%). Although these studies included patients treated in earlier periods, two of them included only patients with proliferative LN, no specification was made regarding CYC dose.17 18 32 Patients treated with CYC in our study had more severe LN, as evidenced by a lower GFR and more severe histological findings in kidney biopsy. Notwithstanding the different treatment patterns in the aforementioned studies, we did not find a difference in response rates between patients who received CYC or MMF, similarly to randomised and observational studies.35
The level of proteinuria at 12 months has been consistently linked to long-term outcomes in LN. Post hoc data from the MAINTAIN and Euro-Lupus Nephritis trials have shown that proteinuria values 0.7 and 0.8 g/day, respectively, had the best predictive value for a SCr<1.0 mg/dL at 7 years.36 37 More recently, the lack of an EULAR/ERA-EDTA response at 12 months was linked to a fivefold increased risk for CKD.32 To this end, we sought to identify baseline factors that are associated with failure to achieve EULAR/ERA-EDTA response at this particular time point. In multivariate analysis, we found that NIH activity and chronicity indices have opposite associations with renal response at 12 months. In accordance with the AMP and other studies,38 39 we identified an increased Chronicity Index as the sole parameter independently associated with failure to reach either CR or PR at 12 months (more than twofold for each unit increase in NIH Chronicity Index). Importantly, increased chronicity in baseline kidney biopsy may account for already irreversible lesions responsible for residual proteinuria that could potentially render a patient unable to be classified as a responder. In this regard, patients with increased chronicity at baseline may by default not be able to reach the proteinuria targets, and may thus need specific treatment targets. A repeat kidney biopsy at 12 months can be most informative in such patients, where the presence of only chronic lesions without histological activity would justify tapering rather than intensification of immunosuppression and a focus on CKD management. However, in our real-life cohort, a repeat kidney biopsy was performed in only 1/19 non-responders at 12 months, thus precluding from further analyses and conclusions.
Early management of LN is crucial for improving disease prognosis, which underscores the importance of maintaining a low threshold for performing a kidney biopsy and prompt referral to specialised centres.40 Notably, in our cohort, among 19 patients who did not achieve a CR, LN was the initial SLE manifestation in four cases. In all patients who did not respond, there was no information regarding delay prior to treatment initiation, however, potential delays in referral to participating study centres cannot be excluded. This consideration is particularly pertinent given that LN may remain clinically silent for extended periods and a kidney biopsy should be considered for patients with low-level proteinuria, despite inactive urine sediment or normal serology, due to the discrepancy between clinical and histological findings.41 42
A common limitation of observational studies in LN is that they focus on treatment regimens and outcomes at strict predefined time points (eg, 6 or 12 months, or further), but often lack a more ‘granular’ analysis of possible subtle changes in treatment regimens between these time points. An important feature of our study is that we also explored modifications of initial (or subsequent) immunosuppressive regimens in between prespecified time points (quarterly for the first year, and semi-annually thereafter). By this, we found that as many as 20% of patients necessitate some kind of modification in their initial regimen. Such treatment modifications ranged from temporary increases in prednisone doses, to switching to, or adding a different immunosuppressive drug. Indeed, more than half of the patients who underwent treatment modification within the first year required a temporary escalation in their daily prednisone dosage. Interestingly, a real-world analysis of patients with LN in Germany indicated frequent use of high-dose glucocorticoids during the first year,20 and a recent study from the US based on claims data, found a similar percentage (24.6%) to our study regarding the switch between induction therapies within the first year.19 Of note, less than 50% of patients who underwent any treatment change ultimately reached CR at 12 months, highlighting again the importance of the choice of optimal immunosuppressive therapy at the start of treatment. Pending the identification of reliable biomarkers that would optimise the selection of patients for early combination therapies, those who require frequent increases in glucocorticoid doses or those at high risk for glucocorticoid side effects due to comorbidities, may be suitable candidates for such treatment regimens.33
Together with the need for early diagnosis, the impact on prognosis that accompanies an increased Chronicity Index at baseline kidney biopsy highlights the importance of adjunct treatment in every patient with LN, aiming to minimise the progression of CKD. Since every patient with LN should be viewed as a patient with CKD, lifestyle modifications, including a low-sodium diet and maintaining optimal blood pressure control, along with adjunct therapies, like renoprotective medications [eg, Renin-angiotensin-aldosterone system (RAAS) blockade and sodium-glucose cotransporter-2 (SGLT2) inhibitors], should be considered for all patients with LN to prevent CKD progression.7 12 43 In our cohort, ACEi/ARB were used in half of the patients (50.7%), however, SGLT2 inhibitors (SGLT2i) use was very limited, possibly because our study was conducted prior to the increasing use of the latter in LN but may also reflect the fact that patients were treated mainly by rheumatology departments.
Our study has certain limitations that need to be acknowledged. Our cohort consists almost exclusively of white patients; thus, our data cannot easily be extrapolated to LN populations of different racial backgrounds, in which kidney involvement is more frequent and occasionally more severe. Moreover, immunosuppressive treatment regimens were far from homogeneous, including also very small numbers of individual drugs or combination therapies, for which no conclusion can be made regarding efficacy. Yet, our study was not designed to test or compare the efficacy of different treatment regimens in LN, but rather to profile the landscape of LN in the modern era. In this regard, we included all cases that fulfilled the inclusion criteria of our study, irrespective of the therapy chosen by physicians in multiple centres. Last, for all patients in our cohort, rheumatologists were the leading physicians, which may have conferred bias regarding the choice of treatment or other parameters. Nevertheless, it is common practice at most, if not at all, centres that participated in the study, to discuss new LN cases in multidisciplinary teams, with the participation of nephrologists and renal pathologists, in order to optimise patient care.38 39 44
In conclusion, in a contemporary real-life study of LN treated with current immunosuppressive regimens, we found high rates of CR and PR at 12 months. Increased Chronicity Index in baseline kidney biopsy is an independent risk factor for a failure to respond. We also found that treatment modifications within the first year are not uncommon, concerning nearly 20% of patients. These subsets of patients may be potential candidates for combination treatments with emerging novel drugs, while data regarding optimal patient selection are awaited.
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
The study protocol was approved by the Institutional Review Board of the participating rheumatology centres. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
We are thankful to the staff physicians of participating Rheumatology Departments in the study for providing care to the patients with SLE and lupus nephritis. We also thank Spyridon Katechis, MD, for a critical review of the statistical analysis.
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
X @george_bertsias
GB and MGT contributed equally.
Contributors MP and MK: acquisition of data, analysis of data, drafting and critical revision of the manuscript. AP, MN, NML, CT, AC, EA, TD, MC, GD, CP, AG and CK: acquisition of data and critical revision of the manuscript. DV, PS, KAB, PPS and PVV: critical revision of the manuscript. DTB, GB and MGT: conception and design of the study, interpretation of data, critical revision of the manuscript. AF: conception and design of the study, acquisition of data, analysis, and interpretation of data, drafting and critical revision of the manuscript, guarantor. All authors have read and approved the final version of the manuscript.
Funding The study received funding by the Hellenic Society of Rheumatology & Professionals Union of Rheumatologists of Greece (Nr. 1077/21.12.21).
Competing interests None declared.
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.