*Abbreviations: FAP, familial amyloid polyneuropathy; IVS, intraventricular septum; LVPW, left ventricular posterior wall; OLT, orthotopic liver transplantation; SAP, serum amyloid P component; TTR, transthyretin.
Hereditary amyloidosis is uncommon but is an important model for studying the pathogenesis of amyloid generally. The most common form is familial amyloid polyneuropathy (FAP*), an autosomal dominant syndrome associated with at least 59 different transthyretin (TTR) mutations (1,2). Wild-type TTR is itself weakly amyloidogenic, a property that manifests predominantly as senile cardiac amyloidosis(3). FAP has an onset at any time from the second decade and is characterized by peripheral and autonomic neuropathy in association with variable visceral and cardiac amyloid involvement. Major foci of FAP associated with the most prevalent TTR variant, Met30, occur in Portugal, Japan, and Sweden, and although phenotypic expression varies, FAP is almost always fatal within 5 to 15 years (1,2).
In light of evidence that the liver is the main source of circulating TTR, orthotopic liver transplantation (OLT) was introduced as a radical treatment for FAP (4). Early follow-up data of transplant patients indicate that variant TTR disappears from the plasma, and although the peripheral neuropathy usually only stabilizes, autonomic function can improve substantially (4-8). The associated splenic and renal amyloid deposits that can be imaged by serum amyloid P component (SAP) scintigraphy gradually regress in most cases(9,10). OLT is now regarded as the treatment of choice for this disease.
Although cardiac amyloidosis is commonly present in FAP, it is often asymptomatic and clinically insignificant, until patients are stressed hemodynamically, for example, during liver transplantation (11,12). In the present study, we serially estimated cardiac amyloid infiltration by echocardiography in 20 patients with hereditary TTR amyloidosis and report the unexpected finding that cardiac amyloidosis in some cases may progress rapidly after liver transplantation.
PATIENTS AND METHODS
Comprehensive echocardiographic and Doppler studies were performed by two experienced operators using a standard sequence of views with either a Toshiba SSH 160A or Hewlett Packard SONOS 1000 ultrasound system, blindly, in 20 patients with variant TTR-associated FAP and in a control group of 10 age- and sex-matched patients who underwent OLT for primary liver disease. The latter comprised hepatitis B- or C-related cirrhosis (six cases), alcoholic liver disease (two cases), and autoimmune chronic hepatitis and fulminant liver failure (one case each). Follow-up echocardiography was performed 1 year after OLT in the patients with liver disease, and annually for up to 3 years in 11 FAP patients, before they were listed or while on the waiting list for OLT. The 14 patients with FAP who had undergone OLT were followed annually for up to 3 years.
The FAP patients who received transplants comprised nine who were heterozygous for the Met30 mutation, three with TTR Pro52, one with TTR Thr84, and one patient with TTR Tyr77 who had a combined liver and heart transplant. The FAP patients who did not receive transplants were heterozygous for TTR Met30 (three cases), Pro52 (three cases), Ala60 (two cases), Tyr77, and Thr84 (one case each), and a single patient was homozygous for TTR Met30. All three TTR Pro52 patients, as well as the single Tyr77 and Thr84 patients, received transplants before the end of the study; however, their clinical course had been monitored for a substantial period before as well as after transplantation. Therefore, they were included in both the OLT and non-OLT groups in the analysis of results. Of the 14 FAP patients who underwent OLT, 7 received tacrolimus (FK506)- and 7 received the cyclosporin-based immunosuppression; in the control group, 6 patients received FK506 and 4 received cyclosporine. A right ventricular cardiac biopsy specimen was obtained from one patient with TTR Thr84 6 months after liver transplantation and was stained with Congo red and anti-TTR antibodies. Measurements of thickness of the left ventricular posterior wall (LVPW) and intraventricular septum (IVS) were subjected to statistical analysis using Student's paired t test. Plasma TTR was estimated by electroimmunoassay before and 6 months after OLT in 10 patients with FAP, after an acute phase response that might have confounded the results was excluded. Quantitative whole body scintigraphy with 123I-labeled SAP was also performed in 10 such patients to evaluate their renal and splenic amyloid deposits (9).
