Renal response status to predict long-term renal survival in patients with lupus nephritis: results from the Toronto Lupus Cohort Authors: Murray Urowitz, Mary E Georgiou, Zahi Touma, Jiandong Su, Juan Pablo Diaz-Martinez, Qinggong Fu, Roger A Levy, Kerry Gairy, Anne MacKinnon, Nicole Anderson, Patricia C Juliao Published: 2024 ABSTRACT ABSTRACTObjectiveTo evaluate modified versions of the Belimumab International Study in Lupus Nephritis (BLISS-LN) belimumab study primary efficacy renal response (mPERR) and complete renal response (mCRR) criteria (excluding mandatory corticosteroid tapering) as predictors of real-world, long-term renal outcomes among patients with lupus nephritis (LN).MethodsThis retrospective, observational study (GSK Study 212866) used deidentified data between 1970 and 2015 from the University of Toronto Lupus Cohort from adults diagnosed with systemic lupus erythematosus and biopsy-proven Class III±V, IV±V or V LN. At 24 months postbiopsy, patients were retrospectively indexed as responders/non-responders based on mPERR (estimated glomerular filtration rate (eGFR) ≤20% below biopsy value/≥60 mL/min/1.73 m2 and urine protein:creatinine ratio (uPCR) ≤0.7 g/day) or mCRR (eGFR ≤10% below biopsy value/≥90 mL/min/1.73 m2 and uPCR ≤0.5 g/day) criteria. The association between index mPERR (primary outcome) or mCRR (secondary outcome) status and long-term (up to 25 years, until censoring or death) renal survival (no progression to end-stage kidney disease (eGFR <30 mL/min/1.73 m2, dialysis or transplant) or death) was assessed.ResultsOverall, 179 patients were included in the analysis (mPERR responders, n=128; non-mPERR responders, n=51). Most patients were female (87.2%); the mean (SD) age was 34.1 (11.3) years.Long-term renal survival was attained for 78.9% of mPERR responders and 60.8% of non-mPERR responders; achieving mPERR was associated with an increased likelihood of long-term renal survival versus not achieving mPERR (log-rank p=0.0119). Overall, 102 patients were mCRR responders, and 77 were non-mCRR responders. Long-term renal survival was attained for 80.4% of mCRR responders and 64.9% of non-mCRR responders; achieving mCRR was associated with an increased likelihood of long-term renal survival than not achieving mCRR (log-rank p=0.0259).ConclusionsAchieving mPERR or mCRR was associated with improved long-term renal survival, highlighting that these statuses are suitable predictors of long-term renal outcomes in patients with LN. WHAT IS ALREADY KNOWN ON THIS TOPIC Lupus nephritis (LN), a severe manifestation of systemic lupus erythematosus, may lead to permanent, irreversible chronic kidney disease and, if not managed, to end-stage kidney disease (ESKD), a condition that is associated with poor disease prognosis. Thus, identifying predictors of long-term renal outcomes, particularly ESKD, would be of great value in LN trials and clinical practice. Lupus nephritis (LN), a severe manifestation of systemic lupus erythematosus, may lead to permanent, irreversible chronic kidney disease and, if not managed, to end-stage kidney disease (ESKD), a condition that is associated with poor disease prognosis. Thus, identifying predictors of long-term renal outcomes, particularly ESKD, would be of great value in LN trials and clinical practice. Renal response is a frequently used treatment target in LN clinical trials, with most studies using proteinuria level and glomerular filtration rate in their definition of renal response; however, there is no consensus on the precise combination and threshold levels for these outcome measures. Renal response is a frequently used treatment target in LN clinical trials, with most studies using proteinuria level and glomerular filtration rate in their definition of renal response; however, there is no consensus on the precise combination and threshold levels for these outcome measures. WHAT THIS STUDY ADDS This retrospective analysis of the Toronto Lupus Cohort evaluated the prognostic value of renal response status measured by Belimumab International Study in Lupus Nephritis (BLISS-LN)-based modified primary efficacy renal response (mPERR) or modified complete renal response (mCRR) definitions on long-term renal outcomes. This retrospective analysis