Renal Transplantation in Systemic Amyloidosis—Importance of Amyloid Fibril Type and Precursor Protein Abundance


Corresponding author: Jennifer H. Pinney,


Renal transplantation remains contentious in patients with systemic amyloidosis due to the risk of graft loss from recurrent amyloid and progressive disease. Outcomes were sought among all patients attending the UK National Amyloidosis Centre who received a renal transplant (RTx) between January 1978 and May 2011. A total of 111 RTx were performed in 104 patients. Eighty-nine percent of patients with end-stage renal disease (ESRD) due to hereditary lysozyme and apolipoprotein A-I amyloidosis received a RTx. Outcomes following RTx were generally excellent in these diseases, reflecting their slow natural history; median graft survival was 13.1 years. Only 20% of patients with ESRD due to AA, AL and fibrinogen amyloidosis received a RTx. Median graft survival was 10.3, 5.8 and 7.3 years in these diseases respectively, and outcomes were influenced by fibril precursor protein supply. Patient survival in AL amyloidosis was 8.9 years among those who had achieved at least a partial clonal response compared to 5.2 years among those who had no response (p = 0.02). Post-RTx chemotherapy was administered successfully to four AL patients. RTx outcome is influenced by amyloid type. Suppression of the fibril precursor protein is desirable in the amyloidoses that have a rapid natural history.


Hereditary Apolipoprotein AI amyloidosis


Hereditary Apolipoprotein AII amyloidosis


Hereditary Fibrinogen amyloidosis


Lysozyme amyloidosis


Bence Jones Protein


Chronic allograft nephropathy


Chronic Kidney disease


Combined liver kidney transplant


Complete response


Difference between the involved and uninvolved light chain


Estimate glomerular filtration rate


End stage renal disease


End stage renal failure




Free light chain


Interquartile range


Leukocyte chemotactic factor 2


National Amyloidosis Centre


No response


Orthotopic liver transplantation


Partial response


Post transplant lymphoproliferative disorder


Renal replacement therapy


Renal transplantation


Serum amyloid A protein


123I-labeled serum amyloid P component.


Amyloidosis describes a group of disorders caused by deposition of amyloid fibrils. There are more than 20 different proteins which can form amyloid fibrils in vivo and form the basis of the modern classification of amyloidosis [1]. Although there is considerable overlap between the clinical syndromes associated with various types of amyloid, seven types of systemic amyloidosis commonly affect the kidneys and cause progressive renal dysfunction. The most commonly diagnosed type, AL amyloidosis, is due to deposition of monoclonal immunoglobulin light chain derived fibrils. In AA amyloidosis, the fibrils are composed of serum amyloid A protein (SAA), an acute phase protein, and in hereditary fibrinogen A-chain (AFib), apolipoprotein A-I (AApoAI), apolipoprotein A-II (AApoAII) and lysozyme (ALys) amyloidosis the fibrils are derived from structurally abnormal ‘amyloidogenic’ protein variants. A newly identified amyloid fibril protein, leukocyte chemotactic factor 2 (LECT2), is associated with renal amyloidosis but appears to run a relatively benign course in most affected patients [2].

Disease natural history and organ involvement vary widely according to the amyloid type. Approximately 50% of patients with AL amyloidosis present with renal dysfunction, which in the absence of disease modifying treatment, typically progresses rapidly to end-stage renal disease (ESRD) and/or death [3, 4]. AA amyloidosis confers a better overall prognosis but is also typically associated with progression to ESRD which can be rapid [5]. AFib, which presents universally with proteinuria, is also characterized by a progressive decline in kidney function to ESRD within 5 years of diagnosis [6]. AApoAI is phenotypically heterogeneous but chronic kidney disease (CKD) is common and typically progresses more slowly to ESRD than AL, AA and AFib [7]. Likewise, ALys usually causes slowly progressive CKD [8, 9].

