This is a retrospective review of patients with AL or ATTR who underwent HT in European, Chinese and US centers (Hospital Universitario de Bellvitge, Spain; University of Bologna, Italy; Peking Union Medical College, China; Hospital Universitario Puerta de Hierro Majadahonda, Spain; National Amyloidosis Center, United Kingdom; University of Pavia, Italy; University of Padua, Italy; Oregon Health & Science University, United States; University of California San Diego, United States; University Hospital of Udine, Italy; University of Trieste, Italy; University Hospitals Birmingham, United Kingdom; University of Florence, Italy; La Sapienza University, Italy; The University of Campania Luigi Vanvitelli, Italy) (figure S1).

Before 2016, histological confirmation through tissue biopsy was required in all cases. After 2016, following the validation of nonbiopsy diagnostic algorithms for ATTR amyloidosis, patients without evidence of monoclonal gammopathy were diagnosed noninvasively using bone scintigraphy, while tissue biopsy was reserved for patients with a detectable monoclonal component to exclude AL amyloidosis.

Regarding bone scintigraphy, different technetium-labeled tracers (technetium-99m pyrophosphate, technetium-99m hydroxymethylene diphosphonate, or technetium-99m 3,3-diphosphono-1,2-propanodicarboxylic acid) were used according to local availability and institutional practice. Similarly, assessment of monoclonal components was performed using serum and urine immunofixation and serum-free light chain assays, in accordance with guideline recommendations at each center.2

Smoldering multiple myeloma and multiple myeloma were diagnosed according to the International Myeloma Working Group (IMWG) criteria,13 which were updated in 2014.14

Data were collected retrospectively via chart review from patients diagnosed between 1996 and 2024. Baseline laboratory values and information regarding extracardiac amyloidosis were recorded. Patients were monitored and treated according to standard institutional protocols before and after HT.

Gastrointestinal involvement was defined by the development of constipation, diarrhea, or both, and/or a history of gastrointestinal bleeding with biopsy verification for amyloid deposition. Hepatic involvement was defined as an alkaline phosphatase level> 1.5 times the upper institutional laboratory limit with liver biopsy confirmation.10 Renal involvement criteria required urinary protein excretion> 0.5g/24h or renal biopsy confirmation.15

Peripheral neuropathy involvement required confirmation by an expert neurologist per center practice or use of medications for neuropathy treatment.

Patients were evaluated for postoperative complications, including postoperative bleeding, renal impairment, rejection, infection, cancer, graft failure, and death. Acute cellular rejection was defined as International Society for Heart and Lung Transplantation grade ≥ 2.16

Endomyocardial biopsies after HT were performed according to the local surveillance protocols of each center, either as part of routine post-transplant follow-up or when clinically indicated. Amyloid deposition in biopsy samples was assessed by Congo red staining, followed by deposit typing using immunofluorescence and/or laser microdissection with mass spectrometry.

Systematic screening with cardiac imaging modalities such as scintigraphy or single-photon emission computed tomography was not performed.

Patients were stratified on the basis of the date of HT into 3 eras: early (1999-2007), intermediate (2008-2015), and late (2016-2024). The years 2008 and 2016 were chosen as cutoff dates as they marked key changes in diagnostic criteria for CA and advances in the treatment of AL and ATTR.2,10 Furthermore, 2008 represents an inflection point in the development of HT for CA, coinciding with the publication of more stringent patient selection criteria addressing extracardiac organ involvement,17 which significantly influenced transplant candidacy. In addition, this period was characterized by major therapeutic advances in AL amyloidosis, particularly the inclusion of proteasome inhibitors such as bortezomib in treatment protocols.18 In addition, in 2021, the human CD38-targeting antibody daratumumab was demonstrated to be safe and effective in the treatment of AL amyloidosis.6

Ethics approval was obtained from the coordinating center, and each participating center confirmed approval from its local institutional review board/ethics committee for the conduct of observational research. The study was conducted in accordance with the Declaration of Helsinki.

The aim of this study was to describe an international multicenter cohort of patients with CA who underwent HT. Specifically, we sought to characterize clinical features at diagnosis and at the time of transplantation, pretransplant support requirements, and post-transplant outcomes and disease progression, with particular attention to the use of disease-modifying therapies.