of the Toronto Lupus Cohort evaluated the prognostic value of renal response status measured by Belimumab International Study in Lupus Nephritis (BLISS-LN)-based modified primary efficacy renal response (mPERR) or modified complete renal response (mCRR) definitions on long-term renal outcomes. The results of this study demonstrate that achieving mPERR or mCRR at 2 years following biopsy-proven LN diagnosis is associated with favourable long-term (up to 25 years) renal survival. The results of this study demonstrate that achieving mPERR or mCRR at 2 years following biopsy-proven LN diagnosis is associated with favourable long-term (up to 25 years) renal survival. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY Data from this study confirm the use of mPERR or mCRR as prognostic measures for long-term renal outcomes in patients with LN. Data from this study confirm the use of mPERR or mCRR as prognostic measures for long-term renal outcomes in patients with LN. Introduction Systemic lupus erythematosus (SLE) is a multisystem autoimmune disease with complex clinical presentation associated with multiple autoantibodies.1 2 Lupus nephritis (LN) is a severe manifestation of SLE that affects approximately 40–50% of patients with SLE.35 If not recognised and treated urgently, LN significantly increases the risk of disease burden and irreversible kidney damage, leading to end-stage kidney disease (ESKD).6 7 Approximately 20% of patients with LN will develop ESKD after 10 years of disease.8 The long-term management of LN, including ESKD and kidney transplant needs, is costly, and patients with LN tend to have a worse health-related quality of life compared with those without LN.9 10 Given the importance of detecting and treating abnormalities in kidney function among patients with LN, renal response is a frequently assessed treatment efficacy outcome in clinical trials; however, various definitions have been used to define renal response to date,1115 indicating a lack of consensus on the precise renal parameter combinations and thresholds to use. The Belimumab International Study in Lupus Nephritis (BLISS-LN) Phase 3 clinical trial (NCT01639339) assessed belimumab in patients with LN included primary efficacy renal response (PERR; defined as urine protein:creatinine ratio (uPCR) ≤0.7 g/day, estimated glomerular filtration rate (eGFR) of ≤20% below the baseline value or ≥60 mL/min/1.73 m2 and no use of corticosteroids) and complete renal response (CRR; defined as uPCR ≤0.5 g/day, eGFR of ≤10% below the baseline value or ≥90 mL/min/1.73 m2 and no use of corticosteroids) as endpoints.16 While up to 43% of patients were able to achieve one or both endpoints after 2 years of treatment,16 further evidence is required to support whether these renal response outcomes can accurately predict long-term renal survival. In a retrospective analysis of real-world data from the Hopkins Lupus Cohort, Cooper Blenkinsopp et al (2022)17 assessed long-term renal outcomes in patients with LN using modified versions of the BLISS-LN PERR (mPERR) and CRR (mCRR) definitions that omitted the requirement for mandatory tapering of corticosteroids (including rescue therapy restrictions), and to allow for a comparison between eGFR and the GFR value taken at biopsy. Classifying patients into mPERR/mCRR responders and non-responders showed clinical relevance for the prediction of long-term renal outcomes, as patients who achieved mPERR/mCRR had significantly reduced risk of chronic renal insufficiency and ESKD or death compared with those who did not.17 However, the high representation of patients with black African ancestry in the Hopkins Lupus Cohort (52.6%)17 calls for a broader real-world evaluation to ensure the robustness and generalisability of the mPERR and mCRR endpoints in predicting long-term renal outcomes across diverse demographic groups. The goal of this study was to build on evidence from the Hopkins Lupus Cohort17 by evaluating the ability of the mPERR and mCRR endpoints to predict real-world, long-term renal outcomes in patients with LN in a separate patient population, the University of Toronto Lupus Cohort. Methods Study design This was a retrospective analysis (GSK Study 212866) of the deidentified data from