Testing Immunosuppression Threshold in Renal Allografts To Extend eGFR

1. Background Kidney transplantation is the treatment of choice in patients with end stage renal disease in need of function replacement therapy, and in whom there is no absolute contraindication to the procedure. The current figures reported in the USA on post-transplant graft survival at one year, are 95% in cases of living donor (LD) recipients and 90% in deceased donor (DD) recipients3; the mean calculated survival of these grafts has increased and is currently 14 years in LD and 11 years in DD recipients. The fact that graft survival hinges on multiple immune and non-immune factors is well recognized; however, the main cause of graft loss is solidly linked to its rejection, with evidence of involvement of the immune humoral response in most cases (specific antibodies against HLA antigens).

The main cause of acute graft dysfunction is acute rejection (AR). Several studies have associated episodes of this entity with chronic structural injury and subsequent functional impairment. Thus, timely diagnostic certainty is pivotal so as to optimize immunosuppressive therapy and preserve graft function. To date, the only tool available to confirm AR is graft biopsy and its established histological diagnostic criteria8. However, this diagnostic tool has limitations resulting mainly from sampling errors and costs. Furthermore, about 10% of patients with normal graft function have evidence of AR in protocol biopsies, so centers that do not use this follow-up modality in kidney transplant recipients (KTR) are unable to document this immune phenomenon. Ideally, an AR non-invasive, diagnostic tool would be of utmost use in KTR follow-up since it would preclude the need for graft biopsies and efficiently support the management of immune suppressors. Therefore, the systematic performance of a highly specific and sensitive assay reflecting AR, would allow for the timely detection of this immune phenomenon compared to the current standard (biopsy). Moreover, the systematic use of a tool with these characteristics would allow stratification of the immunological risk as well as proactive adjustment of immunosuppressor drug dosages, potentially limiting chronic injury and improving the graft's survival.

The background to the proposed study is based on the extensive line of research conducted by Dr. Sarwal et al. They have recently published results10 on monitored pediatric kidney transplant recipients. Biomarkers of gene panels in peripheral blood detected by microarray were discovered in a single institution and subsequently validated by quantitative-PCR in 12 pediatric kidney transplant programs. A total of 367 individual blood samples, each paired to a graft biopsy for blinded centralized classification, were analyzed (115 with acute rejection, 180 with stable kidney graft function, and 72 with other causes of graft injury). Among the genes differentially expressed in microarrays, quantitative-PCR analysis of a set of 5 genes (DUSP1, PBEF1, PSEN1, MAPK9 and NLTR) classified acute rejection with great precision. The 5 gene model has a sensitivity of 91%, specificity of 94%, positive predictive value of 83%, negative predictive value of 97% and a 92% precision when differentiating acute rejection samples from other phenotypes (stable and no acute rejection, no acute rejection/ non-stable). Interestingly, 8/12 samples from patients with borderline rejection on biopsy, were classified as acute rejection with the 5 gene model. The frequent prediction of the phenotype in borderline rejection biopsies suggests that pre-clinical injury in acute rejection could also be identified by quantitative-PCR in peripheral blood samples, thus allowing earlier treatment of these patients.10 The fact that this model predicts acute rejection in both cellular and humoral immune injury modalities, is of utmost relevance. Moreover, these genes in peripheral blood, strongly linked to graft rejection, do not correlate (are independent) with many demographic, clinical, treatment modalities and bacterial/viral infection parameters.

The 5 genes are centered on leukocyte trafficking and T/B cell activation and are mostly expressed in monocytes, activated in the peripheral circulation; they reflect injury response mechanisms to cellular oxidative stress (DUSP1), apoptosis (MAPK9), IL2-dependent cytolytic gene activation (NKTR), increased cell adhesion via the e-cadherin/catenin complex (PSEN1) and smooth muscle vascular cell injury (PBEF1).

The authors pointed out that the study was conducted in a pediatric and young adult population, in whom rejection may be more aggressive due to the disparity between the adult transplanted organ and the pediatric recipient, or as a result of poor treatment adherence in adolescent recipients, leading to a stronger immune response. Thus, the 5-gene model would require further evaluation in order to determine its diagnostic potential of acute rejection in patients of all ages, in larger patient cohorts and in association with different immunosuppressant regimes.

