Efficacy and safety of BCMA-targeted therapies in relapsed or refractory AL amyloidosis: a descriptive pooled analysis

Key Points

• •BCMA-targeted therapies achieved an 83% overall hematologic response rate in patients with heavily pretreated AL amyloidosis.
• •Deep hematologic responses (≥very good partial response) were achieved in ∼70% of CAR-T and BsAb recipients.

Visual Abstract

Abstract

B-cell maturation antigen (BCMA)–targeted therapies have emerged as promising treatment options for patients with relapsed/refractory immunoglobulin light-chain (AL) amyloidosis. However, comprehensive data on their efficacy and safety remain limited. We conducted a systematic descriptive pooled analysis of published studies reporting outcomes of BCMA-directed therapies including chimeric antigen receptor T-cell (CAR-T) therapies, bispecific antibodies (BsAbs), and antibody-drug conjugates (ADCs). The analysis included 256 patients. CAR-T therapy was administered to 89 patients (35%), BsAbs to 82 patients (32%), and ADCs to 85 patients (33%). The pooled overall response rate (ORR) was 83%, with CAR-T demonstrating 92% ORR; BsAbs, 89%; and ADCs, 68%. Deep hematologic responses (≥very good partial response) were achieved in ∼70% of CAR-T and BsAb recipients and 64% of ADC recipients. Minimal residual disease negativity was documented in 75% of evaluable CAR-T patients and in 86% of evaluable BsAb patients with reported data. Organ responses were observed in 30% of evaluable patients, with cardiac responses in 41%. Cytokine release syndrome (CRS) occurred in 74% of CAR-T and 49% of BsAb patients, predominantly grade 1 to 2. Grade ≥3 CRS was rare (9% in CAR-T and 1% in BsAb). Ocular toxicity was the predominant adverse event in ADC recipients (84% with any grade and 29% with grade ≥3). Grade ≥3 infections occurred in 13% of CAR-T, 20% of BsAb, and 6% of ADC patients. BCMA-targeted therapies result in high response rates with manageable safety profiles among patients with heavily pretreated AL amyloidosis, supporting continued investigation and clinical application of these approaches.

Introduction

Immunoglobulin light-chain (AL) amyloidosis is a rare plasma cell dyscrasia characterized by the deposition of misfolded monoclonal ALs in various organs, leading to progressive organ dysfunction and significant morbidity and mortality.¹ Despite advances in treatment, outcomes remain suboptimal for patients with relapsed or refractory disease, particularly those with cardiac involvement, who face a median survival of <6 months without effective therapy.2, 3, 4

Traditional treatment approaches for AL amyloidosis have been adapted from multiple myeloma (MM) regimens, including proteasome inhibitors, immunomodulatory drugs, and high-dose chemotherapy followed by autologous stem cell transplantation.^1,5 The therapeutic landscape for AL amyloidosis was transformed by the ANDROMEDA trial, which established Dara-VCd (daratumumab combined with velcade [bortezomib], cyclophosphamide, and dexamethasone) as the first US Food and Drug Administration-approved regimen for newly diagnosed disease.⁶ In this phase 3 study of 388 patients, Dara-VCd showed superior hematologic complete response (CR) rates compared with VCd alone (53% vs 18%; P < .001). Updated analyses with 61-month median follow-up confirmed an overall survival benefit (hazard ratio, 0.62; P = .0121), with 5-year survival rates of 76% vs 65%.⁷ However, ∼10% to 20% of patients are refractory to initial daratumumab-based therapy, and many patients who initially respond eventually experience disease progression or relapse, creating an urgent need for effective salvage strategies. Moreover, patients with AL amyloidosis often present with significant organ dysfunction, particularly cardiac involvement, which limits tolerance to intensive chemotherapy and excludes many from transplant eligibility.¹

