Long-term survival in Japanese renal transplant recipients with Alport syndrome: a retrospective study

Alport syndrome (AS) is an inherited nephropathy characterized by sensorineural hearing loss and typical ocular abnormalities. It is caused by a hereditary type IV collagen deficiency [1‐3]. In 85% of patients with AS, the collagen deficiency exhibits X-linked inheritance (COL4A5 gene), and the remaining 15% show autosomal recessive and rarely autosomal dominant inheritance (COL4A3 or COL4A4 genes). Histopathologically, these genetic alterations reflect glomerular basement membrane (GBM) thickening and lamellation, leading to focal segmental glomerulosclerosis and resultant global sclerosis and/or hyalinosis [1]. With respect to therapy, several studies have demonstrated the efficacy of angiotensin-converting enzyme inhibitors as the first-choice medications [4, 5], with angiotensin II receptor blockers and spironolactone as alternatives [6]. However, these medications are not always satisfactory, and there are no disease-specific medications. For these reasons, AS commonly progresses gradually from childhood, resulting in end-stage renal disease at a young age [3]. Indeed, according to the United States Renal Data System, the median age at which renal replacement therapy was initiated in AS from 2005 to 2009 was 33.7 years [7].

Patients with AS account for > 1% of all patients undergoing kidney transplantation [8]. Kidney allograft survival in patients with AS is excellent despite the fact that AS recipients can develop anti-GBM disease, which can induce allograft loss at a high rate early after kidney transplantation. These excellent results have been reported in several countries. These reasons were mainly explained by the lack of risk factors for chronic kidney disease (hypertension and diabetes mellitus), the absence of AS recurrence, and the very low rate of anti-GBM disease (only 1.9%) [9]. However, the clinical outcomes of renal allografts in the Japanese population, the vast majority of whom have undergone living related kidney transplantation (LRKTx), have not yet been investigated.

Another aim of this study was to evaluate the recurrence of AS in recipients of kidney transplantation. This topic was investigated for the following three reasons. First, the type IV collagen α3α4α5 heterotrimer in the GBM originates only from podocytes [10]. Second, there is evidence that allograft podocytes are produced from turnover of the recipient’s bone marrow-derived cells [11]. Third, experimental evidence has shown that the transplantation of wild-type bone marrow into irradiated Alport mice results in the recruitment of podocytes, which produce type IV collagen α5 within the damaged glomerulus [12, 13]. We hypothesized that recurrence of AS occurs locally and influences the clinical abnormalities of patients with AS.

To address these clinical and histopathological issues, we investigated the graft survival and histopathological changes in patients with AS as well as the expression of GBM type IV collagen α5 and electron microscopy findings in renal allograft recipients with AS in the Japanese population.

We used the clinical, laboratory, and pathological data from recipients who underwent kidney transplantation at the Department of Urology, Tokyo Women’s Medical University, and the Division of Nephrology and Hypertension, Department of Internal Medicine, The Jikei University School of Medicine, from February 1984 to February 2015. Twenty-one recipients had AS as the primary renal disease leading to end-stage renal disease. Controls included patients who underwent transplantation with end-stage renal disease due to a condition other than AS. The two groups of patients were matched for recipient and donor age, number of renal transplants, presence of donor-specific antibodies, donor resources, and era of transplantation using propensity-score analysis. Endopoints were graft loss or patients death.

