研究背景 原发性膜性肾病(idiopathic membranous nephropathy, IMN)是原发性肾小球疾 病的重要组成部分，其发病比例逐年上升，而环孢素A(cyclosporine A, CSA)是 IMN的重要治疗手段之一。既往研究表明，CSA在IMN中治疗缓解率可达60-90%，但其对肾功能的影响是目前临床应用的最大顾虑。现有研究关于CSA在 IMN中的研究纳入样本较少，对CSA治疗导致的尿蛋白缓解和对肾功能的影响预 测因素探究不足，有关于CSA对IMN的治疗反应、远期预后及相关影响因素尚需 进一步探究。 研究目的 1. 评估IMN治疗中的疗效预测因素。 2. 与其他免疫抑制剂比较，评估CSA在IMN治疗中的有效性及其对肾功能的影响。 3. 探查CSA治疗时影响疗效和肾功能的因素。 4. 探索糖尿病合并IMN与糖尿病肾病(DN)鉴别的临床指标及CSA在其治疗中的作用。 研究方法 回顾性收集了 875名于2004年1月至2015年4月经病理确诊的IMN成人患 者基线及随诊的临床和病理资料，使用Logistic或COX回归模型：1)通过与其他免 疫抑制剂治疗比较，评估CSA治疗IMN对尿蛋白缓解、复发以及估测的肾小球滤 过率(eGFR)较基线值 ≥30%的作用；2)在使用CSA队列中分析IMN尿蛋白缓 解、复发及eGFR较基线值下降 ≥30%的影响因素。同时，也对糖尿病合并IMN的 临床特点和治疗情况进行了回顾和总结。 研究结果 1. 875名入组患者男性占55.4%，平均年龄48.2±14.8岁，肾病综合征(nephrotic syndrome, NS)比例55.7%。接受免疫抑制剂治疗患者总数743人，包括251例 (33.8%)未达到NS患者，主要免疫抑制剂种类为CTX(37.6%)和CSA(31.9%)。随访患者比例90.2%，共发生完全缓解(complete remission, CR) 事件447例及部分缓解(partial remission, PR)事件192例，总缓解率81.0%;复 发事件129例，分别来自PR57例和CR72例；eGFR下降 ≥30%共发生188例。 2. CSA治疗与IMN尿蛋白缓解情况：1)初始治疗前24h-UP >8g是IMN患者 (OR=0.55, p=0.018)和CSA治疗(HR=0.33, p=0.003)首次CR发生的危险因素； 2)联合糖皮质激素治疗(OR=3.71, p=0.001)和RASi治疗(OR=2.35, p=0.002)是 IMN患者首次CR的独立预测因素；3)初始治疗方案选择CSA和CTX在CR 上无显著性异(HR=1.32， p=0.12); 4)治疗1月内出现PR是CSA治疗首次CR 的预测因素(HR=2.76, p=0.004)。 3. CSA治疗与IMN尿蛋白复发情况：1)初始治疗为CSA或CTX在尿蛋白复发 上无显著性差异(OR=1.50, p=0.21); 2)初始治疗前24-UP >8g是CSA治疗复发的独立危险因素(OR=2.71, p=0.014)。 4. CSA治疗与IMN肾功能下降情况：1))使用CSA治疗是eGFR下降≥30%的独 立危险因素(OR=4.40, p<0.001); 2)年龄 ≥49岁(HR=4.16, p=0.006)和CSA治疗 疗程 >12月(HR=3.25, p=0.041)是CSA治疗中eGFR下降 ≥30%的独立危险因素. 5. 糖尿病(Diabetic Mellitus, DM)合并IMN患者的临床特点及治疗情况：1)在糖 尿病合并肾病综合征(NS)的患者中，eGFR >60.3ml/min/1.73m2是病理诊断为 IMN而非DN的重要预测指标；2) CSA和CTX初始治疗对DM+IMN患者首 次CR的影响无显著性差异；3)与CTX治疗相比，CSA治疗eGFR下降 ≥30% 发生的可能性显著较高(p=0.009)。 研究结论 1. 治疗前24-UP>8g是IMN尿蛋白缓解的不利因素和复发的危险因素，IMN的 免疫抑制剂治疗或许应在24h-UP≤ 8g或更早开始。 2. IMN治疗中，糖皮质激素及RASi使用均有利于首次CR发生。 3. CSA初治治疗第1个月内达到PR是将来CR的独立预测因素，提示CSA治疗早期尿蛋白下降幅度或许可用于疗效的预测及后期药物的调整。 4. 作为IMN的初始治疗方案，CSA和CTX在尿蛋白CR及复发上无显著性差异。 5. CSA治疗是肾功能下降的独立危险因素，年龄≥49岁和CSA疗程 >12月可能 是CSA治疗中eGFR下降的影响因素。

Background Idiopathic membranous nephropathy(IMN) is an important glomerulonephropathy leading to nephrotic syndrome(NS). The incidence of IMN has increased rapidly in recent years. Cyclosporine A(CSA) is an established option for treatment in IMN patients, which has achieved cumulative remission in up to 60%-90% of patients. The major concern with CSA remains its propensity to induce renal function impairment. The samples of researches before studying IMN patients treated with CSA were usually small. Few studies had investigated the prognostic factors of the treatment outcomes of CSA in IMN patients. Aims 1. Assess the prognostic factors of treatment effect in IMN patients. 2. Assess the influence of CSA on remission, relapse and renal function through comparison with other immunosuppressants. 3. Study the prognostic factors which can predict remission, relapse and decreased renal function in IMN patients treated with CSA. 4. Study the factors using to distinguish IMN and DN in diabetic mellitus(DM), and to assess the effect of CSA treatment in diabetic patients with IMN. Method We retrospectively collected the clinical information of 875 adult patients who were diagnosed as IMN through renal biopsy. The multivariate regression models were used to 1) assess the relationship between CSA and remission, relapse and renal function impairment in IMN patients compared with other immunosuppresants; 2) find out the prognostic