Transplantation outcomes in patients with primary hyperoxaluria: a systematic review

The primary hyperoxalurias (PHs) are a group of rare inherited metabolic disorders leading to endogenous overproduction of oxalate. Three subtypes have been identified based on the underlying enzyme deficiency. PH type 1 (PH1) accounts for over 80% of all patients [1]. These patients present with kidney stones, nephrocalcinosis, or kidney failure in almost 40% of cases [2]. Eventually, over 70% will develop kidney failure [3]. In patients with advanced chronic kidney disease (CKD) or kidney failure, systemic oxalate storage occurs and causes multi-organ failure. Conservative therapy (e.g. hyperhydration and citrate supplementation to prevent stone formation) is not sufficient in those cases. So far, only liver transplantation can ‘cure’ the metabolic disorder and is therefore generally recommended in PH1 patients with kidney failure. A kidney transplantation is required since oxalate clearance by conventional dialysis cannot match endogenous oxalate production rate and thus will not prevent disease progression [4, 5].

Choosing the right transplantation strategy for each specific patient case remains challenging [4]. The most recent European guidelines recommend either combined liver-kidney transplantation (CLKT) or sequential liver-kidney transplantation (SLKT) in all PH1 patients with CKD stages 4 and 5 [6]. The guideline is reluctant to recommend pre-emptive liver transplantation (PLT), performed prior to the development of kidney failure and meant to prevent the need for dialysis and/or kidney transplantation. The procedure carries a significant mortality rate of 10% [7‐9]. The guidelines further advise against isolated kidney transplantation (KT); it should be considered only for ‘selected adult patients with confirmed evidence of B6 responsiveness’ [6]. In approximately 30% of Western PH1 patients, vitamin B6 effectively lowers hepatic oxalate production. A small subset of patients shows a complete response defined as a normalization of oxalate excretion rate [3]. However, the paucity of data on performing an isolated KT in these patients prevents the guidelines from supporting such a deviation from the general recommendation in PH1 patients. As stated in the guideline, suggestions are based on ungraded statements because of the lack of randomized clinical trials and the rarity of PH1 [6]. It is of great clinical importance to identify the best transplantation strategy, as the entire transplantation procedure is costly and carries significant risks, including potentially fatal postoperative complications and the risk of tissue oxalate mobilization causing recurrent oxalate nephropathy in the kidney graft [6].

To our knowledge, this is the first systematic review of transplantation outcomes in PH. The aim of this systematic review was to compare patient and graft survival rates for different transplantation strategies in order to identify the optimal approach for PH patients. We feel that this will remain a relevant discussion, especially with new emerging therapies that appear to be effective but also may become very costly [10, 11].

We systematically reviewed outcomes of different transplant modalities used in PH1. In total, we identified 51 observational studies on transplantation outcomes in 1201 PH1 patients. Outcomes were mainly reported as survival probabilities; only two studies reported hazard ratios [16, 19]. Out of five high-quality studies, only one study found a statistically significant difference in patient survival, in favour of KT [4]. In this study however, outcomes were not adjusted for year of transplantation or any other factors. In previous decades, CLKT was a ‘hazardous venture’ [21]. There were no significant differences in patient survival at 15 years post-transplantation (78% for CLKT and 60% for KT) according to a multicentre study by Compagnon et al. [16]. The same was observed in a large registry study by Cibrik et al. [15]. Both studies adjusted for several factors including year of transplantation. The risk of death due to complications of the procedure for CLKT seems to have outbalanced the risk of death due to severe oxalosis in KT recipients in the long term.

The substantially higher kidney graft survival for CLKT recipients (87% at 10 years [16]) is expected to be due to the pathophysiology of PH; the devastating kidney graft survival rates for KT (14% at 10 years [16]) can be ascribed to the unabated hepatic oxalate production and release of stored oxalate, resulting in damage to the kidney transplant soon after the procedure. However, genotype and clinical pyridoxine responsiveness are of major importance with regard to the risk of graft failure in these patients but were not reported in all except one study.

In the case series by Lorenz et al., four pyridoxine-responsive patients successfully received a KT in combination with conservative therapy [37]. eGFR was moderately reduced (CKD stage 3) at a median follow-up of 5.2 years (range 0.2–13.9). In a large cohort of non-PH kidney transplant recipients, the distribution of CKD stages 1–5 at 12 months was 2.7, 27.1, 59.4, 10.3 and 0.5%. This was very similar at 5 years and 10 years of follow-up [64]. The idea that KT may be a viable option in this subgroup of (adult) patients who are deemed to be completely responsive to pyridoxine and are expected to have better outcomes [65] has been suggested previously [6, 66]. Despite the good clinical reasoning behind performing a KT in patients who clinically respond to pyridoxine therapy, there is a lack of evidence to support this approach and consequently clinicians opt for a liver-kidney transplant.