Continuous variables are presented as median (interquartile range [IQR]), while categorical variables are presented as counts and percentages. Between-group comparisons were evaluated using Wilcoxon rank sum test and chi-square tests as appropriate. Survival curves were created using the Kaplan-Meier method, and unadjusted survival rates were compared using the log-rank test. Survival analysis was performed using Cox proportional hazards regression models to estimate hazard ratios (HRs) and 95% confidence intervals (95%CIs) for all-cause mortality. Univariable analyses were conducted for the entire cohort and separately for subgroups of patients with AL and ATTR amyloidosis. The proportional hazards assumption was assessed for each model using Schoenfeld residuals. To test the different distribution of amyloidosis subtypes across eras, the chi-square test for independence was used. Statistical analyses were performed using R version 4.3.1 (R Foundation for Statistical Computing, Vienna, Austria).

A total of 113 patients were included: 57 with ATTR (22 with ATTRwt and 35 with ATTRv; table S1) and 56 with AL. Most patients were male (n = 47 [82%] in ATTR; n = 32 [57%] in AL). The median age at transplantation was 60.8 years [IQR, 53.2-64.1 years] in the ATTR group and 52.5 years [IQR, 46.3-59.2 years] in the AL group (table 1, table S2).

The median interval from diagnosis to HT was significantly shorter in AL than in ATTR (7.9 months [IQR, 3.8-16.3 months] in AL vs 31.5 months [IQR, 12.6-44.4 months] in ATTR, P < .001). Similarly, the time from listing to HT was shorter in patients with AL (1.5 months [IQR, 0.6-3.1 months]) than in those with ATTR (4.0 months [IQR, 1.4-7.3 months], P < .001).

Before HT, patients with AL showed more advanced hemodynamic compromise compared with those with ATTR, including a lower cardiac index (1.7 L/min/m2 [IQR, 1.4–2.0 L/min/m2] vs 1.9 L/min/m2 [IQR, 1.6–2.4 L/min/m2]; P = .027) and higher right atrial pressure (15.5mmHg [IQR, 12.0–23.7mmHg] vs 10.0mmHg [IQR, 5.0–17.8mmHg]; P = .004).

Mechanical circulatory support (MCS) was used as a bridge to HT in 25 patients (22%) and included intra-aortic balloon pumps (n = 8), Impella (n = 2), extracorporeal membrane oxygenation (n = 5), left ventricular assist devices (n = 9), and a total artificial heart (n = 1). Left ventricular assist devices were implanted in 9 patients (3 ATTRv, 6 AL), with 6 deaths during follow-up (3 ATTRv, 3 AL).

Intra-aortic balloon pump support was used in 8 patients (4 AL and 4 ATTR), with 3 deaths occurring in AL patients. Two patients (1 ATTR, 1 AL) received an Impella device, with no deaths occurring during follow-up. Extracorporeal membrane oxygenation was required in 5 AL patients, 3 of whom died during follow-up. One patient with AL underwent total artificial heart implantation 1 year before HT and died perioperatively due to massive bleeding.

Among the 57 patients with ATTR amyloidosis (22 ATTRwt and 35 ATTRv), transplantation strategies differed substantially. Overall, 38 patients (67%) underwent isolated HT, whereas 19 (33%) underwent combined organ transplantation. In patients with ATTRv (n = 35), 16 (46%) underwent isolated HT, 18 (51%) underwent combined heart–liver transplantation, and 1 (3%) underwent combined heart–liver–kidney transplantation. In contrast, all patients with ATTRwt (n = 22) underwent isolated HT. The characteristics of patients with ATTRwt and ATTRv are detailed in table S3. All AL patients underwent isolated HT.

Infections were the most common post-transplant complications, occurring in 23 patients with ATTR (40%) and 24 patients with AL (43%) (table 2, table S4). Pneumonia was the most frequent infection, reported in 8 patients with ATTR (35%) and 12 with AL (50%) (table S2). Sepsis occurred in 4 patients with ATTR (17%) and 1 with AL (4.2%). Gastroenteritis was reported in 2 patients with ATTR (8.7%) and 2 with AL (8.3%).

One episode of rejection occurred in 6 patients with ATTR (11%) and 12 with AL (21%), while more than 1 episode occurred in 4 patients with ATTR (7%) and 10 with AL (18%). Antibody-mediated rejection was observed in 2 patients with ATTR (3.5%) and 2 with AL (3.6%). Among the 36 patients who experienced rejection, 11 died. In 3 patients with AL, rejection was considered the cause of death (2 acute cellular rejections and 1 antibody-mediated rejection).

Cardiac allograft vasculopathy was diagnosed in 5 patients with ATTR (8.8%) and 3 with AL (5.4%).