For these reasons, the same group of investigators undertook a large study to identify a transcription profile that would unify the kidney transplant recipient population independently of age, the cause of terminal renal disease, comorbidities, and the type of immune suppression used in different centers in the USA, Mexico and Europe. The final selection of 17 genes included 10 genes in the pediatric population and 7 newly identified genes and allowed the molecular classification of acute rejection in adult and pediatric patients. They correctly predicted the presence of acute rejection and non-acute rejection in the different samples collected from patients in various participating study sites, and in recipients of different ages with no previous data normalization. On the basis of these observations, a score was generated using 14 of the 17 genes, and with Pearson's correlation test, a score was attributed to each detected gene and compared with the genetic profile found in patients with and without rejection, assigning the number +1 or -1. The score ranges between -13 and +13. Among the patients in whom the score predicted a "high risk of rejection" (Acute Rejection Risk-Score ≥ 7), 90.24% were correctly classified as acute rejection, while if the test predicted "a very low risk of acute rejection" (Acute Rejection Risk-Score ≤ -7), 97.7% were correctly classified as non-acute rejection on the basis of the biopsy findings. Furthermore, the average predictive probability of acute rejection was highly significant when comparing acute rejection vs. non-acute rejection in 4 centers (p<0.0001; p=0.002; p<0.0001; p<00001), respectively. This model equally detected acute rejections mediated by antibodies as well as those that were cell-mediated, with a high predictive probability. The prediction of acute rejection was independent of the post-transplant period duration. These 17 genes are included in the KSORT assay (Kidney Solid Organ Rejection Test) and the algorithm generating the risk score is known as the kSORT analysis suit (kSAS).

In a recently published study with data from our Institute, in which the results of biopsies obtained because of post-transplant dysfunction were reviewed between January 2007 and December 2011 (n=223), we observed an AR incidence during the first year post-transplant of 11.8% in live donor KTR and of 17.4% in deceased donor graft recipients. Throughout 2013, the number of kidney transplants performed at the Institute increased to 63 (53% from living donors, 46% from deceased donors). In accordance with our institutional protocol for the follow-up of kidney transplant recipients, we currently obtain protocol graft biopsies on the 3rd and 12th month post-transplant. The protocol biopsies are obtained at established timepoints with the purpose of documenting whether there is any evidence of immunological activity (sub-clinical) or signs of other pathologies that would require modification of the immunosuppressive therapeutic regime. We also obtain a graft biopsy at any point in the post-transplant period if abnormalities in renal function develop, defined as a verified increase in serum creatinine ≥ 25% over previous baseline values, and in the absence of other evident pathologies such as obstructive processes of the urinary tract, urinary tract infections, dehydration, supra-therapeutic blood levels of the used calcineurin inhibitor.

As of May 2013, the date in which the 3rd month protocol biopsy was established, we have documented an incidence of sub-clinical acute rejections, including borderline and major rejections according to the Banff classification, of 52% (10/40 biopsies revealed acute cellular or humoral rejection and 11/40 had borderline injury). This number is unprecedented at the Institute because protocol biopsies were not previously obtained on the 3rd month.

Currently, all KTR at the Institute receive an induction therapy modality (Thymoglobulin or Basiliximab), depending on their individual immunological risk and the origin of the graft. Hence, all live donor KT recipients with a PRA <30% and no HLA donor specific antibodies (DSA) are treated with Basiliximab (anti-IL2R monoclonal antibody; 2 doses of 20mg each); the exceptions are KTR sharing 2 haplotypes with their donor and that receive no induction therapy. All deceased donor graft recipients (independently of the PRA %), live donor KTR with positive donor-specific HLA antibodies (independently of the PRA %) or with a PRA ≥30%, are induced with Thymoglobulin (rabbit polyclonal antibody preparation, at a total dose of 4.5mg/Kg). All patients must have a lymphocyte crossmatch test (T/B), a negative CDC-AHG and those with DSA (Single antigen, Luminex) also require a negative crossmatch by flow cytometry.

Biopsy-proven acute rejection events are treated with pulses of methylprednisolone for 3 consecutive days. Patients with steroid-resistant acute cellular rejection (biopsy-proven), are administered Thymoglobulin (1.5 mg/Kg/body weight, for 7 days). Patients with antibody-mediated acute rejection (humoral rejection) are treated with 3-5 plasmapheresis (PP) sessions, 100 mg IV Ig per Kg body weight after each PP session and rituximab (anti-CD20 monoclonal antibody) at the end of all PP sessions.

The standard initial and maintenance immunosuppressive regime is Tacrolimus, Mofetil Mycophenolate and prednisone. The initial Tacrolimus doses are aimed to obtain blood trough levels of 10 to 15 ng/mL, until the 3rd month post-transplant; subsequently and in the absence of AR during the first trimester (clinical or by protocol biopsy), the trough levels are maintained between 8 and 10 ng/mL throughout the first year post-transplant. The maintenance dose of prednisone is 5 mg daily and that of Mofetil Mycophenolate is 1-1.5 g daily.

II. Justification In previous paragraphs, we described the KSORT test's high sensitivity and specificity in the detection of acute rejection. It is now necessary to attempt its transfer into the clinical setting and to help support treatment decisions, such as increasing immune suppression based on a positive test result. The impact of this maneuver on efficacy and safety outcomes will be pivotal in order to incorporate this scrutinizing method into daily clinical practice. Our Institute has transplanted over 60 patients in the last few years, so conducting this study in our center is feasible.

III. Hypothesis and aims Hypothesis Adding the kSORT genetic panel to the post-transplant follow-up will decrease the incidence and severity of acute rejection as well as the chronicity indices in the yearly protocol biopsy, and will improve the calculated glomerular filtration rate after a follow-up of two years.

Aims:

To evaluate the safety and efficacy of adding the kSORT test to the surveillance of patients after kidney transplantation.