None of the patients with chronic liver disease had echocardiographic evidence of cardiac disease at baseline or at follow-up 1 year after OLT(Fig. 1). Serial echocardiograms in the 11 FAP patients who did not receive transplants showed changes consistent with modest progression of amyloid in 4 cases and no progression in the 7 others; the maximum increment in thickness of the LVPW and IVS was 2-3 mm. Mean values of LVPW and IVS thickness were 13.7 mm and 14.6 mm, respectively, at baseline and 14.0 mm and 14.9 mm at follow-up (P=NS). In contrast, substantial progression of the amyloidotic changes at echocardiography was seen in 3 of the 14 FAP patients who had undergone OLT. Two of these patients, both of whom later died suddenly, had TTR Pro52; the other had TTR Thr84 and remains in controlled cardiac failure. Progressive changes at echocardiography included an increased restrictive pattern on transmitral Doppler, a qualitative increase in myocardial echogenicity, and increased thickness of the valves and ventricular walls. Mean IVS thickness in the four non-Met30 cases was 15.2 mm before OLT, increasing to 21.5 mm(P<0.05) within 2 to 12 months (mean: 5.2 months) after OLT(Fig. 2). There was no evidence of cardiac amyloid either before or after OLT in six patients with TTR Met30, and the three other patients with this variant who did have evidence of cardiac amyloid before liver replacement showed substantial improvement afterward. In contrast to the echocardiographic findings, SAP scintigraphy and clinical outcome were stable or improved in every FAP patient who underwent OLT. Total plasma TTR levels increased by up to 60% after OLT in some cases and fell in others, but it did not correlate with the course of the disease(Table 1). Histological examination of a right ventricular biopsy specimen obtained 6 months after OLT from the single patient with TTR Thr84 showed dense infiltration by TTR amyloid and no other pathology. The right ventricle of this patient appeared severely amyloidotic on echocardiography at the time of biopsy: the wall thickness had doubled as compared with a study just before OLT. No echocardiographic abnormalities developed in the cardiac graft that was performed simultaneously with OLT in the patient with TTR Tyr77 (Fig. 2), and no amyloid deposits were identified on routine posttransplant endomyocardial biopsies. None of the patients had hypertension or any identifiable cause of restrictive cardiomyopathy other than amyloidosis.
The present findings show that the cardiomyopathy associated with FAP progresses at an accelerated rate after liver transplantation in some patients. The evidence that this is due to ongoing amyloid deposition, despite the stabilization or regression of amyloid in other sites, is compelling: the full echocardiographic picture is typical (13,14); myocardial histology in the most severely affected case showed TTR amyloid infiltration and no other pathology, and no patient had hypertension or any other identifiable cause of restrictive cardiomyopathy. Progressive cardiac disease occurred only in patients who already had substantial cardiac amyloidosis before OLT and only in cases associated with the TTR Pro52 and Thr84 variants. Patients with TTR Met30, the most prevalent TTR mutation, either had no echocardiographic evidence of amyloid before or after OLT or showed remarkable improvement after the procedure. Our data support a recent report by Dubrey et al.(15) suggesting that patients who have undergone liver transplantation for FAP may continue to develop wall thickening after surgery. The phenomenon was observed among patients with variant TTR Gly42- and Pro36-associated FAP, all of whom had evidence of cardiac amyloidosis before surgery, but was not seen in the TTR Met30-associated FAP cases. The present study, in addition to following all patients who received transplants for FAP, included an age- and mutation-matched group of patients with FAP who had not yet been treated with transplantation and who were investigated with serial echocardiography for up to 3 years, with the changes observed being nonsignificant. A third group of patients who received transplants for primary liver diseases and who were treated with similar immunosuppressive regimens was also included in our study as a control group. Although tacrolimus (FK506) has been reported to cause myocardial pseudohypertrophy in a small number of children, this phenomenon has not been confirmed in adults(16-18), and no such changes were evident in our control group of patients who had received transplants for chronic liver disease.