The merits of SLKT as compared to CLKT are not evident, mainly due to the small number of studies comparing both strategies (maximum 20 SLKT procedures [20]). Sequential procedures were performed in patients with severe systemic oxalosis [46] and small infants [27]. Even at the age of 4 months, an infant successfully underwent liver transplantation and is now awaiting a kidney transplant [27]. Also, SLKT has been performed safely with organs retrieved from a single living donor [46, 55]. In that case, this strategy attains the immunological advantage of a CLKT. However, in most cases, two donors are needed for a SLKT procedure [67]. Very few cases of SLKT have been reported and even fewer reports of donor outcomes exist. Therefore, the guidelines do not favour either CLKT or SLKT and advise on a simultaneous or sequential procedure according to the patient’s condition, local facilities and preferences [6].

Pre-emptive liver transplantations were not widely performed, but case series reported very high patient survival rates, up to 100% after 10 years of follow-up [25]. However, the European Liver Transplant Registry (ELTR) reported a 1-y mortality rate of 16% for 258 PH patients who underwent liver transplantation between 2001 and 2016 (presumably combined with a kidney transplant in most cases) [68]. Even while children affected with metabolic disorders are known to achieve the best outcomes [68], their mortality rates remain considerably high, and this holds true for transplantations performed in the past two decades. Death due to long-term complications of chronic usage of immunosuppressive medication should be added onto that. A review by Kemper et al. included nine patients who received a pre-emptive liver transplant of whom four patients required either a second liver transplant or a kidney transplant during follow-up [69]. Yet, differences in follow-up duration hamper a valid comparison between studies. The current guidelines do therefore not recommend this approach considering the ethical dilemma of performing a risky procedure in a patient who could remain stable for many years with conservative treatment only [6].

The most important limitations of this systematic review are due to the observational nature of the included studies, in which confounding by indication plays a role. Even in the few high-quality studies that did correct for confounders, residual confounding cannot be excluded. A meta-analysis could not be performed since survival probabilities were reported with various follow-up durations. Recently, methods have been developed to reconstruct time-to-event data from published Kaplan–Meier curves [70]. However, the few high-quality studies differed in patient population in terms of period of time, country and pre-transplant care, to the extent that an attempt to pool these heterogeneous data was considered inappropriate. Additionally, a comparison between CLKT and KT in pyridoxine-responsive patients was not feasible due to the lack of reporting of genotypes. Determinants of graft failure or death are rarely studied in PH patients; only Cibrik et al. assessed the influence of covariates and found multiple transplants, recipient ethnicity, panel reactive antibody, cold ischemic time and donor age as significant risk factors for death-censored kidney graft survival [15]. Furthermore, some included studies based their diagnosis on clinical phenotype, not liver biopsy or mutation analyses. Even relatively recently performed high-quality studies included patients whose diagnoses were based on metabolites only [4, 16]. Also, publication bias is likely to play a role in our review. This is rather true because researchers tend to publish nice-ending small studies of their transplantations, especially considering pre-emptive liver transplantations, which could explain the 100% 1-y survival rate in studies solely describing outcomes of this type of transplant. As a final limitation, we cannot exclude that there was any overlap of included patients. We excluded studies that evidently described the same patient cohort in a previous period of time, but individual patients may have been registered in more than one registry and thus described in more than one study. It is unlikely that this would concern a substantial number of studies since most included studies were single-centre studies.

The findings of this systematic review suggest that a combined or sequential liver-kidney transplantation has to be recommended as the first choice for treatment of PH1. This conclusion is however based on a relatively small number of cohort studies and registry studies that were of relatively good quality, which do not capture important patient characteristics such as genotype. Due to the rareness of this disease and the impossibility of performing randomized controlled trials, a well-maintained international registry is crucial for comparing outcomes for different transplantation strategies. This is needed to demonstrate possible merits of SLKT. In particular, there is a great need for studies investigating the possibility of KT in pyridoxine-responsive patients. The spectrum of therapeutic options to treat PH is expected to be expanded in the near future: medications comprising small interference RNA are emerging. Indeed, the investigational product Oxlumo has recently been approved for all ages by both the EMA and FDA as the first pharmaceutical treatment for PH1 [71, 72]. Preliminary data by Alnylam pharmaceuticals show that urinary oxalate excretion is effectively lowered with 65% and 72% mean reduction relative to baseline, in adults and children, respectively. Promising results are presented by Dicerna pharmaceuticals as well; due to the different mechanisms of action, these clinical trials also include patients with primary hyperoxaluria type 2 and 3 (clinicaltrialsgov, NCT number 03847909). These new medications will likely obviate the need for a liver transplant, but at first this will not be available for everyone. In addition, a kidney transplant will remain inevitable for patients who have already proceeded to CKD stage 5 [73]. This systematic review provides an overview of transplantation approaches in order to contribute to evidence-based decision-making. Yet, it also identifies the knowledge gap concerning outcomes of kidney transplants in pyridoxine-responsive patients, who might benefit from an isolated kidney transplant even in this new era. The rarity of PH1 should encourage close cooperation between expert PH centres to fill that gap and identify the optimal transplantation strategy for individual patients that will further enhance survival and quality of life of PH1 patients.