Validation of Prognostic Index for Allograft Outcome in Kidney Transplant Recipients With Transplant Glomerulopathy

Transplant glomerulopathy (TG) is a histological lesion of kidney allograft characterized by thickening or duplication of glomerular basement membrane, double contour formation, and mesangial interposition seen on light and electron microscopy.1Remport A. Ivanyi B. Mathe Z. et al.Better understanding of transplant glomerulopathy secondary to chronic antibody-mediated rejection.Nephrol Dial Transplant. 2015; 30: 1825-1833Crossref PubMed Scopus (30) Google Scholar It is commonly associated with chronic antibody-mediated rejection (cAMR) and is often attributed to chronic microvascular injury.1Remport A. Ivanyi B. Mathe Z. et al.Better understanding of transplant glomerulopathy secondary to chronic antibody-mediated rejection.Nephrol Dial Transplant. 2015; 30: 1825-1833Crossref PubMed Scopus (30) Google Scholar It has an extremely poor prognosis, resulting in kidney allograft failure within a year after diagnosis in a large number of affected patients.1Remport A. Ivanyi B. Mathe Z. et al.Better understanding of transplant glomerulopathy secondary to chronic antibody-mediated rejection.Nephrol Dial Transplant. 2015; 30: 1825-1833Crossref PubMed Scopus (30) Google Scholar It is estimated that approximately 5000 allografts are lost each year in the United States, primarily from TG and cAMR.2Loupy A. Hill G.S. Jordan S.C. The impact of donor-specific anti-HLA antibodies on late kidney allograft failure.Nat Rev Nephrol. 2012; 8: 348-357Crossref PubMed Scopus (242) Google Scholar There are several potential options to treat TG, including plasmapheresis (PLEX), i.v. Ig, rituximab, bortezomib, and tocilizumab, or a combination thereof.3Abreu R. Carvalho F. Viana H. et al.Morphologic patterns and treatment of transplant glomerulopathy: a retrospective analysis.Clin Transplant. 2017; 31Crossref PubMed Scopus (6) Google Scholar, 4Wan S.S. Ying T.D. Wyburn K. et al.The treatment of antibody-mediated rejection in kidney transplantation: an updated systematic review and meta-analysis.Transplantation. 2018; 102: 557-568Crossref PubMed Scopus (57) Google Scholar, 5Moreso F. Crespo M. Ruiz J.C. et al.Treatment of chronic antibody mediated rejection with intravenous immunoglobulins and rituximab: a multicenter, prospective, randomized, double-blind clinical trial.Am J Transplant. 2018; 18: 927-935Crossref PubMed Scopus (64) Google Scholar, 6Eskandary F. Regele H. Baumann L. et al.A randomized trial of bortezomib in late antibody-mediated kidney transplant rejection.J Am Soc Nephrol. 2018; 29: 591-605Crossref PubMed Scopus (103) Google Scholar, 7Choi J. Aubert O. Vo A. et al.Assessment of tocilizumab (anti-interleukin-6 receptor monoclonal) as a potential treatment for chronic antibody-mediated rejection and transplant glomerulopathy in HLA-sensitized renal allograft recipients.Am J Transplant. 2017; 17: 2381-2389Crossref PubMed Scopus (144) Google Scholar It is essential to determine the risk of allograft failure, using prognostic tools, in individual patients before subjecting patients to intensified immunosuppressive treatment measures, which have questionable benefits. There have been multiple risk stratification tools or indices developed in the last decade, with limitations and concerns regarding practical application of such tools. Some examples include a prognostic index developed by Patri et al. and the iBox risk prediction scoring developed by the Paris group.8Patri P. Seshan S.V. Matignon M. et al.Development and validation of a prognostic index for allograft outcome in kidney recipients with transplant glomerulopathy.Kidney Int. 2016; 89: 450-458Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar,9Loupy A. Aubert O. Orandi B.J. et al.Prediction system for risk of allograft loss in patients receiving kidney transplants: international derivation and validation