the University of Toronto Lupus Cohort in Canada (figure 1). The Toronto Lupus Cohort is a prospective, longitudinal study collecting information since 1970 on SLE disease activity, organ damage and health-related quality of life in patients with SLE, and it is continually updated with patient sociodemographic data, medical and reproductive history, comorbidities, SLE complications and treatments. Study design. Due to the observational nature of this study, a window of ±3 months was permitted for the inclusion of renal function laboratory data around each of these dates for each patient. ESKD, end-stage kidney disease; mPERR, modified primary efficacy renal response. The study period was from the Toronto Lupus Cohort database initiation through to December 2015, and patients were identified from the Toronto Lupus Cohort database based on biopsy-proven LN diagnosis. It is noteworthy that the study was conducted prior to the 2020 Food and Drug Administration approval of belimumab as the first biologic for the treatment of LN18 and, thus, no patients were receiving biologics as part of their treatment regimen. Baseline was defined as the date of the kidney biopsy. Patients were indexed after a 24-month window postbaseline (ie, index date) and retrospectively categorised according to their renal status achieved based on mPERR and mCRR criteria (responder vs non-responder). To be classed as an mPERR responder at index, patients had to have an eGFR ≤20% below biopsy values (or ≥60 mL/min/1.73 m2; calculated from the four-variable Modification of Diet in Renal Disease Study formula19 as follows: GFR=186.3×serum creatinine−1.154×Age−0.203×1.212 (if African American)×0.742 (if female); where serum creatinine was measured in mg/dL, age in years and GFR expressed in mL/min/1.73 m2) and one of the following (ordered hierarchically): 24-hour urine protein measure ≤0.7 g/24 hours, uPCR ≤0.7 g/day or urine protein dipstick score ≤30 mg/dL. To be classed as an mCRR responder at index, patients had to have an eGFR ≤10% below biopsy values (or ≥90 mL/min/1.73 m2) and one of the following (ordered hierarchically): 24-hour urine protein measure <0.5 g/24 hours, uPCR <0.5 g/day or urine protein dipstick score ≤30 mg/dL. Patients not meeting the respective criteria were classed as non-responders. Patients were followed for up to 25 years postindex or until censoring (outcome event, loss to follow-up or end of study dataset) or death. Additional renal function assessments were made postbaseline during designated 6- and 36-month windows. Further information on the windows for each time period is detailed in online supplemental table 1. Patients Eligible patients with SLE had to have been ≥18 years of age at baseline and met the SLE European Alliance of Associations for Rheumatology (EULAR)/American College of Rheumatology (ACR) 2019 classification criteria,1 20 21 with biopsy-proven Class III, III/IV, IV, IV/V or V LN. Patients were required to have ≥1 serum creatinine measure within the baseline and 24-month windows along with ≥1 uPCR measure, ≥1 24-hour urine protein measure or ≥1 urine dipstick score. Patients were also required to have attended ≥1 clinic visit postindex. Patients were excluded if they had ESKD progression preindex. ESKD was defined as eGFR <30 mL/min/1.73 m2 (chronic kidney disease (CKD) stage 4 or 5) or receipt of renal replacement therapy (dialysis or transplantation) as per criteria outlined by Cooper Blenkinsopp et al.17 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. Outcomes The primary outcome was to assess the association between mPERR at index and long-term renal survival, defined by the absence of ESKD or death. The key secondary outcome was to assess the association between mPERR at index and survival without risk of mild CKD or worse during the follow-up period, defined as a new occurrence of GFR <60 mL/min/1.73 m2 (equivalent to CKD stage 3a, 3b, 4 or 5) on ≥2 consecutive occasions ≥3 months apart. Patients with mild CKD or worse prior to index were excluded from the key secondary outcome analysis. Other secondary outcomes included the association between mCRR at index and long-term renal survival, and the association between mCRR at index and survival without risk of mild CKD or