Outcomes of the Re-classification of WHO Class IV Lupus Nephritis Using the ISN / RPS Classification : a Single Center Retrospective Observational Study from the JUDE Study

Objective: The International Society of Nephrology-Renal Pathology Society (ISN/RPS) 2003 classification of lupus nephritis (LN) was designed to provide beneficial pathologic information relevant to the renal outcome. We conducted a retrospective observational study to investigate the baseline characteristics and renal response to treatment for patients with LN, comparing the World Health Organization (WHO) and ISN/RPS classification systems. Materials: A total of 39 Japanese patients (2 men; 37 women) with LN who underwent percutaneous needle renal biopsy between 1998 and 2012 were evaluated. Methods: Renal biopsy samples were classified using the 1995 WHO and 2003 ISN/RPS criteria. Results: Among WHO class IV patients, a higher number of patients reclassified into ISN/RPS class III achieved complete response to treatment compared to those reclassified into class IV at 6 months follow-up. Twenty patients in WHO class IVc were reclassified into ISN/RPS classes III, III+V, IV-S, IV-S+V, IV-G and IV-G+V. No patients developed end-stage renal failure requiring renal replacement therapy. Conclusions: The results suggest that the ISN/RPS classification system is more advantageous in predicting renal outcome and guiding treatment, especially for those previously classified with WHO class IVc LN.


Introduction
Systemic lupus erythematosus (SLE) is a clinically and serologically diverse multisystem autoimmune disease 1) .SLE may affect any organ, but mainly affects the skin, joints, blood cells, nervous system, and kidneys.Renal involvement is the most important risk factor associated with a poor prognosis in SLE.Lupus nephritis (LN) occurs in approximately 50-60% of patients with SLE 2) 3) .LN plays a key role in the prognosis of SLE and nearly 10% of patients with LN develop end-stage renal failure requiring dialysis or renal transplantation 4) -6) .Moreover, LN and the chronic use of corticosteroids and immunosuppressive agents contribute significantly to mortality.The world Health Organization (WHO) classification for LN was introduced in 1974, defined in 1982, and revised in 1995.The evolution of the WHO classification did not satisfy needs as the different categories and diagnostic terminology required clarification.Thus, the International Society of Nephrology/Renal Pathology Society (ISN/RPS) 2003 classification of LN was designed 7) .The most common classswitching events have been reported to be from WHO class III to ISN-RPS class IV 8) .Moreover, using this new classification system for LN, numerous studies have examined the differential outcomes of classes IV-S and IV-G.
In the present study, we conducted a retrospective observational study to investigate the baseline characteristics and the renal response to treatment, comparing the WHO and ISN/RPS classification systems.Our data demonstrate that the ISN/RPS classification system is more informative in predicting renal outcome, particularly for WHO class IVc.

Patients
This study received ethical approval from the institutional ethics committee of Juntendo University (No. 20-72-2).The study participants provided informed consent, and the study was conducted in accordance with the 2008 Declaration of Helsinki.SLE patients who had a visiting history were entered into the Juntendo University Database of Erythematosus (JUDE) at the Department of Internal Medicine and Rheumatology, Juntendo University Faculty of Medicine.Patients with LN who underwent percutaneous needle renal biopsy between 1998 and 2012 at the Juntendo University Hospital were included.All patients met the 1982 American College of Rheumatology (ACR) revised criteria for the classification of SLE 9) .Patients were required to have a follow-up period >6 months and adequate renal biopsy samples for histological diagnosis, including >10 glomeruli.Renal re-biopsy was performed in two patients with a clinical relapse of nephritis.For patients with repeated biopsies, the samples and clinical records at the time of the first biopsy were used in the analyses.End-stage renal disease (ESRD) was defined as the need for introducing hemodialysis, peritoneal dialysis, or renal transplantation.

Histological diagnosis
Renal biopsy samples were histologically re-classified according to the 1995 WHO classification system and the ISN/PPS 2003 criteria.The time of LN diagnosis, defined as the first diagnostic renal biopsy, was the starting point for determining the follow-up period.The histological diagnosis was reviewed by four doctors including one pathologist, one rheumatologist, and two nephrologists, who were blind to the clinical information, respectively.