New-onset neuropathy was diagnosed in 5 patients with ATTRv (14%), 1 with ATTRwt (4.5%), and 6 with AL (11%). The median time from HT to neuropathy onset was 4 months [IQR, 1.9-19.2 months] in ATTR and 32 months [IQR, 15.7-84.7 months] in AL. Progression of preexisting neuropathy occurred in 12 patients with ATTRv (34%), 1 with ATTRwt (4.5%), and 4 with AL (7.1%), after a median time of 14.8 months [IQR, 4-72 months] in ATTR and 32.2 months [IQR, 25.7-42.5 months] in AL. Patients with ATTR had a significantly higher risk of new-onset neuropathy and neuropathy progression compared with those with AL (log-rank test, P=.044).

Amyloid deposits in endomyocardial biopsies were detected in 2 patients with ATTRv (3.5%; variants p.Val142Ile and p.Val142del) after a median time of 67 months [IQR, 40.7-93.5 months] and in 5 patients with AL (8.9%) after a median time of 21 months [IQR, 5-129 months] from HT.

Only one patient carrying the p.Ile88Leu variant developed carpal tunnel syndrome and lumbar spinal stenosis after HT.

During a median follow-up of 4.6 years [IQR, 1.9–8.4 years] in the ATTR group and 5.4 years [IQR, 1.6–8.3 years] in the AL group, 16 patients with ATTR (28%) and 22 patients with AL (39%) died. There was no significant difference in overall survival between AL and ATTR (log-rank test, P=.19), with unadjusted 1-year, 3-year, and 5-year survival rates of 93%, 84%, and 80% in the overall cohort, 91%, 81%, and 77% in the AL group, and 94%, 86%, and 83% in the ATTR group (figure 1). Sepsis was the most common cause of death, accounting for 6 out of 16 deaths (38%) in the ATTR group and 5 out of 22 (23%) in the AL group. Four sepsis-related deaths occurred within 3 months after HT, and 7 occurred more than 36 months after HT.

Figure 1.

Central illustration. Unadjusted time-to-event analysis of post-heart transplantation mortality. AL, light chain amyloidosis; ATTR, transthyretin amyloidosis; ATTRv, variant transthyretin amyloidosis; ATTRwt, wild-type transthyretin amyloidosis; HT, heart transplantation; US, United States.

(0.73MB).

Survival rates were comparable to those reported in 2 previously published cohorts of patients with CA9,16 and to those of patients without CA from the same institutions (figure 1).

Among patients with amyloid deposits, the patient with the p.Val142del variant was treated with diflunisal after HT and died 120 months after HT due to neuropathy progression. One patient with AL and multiple myeloma, who did not achieve hematologic response before or after HT, died due to disease progression 34 months after HT.

Before HT, 4 patients with ATTRv and 5 with ATTRwt were on tafamidis (table 3). The median time from tafamidis initiation to HT was 16.3 months [IQR, 6.8-27.8 months] in patients with ATTRv and 13.8 months [IQR, 10.9-27.6 months] in patients with ATTRwt. Four patients (2 ATTRwt and 2 ATTRv) were on tafamidis before and after HT. One patient with ATTRv was on patisiran, initiated 5.7 months before transplantation.

After HT, 5 patients with ATTRv and 2 with ATTRwt were treated with tafamidis, with a median initiation time of 15.3 months [IQR, 1.7-58.3 months] for ATTRv and 4.9 months [IQR, 2.4-7.3 months] for ATTRwt. Eight patients with ATTRv received patisiran after HT (1 of whom had already been on patisiran before HT), with a median initiation time of 83.1 months [IQR, 5.6-93.3 months]. Three ATTRv patients were treated with diflunisal post-HT, initiated at a median time of 74.2 months [IQR, 69.5-86.0 months] after HT.