worse during the follow-up period (as defined above). Statistical analysis Sample size calculations for the primary analysis were conducted using PASS software (2019, V.19.0.1), with parameters set for a one-sided log-rank test with 80% power, alpha=0.05, no dropouts and total survival time of 60 months. Various scenarios of mPERR response rates and differential survival outcomes were considered as additional attrition was likely to occur due to study inclusion requirements. Based on the final sample size, the study was powered to detect a 20% difference in survival between those with and without mPERR. The association between mPERR and mCRR status and long-term renal survival outcomes were compared using Kaplan-Meier plots and log-rank tests. Cox proportional hazards models were used to evaluate confounding effects on the association between mPERR or mCRR status and long-term survival outcomes. Confounders were defined as variables that changed the HR by >10%. The following confounders were tested for: International Society of Nephrology (ISN) Class, early response, age at biopsy date, biopsy date, gender, race, Systemic Lupus International Collaborating Clinics (SLICC) Damage Index Score (Day 0 window), immunosuppressants, eGFR at Day 0, urine protein at Day 0, SLE Disease Activity Index (SLEDAI) score and year of biopsy. The following confounders were excluded from consideration for the multivariate model due to missing data: Body Mass Index, SLE treatments (antimalarials, ACE inhibitors, angiotensin receptor blockers, steroids, rituximab, belimumab), diabetes and hypertension. The proportion of patients who achieved mPERR at the indicated time periods was calculated with 95% CIs. Results Patient baseline characteristics Overall, 179 patients met the eligibility criteria and were included in this analysis (online supplemental figure 1). The median follow-up in the cohort recruited between 1970 and 2015 was approximately 12 years. At index, 71.5% of patients were classed as mPERR responders and 28.5% classed as non-responders. Patient characteristics at baseline are presented in table 1. For the overall population, most patients (87.2%) were female, and the mean (SD) age was 34.1 (11.3) years. Few patients had a mixed LN class (7.3%), with a relatively balanced distribution of patients across LN Class III (28.5%), IV (36.9%) and V (27.4%) categories. Patients had a mean (SD) Systemic Lupus International Collaborating Clinics/ACR Damage Index (SDI) and SLE Disease Activity Index 2000 (SLEDAI-2K) score of 0.5 (1.0) and 15.2 (8.2), respectively, with 73.7% having an SDI score of 0. Of 179 patients, 84.9% had an SLEDAI-2K score ≥8, with 62.0% (n=111/179) presenting ≥4 points in the renal component at baseline. The most common medication used prior to biopsy (baseline) was corticosteroids (93.3%) followed by antimalarials (52.5%) and immunosuppressants (50.3%) (table 1). Table 1Patient baseline characteristicsCharacteristicOverall (N=179)mPERR statusResponders (n=128)Non-responders (n=51)Age (continuous), years, mean (SD)34.1 (11.3)34.1 (11.2)34.3 (11.7)Female, n (%)156 (87.2)117 (91.4)39 (76.5)Race, n (%) White112 (62.6)80 (62.5)32 (62.7) Asian27 (15.1)17 (13.3)10 (19.6) Black25 (14.0)20 (15.6)5 (9.8) Other15 (8.4)11 (8.6)4 (7.8)BMI (continuous)*, kg/m2, mean (SD)N=6626.1 (12.4)n=4924.0 (9.5)n=1732.3 (17.3)Decade of biopsy, n (%) 1989 or earlier43 (24.0)24 (18.8)19 (37.3) 1990–199928 (15.6)19 (14.8)9 (17.6) 2000–200965 (36.3)52 (40.6)13 (25.5) 2010–201943 (24.0)33 (25.8)10 (19.6)LN Class†, n (%) Class III51 (28.5)43 (33.6)8 (15.7) Class IV66 (36.9)43 (33.6)23 (45.1) Class V49 (27.4)31 (24.2)18 (35.3) Mixed (III±V, IV±V)13 (7.3)11 (8.6)2 (3.9)Serum creatinine (continuous)†, mg/dL, mean (SD)0.95 (0.34)0.96 (0.36)0.93 (0.28)eGFR†, mL/min/1.73 m2, mean (SD)94.4 (94.7)88.4 (37.3)109 (168) <60 at baseline, n (%)36 (20.1)29 (22.7)7 (13.7) ≥60 at baseline, n (%)143 (79.9)99 (77.3)44 (86.3)Urine protein levels, g/24 hours,†‡ mean (SD)§N=1672.5 (2.6)n=1182.5 (2.7)n=492.4 (2.4) ≥0.5 at baseline, n (%)148 (82.7)103 (80.5)45 (88.2) >0.7 at baseline,§ n (%)125 (74.9)84 (71.2)41 (83.7) >6.0 at baseline, n (%)14 (8.4)11 (9.3)3 (6.1)SDI score†, mean (SD)0.5 (1.0)0.4 (0.9)0.6 (1.2) 0 at baseline, n (%)132 (73.7)95 (74.2)37 (72.5)SLEDAI-2K score†, mean (SD)15.2 (8.2)16.0 (8.7)13.3 (6.4) ≥6 at baseline, n (%)163 (91.1)115 (89.8)48 (94.1) ≥8 at baseline, n (%)152 (84.9)109 (85.2)43 (84.3)Hypertension¶, n (%)92 (51.4)67 (52.3)25 (49.0)Diabetes¶, n (%)10 (5.6)9 (7.0)1 (2.0)Corticosteroids, n (%)167 (93.3)120 (93.8)47 (92.2)Any immunosuppressants, n (%)90 (50.3)64 (50.0)26 (51.0) Azathioprine57 (31.8)42 (32.8)15 (29.4) Mycophenolate49 (27.4)38 (29.7)11 (21.6) Cyclophosphamide7 (3.9)7 (5.5)0 Methotrexate8 (4.5)6 (4.7)2 (3.9) Cyclosporine5 (2.8)3 (2.3)2 (3.9)Antimalarials, n (%)94 (52.5)68 (53.1)26 (51.0)Baseline was defined as the date of the kidney biopsy.*Any time pre-index (Dday 730 (24 months]) post-baseline);.†cClosest data point to baseline with a time window of 3 months pre-baseline and post-baseline;.‡Or equivalent;.§Excludes patients with only urine protein dipstick, as actual proteinuria level could not be obtained with this assay.¶Closest data point to baseline with a time window of 24 months prebaseline.BMI, Body Mass Index; eGFR, estimated glomerular filtration rate; LN, lupus nephritis; mPERR, modified primary renal efficacy response; SDI, Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index; SLEDAI-2K, SLE Disease Activity Index 2000 Patient baseline characteristics Baseline was defined as the date of the kidney biopsy. Any time pre-index (Dday 730 (24 months]) post-baseline);. cClosest data point to baseline with a time window of 3 months pre-baseline and post-baseline;. Or equivalent;. Excludes patients with only urine protein dipstick, as actual proteinuria level could not be obtained with this assay. Closest data point to baseline with a time window of 24 months prebaseline. BMI, Body Mass Index; eGFR, estimated glomerular filtration rate; LN, lupus nephritis; mPERR, modified primary renal efficacy response; SDI, Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index; SLEDAI-2K, SLE Disease Activity Index 2000 Association between mPERR status at index and long-term renal survival during follow-up Achieving mPERR at index was associated with an increased likelihood of long-term renal survival versus not achieving mPERR (log-rank p=0.0119; figure 2A), with 78.9% of mPERR responders having long-term renal survival (mean (SD) follow-up time: 11.0 (8.5) years) compared with 60.8% of mPERR non-responders (mean (SD) follow-up time: 9.5 (8.4) years). The Cox regression analysis did not find any confounders apart from eGFR <60 mL/min/1.73 m2 at baseline (online supplemental table 2). After adjusting for eGFR at baseline (ie, biopsy date or closest datapoint to biopsy), mPERR responders had a significantly lower risk of ESKD progression or death than mPERR non-responders during the follow-up period (HR (95% CI) 0.44 (0.24, 0.79); p=0.0063; online supplemental table 2). Kaplan-Meier plot and log-rank tests to compare long-term renal survival according to mPERR (A) and mCRR (B) status at index. Index was defined as the 24 months after (day 730 (±3 months)) the date of the kidney biopsy. Long-term renal survival was defined as eGFR <30 mL/min/1.73 m2 (CKD stage 4 or 5), dialysis, transplant or death. CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; mCRR, modified complete renal response; mPERR, modified primary efficacy renal response. Association between mPERR at index and long-term renal survival without risk of mild CKD or worse during follow-up Achieving mPERR at index was associated with an increased likelihood of long-term renal survival without risk of mild CKD or worse versus not achieving mPERR (n=154; log-rank p<0.0001; figure 3A), with 86.4% of mPERR responders not having mild CKD (mean (SD) follow-up: 10.6 (8.6) years) during follow-up compared with 56.8% of mPERR non-responders (mean (SD) follow-up: 7.4 (8.0) years). In the Cox regression analysis, after adjusting for eGFR at baseline (ie, biopsy date or closest data point to biopsy), achieving the mPERR remained significantly associated with an increased likelihood of long-term renal survival without CKD than mPERR non-responders (HR (95% CI) 0.22 (0.11, 0.45); p=0.00002; online supplemental table 3). Kaplan-Meier plot and log-rank tests to compare long-term renal survival without risk of mild CKD or worse according to mPERR (A) and mCRR (B) status at index. Index was defined as the 24 months after (day 730 (± 3 months)) the date of the kidney biopsy. Long-term renal survival without risk of mild CKD or worse was defined as a new occurrence of GFR <30 mL/min/1.73 m2 (CKD stage 4 or 5) on ≥2 consecutive occasions ≥3 months apart. CKD, chronic kidney disease; GFR, glomerular filtration rate; mCRR, modified complete renal response; mPERR, modified primary efficacy renal response. Association between mCRR status at index and long-term renal survival during follow-up At index, of the 128 patients who were mPERR responders, 79.7% were also classified as mCRR responders, with a sequential increase in the cumulative frequency of patients meeting these criteria through month 36 (59/115; 51.3%), month 48 (81/105; 75.0%) and month 60 (82/101; 81.2%) postbaseline. Achieving mCRR at index was associated with an increased likelihood of long-term renal survival versus not achieving mCRR (log-rank p=0.0259; figure 2B), with 80.4% of mCRR responders (mean (SD) follow-up time: 10.8 (8.0) years) having long-term renal survival without ESKD progression or death compared with 64.9% of mCRR non-responders (mean (SD) follow-up time: 10.2 (9.1) years). The Cox regression analysis did not find any confounders apart from eGFR <60 mL/min/1.73 m2 at baseline (online supplemental table 4). After adjusting for eGFR at baseline (ie, biopsy date or closest datapoint to biopsy), mCRR responders had a significantly lower risk of ESKD progression or death than mCRR non-responders (HR (95% CI) 0.45 (0.24, 0.82); p=0.0098; online supplemental table 4). Association between mCRR at index and long-term renal survival without risk of mild CKD or worse during follow-up Similar to mPERR results, a comparable trend was observed when assessing mCRR responders versus mCRR non-responders (n=154; log-rank p=0.0016; figure 3B), with 86.0% of mCRR responders having long-term renal survival without risk of mild CKD or worse (mean (SD) follow-up: 10.7 (8.3) years) compared with 67.6% of mCRR non-responders (mean (SD) follow-up: 8.4 (8.8) years). The Cox regression analysis did not find any confounders apart from race (online supplemental table 5). After adjusting for race in this model, achieving the mCRR remained significantly associated with an increased likelihood of long-term renal survival without CKD (HR (95% CI) 0.31 (0.15, 0.64); p=0.00148; online supplemental table 5). Discussion LN represents a severe manifestation of SLE that, if not managed appropriately, may lead to CKD and ESKD.22 Long-term preservation of kidney function, prevention of renal flares and damage, as well as the improvement of health-related quality of life, are the key goals of LN management.23 Therefore, identifying predictors of long-term kidney function may allow for the implementation of treat-to-target approaches and better outcomes for patients with LN.13 23 24 In this real-world, retrospective study in patients with LN, we showed that achieving mPERR and mCRR was associated with a greater likelihood of long-term renal survival, highlighting these renal response outcomes as predictors of long-term renal survival. Consistent with the data reported by Cooper Blenkinsopp et al,17 we showed that mPERR was achieved 24 months after kidney biopsy by approximately three-quarters of patients with LN receiving standard therapy. Although achieving both mPERR and mCRR was associated with a significantly decreased risk of ESKD progression or death and a significantly enhanced probability of long-term renal survival, the mPERR definition criteria are less stringent than the mCRR criteria, making mPERR more achievable. This was evident from the greater proportion of patients achieving mPERR compared with those achieving mCRR in this study. Therefore, our study underscores the suitability of mPERR as an endpoint for predicting long-term renal outcomes. Interestingly, in our study, the clinical characteristics of mPERR non-responders at 24 months following a kidney biopsy highlight that the patients with lower disease control at 24 months have poorer long-term renal clinical outcomes compared with mPERR responders. This finding appears to be consistent across a diverse patient demographic with similar