Conflict of interests: Speakers fees received from Novartis for Tuberous Sclerosis Presentation Brisbane 2013. Airfares received from ANHCA for presentation at ANHC Melbourne 2013.
Atypical HUS associated with severe, unexpected antibody-mediated rejection post kidney transplant
Article first published online: 14 MAR 2014
© 2014 Asian Pacific Society of Nephrology
Special Issue: Renal Transplant Case Series 2013. Guest Editor: Peter Kerr. Publication of this supplement has been supported through an independent educational grant from Novartis Australia
Volume 19, Issue Supplement S1, pages 22–26, April 2014
How to Cite
Stevenson, S., Mallett, A., Oliver, K., Hyland, V., Hawley, C., Malmanche, T. and Isbel, N. (2014), Atypical HUS associated with severe, unexpected antibody-mediated rejection post kidney transplant. Nephrology, 19: 22–26. doi: 10.1111/nep.12195
- Issue published online: 14 MAR 2014
- Article first published online: 14 MAR 2014
- Accepted manuscript online: 27 JAN 2014 05:57AM EST
- Manuscript Accepted: 18 DEC 2013
- alternative complement pathway;
- antibody-mediated rejection;
- atypical HUS;
- haemolytic anaemia;
- kidney transplantation;
- microangiopathic transplant nephrectomy
We present a case of an unsensitized patient with end-stage kidney disease secondary to atypical haemolytic uremic syndrome (aHUS) with mutations in CD46/MCP and CFH who developed severe, intractable antibody-mediated rejection (ABMR) unresponsive to therapy post kidney transplantation. There were no haematological features of thrombotic microangiopathy. The patient received standard induction therapy and after an initial fall in serum creatinine, severe ABMR developed in the setting of urosepsis. Despite maximal therapy with thymoglobulin, plasma exchange and methylprednisolone, rapid graft loss resulted and transplant nephrectomy was performed. Luminex at 4 weeks showed a new DSA and when repeated after nephrectomy showed antibodies to each of the 5 mismatched antigens with high MFI. The rate of recurrence of disease in patients with aHUS referred for transplantation is 50% and is associated with a high rate of graft loss. It is dependent in part on the nature of the mutation with circulating factors CFH and CFI more likely to cause recurrent disease than MCP which is highly expressed in the kidney. There is increasing interest in the role of complement in the development and propagation of ABMR via terminal complement activation. This case suggesting that dysregulation of the alternative complement pathway within the transplant kidney may have contributed to the severe AMR. Very little is known about the impact of complement dysregulation and the development of anti HLA antibodies however the strength of HLA antibody formation was prominent in this case.
Atypical haemolytic uremic syndrome (aHUS) is a rare disease characterized by activation and dysregulation of the alternative complement pathway resulting in microangiopathic haemolytic anaemia, thrombocytopenia and microvascular occlusion causing organ impairment. The laboratory abnormalities may include an abnormal peripheral blood smear with schistocytes, reticulocytosis and thrombocytopenia; elevated creatinine and serum lactate dehydrogenase (LDH). The identification of case clusters within families give biological plausibility to a genetic predisposition coupled with an inciting event such as sepsis or pregnancy. In 40–60% of cases there is a mutation in genes encoding for regulatory proteins of the alternative complement pathway (including membrane cofactor protein (CD46/MCP), Factor H (CFH) and Factor I (CFI), Factor H related proteins (CFHR1-5), C3, complement factor B and thrombomodulin). Therefore, when the complement system is activated, these genetic defects of the regulatory proteins are associated with defective protection of the endothelial cell surface. More C3b reaches the cell surface leading to higher levels of terminal complement activation, with further endothelial injury, ongoing stimulation of the coagulation cascade and thrombotic microangiopathy results. Among patients with end-stage kidney disease (ESKD) secondary to aHUS referred for transplantation, the rate of recurrence of disease is 50%. Recurrent disease usually occurs early post transplant and is associated with a high rate of graft loss. The rate of disease recurrence depends to some extent on the nature of the mutation with those involving the circulating factors CFH and CFI more likely to cause recurrent disease. The lower rate of recurrence of MCP associated disease may be explained in part, by the finding that MCP is highly expressed in the kidney and allograft transplantation should restore near normal levels.
Complement also has an important role in the pathogenesis of antibody-mediated rejection (ABMR) with initiation of the classical complement pathway by alloantibody, activation of C3 and subsequent graft injury mediated by C5b-9 membrane attack complex. We present a case of an unsensitized patient with ESKD secondary to aHUS with mutations in CD46/MCP (104G>A) and CFH (3590T>C) who developed unexpected, severe and intractable ABMR post transplant suggesting that the dysregulation of the alternative complement pathway may have been a contributing factor.