factors of remission, relapse and renal function impairment in patients treated with CSA. We also studied the clinical manifestations and treatment effects of IMN against a diabetic background. Result 1. There were in total 875 patients in our study with an average age of 48.2±14.8 years old, of whom 485 were males (55.4%). Over half of these patients were suffered from NS at baseline. Immunosuppressants were used in 743 patients, including 251 patients who did not fulfill the diagnostic criteria of NS. Cyclophosphamide(CTX) and CSA were two major immunosuppressive options for IMN patients. 789 patients were followed up, with a culmulative remission rate of 81.0%, of them 447 and192 achieving complete remission(CR) and patial remission(PR), respectively. 129 patients relapsed after entering remission, and the estimated glomerular filtration rate(eGFR) of 188 patients had decreased by at least 30% during the follow-up period. 2. The relationship between CSA treatment and remission of IMN patients. 1) 24-hour urine total protein (24h-UP)>8g was an independent risk factor for failure in achieving CR in IMN patients(OR=0.55, p=0.018) and those treated with CSA(HR=0.33, p=0.003). 2) Corticosteroids(OR=3.71, p=0.001) and RASi(OR=2.35, p=0.002) can predict the initial CR in IMN patients. 3) Compared with CTX, CSA was not significantly inferior in achieving CR(HR=1.32, p=0.12); 4) Reaching PR in the first month is a predictor of further CR in patients treated with CSA(HR=2.76, p=0.004). 3. The relationship between CSA treatment and relapse in IMN patients. 1) The relapse rates did not differ significantly between pateints treated with CTX and those with CSA(OR=1.50, p=0.21). 2) 24-UP>8g was an independent predictor for relapse(OR=2.71, p=0.014). 4. The relationship between CSA treatment and descending renal function in IMN patients. 1) CSA treatment was an independent risk factor of eGFR decline(OR=4.40, p<0.001). 2) Age over 49 years old(HR=4.16, p=0.006) and the duration of CSA treatment exceeding 12 months(HR=3.25, p=0.041) were independently associated with risks of eGFR decrease over 30% at the 24th month. 5. The clinical characteristics and treatment of IMN in diabetic mellitus patients. 1) eGFR>60.3ml/min/1.73m2 may be an indicator of IMN in diabetic patients suffered from NS. 2) No significant difference was found between CSA and CTX treatment in achieving CR in IMN patients with diabetes. 3) Compared with those using CTX, IMN patients treated with CSA may be easier to see a 30% drop in eGFR against the background of diabetic millitus(p=0.009). Conclusion 1. We suggest to begin immunosuppressive treatment before 24h-UP reaching 8g because it meant poor reaction to treatment and high probability of relapse. 2. Corticosteroids and RASi could predict the remission of proteinuria. 3. We found that achieving PR in the first month may be a predictor of further CR in patients treated with CSA. Therefore, the declining magnitude of 24h-UP in the first month might be an indicator of treatment effect and adjustment. 4. The remission rates and relapse rates did not differ significantly between patients treated with CTX and CSA. 5. Using CSA was an independent factor indicating eGFR decreasing by at least 30%. Age over 49 years old and the treatment duration of CSA exceeding 12 months were two prognostic factors of eGFR declining over 30%.