Lately orthotopic liver transplantation has been performed in nearly a hundred patients with FAP, and it remains the only treatment for this otherwise fatal disease. Clinical follow-up studies have shown that the neuropathy usually stabilizes or improves (4,6-8). Similarly favorable results in our patients were supported by serial SAP scintigraphy, a specific nuclear medicine technique that permits quantitative in vivo monitoring of visceral amyloid deposits(10,19). The SAP scan results mirrored the overall neurological progress of our patients very closely, showing that the identifiable visceral amyloid deposits-that is, those in the kidneys and/or spleens-either remained stable after OLT or gradually regressed. The rate of regression differed substantially between patients, and an individual's capacity to turn over their amyloid deposits is clearly a factor that might influence the course of their cardiac amyloidosis after OLT. Notably, all three TTR Met30 patients whose echocardiogram results improved after OLT showed scintigraphic evidence of regression of amyloid in other sites. Unfortunately, neural amyloid deposits are beyond the resolution of SAP scintigraphy, and evaluation of the heart is impaired by its motility, blood pool content, and the slow passage of labeled SAP across its non-fenestrated capillary endothelium. However, the objective demonstration that visceral amyloid in FAP can regress after treatment in accordance with the findings of SAP scintigraphy in AA (reactive), AL (immunoglobulin light chain), and dialysis amyloidosis supports the idea that amyloid deposits generally exist in a state of dynamic turnover (19-22).
The most plausible explanation of our present observations is deposition of normal wild-type TTR amyloid on a template of amyloid derived from the TTR variant, at a rate that exceeds the individual patient's capacity to mobilize amyloid from this particular site. Wild-type TTR is inherently amyloidogenic, but this property is weak and only manifests in vivo as senile systemic amyloidosis in 25% of very elderly individuals. Deposition occurs mainly in the heart and is otherwise very modest. Although senile TTR amyloidosis does not occur before the seventh decade, individuals can have substantial cardiac deposits by the eighth decade, which suggests that its accumulation can be rapid after the process has begun (23,24). Conversion of normal TTR to its amyloid conformation may therefore be promoted by a template of TTR amyloid. This hypothesis would accommodate the deposition of wild-type TTR amyloid after OLT in the hearts of FAP patients with preexisting cardiac amyloidosis and the lack of it among patients whose echocardiograms were normal beforehand.
Evidence that FAP amyloid deposits in other anatomical sites remain stable or regress after OLT, along with the heart being the principal target for wild-type TTR amyloidosis, suggests that amyloidogenic processing of TTR in the myocardium differs from that in other tissues. Intriguingly, the fibril subunit protein isolated from the hearts of patients with TTR amyloidosis show a preponderance of C-terminal TTR fragments, cleaved at positions 45, 48, and 51 (3,25). It is not known how or when this cleavage occurs, but the absence of variant amino-acid sequence in cardiac amyloid fibrils formed from TTR Met30 may account for the different, favorable course of such patients compared with those in whom the variant sequence is present in the fibrils. The accelerated rate at which the cardiac disease progressed after OLT in the FAP patients with TTR Pro52 and Thr84 suggests that homogeneous wild-type TTR can even be incorporated as fibrils onto the amyloid template in these cases more efficiently than a mixture of wild-type and variant protein.
The clinical implications that arise from these observations will require further follow-up studies. It is reassuring that cardiac involvement improved or remained stable after OLT in all of the patients with TTR Met30, the most common TTR variant, both in the present series and the study from Dubrey et al. (15). Indeed, the progressive cardiomyopathy we observed may be a mutation-dependent phenomenon; fortunately, the prevalence of both TTR Pro52 and Thr84, as well as Gly42 and Pro36, is extremely low. The outcome of the unique TTR Tyr77 variant-associated FAP patient who underwent combined heart and liver transplantation and never developed amyloid on the healthy cardiac graft gives grounds for optimism for other FAP patients who have severe cardiac amyloidosis and further supports the hypothesis that a template of variant TTR amyloid in the heart is probably the sine qua non for accumulation of wild-type TTR amyloid following OLT.
Acknowledgments. This work was supported in part by grants to M.B. Pepys and P.N. Hawkins from the Medical Research Council and The Wellcome Trust.
1. Saraiva MJM, Birken S, Costa PP, Goodman DS. Amyloid fibril protein in familial amyloid polyneuropathy, Portuguese type: definition of molecular abnormality in transthyretin (prealbumin). J Clin Invest 1984; 74: 104.
2. Benson MD, Uemichi T. Transthyretin amyloidosis. Amyloid: Int J Exp Clin Invest 1996; 3: 44.
3. Westermark P, Sletten K, Johansson B, Cornwell GG. Fibril in senile systemic amyloidosis is derived from normal transthyretin. Proc Natl Acad Sci USA 1990; 7(7): 2843.
4. Holmgren G, Ericzon BG, Groth CG, et al. Clinical improvement and amyloid regression after liver transplantation in hereditary transthyretin amyloidosis. Lancet 1993; 341: 1113.