Clinical data
The collection of clinical data was performed in July 2013.Clinical information and laboratory data were obtained every 6-12 months during the follow-up period and included serum complement levels (C3, C4), complement hemolytic activity (CH50), anti-DNA antibodies, serum creatinine (sCr), serum albumin, and proteinuria.Urinary protein was assessed using a dipstick test or the collection of urine for 1 day.Nephrotic syndrome was defined as an increase in the urinary protein ratio ≥3.5 g/gCr or 3.5 g/day and a decrease in serum total protein to ≤6.0 g/dl, or serum albumin ≤3.0 g/dl.The estimated glomerular filtration rate (eGFR) at the renal biopsy was compared with that at the last examination.The eGFR was calculated using formulas for the Japanese population as defined by the Japanese Society of Nephrology (Male, eGFR = 194 x sCre -1.094 x age -0.287 ; Female, eGFR = 194 x sCre -1.094 x age -0.287 x 0.739) 10) .

Definitions
Renal response at 6 months was classified as a complete response (CR), partial response (PR), or no response (NR).CR was defined as sCr<1.4mg/dl, urine protein <500 mg/day or trace by dipstick test, the absence of activity urinary sediment (≤10 red blood cells (RBC) /high power field (HPF)), and no casts.PR was defined as sCr<1.4mg/dl and urinary protein decreased by 50% or more from baseline or 1+ to 2+ by dipstick test.NR was defined as a deterioration in renal function exclusive of other causes (such as sepsis or nephrotoxic agents), a reduction in urinary protein with failure to reach PR, or the persistence of active urinary sediments (<10 RBC/HPF) 8) .Renal flare was defined as an increase in urinary protein of 1 g/gCr associated with an increase in serological activity.
Renal function at the last follow-up was classified into 3 groups based on the eGFR: last eGFR <-20% of the first eGFR at the renal biopsy, -20% first eGFR <last eGFR <+20% first eGFR, +20% first eGFR <last eGFR.

Statistical analysis
Statistical analyses were performed using R (Foundation for Statistical Computing).Baseline characteristics are given as the proportion and mean±standard deviation.Intergroup comparisons were performed using the Kruskal-Wallis test.Two-group comparisons were performed using two-sample t tests.A p-value of <0.05 was considered statistically significant.

Patient profile
Patient profiles are shown in Table -1.Of the 39 patients, 37 (94.9%) were women.The mean age was 30.6 years at SLE diagnosis and 35.1 years at renal biopsy.The median observation period was 36 months.Twenty-three patients (59.0%) had already been treated for SLE by the time of renal biopsy.

Pathological classification
The histological re-classifications of the renal

Clinical parameters of ISN/RPS class III and IV
Clinical data at renal biopsy, stratified by ISN/RPS class III and IV, is summarized in Table-3.In all cases of ISN/RPS class III and IV, renal function was preserved.Urinary protein was significantly lower in class III group (n = 4; III(A) and III(A/C)) compared to that in class IV group (n = 14; IV-S (A/C), IV-G (A), and IV-G (A/C)) (p<0.05, Figure -1).Especially, urinary protein was significantly lower in class III (A/C) group compared to that in class IV-G(A/C) group (p<0.05).Although renal function was maintained in both classes, the patients in Class IV (S(A)+G (A/C))+V had the highest urinary protein (3.6 g/gCr) but also had lower anti-DNA antibodies (53.8 IU/ml).In addition, the percentage of patients with nephrotic syndrome in class IV-S (A) +V, the percentage was 100%.

Renal response to treatment and other followup data
The treatment regimen for WHO class IVc during the first year after renal biopsy was assessed.Table -  ＊ reclassified into ISN/RPS class IV, 69.2% received immunosuppressive agents after renal biopsy as follows: azathioprine, 3; intravenous cyclophosphamide (IVCY), 5; tacrolimus, 2; rituximab, 1; and oral cyclophosphamide, 1.Among the 5 patients in ISN/RPS class IV who received IVCY, the renal response was a CR in 2 cases, PR in 2 cases, and NR in 1 case.For all ISN/RPS classes, the serum creatinine level was almost unchanged (ΔsCr= sCre at the last follow up -sCre at the biopsy) and was as follows: ISN/RPS class III, 0.01; III+V, 0.08; IV, -0.03; and IV+V, -0.13.