DS is a 40-year-old mother of 2 living in a regional centre who underwent deceased donor renal transplantation on 26 November 2011 for ESRD secondary to biopsy proven aHUS that occurred post partum.
DS had three pregnancies with high complication rates, the first ending in miscarriage and the second complicated by preeclampsia. The third pregnancy was characterized by hypertension and proteinuria at 26 weeks, and gestational diabetes. She was induced for preeclampsia at 34 weeks and delivered by caesarean section. Post partum she became increasingly unwell and at six weeks post partum was found to be in acute renal failure with thrombotic microangiopathy (TMA). Renal biopsy confirmed vascular and glomerular changes typical of aHUS. She underwent plasma exchange that was unsuccessful and was commenced on haemodialysis. There was no recovery of renal function. There is no family history of kidney disease or aHUS.
DS spent 5 years on dialysis before being listed for transplantation. Peritoneal dialysis had failed and she had significant vascular access problems with recurrent thromboses. She was counselled regarding the risk of recurrent aHUS and graft loss post transplant. DS proceeded to renal transplant (brain death donor[DBD]) on 26 November 2011. There was a 5 HLA mismatch; she was unsensitized with Luminex class I and II negative screens pretransplant. Both donor and recipient were CMV/EBV positive and she received standard induction therapy with basiliximab and maintenance tacrolimus, mycophenolate mofetil and prednisone. The operation was uncomplicated and implantation biopsy showed acute tubular necrosis (ATN) and mild arteriosclerosis. Early graft function was good with a rapid fall in serum creatinine from 700 to 110 μmol/L (see Fig. 1). She developed urosepsis with Proteus mirablis and Klebsiella oxytoca bacteraemia on day 5 and was treated with intravenous antibiotics.
On day 14 her serum creatinine rose to 173 μmol/L, with no evidence of TMA. Peripheral blood film showed no schistocytes or reticulocytosis, Hb was stable at 73 g/L, platelets 161 × 109, WCC 8.1 × 109/L with lymphopenia (0.54 × 109/L) but a normal neutrophil count (7.01 × 109/L); LDH was 154 U/L. She was treated with pulse methylprednisolone over the weekend prior to a biopsy. On day 16 a renal biopsy showed severe vascular rejection (v3), moderate micro vascular inflammation (g2, ptc2), acute tubular necrosis and moderate tubulointerstitial inflammation (i2, i2) (Fig. 2a). C4d was negative in peritubular capillaries and glomerular capillaries but appeared to stain arteriolar endothelium. She underwent plasma exchange and was commenced on thymoglobulin and IVIG. Tacrolimus was withheld for 5 days during thymoglobulin treatment. Prior to instigating thymoglobulin, tacrolimus levels had been therapeutic and ranged between 3.8 and 9.2 μg/L. Levels were unrecordable during the period it was withheld and remained therapeutic during the remainder of treatment.
Her renal function continued to deteriorate and repeat biopsy on day 25 showed persisting V3 and ATN with slightly worse micro vascular inflammation (g3, ptc2) and tubulointerstitial inflammation (i3, t3) (Fig. 2b). C4d staining patterns remained the same.
Anti-C5 antibody therapy was not available. DS had been doing very poorly on dialysis pre-transplant and was very keen to pursue all avenues of treatment. In this setting of severe, treatment refractory rejection a splenectomy was performed and she was continued on plasma exchange. After some initial improvement, her creatinine continued to rise and a progress biopsy at 5 weeks was remarkable for cortical necrosis and interstitial haemorrhage (Fig. 2c). V3 was still present, as was severe tubulointerstitial inflammation (i3, t3). Mild tubular atrophy was thought to be present but it was difficult to assess the amount of interstitial fibrosis. No transplant glomerulopathy was evident. Very focal, weak C4d positivity was noted in peritubular capillaries; arteriolar wall staining was again noted.
Six weeks post transplant she developed P. mirablis line sepsis and repeat biopsy showed ongoing rejection and more scarring than previously. Her creatinine had risen to 497 μmol/L and emergency dialysis was required for pulmonary oedema. In the setting of uncontrolled rejection on maximal treatment it was considered futile to continue and graft nephrectomy was performed on day 50 post transplant (Fig. 2d).