5. Holmgren G, Steen L, Ekstedt J, et al. Biochemical effect of liver transplantation in two Swedish patients with familial amyloidotic polyneuropathy (FAP-Met30). Clin Genet 1991; 40(3): 242.
6. Lewis WD, Skinner M, Simms RW, Jones LA, Cohen AS, Jenkins RL. Orthotopic liver transplantation for familial amyloidotic polyneuropathy. Clin Transplant 1994; 8 (2 Pt 1): 107.
7. Suhr OB, Holmgren G, Steen L, et al. Liver transplantation in familial amyloidotic polyneuropathy: follow-up of the first 20 Swedish patients. Transplantation 1995; 60(9): 933.
8. Ericzon BG, Suhr O, Broome U, et al. Liver transplantation halts the progress of familial amyloidotic polyneuropathy. Transplant Proc 1995; 27(1): 1233.
9. Hawkins PN, Lavender JP, Pepys MB. Evaluation of systemic amyloidosis by scintigraphy with 123I-labeled serum amyloid P component. N Engl J Med 1990; 323: 508.
10. Tan SY, Pepys MB, Hawkins PN. Treatment of amyloidosis. Am J Kidney Dis 1995; 26(2): 267.
11. Eriksson P, Boman K, Jacobsson B, Olofsson BO. Cardiac arrhythmias in familial amyloid polyneuropathy during anaesthesia. Acta Anaesthesiol Scand 1986; 30(4): 317.
12. Suhr OB, Wiklund U, Eleborg L, et al. Impact of autonomic neuropathy on circulatory instability during liver transplantation for familial amyloidotic polyneuropathy. Transplantation 1997; 63(5): 675.
13. Eriksson P, Backman C, Eriksson A, Eriksson S, Karp K, Olofsson BO. Differentiation of cardiac amyloidosis and hypertrophic cardiomyopathy: a comparison of familial amyloidosis with polyneuropathy and hypertrophic cardiomyopathy by electrocardiography and echocardiography. Acta Med Scand 1987; 221(1): 39.
14. Eriksson A, Olofsson BO, Eriksson P. Heart valve involvement in familial amyloidosis with polyneuropathy. Pathol Res Pract 1986; 181(5): 563.
15. Dubrey SW, Davidoff R, Skinner M, Bergethon P, Lewis D, Falk R. Progression of ventricular wall thickening after liver transplantation for familial amyloidosis. Transplantation 1997; 64: 74.
16. Dhawan A, Mack DR, Langnas AN, Shaw BW Jr, Vanderhoof JA. Immunosuppressive drugs and hypertrophic cardiomyopathy [Letter]. Lancet 1995; 345: 1644.
17. Natazuka T, Ogawa R, Kizaki T, et al. Immunosuppressive drugs and hypertrophic cardiomyopathy [Letter]. Lancet 1995; 345: 1644.
18. Atkison P, Joubert G, Barron A, et al. Hypertrophic cardiomyopathy associated with tacrolimus in paediatric transplant patients. Lancet 1995; 345: 894.
19. Hawkins PN. Studies with radiolabeled serum amyloid P component provide evidence for turnover and regression of amyloid deposits in vivo. Clin Sci 1994; 87: 289.
20. Hawkins PN, Richardson S, Vigushin DM, et al. Serum amyloid P component scintigraphy and turnover studies for diagnosis and quantitative monitoring of AA amyloidosis in juvenile rheumatoid arthritis. Arthritis Rheum 1993; 36(6): 842.
21. Hawkins PN, Richardson S, MacSweeney JE, et al. Scintigraphic quantification and serial monitoring of human visceral amyloid deposits provide evidence for turnover and regression. Q J Med 1993; 86: 365.
22. Tan SY, Irish A, Winearls CG, et al. Long term effect of renal transplantation on dialysis-related amyloid deposits and symptomatology. Kidney Int 1996; 50(1): 282.
23. Cornwell GG, Murdoch WL, Kyle RA, Westermark P, Pitkanen P. Frequency and distribution of senile cardiovascular amyloid: a clinicopathologic correlation. Am J Med 1983; 75(4): 618.
24. Cornwell GG, Johnson KH, Westermark P. The age related amyloids: a growing family of unique biochemical substances. J Clin Pathol. 1995; 48(11): 984.
25. Gustavsson A, Jahr H, Tobiassen R, Jacobson DR, Sletten K, Westermark P. Amyloid fibril composition and transthyretin gene structure in senile systemic amyloidosis. Lab Invest 1995; 73(5): 703.