study.BMJ. 2019; 366: l4923Crossref PubMed Scopus (66) Google Scholar It is imperative to develop tools based on easily available clinical and histopathological factors to predict allograft failure in patients with TG. This has been previously demonstrated by Patri et al.8Patri P. Seshan S.V. Matignon M. et al.Development and validation of a prognostic index for allograft outcome in kidney recipients with transplant glomerulopathy.Kidney Int. 2016; 89: 450-458Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar However, the main limitation of the Patri et al. study was its external validation, being conducted only in a French cohort that is not representative of the U.S. population. The aim of our study was to externally validate a previously developed TG prognostic score by the Patri et al.8Patri P. Seshan S.V. Matignon M. et al.Development and validation of a prognostic index for allograft outcome in kidney recipients with transplant glomerulopathy.Kidney Int. 2016; 89: 450-458Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar with a cohort of kidney transplant recipients of whom a majority are African Americans. Our hypothesis was that this transplant score has an excellent discrimination statistic in our cohort and can be used for prediction in these patients. Of the 38 recipients, 16, 14, and 8 had high-risk, intermediate-risk, and low-risk scores, respectively (Figure 1). As shown in Table 1, the mean age at the time of biopsy was 41 ± 17 years, 66% were male, and 61% were African American. The recipients with higher TG scores were significantly younger and also had worse graft function and proteinuria at the time of diagnosis (Table 1). The distribution of the histopathological features in the entire group is shown in Figure 2, and the distribution between different risk groups is shown in Supplementary Figures S1−S3.Table 1Baseline characteristics of the patientsCharacteristicsTotal cohort (N = 38)Low-risk group (n = 8)Intermediate-risk group (n = 14)High-risk group (n = 16)P valueDemographicsAge, yr, mean (SD)41 (17)46 (21)49 (12)30 (15)0.008Sex, n (%)0.564 Male25 (66)6 (75)10 (71)9 (56) Female13 (34)2 (25)4 (29)7 (44)Race, n (%)0.081 White13 (34)5 (63)2 (14)6 (38) African American23 (61)2 (25)12 (86)9 (56) Asian2 (5)1 (13)01 (6)Marital status, n (%)0.223 Divorced1 (3)001 (6) Married15 (39)5 (63)5 (36)5 (31) Single21 (55)2 (25)9 (64)10 (63) Widowed1 (3)1 (13)00Insurance, n (%)0.331 Medicare27 (71)5 (63)12 (86)10 (63) Tenncare2 (5)002 (13) Other9 (24)3 (38)2 (14)4 (25)Comorbidities, n (%) Diabetes17 (46)3 (43)9 (64)5 (31)0.191 Hypertension37 (97)8 (100)14 (100)15 (94)0.494 CAD3 (8)0 (0)1 (7)2 (13)0.559Time between transplantation and biopsy, d, median (IQR)2051 (1123−4602)1286 (1098−3720)1961 (674−3449)3383 (1927−6225)0.059Dialysis vintage, d, median (IQR)1002 (454−2685)1074 (594−2618)533 (431−2752)1215 (613−2324)0.790Laboratory parametersSerum creatinine, mg/dl, mean (SD)2.74 (1.12)1.99 (0.66)2.53 (0.84)3.30 (1.27)0.006UPCR, median (IQR)1.96 (1.02−4.30)0.55 (0.27−1.00)1.48 (1.02−1.92)4.95 (3.84−6.97)<0.001Body mass index, kg/m2, mean, (SD)27.5 (7.3)26.6 (12.4)28.8 (4.9)26.6 (5.3)0.239Maintenance immunosuppressive therapy, n (%) Tacrolimus36 (95)8 (100)14 (100)14 (88)0.234 Cyclosporin3 (8)1 (13)02 (13)0.387 mTOR inhibitors1 (3)001 (7)0.494 Prednisone36 (95)8 (100)14 (100)14 (88)0.234 Mycophenolate mofetil25 (66)4 (50)11 (79)10 (63)0.372 Mycophenolic acid23 (61)6 (75)10 (71)7 (44)0.194 Azathioprine3 (8)1 (13)0 (0)2 (13)0.387Treatment for TG, n (%) Received treatment26 (68)7 (88)9 (64)10 (63)0.424 PLEX22 (58)6 (75)8 (57)8 (50)0.503 i.v. Ig25 (66)6 (75)9 (64)10 (63)0.822 Rituximab2 (5)1 (13)0 (0)1 (6)0.483 Bortezomib3 (8)1 (13)2 (14)0 (0)0.303 Thymoglobulin9 (24)2 (25)4 (29)3 (19)0.815 Tocilizumab5 (13)3 (38)2 (14)0 (0)0.037 Steroid29 (76)5 (63)13 (93)11 (69)0.176 ACEI/ARB28 (74)5 (63)11 (79)12 (75)0.704ACEI, angiotensin-converting enzyme