In both AL and AA amyloidosis, suppression of the fibril precursor protein alters disease natural history [5, 10, 11]. Chemotherapy to suppress production of amyloidogenic monoclonal immunoglobulin light chains in AL amyloidosis [12], and antiinflammatory therapy to suppress production of SAA in AA amyloidosis [5] improve renal outcomes and patient survival. Fibrinogen A-chain is synthesized by the liver alone [13] and approximately 50% of plasma apolipoprotein A-I is liver derived [14]. Orthotopic liver transplantation (OLT) therefore results in complete replacement of variant amyloidogenic fibrinogen in AFib and partial replacement of variant apoAI in AApoAI by the respective normal protein and thus halts or slows amyloid accumulation [7, 15, 16]. It is not currently possible to alter the natural history of ALys.

Renal transplantation remains a contentious issue in patients with systemic amyloidosis. This is predominantly due to fears of early allograft loss from recurrent amyloid and poor outcomes related to progressive extrarenal amyloidosis. Concerns that immunosuppression could increase the malignant potential of low-grade plasma cell dyscrasias have also been expressed although remain largely unfounded. Patient survival in AL amyloidosis has improved in association with an ever increasing armamentarium of available chemotherapeutic agents [17] and similarly, there are new therapeutic options for patients with chronic inflammatory diseases [18]. These advances mean that there are more ESRD patients with stable extrarenal AL and AA amyloid deposits. A recent study of renal transplantation in AA amyloidosis found that 14% of patients developed recurrent disease in their graft which was associated with an increased risk of death; it was therefore hypothesized that there may be an association between SAA concentration and recurrent disease although this has yet to be systematically tested [19].

Here we report the outcome of renal transplantation in 104 consecutive patients with different forms of systemic amyloidosis evaluated at the UK National Amyloidosis Centre (NAC). Our findings highlight both the relationship between the natural history of the disease and outcome following renal transplantation and how suppression of the amyloid fibril precursor protein affects recurrence of amyloid in the graft.

Materials and Methods


All patients who received either a RTx or LKT between January 1978 and May 2011 were identified from the NAC database. The amyloid type was confirmed in each patient by review of histology [20] and where appropriate, by additional genetic analysis, as previously described [21]. The diagnosis of amyloidosis had been established prior to renal transplantation in 91 patients and only after renal transplantation in 13 cases.

Patients with AL amyloidosis were reviewed at the NAC every 6 months and those with AA and hereditary amyloidosis were reviewed annually.

All patients were managed in accordance with the declaration of Helsinki. Informed patient consent and institutional review board approval from the Royal Free Hospital Ethics Committee were obtained for this study.

Assessment of fibril precursor protein abundance

Among patients with AL amyloidosis, markers of the underlying clonal dyscrasia (serum free light chains, serum paraprotein and urinary Bence Jones proteins) were monitored at each NAC attendance. In addition, serum free light chain concentration was analyzed monthly between appointments. The clonal response was defined according to the minimum response achieved after renal transplantation (e.g. a complete clonal response must have been achieved throughout follow-up) by previously published criteria [22] adapted for use in CKD [12] as follows; complete response (CR) was a normal serum free light chain (FLC) ratio and no detectable serum paraprotein or Bence Jones protein (BJP), no response (NR) was a dFLC concentration (difference between the involved and uninvolved light chain) of >50% of the pretreatment value, and partial response (PR) was a dFLC concentration of <50% the pretreatment value in the absence of CR.

Among patients with AA amyloidosis response was defined according to serial SAA concentration, measured at each NAC appointment and at monthly intervals in between [5]. Median SAA concentration was calculated for each year after renal transplantation and response was defined according to the minimum response after renal transplantation as follows; complete response was a median SAA <10 mg/L, PR was defined as median SAA from 10 to 50 mg/L and NR >50 mg/L, as previously described [11].

Fibrinogen is made exclusively by the liver and the amyloidogenic variant protein disappears from the plasma after liver transplantation in AFib [13]. Liver transplantation in AApoAI results in a 50% reduction in the plasma concentration of the amyloidogenic apoAI variant [14].