Among patients with AL, 40 (72%) had lambda light chain plasma cell dyscrasia. The most common underlying hematological disorder was multiple myeloma, diagnosed in 38 patients (69%), followed by smoldering myeloma in 10 (18%) (table 4). Twenty-one patients (42%) underwent autologous stem cell transplantation. Two patients underwent HT before initiation of chemotherapy. Chemotherapy was started a median of 8.54 months [IQR, 5.61-18.16 months] before HT and was administered for a median duration of 5.9 months [IQR, 3.15-10.01 months]. As first-line therapy, 37 patients (70%) received a bortezomib-based regimen, 15 (28%) received a daratumumab-based regimen, and 1 patient (2%) received upfront autologous stem cell transplantation. Before HT, the hematologic response was classified as no response in 11 patients (21%), partial response in 11 (21%), very good partial response in 9 (17%), and complete response in 22 (42%). After HT, chemotherapy was initiated a median of 5.65 months [IQR, 1.84-23.81 months] after HT and continued for a median duration of 9.37 months [IQR, 2.01-17.80 months]. Two patients received maintenance therapy with daratumumab. The post-transplant hematologic response was classified as no response in 8 patients (15%), partial response in 9 (17%), very good partial response in 7 (13%), and complete response in 29 (54%). Thirty-three patients (65%) received 1 line of therapy, 9 (18%) received 2 lines, and 5 (9%) received 3 lines (figure 2).

Figure 2.

Survival analysis of the study cohort compared with amyloid and nonamyloid patient cohorts.

(0.27MB).

In the overall population, increased mortality risk was observed in patients who received advanced circulatory support (ventricular assist device, intra-aortic balloon pump, Impella, extracorporeal membrane oxygenation, or artificial heart; HR, 5.11; 95%CI, 2.37-11.0) and those with greater systemic involvement (1 involved organ system vs no systemic involvement: HR, 2.49; 95%CI, 1.01-6.12; ≥ 2 involved organ systems vs no systemic involvement: HR, 3.55; 95%CI, 1.44-8.75) (table S5). In subgroup analyses, advanced mechanical support remained associated with higher mortality in both patients with AL (HR, 5.59; 95%CI, 2.07-15.1) and those with ATTR (HR, 4.05; 95%CI, 1.08-15.2). Systemic involvement of ≥ 2 organ systems was associated with increased mortality only in patients with ATTR.

The distribution of amyloidosis subtypes varied by transplant era (figure 3). The proportion of patients with AL showed a nonmonotonic pattern that increased from the 1999 to 2007 period to the 2008 to 2015 period and then declined between 2016 and 2024. Over time, the proportion of patients with ATTRv decreased, while that of patients with ATTRwt increased. Among transplant eras, there was a trend toward improved survival in more recent periods (log-rank P = .053, figure S2).

Figure 3.

Distribution of amyloidosis type by transplant era. ATTR, transthyretin amyloidosis; ATTRv, variant transthyretin amyloidosis; ATTRwt, wild-type transthyretin amyloidosis.

(0.19MB).

Despite presenting with more advanced hemodynamic compromise, patients with AL had survival outcomes similar to those with ATTR. This finding likely reflects improved patient selection, younger age, and fewer comorbidities in the AL group. Earlier cohorts (HT between 2001 and 2007) reported worse post-transplant survival in AL compared with ATTR,8 whereas the results of recent studies are consistent with our findings.10,11 Improvements are likely attributable to advances in plasma cell-directed therapies6 and stricter pretransplant screening protocols to exclude significant extracardiac involvement.7,9

Overall, survival among transplant recipients with CA has improved over time and is now comparable to that of nonamyloid transplant populations.9–11 This positive trend is likely attributable to growing clinical experience in the management of advanced heart failure in CA, advances in AL-targeted therapies, and refinements in immunosuppression.7

Infections were the most common post-transplant complications, affecting 42% of patients, with similar rates in AL and ATTR, despite the potential added immunosuppressive burden from chemotherapy in AL. These findings are consistent with prior reports. For example, in a Stanford University cohort,16 infections were reported in 61% of patients (total n = 31), and pneumonia was the most common.

Extracardiac progression is a major post-transplant challenge. Neurological progression (new-onset or worsening) occurred frequently (29% in ATTR and 14% in AL), as did gastrointestinal involvement (14% in both AL and ATTR). In contrast, only 1 patient with ATTRv developed carpal tunnel syndrome and lumbar spinal stenosis after HT. Comparable findings have been reported in a cohort of 12 patients from Columbia University.10 Neurological progression occurred in 1 patient with ATTRwt and in 3 with ATTRv; gastrointestinal involvement was observed in 1 patient with ATTRwt and 4 patients with ATTRv. Interestingly, 4 patients developed new or relapsed carpal tunnel syndrome or lumbar spinal stenosis. Similarly, among 7 patients with ATTRwt who received transplants at the Mayo Clinic,21 1 had neurological progression, 2 had symptomatic gastrointestinal involvement, and 2 developed carpal tunnel syndrome requiring surgical release. These findings underscore the progressive multisystem nature of amyloidosis and highlight the importance of continued multiorgan monitoring and the potential role of disease-modifying therapies, even after HT.10,21

