Longitudinal study of human polyomaviruses viruria in kidney transplant recipients

Human Polyomaviruses (HPyVs) are small, non-enveloped DNA viruses [1]. HPyVs family includes 14 polyomaviruses [2] with JC Polyomavirus (JCPyV) and BK Polyomaviruses (BKPyV) representing the most extensively investigated ones. The prevalence of HPyVs infection in the general population is very high, with seroprevalence rates ranging from 60 to 100%, depending on the series [3]. It is known that HPyVs can establish latency in kidney tubular epithelial cells [4‐6]. Transient or chronic immunosuppression are risk factors for viral reactivation [7]. JCPyV is the causative agent of progressive multifocal leukoencephalopathy (PML)[8], and BKPyV is etiological agent of Polyomavirus-associated Nephropathy (PVAN) [9]. Anti-rejection prophylaxis after kidney transplantation (KTx) is considered as the main trigger of latent HPyVs reactivation [10], but all these viruses may also be transmitted through the allograft [11‐14]. Polyomavirus-induced diseases can increase morbidity and mortality in solid organ transplant recipient [15]. In KTx, BKPyV nephropathy, occurring in 1–10% of KTx [16], are associated with renal dysfunction and premature allograft loss, and, to date, no therapeutical strategies are still approved. Whilst the role of BKPyV and PVAN in KTx is well recognized, the link between JCPyV infections and renal allograft dysfunction remains unclear.

The aim of this prospective observational study with 18 months of follow-up was to assess HPyVs prevalence and replication pattern in renal biopsies and urine samples of a cohort of consecutive KTx donor-recipient pairs. Risk factors for post-transplant HPyVs infection or reactivation were also investigated, particularly focusing on donor and recipient HPyVs-specific immunization and recipient immunosuppression.

KTx is the best therapeutic option for ESRD. Nevertheless, the need for chronic immunosuppression exposes transplant recipients to serious complications such as drug-related toxicity, cardiovascular disease, infections, and malignancy [23‐26]. Among the others, HPyV infections represent as a major cause of post-transplant morbidity and premature allograft loss [27, 28].

Despite recent advances in the understanding of HPyVs epidemiology and biology [17, 22], the origin and the natural history of BKPyV and JCPyV infections in KTx recipients have not been clarified. In particular, there is a lack of information regarding HPyVs behavior and reciprocal interactions in asymptomatic patients. In the present study, we have prospectively assessed JCPyV and BKPyV urinary replication in a cohort of KT recipients. Viruria of donors and pre-transplant intra-allograft replication were also evaluated.

We found that JCPyV viruria was particularly frequent, in both donors and recipients. Moreover, we observed that the proportion of donors and recipients with JCPyV urinary replication was significantly higher than BKPyV. These results are overall in line with previously published data [29‐31]. However, some authors have reported an higher prevalence of BKPyV viruria, particularly among recipients [32‐35]. Such discrepancy may be due to several reasons including different geographical or ethnic characteristics or it may reflect the fact that BKPyV is more frequently associated with clinical manifestations than JCPyV. In this regard, it is worth mentioning that, in our series, samples collection was not clinically-driven, but preliminarily planned as a part of the study follow-up.

As reported by other groups, JCPyV urinary replication was significantly more frequent in patients receiving a kidney from a donor with JCPyV viruria at the time of organ procurement [11, 12, 36], 13]. In a recent study, we have demonstrated that there is a very strong concordance between donor and recipient JCPyV genotypes [17]. This observation suggests that peri-transplant recipient JCPyV viruria is mostly secondary to a direct virus transmission from the donor to the recipient via the allograft [17]. The fact that the median time from transplant to first JCPyV urinary excretion was only about one day certainly supports this hypothesis [34, 36, 37].

The pattern of JCPyV urinary replication during the post-transplant follow-up was similar to those described in other prospective studies on KT recipients or healthy subjects and overall suggests that the virus can remain latent in renal tubular epithelial cells, easily reactivating in case of transient or persistent states of immunosuppression [32, 35, 38‐41].

Given the small sample size and the short follow-up, no episodes of PML were recorded. Also, no correlations between JCPyV viruria and allograft function or other adverse outcomes could be detected. Therefore, our experience further confirms that there is no need for systematic JCPyV screening before and after transplant.

The prevalence of BKPyV viruria among our recipients was lower than expected [34]. Furthermore, the median time from transplant surgery to the first recorded episode of viruria was longer than those described by previously published studies [37, 42], but in line with the one recently reported by Querido and colleagues [35]. The prevalence of BKPyV viruria and BKPyV urinary load were highest around T4, with a steady decrease during the rest of the follow-up. For instance, no cases of BKPyV-related allograft dysfunction were recorded. Rejection rate was also extremely low. These data seem to suggest that BKPyV urinary replication is not primarily determined by induction or early maintenance immunosuppressive therapies, but more likely by the chronic burden of immunosuppression. The difference between our results and other reports is difficult to explain [34]. The fact that our population included many living-donor and low-immunological risk recipients (requiring a lower degree of immunosuppression) may have played a role. As a matter of fact, our experience confirms that early screening for BKPyV infection may not be appropriate in standard-risk KT [22].

Simultaneous BKPyV viruria in both donor and recipient was quite uncommon compared to JCPyV. Certainly, the paucity of events recorded prevents any meaningful conclusion. Nevertheless, we have already shown with sequencing analysis that, in this studied cohort, the concordance between donor and recipient BKPyV strains was rather low, thus speculating that at least some episode of post-transplant BKPyV replication is likely due to reactivations of latent viruses in the recipients, while some other reactivation episodes might be due to the donor transmission [17].

Analyzing the distributions and the entities of JCPyV and BKPyV urinary replications in donors and recipients, we observed that the subjects with JCPyV viruria were less prone to develop BKPyV viruria. As proposed by Saundh et al. or Cheng and colleagues, this finding suggests that JCPyV may interfere with BKPyV replication [34]. The possible mechanisms underlying this phenomenon are undetermined. The high genomic similarity of the antigenic proteins of JCPyV and BKPyV could promote a sort of immune cross protection [43‐45] or, more likely, JCPyV could compete for the very same cellular replication machinery as BKPyV [34]. Future studies should include baseline serological evaluation of donors and recipients, and possibly focus on this interesting interaction.

Our study was clearly underpowered to detect any significant association between baseline immunosuppression and the risk of post-transplant HPyVs viruria. However, as described in larger populations, the prevalence of JCPyV and BKPyV viruria in patients receiving basiliximab or rATG was similar [22, 46]. Remarkably, we found that recipients on CMV universal prophylaxis were less likely to exhibit HPyVs viruria than controls. Although we cannot provide any matching information regarding HPyVs and CMV replication in blood or urine, these data confirm a possible relationship between CMV and BKPyV infection [22, 47].

The impact of transient or prolonged HPyVs viruria on transplant-related outcomes is still controversial. As previously discussed, we did not observe any episode of PML or PVAN. Prospective assessment of HPyVs viruria and SCr also failed to demonstrate a possible effect of HPyVs urinary replication on allograft function [48‐50].

We recognize that our study has several limitations including a relatively small sample size, the lack of information regarding baseline HPyVs serological status, no systematic screening for post-transplant HPyVs viraemia, and no follow-up allograft histology. Nevertheless, our findings will serve as a basis for future research investigating the correlation between chronic exposure to specific immunosuppressive agents and HPyVs urinary replication. Due to the lack of approved antiviral therapies, if our results will be confirmed, the late screening for BKPyV infection after KT may improve the clinical practice impact in these patients.