Serum lysozyme concentration changes with GFR and is associated with allograft function following renal transplantation [23]. It is not known what proportion of variant lysozyme remains in the plasma after solid-organ transplantation. Patients with ALys were therefore excluded from the analysis of outcome by precursor protein response.

Assessment of amyloid load and recurrent disease

Serial whole-body anterior and posterior scintigraphic imaging after administration of 123I-labeled serum amyloid P component (SAP) was undertaken at each clinic visit to establish baseline and change in whole-body amyloid load, as previously described [24]. SAP scintigraphy has previously been validated as a tool for assessment of recurrent graft amyloid [25]. Recurrent amyloid in the graft was defined by abnormal uptake of tracer in the transplant on SAP scintigraphy ([23] cases) and was corroborated by histology showing amyloid in seven such cases. Patients underwent renal biopsies where clinically indicated. All cases that had recurrent amyloid on biopsy were associated with abnormal uptake of tracer within the allograft on SAP scintigraphy.

Assessment of organ function

Renal allograft function was assessed at each NAC appointment by measurement of serum creatinine, MDRD eGFR, measured creatinine clearance and 24 h urine protein. Graft failure was defined as the date of recommencement of renal replacement therapy (RRT).

Statistical analysis

Results were expressed as median and interquartile range (IQR) or percentage. Patients were censored at their last NAC clinic visit and patient and graft survival were estimated by Kaplan–Meier analyses. Graft survival was noncensored for death; whereby death with a functioning graft was classified as graft loss. The log-rank test was used to compare the difference in stratified Kaplan–Meier survival analyses. Statistical analysis was with the Man–Whitney U-test as all data analyzed were nonparametric (Graph Pad Prism version 5, Graph Pad, San Diego, CA, USA).


One hundred and four patients with systemic amyloidosis received a total of 111 renal transplants including 10 LKTs and one combined cardiac and renal transplant. Sixty patients (58%) were male and 29 (28%) patients received grafts from live donors. A small proportion (12%) of patients received preemptive renal transplants, most commonly those with ALys where the rate of renal decline is typically slow (Table 1). A variety of immunosuppression regimens were used in accordance with local protocols (Table 1). Transplant survival differed between the different amyloid types and there appeared to be an association between natural history of disease and transplant survival such that the amyloidosis syndromes associated with gradual loss of native GFR, ALys and AApoAI, were associated with longer transplant survival than those known to cause rapid GFR loss, AL, AA and AFib (Figure 1, log rank p = 0.03). Amyloid did not recur in the renal allograft of any patient who achieved complete suppression of the precursor protein throughout the duration of their follow-up despite prolonged follow-up after transplantation in certain patients (Figure 2).

Table 1. Baseline characteristics and outcome of patients who underwent renal transplantation (N = 104)
  No. of AL patients N (%)No. of AA patients N (%)No. of AFib patients N (%)No. of AApo AI patients N (%)No. of ALys patients N (%)
  1. a

    One combined cardiac and renal transplant.

  2. b

    Three patients received renal transplants before they had baseline SAP scintigraphy.

  3. c

    AA amyloidosis patients only.

  4. d

    AL amyloidosis patients only.