inhibitor, ARB, angiotensin receptor blocker; CAD, coronary artery disease; IQR, interquartile range; mTOR, mammalian target of rapamycin; PLEX, plasma exchange;TG, transplant glomerulopathy; UPCR, urine protein:creatinine ratio. Open table in a new tab Figure 2Histopathological characteristics of kidney transplant recipients with transplant glomerulopathy.View Large Image Figure ViewerDownload Hi-res image Download (PPT) ACEI, angiotensin-converting enzyme inhibitor, ARB, angiotensin receptor blocker; CAD, coronary artery disease; IQR, interquartile range; mTOR, mammalian target of rapamycin; PLEX, plasma exchange;TG, transplant glomerulopathy; UPCR, urine protein:creatinine ratio. The primary outcome of interest was graft loss within 3 years following the diagnosis of TG. During a median follow-up period of 0.83 years (minimum−maximum, 0.04−8.15 years), a total of 21 (55%) graft losses occurred (crude incidence rate, 368/1000 patient-years; 95% confidence interval [CI], 240−564). Figure 3a shows the graft survival probability over time. Close to 75% of our recipients lost their graft within 3 years after diagnosis of TG (Figure 3a). The crude mortality rate was significantly different between groups, as shown in Figure 3B. The lowest incidence rate of graft loss (n = 2; 25%) occurred (crude incidence rate, 130 per 1000 patient-years; 95% CI, 33−522) in the low-risk group; a total of 7 (50%) graft losses occurred (crude incidence rate, 233/1000 patient-years; 95% CI, 111−490) in the intermediate-risk group; and the highest incidence rate of graft loss (n = 12, 75%) occurred (crude incidence rate, 1019/1000 patient-years; 95% CI, 579−1794) in the high-risk group (P = 0.0025). Compared to patients with a low-risk TG score, patients with an intermediate-risk TG score had similar risk of graft loss over time (hazard ratio, 1.64; 95% CI, 1.640.33−8.06), whereas recipients with a high-risk TG score had significantly higher risk of graft loss (hazard ratio, 6.69; 95% CI, 1.39−32.23) using an unadjusted Cox proportional risk regression model. The incidence rate of graft loss was similar (P = 0.914) between recipients who received antirejection treatment for TG (n = 13; 50%; crude incidence rate, 382/1000 patient-years; 95% CI, 222−657) versus recipients who did not receive treatment (n = 8; 67%; crude incidence rate, 347/1000 patient-years; 95% CI, 174−694), as shown in Supplementary Figure S4. The Harrel c-index, which is the measure of discrimination, was 0.69, which indicates good discrimination of the model. Figure 4 shows the receiver operating characteristic curve of the transplant glomerulopathy prediction score for using 1-year graft loss as the gold standard outcome with an area under the curve of 0.80. Supplementary Table S1 presents a detailed report of sensitivity, specificity, positive, and negative likelihood ratio of different cut points. In this retrospective, single-center, observational study, we have externally validated the TG prognostic index score developed by Patri et al.8Patri P. Seshan S.V. Matignon M. et al.Development and validation of a prognostic index for allograft outcome in kidney recipients with transplant glomerulopathy.Kidney Int. 2016; 89: 450-458Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar in a cohort of kidney transplant recipients largely comprising African American individuals. In our cohort, we used the TG score to stratify the patients, and were able to show that this stratification has acceptable discrimination and calibration statistics, therefore enabling accurate prediction of their graft outcomes. It has been previously shown that African Americans have worse renal allograft survival compared to patients who are not African American.S1 Stratification of these patients is necessary to decide which patients should be exposed to further aggressive immunosuppressive treatments. A data-driven archetypes approach can refine the