Discussion
To best guide clinical treatment and predict the prognosis of LN, determining the correlation between clinical manifestations and pathological characteristics is critical.Therefore, a new LN classification system with clearer definitions was of the total number of glomeruli) while class IV involves diffuse glomerulonephritis (≧50% of the total number of glomeruli).One of the most notable changes between the WHO and ISN/RPS classification systems is that diffuse LN (WHO class IV) was subdivided into diffuse LN with segmental (class IV-S, including focal segmental necrotizing lesions in less than 50% of glomeruli) or global involvement (class IV-G, including focal segmental necrotizing lesions in more than 50% of glomeruli).Numerous reports exist regarding potential differences in renal outcomes between classes IV-S and IV-G 11) -13) .In the present study, changes in eGFR from the renal biopsy to the last examination did not differ between classes IV-S and IV-G.Hill et al. found that the pathological tissue in class IV-G (n = 31) had greater overall immune deposits and subendothelial deposits on immunofluorescence, and greater hyaline deposits on light microscopy, compared to those in class IV-S (n = 15) 11) .Furthermore, there were tendency for worse proteinuria in class IV-G.On the other hand, the tissue in class IV-S presented with proportionally greater glomerular fibroid necrosis and a lack of correlation with the extent of immune deposits.The authors concluded that the pathogenesis may qualitatively differ between classes IV-G and IV-S 11) .
Yokoyama and coworkers reported that patients in ISN/RPS classes IV-S or IV-G at the initial biopsies showed a higher rate of ESRD as compared to that for other groups (40.9% vs 2.6%, respectively).Furthermore, the mean 50% renal survival time for patients in classes IV-S and IV-G was 95±22 months and 214±35 months, respectively.However, a lack of significant differences between classes IV-S and IV-G has been reported 13) .Mittal and coworkers reported that the percentage of glomeruli with cellular crescents was greater in the ISN/RPS class IV-S group (n = 11) while a wire loop lesion was more common in the IV-G group (n = 22) 14) .However, no significant group differences in outcomes were detected.Similarly, Hwang and coworkers reported that among 43 patients with WHO class III LN, 12 patients were reclassified into ISN/RPS class IV (class IV-S, n = 9; class IV-G, n = 3) 8) .At one-year follow-up, there were no group differences in the response to treatment.These results suggest that distinctive outcomes for classes IV-S and IVs-G still remain to be ascertained.
Class-switching events from WHO class III to ISN/RPS class IV have been reported to be the most common class-switching event 15) .A point worthy of special mention regarding the present study is that patients in WHO class IVc were reclassified into 6 different classes.These data suggest the importance of adequately assessing endocapillary and/or extracapillary proliferation, and glomerular basement membrane when using the ISN/RPS classification.Moreover, the ISN/RPS classification system was more advantageous in predicting the outcome and had a higher correlation with clinical and pathological presentation compared to that for the WHO classification system, especially for those in WHO class IVc.Although we clearly determined the clinical significance of the ISN/RPS classification system in comparison with the WHO classification system, we were not able to completely predict the renal prognosis using only the ISN/RPS class as the outcomes sometimes differed among cases in the same class.Thus, new pathological parameters which predict the prognosis of LN more accurately than the ISN/RPS class are needed.
In the present study, the renal response to treatment was comparatively poor in ISN/RPS class IV.Treatment using cyclophosphamide combined with glucocorticoid has been the standard induction therapy for WHO Class IV LN 16)-18) .However, cyclophosphamide was administered to only 35% of the patients in ISN/RPS class IV.The less frequent use of cyclophosphamide may be the cause of the poor renal response in ISN/RPS class IV.In regard to treatment, recently mycophenolate mofetil (MMF) became available in Japan.The ACR and Asian Lupus Nephritis Network guidelines recommend either IVCY regimen or MMF combined with glucocorticoids for induction in patients with class III and IV lupus nephritis 19) 20) .In Asian individuals, a reduced dose of MMF (2 g daily) is recommended 20) .The pooled data from literature and anecdotal experience seem to show that MMF is inferior to cyclophosphamide in the long term preservation of renal function 21) .Therefore, MMF should consider as a choice of the treatment.The goals of lupus nephritis treatment are to induce remission, reduce flares, preserve kidney function and minimize vascular or treatment related complications.Although aggressive management may effectively reduce the activity of LN, the risk of a life-threatening infection may be increased.The main cause of death in SLE patients is infection 22) .Therefore, attention should be paid to not only the treatment, but also to infections at the time of treatment.However, no patients developed ESRD or death in the present study.
This study has several limitations, mainly stemming from the retrospective design, single-center, small sample size, and short observational period.The baseline clinical characteristics in the sample were quite variable, and not all the induction therapies were standard regimens.The IVCY doses were reduced in some patients because of safety concerns.Some patients refused the use of IVCY.Therefore, the results of outcome in this study may not be applicable to the other population that received the standard regimens.