Total number of patients 254319143
Total number of grafts 254621163
Total number of combined liver kidney transplants009210
SexMale9 (36)27 (62)12 (63)10 (71)2 (66)
Age a transplant (years)Median (IQR)60 (52–63)37 (29–48)59 (56–61)49 (37–56)45 (32–62)
DonorsLive5 (20)16 (35)2 (9.5)3 (14)3 (100)
PreemptiveYes1 (4)3 (6.5)5 (24)1 (6.25)2 (66)
Time from diagnosis to ESRF (years)Median (IQR)1.1 (0.0–3.6)1.4 (0–5.8)1.06 (0.03–1.6)0.99 (0–27.4)10.6 (0–14.6)
Time from ESRF to first transplant (years)Median (IQR)2.3 (1.1–5.0)1.5 (0.9–3.0)1.3 (0–2.8)2.5(0.5–3.2)0 (0–0.2)
Amyloid load atSmall7 (28)7 (16)12 (63)2 (14.5)0 (0)
presentationModerate8 (32)26 (60)6 (31.5)4 (28.5)1 (33)
 Large7 (28)10 (23)1 (5.5))8 (57)2 (67)
 Missing23 (12)0 (0)0 (0)0 (0)0 (0)
Other organ involvementCardiac5 (20)0 (0)0 (0)1 (7.1)Cardiac
at presentationLiver12(48)11 (25.5)1 (5.2)12 (85.7)3 (100)
 Spleen21 (87.5)43 (100)17 (89.4)14 (100)2 (66.6)
 Adrenals1 (4)7 (16.2)1 (5.2)0 (0)0 (0)
Underlying disease3Hereditary fever syndrome 10 (23.2)   
 Inflammatory bowel disease 9 (20.9)   
 Inflammatory arthritis 16 (37.2)   
 Recurrent infection 5 (11.6)   
 Castleman's 2 (4.65)   
 Unknown 1 (2.32)   
Number of lines of03 (12)    
treatment4112 (48)    
 26 (24)    
 32 (8)    
 42 (8)    
ImmunosupressionCNI/Pred/MMF7 (28)9 (20.9)5 (26.3)6 (42.8)2 (66.6)
regimenCNI/Pred5 (20)9 (20.9)3 (15.7)3 (21.4)0 (0)
 CNI/MMF2 (8)2 (4.6)1 (5.2)1 (7.1)1 (33.3)
 Aza/Pred1 (4)7 (16.2)2 (10.5)1 (7.1)0 (0)
 MMF/Pred0 (0)1 (2.3)0 (0)0 (0)0 (0)
 CNI only2 (8)3 (6.9)0 (0)0 (0)0 (0)
 Sirolimus/Pred/MMF2 (8)3 (6.9)1 (5.2)0 (0)0 (0)
 Sirolimus only1 (4)0 (0)0 (0)0 (0)0 (0)
 Pred only0 (0)0 (0)0 (0)1 (7.1)0 (0)
 Unknown5 (20)9 (20.9)7 (36.8)2 (14.2)0 (0)
Grafts with recurrentNumber7 (28)9 (19.5)7 (33.3)3 (18.75)0 (0)
amyloidMissing2 (8)4 (8.6)0 (0)2 (12.5)0 (0)
Follow-up posttransplantMedian (IQR)4.6 (2.6–6.9)5.1 (3.4–11.6)3.3 (0.5–6.5)9.8 (3–13.7)2.9 (0.9–6.8)
Number of deaths 13 (52)16 (37.2)7 (36.8)1 (7.1)0 (0)
Figure 1.

Renal transplant survival in years, noncensored for death stratified by disease natural history. Median survival in apolipoprotein A-I and lysozyme amyloidosis (slow natural history) was significantly longer than AL, AA and fibrinogen amyloidosis (fast natural history) (median survival 13.1 years vs. 8.3 years; p = 0.03).

Figure 2.

Time to recurrent amyloid in all patients by precursor protein abundance. *Patients with combined liver kidney transplant in Fibrinogen amyloid are defined as CR and those with renal transplant NR and patients with Apolipoprotein AI amyloidosis are defined as PR.

AL amyloidosis

Twenty five of 246 (10.2%) patients with AL amyloidosis who received RRT underwent renal transplantation, among whom median time to ESRF from diagnosis of amyloidosis was 1.1 (IQR 0–3.6) years, and median time to transplantation from ESRF was 2.3 (IQR 1.1–5.0) years. Median follow-up from transplantation was 4.6 (IQR 2.6–6.9) years. Thirteen (52%) patients died, most commonly from infection (6 patients), with 1 patient each dying from GI blood loss and cardiac decompensation. The remainder (5 patients) died from unknown causes. No patient who died from infection was receiving chemotherapy at the time. At the time of transplantation, amyloid was present in the spleen, liver and autonomic nerves in 21, 12 and 2 patients respectively. Five patients had cardiac amyloidosis deemed ‘mild’ on the basis of their echocardiogram. ECOG performance status was <2 in every case. The presence of extrarenal involvement by amyloid did not significantly influence patient survival (data not shown).