diagnostic and prognostic features associated with TG; however, this might be difficult to use at the bedside, and it was developed based on French and Canadian patients.S2 The prognostic score developed by Patri et al. for TG was developed and validated previously in a non−African American majority cohort, and it needed to be validated in this high-risk population. In our cohort, the TG score showed acceptable discrimination and calibration statistics. We found no statistical difference in the incidence of graft loss in the low-risk and intermediate-risk group, which was seen in the developmental cohort. There are several potential explanations why our result was different from that in the original developmental and validation cohort.8Patri P. Seshan S.V. Matignon M. et al.Development and validation of a prognostic index for allograft outcome in kidney recipients with transplant glomerulopathy.Kidney Int. 2016; 89: 450-458Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar First, this could have been due to our small sample size. Second, in our cohort there was a high number of African Americans, which might explain the observed differences. Third, the treatment pattern and practice might be different in our center from those in the original centers; however, in both our cohort and the original cohorts, the graft survival rate was similar in patients who received treatment versus those who did not.8Patri P. Seshan S.V. Matignon M. et al.Development and validation of a prognostic index for allograft outcome in kidney recipients with transplant glomerulopathy.Kidney Int. 2016; 89: 450-458Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar Finally, differences in immunological risk and treatment adherence might be contributing factors. We found no difference in graft outcomes in the treatment group versus the no-treatment group, similarly to the original cohorts.8Patri P. Seshan S.V. Matignon M. et al.Development and validation of a prognostic index for allograft outcome in kidney recipients with transplant glomerulopathy.Kidney Int. 2016; 89: 450-458Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar In our cohort, we found that patients in the high risk group were younger than those in the low-risk and intermediate-risk groups. This observation echoes the previous findings where younger age has been associated with increased risk of renal allograft rejection.S3 Our study has several limitations. First, our sample size was small, substantially lower than in the original paper,8Patri P. Seshan S.V. Matignon M. et al.Development and validation of a prognostic index for allograft outcome in kidney recipients with transplant glomerulopathy.Kidney Int. 2016; 89: 450-458Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar with relatively fewer patients in the low-risk group; however, we were still able to perform statistical comparison in this group, although our analysis was most likely underpowered. Second, we did not have a standardized approach to the treatment of TG; however, we did not see any difference in the treatment arm versus the no-treatment arm. Finally, we did not have information about the proximal cause of graft failure, and the follow-up time was only 3 years. In conclusion, the transplant glomerulopathy prognostic index score developed by Patri et al.8Patri P. Seshan S.V. Matignon M. et al.Development and validation of a prognostic index for allograft outcome in kidney recipients with transplant glomerulopathy.Kidney Int. 2016; 89: 450-458Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar showed an acceptable discrimination and calibration statistic in an independent U.S. cohort largely comprising African Americans. All the authors declared no conflict of interest. We thank Kenton Wong for proofreading our manuscript. Download .pdf (.77 MB) Help with pdf files Supplementary File (PDF)

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.