results reported in the Hopkins Lupus Cohort,17 highlighting that patients with lower disease control may be more likely to develop renal function deterioration over time. This reinforces the need to identify patients with lower disease control as early as possible to allow for more effective treatment intervention and prevention of kidney failure.6 13 23 As current EULAR recommendations for response to therapy for LN suggest reaching <500–700 mg protein/24 hours at 12–24 months,22 examining mPERR/mCRR achievement at earlier timepoints will be of interest for future research to identify the optimal timing for the assessment of remission. The findings of this study emphasise the potential generalisability of mPERR status as a predictor of long-term renal survival in patients with LN, with and without mild CKD progression. These data expand on previous findings over a median 6-year follow-up period17 to show that mPERR can predict long-term renal survival, encompassing definitions from mild CKD to ESKD or death. Additionally, the current study was based on a more diverse demographic population, increasing the robustness and generalisability of the mPERR and mCRR endpoints in predicting long-term renal outcomes results in a real-world setting. Our data also support mCRR as a measure for predicting long-term renal survival, despite being a more stringent set of criteria than the mPERR; over an approximate 11-year follow-up period, mCRR responders had a 55% greater likelihood of long-term renal survival, which increased to 69% when defining long-term renal survival by the absence of mild CKD disease progression. These findings support those from an analysis of the Hopkins Lupus Cohort, where responders meeting the same criteria were significantly less likely to experience progression to mild CKD, end-stage renal disease or death over time than non-responders.25 Our study is being reproduced in Hong Kong (GSK Study 216940) to provide equivalent data in Asia, with the aim to provide further evidence on whether both mPERR and mCRR are achievable measures for predicting long-term renal survival across a diverse demographic of patients with LN receiving standard therapy.26 Considering the real-world nature of this study, there are some associated limitations to acknowledge. First, as this study used data from the long-standing Toronto Lupus Cohort database, where data have been collected since 1970, any changes in clinical management and laboratory testing may have introduced bias over time. However, as the focus of this analysis is on the ability of mPERR or mCRR to predict renal survival, we do not believe that how or when the outcome was achieved should affect the predictive ability. Additionally, biopsy data were explored for possible temporal confounding in the regression model, with no significant confounding observed. There was a large proportion of missing data from comorbidities at baseline as data capture may have occurred prior to the enrolment into the Toronto Lupus Cohort; thus, the potential confounding effect on study objectives could not be further assessed for these variables. In addition, as this study was observational in nature, no treatment restrictions were implemented, which may have influenced outcomes over time. Also, due to the semiquantitative nature of urine dipstick measurements, available data from patients with only this measurement are limited, and the available values from a small number of patients who had only urine dipstick measures were not in the range that could have interfered with quantitative comparison. Finally, the results beyond 10 years of the follow-up period should be interpreted with caution owing to the small patient sample sizes for these time periods. Regardless of these limitations, our study adds to the body of literature supporting the use of mPERR and mCRR as prognostic measures for long-term renal survival, and their implementation in routine clinical practice may enable earlier treatment intervention and avoidance of irreversible kidney damage and ESKD for patients with LN. Future research examining earlier timepoints for mPERR and mCRR achievement may be worth investigating to obtain the optimal time for remission assessment. supplementary material