Median graft survival was 5.8 years, 5- and 10-year graft survival was 74% and 25%, respectively. Two patients lost their grafts, one to chronic allograft nephropathy (CAN) after 2.9 years and one from recurrent transplant pyelonephritis and obstructive nephropathy after 0.9 years. No grafts failed from recurrent amyloid, despite the presence in 7 (28%) patients of renal allograft amyloid, diagnosed a median of 5.9 (IQR 3.8–6.3) years after transplantation.

Twenty-two patients (88%) received chemotherapy in total. Clonal response at the time of transplantation was variable; 5 (20%) patients were in a clonal CR, 13 (52%) were in a PR, 3 (12%) had not achieved any clonal response or had not been treated prior to renal transplantation and 4 (16%) were not evaluable for clonal response (Table 2). Among 21 patients who were evaluable for clonal response, 2 died within 1 year of RTx, 18 were followed for more than 1 year post-RTx, and one was lost to follow-up (Table 3). There was no significant difference in renal allograft survival between patients who were in CR at the time of transplantation and those in PR although patients who had not achieved at least a PR prior to transplantation had significantly worse graft survival (5.3 vs. 8.9 years; p = 0.02) (Figure 3). Five patients (28%) had a clonal relapse following transplantation after a median of 2.0 years for which four patients received further chemotherapy, three achieving a subsequent CR and one a PR. No patient developed symptomatic myeloma during follow-up.

Table 2. Fibril precursor protein response prior to renal transplantation
Type of amyloid (precursor protein)Abundance of precursor proteinNo. (%)
  1. 1In patients with AL amyloidosis; CR, complete response is defined as normal free light chain ratio, no paraprotein or Bence Jones proteins, partial response (PR) is defined as dFLC response of >50% and no response (NR) a dFLC of < 50%. (dFLC: the difference between the involved and uninvolved light chain).

AL (serum freeClonal response atCR5 (20)
 light chain)1 transplantationPR13 (52)
  NR3 (12)
  unknown4 (16)
AA (serumMedian SAA (mg/L)<1013 (30.2)
 amyloid A during 6 months10-5014 (32.5)
 protein) prior to>504 (9.3)
  transplantationunknown12 (27.9)
AFib (variantCLKTComplete9 (47.3)
 fibrinogen)  removal 
 Kidney onlyNo reduction10 (52.6)
AApoAI (variantCLKT50% reduction2 (12.5)
 ApoAI)Kidney onlyNo reduction14 (87.5)
ALys (variantKidney onlyNo reduction3 (100%)
Table 3. Response to chemotherapy in 18 evaluable patients with AL amyloidosis
Patient no.Response at TxNumber of years followed- up after transplantDead/aliveGraft loss/eGFR at last visitRelapse Y/N (year after Tx patient relapsedChemotherapy after relapseClonal response after further treatmentLast known clonal responseRecurrent amyloid in graft at last visit (yes/no)
  1. *Patient 7 was not treated with chemotherapy until progressive disease and recurrence in the graft was established 7 years after transplantation.

  2. CVAD = cyclophosphamide, vincristine, doxorubicin and dexamethasone; CR = complete response; VGPR = very good partial response; PR = partial response; NR = no response.

1VGPR6Dead33No  VGPRyes
2CR8Dead18No  CRno
3VGPR8Alive49Yes (2)Intermediate dose Melphalan 4 cyclesCRCRno
4VGPR12Alive37Yes (3)CVAD 3 cyclesCRCRyes
5CR6Alive39No  CRno
6VGPR2Alive55Yes (1)Bortezomib/dexamethasone 3 cyclesCRCRno
7NR5Dead30No (7*)Melphalan/dexamethasone 8 cyclesPRPRyes
8CR4Alive41No  CRNo
9VGPR4Dead78No  VGPRYes
10PR3Dead48Yes (1)Nil NRNo
11NR3Dead42No  VGPRNo
12VGPR1Dead58Yes (2)Cyclophosphamide/dexamethasone 5 cyclesVGPRVGPRNo
13NR4Alive45No  VGPRNo
14VGPR4Alive48No  VGPRNo
15PR8Alive33No  VGPRNo
16VGPR5Dead37No  VGPRNo
17CR4Alive45No  CRNo
18CR5Alive26No  CRno
Figure 3.