B-cell maturation antigen (BCMA) has emerged as a highly promising therapeutic target.⁸ BCMA plays a crucial role in plasma cell survival and proliferation, making it an ideal target for immunotherapy-based therapies. The success of BCMA-directed therapies in MM, including chimeric antigen receptor T-cell (CAR-T) therapies, bispecific antibodies (BsAbs), and antibody-drug conjugates (ADC), has prompted investigation of these modalities in AL amyloidosis.9, 10, 11, 12, 13

Although early clinical experience with BCMA-targeted therapies for AL amyloidosis has been encouraging, published data remain scattered across case series, single-center experiences, and early-phase clinical trials. The heterogeneity among patient populations, treatment regimens, response criteria, and follow-up duration makes it challenging to draw comprehensive conclusions about the comparative efficacy and safety of these approaches. Recent reviews have also highlighted the growing interest in BCMA-directed approaches for this indication.¹⁴ Our study aims to comprehensively evaluate the efficacy and safety of BCMA-directed therapies among patients with relapsed or refractory AL amyloidosis to inform clinical decision-making and identify areas requiring further investigation.

Methods

We conducted a descriptive pooled analysis of all BCMA-based therapies used in AL amyloidosis treatment, with reported data in either abstract format or full publication. Eligible study types included clinical trials (phase 1, 2, or 3), real-world retrospective cohort studies, and case reports/series. Relapsed disease was defined as disease recurrence after a previous response; refractory disease was defined per the included studies’ own criteria, which generally required progression on or within 60 days of the last line of therapy. Triple-class refractory disease was defined as refractory to a proteasome inhibitor, immunomodulatory agent, and anti-CD38 antibody. When multiple publications reported overlapping cohorts, data from the most recent or most complete publication at the time of data cutoff were used. This study was designed as a descriptive pooled analysis, not a formal systematic review or meta-analysis, and accordingly formal meta-analytic pooling techniques were not applied. This study was not registered on the International Prospective Register of Systematic Review. The study was deemed exempt from review by the institutional review board of the University of Texas Southwestern Medical Center because data were publicly available.

We performed a comprehensive systematic literature search of PubMed, MEDLINE, Embase, and conference abstracts from major hematology meetings (American Society of Hematology, American Society of Clinical Oncology, European Hematology Association, and International Myeloma Society) from January 2020 through December 2025. Search terms included combinations of “AL amyloidosis,” “light chain amyloidosis,” “BCMA,” ”CAR-T,” “chimeric antigen receptor,” “bispecific antibody,” “teclistamab,” “elranatamab,” “belantamab mafodotin,” “idecabtagene vicleucel (ide-cel),” and “ciltacabtagene autoleucel (cilta-cel).” Medical Subject Heading terms were used because they are universally used for indexing life science journal articles.

Extracted variables included study characteristics (author, year, journal, and study design), patient demographics and baseline characteristics (age, sex, involved light chain, previous lines of therapy, triple-class refractoriness, daratumumab exposure, organ involvement, and cardiac biomarkers), treatment details (specific BCMA-directed agent), hematologic response outcomes (overall response rate [ORR], CR, very good partial response [VGPR], partial response, and measurable residual disease [MRD] status), organ response rates (cardiac, renal, hepatic, and other organs), response kinetics and durability, safety outcomes (cytokine release syndrome [CRS], immune effector cell–associated neurotoxicity syndrome [ICANS], infections, organ toxicity, and cytopenias), and survival data. Discrepancies in data extraction were resolved through discussion and consensus.