Luminex at 4 weeks showed a new donor specific antibody (DSA) to DR 52 (MFI 1094) however when repeated in 2013 showed antibodies to each of the 5 mismatched antigens in the graft with MFI ranging between 8000–15 000.
DNA was extracted and sent for analysis at the Immunology Laboratory, Hunter Area Pathology Service, Newcastle, Australia and analysed using a Fluidigm microchamber chip for the first round of nested PCR and sequencing using Massively Parrallel Sequencing (‘nextgen’) on Illumina Miseq. Variants in CD46/MCP, CFH and CFI were assessed using phenotype prediction models (SIFT, Polyphen2, Align, MutationTaster), publically available genome data (1000Genome Project), mutation registries and past publications. Likely pathogenic single nucleotide polymorphisms were identified in CD46/MCP (104G>A, C35Y)) and CFH (3590T>C, V1197A). Further variants of uncertain though potential pathogenic significance were also found in both CD46/MCP (565T>G, T189D) and CFH (3226C>G, Q1076E; 3572C>T, S1191L). Further confirmatory testing is awaited.
In summary, a DBD renal transplant for ESRD secondary to aHUS was performed. After good early graft function intractable ABMR developed that was unresponsive to aggressive therapy with high dose methyl prednisone, anti-thymocyte globulin and plasma exchange and resulted in rapid graft loss and transplant nephrectomy. Of note, at no stage were any haematological features of thrombotic microangiopathy demonstrable, with LDH and haptoglobin in the normal range and no significant thrombocytopenia or schistocytes present.
This case illustrates an interesting manifestation of severe ABMR causing rapid loss of the renal allograft in a previously unsensitized patient, thought to be secondary to dysregulated complement activation in the setting of genetic polymorphisms previously found to be associated with aHUS. Our patient was demonstrated to have a combined mutation to both CFH and MCP. Combined mutations have been reported in approximately 3% of patients. CFH blocks the formation of C3 convertase and accelerates its breakdown. CFH can also bind to negatively charged molecules within the kidney to regulate the activation of complement on the cell surface. The surface of glomerular endothelium shows high levels of MCP expression where it provides additional cofactor activity for CFI. Wild-type MCP should have been present in the donor kidney and the donor did not undergo MCP genotyping. It is of interest the recipient of the partner kidney also developed ABMR/TMA to a less severe degree, unfortunately neither the donor or the second recipient was tested for complement mutations.
Post-transplant focus is usually on the risk of recurrent aHUS. The risk depends on the genetic abnormality involved and is higher in patients with CFI and CFH mutations and may be up to 50–100% in these groups compared with 15–20% in the group with MCP mutations.[4-6] It has been shown that 50% of patients with confirmed aHUS have recurrent disease in the graft after transplant, and of these 90% progress to graft failure.[4, 6] Although there is increasing interest in the role of complement in the development and propagation of acute antibody-mediated renal allograft rejection via terminal complement activation very little is known about the incidence of AMR in patients with aHUS, who would theoretically be at increased risk. Interesting to note, in the study by Le Quintrec, that 60% of patients with recurrent aHUS had rejection. Same group demonstrated that 30% of patients with de novo TMA post transplant had a mutation in CFH or CFI. Very little study has been done on the impact of complement dysregulation on the development of anti HLA antibodies however the strength of the HLA antibody formation was striking in this case.
Of interest is the case report by Noone et al. of a patient with ESKD secondary to spina bifida whose first graft was lost due to acute rejection and who was subsequently highly sensitized. The patient received a second transplant following a desensitization protocol with a graft to which she had 3 low titre DSA. She developed early oliguric renal failure, severe TMA that was unresponsive to standard therapy and significant increases in antibodies to the mismatched class I and II antigens. She was treated with 2 doses of eculizumab with good effect with rapid normalization of her platelets and creatinine. Subsequent renal biopsy demonstrated ABMR. Complement factor H related protein 3/1 deficiency was subsequently demonstrated.
Unfortunately for our patient we were unable to access eculizumab which is a promising new therapy and has been used pre-emptively in patients with aHUS undergoing transplant, and also after the identification of recurrence or rejection post transplant. Eculizumab is a recombinant humanized monoclonal IgG2/4 antibody that binds specifically to complement protein C5 inhibiting its cleavage to C5a and C5b preventing the formation of the terminal complex (MAC). The role of this expensive medication in transplantation requires further study.[9, 10]
In summary, this case illustrates that genetic abnormalities in the complement regulatory proteins may be associated with severe, uncontrolled antibody-mediated rejection and contribute to the poor graft outcome in patients with aHUS in the absence of haematological changes of TMA.