Graft survival, noncensored for death in patients with AL amyloidosis according to clonal response at the time of transplantation.

AA amyloidosis

Forty-three of 128 (33.6%) patients with AA amyloidosis who required RRT underwent renal transplantation, receiving a total of 46 grafts. The inflammatory diseases underlying AA amyloidosis included inflammatory arthritis in 16 (37.2%) patients, hereditary periodic fever syndromes in 10 (23.2%), inflammatory bowel disease in 9 (20.9%), recurrent infection in 5 (11.6%), Castleman's disease in 2 (4.7%) and was undefined in one patient. Median time from diagnosis to ESRF was 1.4 (IQR 0–5.8) years, and median time from ESRF to transplantation was 1.5 (IQR 0.9–3.0) years. Median follow-up from transplantation was 5.1 (IQR 3.4–11.6) years. Sixteen (37.2%) patients died, most commonly from infection (6 cases). One case each died from posttransplant lymphoproliferative disorder (PTLD), cervical malignancy, infarction of the transplant and bowel perforation and six from unknown causes.

Median estimated graft survival noncensored for death was 10.3 years and 8 (18.6%) grafts had failed at censor. Five and 10-year graft survival was 86% and 59% respectively. There was no significant difference in renal allograft survival between patients stratified by amyloid load or involvement of the liver, the latter being a measure of advanced AA amyloidosis [26]. Nine patients (19.5%) had recurrent amyloid in their renal allograft diagnosed a median of 5.3 (IQR 2.0–7.5) years after transplantation; two such cases lost their grafts due to recurrent amyloid. Three patients received a second RTx after failure of the first: one from primary non function whose second renal allograft lasted 17 years; one after 24 years from CAN whose second transplant is functioning well after 1 year; and one from recurrent amyloid after 8 years (6 years after amyloid was first identified in the renal allograft) who died from sepsis 1 month after the second RTx.

Twenty-nine patients had serial SAA monitoring in the 6 months prior to transplantation. Graft survival noncensored for death was 14.5 years in patients with a median SAA value of <10 mg/L, and 7.8 years in those with a median SAA of >10 mg/L (p = ns). Median SAA concentration was significantly higher among those with recurrent amyloid in their graft compared to those in whom amyloid did not recur (p = 0.04) (Figure 4).

Figure 4.

Median serum amyloid A (SAA) protein concentration measured during the 6 months prior to diagnosis of recurrent amyloid compared to the last 6 months of follow-up in patients without recurrent amyloid.

Hereditary fibrinogen A-chain amyloidosis (AFib)

Nineteen of 51 (37.2%) patients with AFib who required RRT underwent renal transplantation receiving a total of 21 renal allografts including 9 patients who received CLKT. One patient received CLKT following failure of an initial isolated renal transplant after 5.6 years; CLKT was the initial transplant procedure among the remainder. Median time to ESRF from clinical presentation was 3.16 (IQR 1.5–7.8) years and median time from ESRF to transplantation was 1.4 (IQR 0–2.8) years. Four of 10 patients with RTx died, two from malignancy, one following a GI bleed and one from an unknown cause. Among nine patients who received CLKT, three (33.3%) died in the early postoperative course from perioperative complications.