Conclusion
We retrospectively analyzed 39 subjects with biopsy-proven LN and found that the new ISN/RPS 2003 classification system provided beneficial pathologic information relevant to the short-term renal outcome and the optimal therapy preventing ESRD and/or death in patients with LN.The revised ISN/RPS classification appears to be particularly more useful in prognostic and use of optimal treatment agents for cases previously classified as WHO class IVc.

Figure- 1
Figure-1 Urinary protein levels compared to ISN/RPS class III and class IV IV WHO class IV LN had the highest rate (35.9%; 22/39).Of the 22 patients in WHO class IV, 3 (13.6%)were reclassified into ISN/RPS class III while 19 (86.4%) remained within ISN/RPS class IV (Figure-2).Of the 19 patients remaining in ISN/RPS class IV, 2 patients were reclassified into class IV-S, 12were into class IV-G, and 5 were into class IV+V.Among the 2 patients reclassified into ISN-RPS class IV-S, 1 CR and 1 PR were achieved.In contrast, among the 12 patients reclassified into ISN-RPS class IV-G, 7 CR, 4 PR, and 1 NR were achieved.Among the 22 patients in WHO class IV, 20 were specifically classified as WHO class IVc.Among these 20 patients, 2 patients (10%) were reclassified into ISN/RPS class III (class III, N = 1; class III+V, N = 1) while 18 were reclassified into ISN/RPS class IV (class IV-S, N = 1; class IV-S+V, N = 1; class IV-G, N = 12; and class IV-G+V, N = 4).The efficacy of treatment was lower for the newly reclassified patients in ISN/RPS class IV or IV+V (0 NR among 3 total in class IV-S and 2 NR among 16 total in class IV-G) compared to that for newly reclassified patients in ISN/RPS class III or III+V (0 NR among 3 total).This statement is still valid after excluding patients with ISN/RPS combined phenotypes (IV+V or III+V)(Figure-3).

Figure- 3 Figure- 2
Figure-3 The renal response to treatment at 6 months after treatment is shown for patients in WHO class IVc WHO: World Health Organization, ISN/RPS: International Society of Nephrology/Renal Pathology Society, CR: complete response, PR: partial response, NR: no response

Table - 1
Baseline characteristics at renal biopsy

Table - 2
Re-classifications of the renal biopsy samples according to the 1995 WHO classification and the ISN/PPS 2003 criteria biopsy samples according to the 1995 WHO classification and the ISN/PPS 2003 criteria are shown inTable-2.The ISN/RPS classification involves six phenotypic types; however, in this study, some patients presented with combined types, such as III+V-type LN and IV+V-type LN.

Table - 3
4 shows the detailed individual clinical background and treatment data of 20 patients with WHO class IVc LN.Prednisolone was prescribed in all WHO class IVc patients.In addition, 13 patients in WHO class IVc received immunosuppressive agents after renal biopsy.All patients in reclassified into ISN/RPS classes III and III+V received tacrolimus.Of the 13 cases in Clinical data of ISN/RPS Class III and IV at renal biopsy Data are expressed as mean±S.D. or percentage.s-Cre; serum creatinine, u-Pro; urinary protein, s-Alb; serum albumin, NS; nephrotic syndrome, normal range of C3, C4, CH50 and anti-DNA antibody are 69-128, 14-36, 25-54 and <6.0 respectively.＊ vs class IV, ＊＊ vs class IV-G (A/C) ＊ , ＊＊ p<0.05

Table - 4
7)yashita, et al: Re-classification of WHO class IV lupus nephritis using the ISN/RPS classification Individual data of patients with WHO class IVc reclassified according to ISN/RPS classification proposed to accommodate clinicopathological insights by the ISN/RPS7).Importantly, the ISN/RPS 2003 classification is much easier for the diagnosis of LN with little discrepancy in the judgment of classes III and IV.Class III involves focal glomerulonephritis (<50%