This study is the first to demonstrate patient and graft survival in the Japanese population of patients with AS. We retrospectively reviewed the long-term outcomes of 21 patients with AS who underwent kidney transplantation with a mean follow-up duration of 83 months and found that the cumulative patient and graft survival rates in the AS group were not significantly different from those of a well-matched control group (Figs. 1 and 2). Our data as well as those reported by Yilmaz from Turkey showed no difference between patients with and without AS [15]. In contrast, Temme showed better graft survival (HR, 0.75; 95% CI, 0.60–0.93; p = 0.008) and patient survival (HR, 0.46; 95% CI, 0.30–0.59; p = 0.001) in patients with AS using the ERA-EDTA Registry data [7]. The only difference in the patients’ characteristics was the rate of deceased donors (63.0% in the ERA-EDTA data, 32.0% in Yilma’s data, and 9.5% in our data). Notably, the author demonstrated that the patient and graft survival rates were better in deceased donors than in controls, but not in living donors. This finding suggests that living donors with AS could negatively affect patient and graft survival. In line with this, Gross from Germany observed six cases of living donor kidney transplantation in AS from relatives with mild urinary abnormalities and found that three of the six donors developed new-onset hypertension and two of the six donors developed new-onset proteinuria. The author pointed out that, genetically, AS may also affect other family members, and physicians should be aware of an increased risk of renal failure in both the donor and the recipient [16]. In this context, in countries with a donor shortage, such as Japan (in which almost 90% of transplantations are living donor kidney transplantations), careful evaluation and monitoring of both donors and recipients with AS are necessary. In the present study, 13 of 21 patients received kidneys from a father, suggesting that most donations were from an unaffected father with X-linked (COL4A5 gene) AS. However, we found at least two patients (Cases 12 and 16) with a relatively thin GBM on electron microscopy, suggesting possible transmission of thin basement membrane disease from autosomal-recessive (COL4A3 or COL4A4 gene) affected relatives [17, 18]. For example, in Case 16, we confirmed that type IV collagen α5 staining of the recipient’s native kidney biopsy revealed positive in the basement membrane of Bowman’s capsule but negative in GBM, suggesting autosomal-recessive AS associated with COL4A3/4 gene abnormality. Of note, previous report showed that about 40% of TBMD patients have heterozygous mutations at the COL4A3/4 locus [19]. Therefore, the donor in Case 16 could be heterozygous carrier with autosomal-recessive AS, resulting in TBMD shown in baseline allograft biopsy. These two patients had stable kidney function and did not show proteinuria or hematuria. These results suggest that living donor kidney transplantation from relatives with AS is acceptable if the donors and recipients are carefully evaluated. Another important factor associated with patient and graft outcomes is the occurrence of anti-GBM disease. Anti-GBM disease induces graft failure in the early stage of kidney transplantation in recipients with AS, but the incidence is generally very low at 1.9% [9]. Fortunately, no recipients in the AS group developed anti-GBM disease within the observation. Among patients with AS, the risk factors associated with graft loss in patients other than living donors and those with anti-GBM disease have not been fully evaluated. In the present study, 6 of 21 (28.6%) recipients with AS developed graft loss. Of these, four (66.6%) were due to chronic rejection, one (16.7%) was due to noncompliance with medication, and another was of unknown cause. These results are comparable with the causes of graft loss in the matched control group.

Another main finding of this study is that no patients developed recurrence of AS. To the best of our knowledge, this is the first study to evaluate the recurrence of AS in detail using histopathological methods. Generally, recurrent glomerulonephritis or glomerulopathy may cause graft loss in kidney transplantation, but most nephrologists believe that patients with AS do not develop recurrence. However, several recent clinical and experimental studies have implied that GBM type IV collagen originates from podocytes recruited from the recipient’s bone marrow-derived cells, suggesting the possibility of recurrence of AS. For example, Abrahamson clearly demonstrated that α3α4α5 chains are produced only by podocytes, as evidenced by immunoelectron microscopic examination [10]. In human kidney transplantation research, Becker showed that recipient-derived podocytes were found in 4 of 8 biopsies, representing 3 of 6 patients, and that 5 of the 740 podocytes examined in the female-donated allograft were male-derived, using fluorescence in situ hybridization for the X and Y chromosomes [11]. In addition, Sugimoto et al. reported that bone marrow transplantation from wild-type mice to Alport model mice (COL4A3 −/−) improved the symptoms of AS and the level of protein produced from bone marrow-derived cells [12]. Therefore, we initially hypothesized that the podocytes of allografts can be replaced by recipient-derived podocytes that cannot synthesize α3α4α5 chains, resulting in deficiency of the GBM of allograft. However, we found that the structure of the GBM remained normal and that no deficiencies of the α5 chains in the GBM occurred, even long after kidney transplantation, suggesting that AS does not recur histologically. We speculated following two mechanisms for this phenomenon:1) a very small number of donor derived podocytes were replaced by recipient derived ones [11], or 2) podocytes injury is rare in recipients with AS, and both podocytes and GBM turnover was not evident [20‐22]. Additionally, only few recipients in Alport group showed hematuria in the latest examination (11.1%, Table 1), which further strengthen no histological recurrence of AS because electron microscopy might underestimate the presence of GBM changes of AS due to sampling bias.

The strengths of this study include the selection of controls (propensity-score analysis with matching of six factors: recipient and donor age, number of transplants, donor-specific antibody positivity, donor resources, and era of transplantation) and longer observational periods than in previous reports (99.5 vs. 75.4 months) [15]. Limitations include the retrospective design, small sample size due to the rarity of AS, lack of genetic information, and performance of few allograft biopsies because the allograft function was relatively good, even long after kidney transplantation, in recipients with AS.

Our data demonstrate that graft survival in patients with AS is similar to that in patients with other primary diseases in the Japanese population. Careful selection of living donors is necessary for the successful management of kidney transplantation in patients with AS. Neither deficiency of the α5 chains (type IV collagen) nor GBM abnormalities were detected in the kidney allograft biopsies, indicating that AS recurrence was not present or was very limited in recipients with AS. Further pathological analysis with a longer follow-up to evaluate the possible recurrence of AS and transmission of thin basement membrane disease from affected relatives will contribute to a better understanding of kidney transplantation in patients with AS.