Median graft survival among patients who received isolated RTx (defined as NR) was 7.3 years compared to 6.4 years in those who received LKT (defined as CR) (p = ns). Five and 10-year graft survival was 85% and 30% in patients with an isolated RTx, and 63% and 31% in those with CLKT. Recurrent amyloid was identified in the renal allografts of seven patients, all of whom had isolated RTx, after a median of 4.9 (IQR 4.3–6.0) years. No patient with CLKT developed renal allograft amyloid. Four isolated RTx failed, three from recurrent amyloid and another from primary nonfunction. One patient was discovered to have renal allograft amyloid at the time of graft failure, and graft loss occurred 1.2, and 2.5 years after discovery of amyloid in the remaining two cases. One CLKT patient lost their renal allograft after 6.5 years from CAN; none developed recurrent amyloid.

Hereditary apolipoprotein A-I amyloidosis (AApoAI)

Fourteen of 16 (87.5%) patients with AApoAI who required RRT underwent renal transplantation receiving a total of 16 renal allografts including two patients who received CLKT, one who was planned for CLKT but received the liver transplant alone due to intraoperative donor complications followed by a RTx 3 years later, and one who received a combined heart and renal transplant. Median time from clinical presentation to ESRF was 5.9 (IQR 4.2–10.8) years. Median time to transplantation from ESRF was 2.5 (IQR 0.5–3.2) years. One patient died during follow-up from an unknown cause 13 years after renal transplantation.

Median graft survival noncensored for death was 13.1 years; significantly longer than patients with AA, AL and AFib amyloidosis (p = 0.02). Five- and 10-year graft survival was 100% and 77% respectively. Three patients (18.8%), all of whom had received isolated RTx, were found to have recurrent amyloid in their grafts after a median of 3.5 (IQR 2.8–17) years. Graft loss due to recurrent amyloid occurred in one case after a further 2 years of follow-up with the other two cases losing their grafts after 8.8 and 12.5 years from chronic allograft nephropathy.

Hereditary lysozyme amyloidosis (ALys)

Three of three patients with ALys who required RRT received renal transplants, all from living donors. Two renal transplants were preemptive and the third patient only received dialysis for 2 months prior to RTx. Decline in native renal function was very slow; the median time from diagnosis to ESRF was 10.6 (IQR 0–14.6) years. All three grafts were functioning well 0.9, 2.9 and 6.2 years after RTx without evidence in any case of amyloid recurrence.


This study details outcome associated with RTx in different types of renal amyloidosis and in particular, highlights the importance of the natural history of particular types of amyloidosis disease on outcome. The slowly progressive nature of ALys [9] and AApoAI [7] is well established and corroborated by a mean rate of GFR loss among those with nephropathy visiting our center of only 4.7 mL/min/year (95% CI 0.9–8.6 mL/min/year). Despite the presence of extensive liver amyloid in most patients with these types of hereditary amyloidosis at the time of renal transplantation, and the persistent and undiminished production of the amyloid fibril precursor protein after the RTx procedure in all but two CLKT recipients, median estimated graft survival was 13.1 years with only 1/19 grafts lost to recurrent amyloid. There was only one death after median follow-up from transplantation of 8.8 (IQR 1.9–12.5) years. Furthermore, nearly all (17 of 19) patients with these types of amyloidosis who reached ESRD actually underwent renal transplantation.