In this cohort, amyloid deposition in the cardiac allograft was found in 2 patients with ATTRv, of whom 1 died due to neuropathy progression. Similarly, in the Mayo Clinic cohort,9 1 patient with ATTRv showed amyloid deposition in the allograft and died from cardiac arrest 8 years post-HT. In addition, 5 patients with AL had evidence of allograft involvement; 1 patient without hematologic response before and after HT died due to disease progression. In contrast, over a 3-year follow-up, out of 51 patients with CA (13 AL, 38 ATTR) at the Cedars-Sinai Smidt Heart Institute,19 no amyloid recurrence was detected in endomyocardial biopsies, suggesting that allograft deposition may take more time to develop. These observations reinforce the importance of achieving and maintaining deep hematologic response. With the introduction of novel agents such as daratumumab, outcomes are expected to improve in future cohorts.

In this study, only a minority of patients with ATTR received disease-modifying therapies either before or after transplantation. Even when prescribed, therapy was often initiated shortly before or long after HT. In the Columbia cohort,10 no patients were on disease-modifying therapy before HT, and 2 patients were started on tafamidis and 1 on patisiran for neurological progression. The importance of specific therapy has been underscored by recent evidence. For instance, among patients with ATTRv amyloidosis experiencing polyneuropathy progression after liver transplantation, patisiran significantly improved neuropathy severity, quality of life, and autonomic symptoms.22 The underuse of these drugs in our cohort likely reflects historical limitations in drug availability and reimbursement policies across countries, particularly in earlier eras. Broader use of targeted therapies may reduce recurrence risk and delay multisystem progression.

When stratified into early (1999-2007), intermediate (2008-2015), and late (2016-2024) eras, we observed a trend toward improved outcomes in recent years. The increasing use of the noninvasive diagnostic algorithm2 has contributed to an increase in the number of transplant recipients with ATTRwt, while the number of patients with ATTRv declined, likely reflecting the growing use of disease-modifying therapies in that subgroup.

The expanding availability of disease-modifying therapies is expected to reduce the need for HT in CA. However, for patients requiring transplantation, optimized integration of targeted therapies may prevent recurrence, prolong graft survival, and improve long-term outcomes.

This study represents one of the largest international multicenter cohorts of HT in CA introduces a 3-era temporal stratification, providing historical insight into the evolution of transplant practices and outcomes.

The retrospective design inherently introduces the risk of selection bias and unmeasured confounders. The sample size was relatively small, and the number of patients with long-term follow-up was limited. Important granular clinical data were not uniformly available among centers. Specific details regarding the type of temporary MCS (eg, Impella CP vs 5.5), induction immunosuppressive therapy, and pretransplant sensitization status were inconsistently captured. Given that infectious complications were a leading cause of morbidity and mortality in this cohort, the absence of detailed data on induction regimen limits our ability to fully characterize risk factors for post-transplant infections. Our study reflects real-world data, and specific institutional protocols regarding the selection and exclusion criteria for multiorgan transplantation were not available. Consequently, the absence of multiorgan transplants in the AL amyloidosis cohort may reflect a selection bias toward patients with isolated cardiac involvement or center-specific practices regarding systemic disease burden.

Institutional protocols for transplant eligibility, including criteria for multiorgan transplantation, were not uniformly available.

Although all diagnoses were established according to internationally accepted criteria valid at the time of diagnosis, differences in diagnostic algorithms and imaging availability may have contributed to heterogeneity. Access to disease-modifying therapies varied among countries and eras due to differences in drug approval, reimbursement policies, and local prescribing practices. The specific reasons for nonprescription were not captured. Consequently, treatment exposure was heterogeneous and may have influenced outcomes independently of disease characteristics, particularly in temporal comparisons.

Another important consideration is the heterogeneity among participating centers. Differences in referral patterns, patient selection criteria, and management strategies, including the use and timing of MCS, transplantation protocols, and disease-modifying therapies, may have influenced the observed outcomes. This variability reflects real-world clinical practice but may limit the comparability of patients among centers and should be taken into account when interpreting our findings. In addition, surveillance for amyloid recurrence was not standardized among centers and relied on local clinical practice. Routine imaging screening with scintigraphy or single-photon emission computed tomography was not performed, which may have led to underdetection of subclinical amyloid recurrence.