These findings contrast AL, AA and AFib which are known to be associated with a more rapid disease natural history reflected by greater rates of GFR loss [6, 12]. Until relatively recently there was a reluctance among nephrologists to proceed with renal transplantation in AA and AL amyloidosis due to the well-documented risk of allograft failure from recurrent amyloid combined with the risk of death from progressive extrarenal amyloidosis [27-29]. Furthermore, amyloid typically recurs within renal transplants in AFib leading to graft failure between 5 and 12 years after RTx [6]. The natural history of AA, AL and AFib are modifiable however, and studies have shown that suppression of the fibril precursor protein concentration in both AA [5] and AL [12] amyloidosis is associated with improved patient and renal survival. It stands to reason therefore, that the concentration of the fibril precursor protein is also likely to influence outcomes following renal transplantation. In AFib the precursor protein is made in the liver and liver transplantation therefore removes the amyloidogenic protein thus preventing ongoing amyloid accumulation [30, 31]. This study demonstrates good transplant outcomes among carefully selected patients with AA and AL amyloidosis such that only about one third of ESRD patients with AA amyloidosis and ∼10% of ESRD patients with AL amyloidosis underwent renal transplantation. The median graft survival was 10.3 and 5.8 years in AA and AL patients in this series respectively. Amyloid was detected in the renal allografts of 9/43 (21%) and 7/25 (28%) patients with AA and AL amyloidosis respectively, but only two of these 68 grafts failed from recurrent amyloid despite up to 25 years of follow-up. Graft outcomes were better among patients with suppression of the fibril precursor protein concentration prior to renal transplantation, and risk of amyloid recurrence in AA amyloidosis was higher among patients with elevated SAA levels. To what extent the fibril precursor protein needs to be suppressed in any particular individual in order to prevent recurrence in the renal allograft is not known. In this series there was no difference in renal allograft survival between AL amyloidosis patients who had achieved a clonal CR or clonal PR, although outcomes were poorer among those who had not achieved any clonal response prior to renal transplantation. Five (28%) patients experienced a clonal relapse following renal transplantation, and all four of those who received chemotherapy achieved at least a partial clonal response following further treatment, with no detrimental effect on renal allograft function. Based on these data, our recommendations would be to try and achieve at least a partial clonal response in AL amyloidosis prior to listing for renal transplantation but that the absence of a complete clonal response should not preclude listing, particularly since it is usually possible to achieve a deeper clonal response with further chemotherapy once a patient has received a RTx. Our practice is to list patients immediately once a clonal response has been obtained.

The different amyloid types are associated with different patterns of organ involvement [32]. Furthermore, there is considerable heterogeneity in organ involvement, even within the same amyloid type. In this series, we did not find any relationship between patient survival and pattern of extrarenal organ involvement, nor total body amyloid burden. The presence of extrarenal amyloid should not therefore exclude patients from being listed for RTx. We believe that as long as ECOG performance status and exercise tolerance are preserved, thereby excluding patients with substantial cardiac and autonomic nerve involvement by amyloid, RTx should be considered, particularly if the extrarenal disease is stable in the context of a clonal response. It is notable however, that very few patients in this series had cardiac amyloidosis which is known to confer a poor prognosis in AL amyloidosis [33] and is a rare complication of advanced AA amyloidosis [34]. Acute cardiovascular events were recently reported to be a significant cause of death among renal transplant recipients with AA amyloidosis [19]. There is little doubt that all patients with systemic amyloidosis should undergo an extensive cardiovascular assessment including specialized tests for the presence and degree of cardiac amyloidosis prior to renal transplantation.

In summary, progression of extrarenal amyloid and risk of recurrent disease in the renal allograft have, until recently, been major barriers to renal transplantation in systemic amyloidosis. No patient in this series whose precursor protein was completely suppressed, regardless of the amyloid type, developed recurrent disease in their graft. In diseases such as AL and AA amyloidosis where the precursor protein can be modified, it is important to monitor precursor protein concentration pre- and postrenal transplantation. In AFib the only currently available treatment to suppress precursor protein concentration is liver transplantation, and the high risk of early perioperative mortality with CLKT must be balanced against elimination of risk of recurrent amyloid disease in the allograft further down the line; our practice is to consider CLKT in younger and otherwise fitter AFib patients [6]. In the hereditary amyloidoses with a slow natural history such as ALys and AApoAI, there is no need to suppress the precursor protein supply in order to prevent amyloid recurrence in the renal allograft since, in keeping with the disease natural history, this occurs extremely slowly [7, 9]. When considering any patient with systemic amyloidosis for renal transplantation, a search for the presence of cardiac amyloidosis is mandatory.


We thank our colleagues for referring and caring for the patients; A. Hughes, E. Pyart, D. Gopaul and D. Hutt for their technical and clinical support at the National Amyloidosis Centre. In particular we thank Dr. A. Stangou and Dr. J. O'Beirne for their peritransplant care of the subset of patients who underwent liver transplantation.


The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.