Descriptive statistics were used to summarize patient characteristics and outcomes. Categorical variables were reported as frequencies and percentages. Continuous variables were reported as medians and ranges. Response rates were calculated as the proportion of evaluable patients achieving each response category. Pooled analyses were performed by therapeutic class (CAR-T, BsAb, and ADC) and for the entire cohort. Given the heterogeneity in study designs, follow-up duration, and patient populations, formal meta-analytic techniques were not applied. Instead, we present descriptive pooled data to provide a comprehensive overview of outcomes across studies. R statistical software version 4.0.5 (R Project for Statistical Computing) was used to perform the pooled analysis.¹⁵

Results

Study characteristics and patient population

A total of 256 evaluable patients were included in our analysis.16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35 These patients were drawn from 20 eligible studies, comprising 8 clinical trials (7 phase 1/2 and 1 phase 2 trials) and 12 real-world retrospective studies (supplemental Table 1; supplemental Figure 2). Among the 20 studies, 7 enrolled CAR-T cohorts, 8 enrolled BsAb cohorts, and 8 enrolled ADC cohorts (3 studies contributed data to the supplementary study-level figures for the CAR-T and BsAb categories). The median number of previous lines of therapy was not uniformly reported across all studies; where available, the values ranged from 3 to 5 previous lines of therapy. The distribution by therapeutic class was as follows: 89 patients (34.8%) received CAR-T therapy, 82 patients (32.0%) received BsAbs, and 85 patients (33.2%) received ADCs (supplemental Table 1). CAR-T therapies included ide-cel (n = 13), cilta-cel (n = 31), ARI0002h (n = 1), NEXICART-2/NXC-201 (n = 20), and HBI0101 (n = 16). Among BsAb-treated patients, teclistamab was the most commonly used agent (n = 55), followed by elranatamab (n = 15) and etentamig (n = 12). All ADC-treated patients received belantamab mafodotin. Median follow-up duration ranged from 1.6 to 19 months across studies, with considerable heterogeneity reflecting both the timing of publication and study design.

Baseline characteristics showed a heavily pretreated population with advanced disease (Table 1). Most patients had received multiple previous lines of therapy, with high proportions having triple-class refractory disease and previous daratumumab exposure. Among evaluable patients, triple-class refractoriness ranged from 69% to 88% across CAR-T studies, and previous BCMA-directed therapy exposure varied by cohort. Organ involvement was common, with 53.5% of the entire cohort having renal involvement, 63.6% having cardiac involvement, and many patients having multiple organ involvement.

Table 1.

Baseline patient characteristics

Characteristic	CAR-T (n = 89)	BsAb (n = 82)	ADC (n = 85)	All (N = 256)
Renal involvement, n (%)	34 (38)	43 (59)	60 (71)	137 (54)
Cardiac involvement, n (%)	48 (53)	59 (81)	57 (67)	163 (64)
Triple-class refractory, %	69-88	NR	NR	Variable
Previous daratumumab, n (%)	70-100	71-100	72-82	Variable
Median NT-proBNP, pg/mL	1486	2278	1683	∼1815
Median dFLC, mg/L	48-329	20.5-159	117-870	Variable

dFLC, difference between involved and uninvolved free light chain; NR, not reported; NT-proBNP, N-terminal pro-B-type natriuretic peptide.

Median baseline cardiac biomarkers showed significant cardiac involvement in many patients. Median N-terminal pro-B-type natriuretic peptide levels ranged from 211 to 4323 pg/mL across studies reporting this parameter (Table 1). Detailed study-level characteristics and outcomes are presented in (supplemental Table 2).

Hematologic response rates

The pooled hematologic ORR across all 256 patients was 83.2% (213/256), showing substantial efficacy of BCMA-targeted therapies in this heavily pretreated population. Response rates varied by therapeutic class (Table 2; Figure 1). CAR-T therapy was associated with the highest ORR at 92.1% (82/89 patients), followed by BsAbs at 89.0% (73/82 patients) and ADCs at 68.2% (58/85 patients).

Table 2.

Hematologic response rates by therapeutic class

Class	n	ORR, n (%)	CR, n (%)	VGPR, n (%)	≥VGPR, n (%)	PR, n (%)	MRD Neg, n (%)
CAR-T	89	82/89 (92)	50/63 (79)	11/63 (17)	61/63 (97)	2/82 (2)	23/30 (77)
BsAb	82	73/82 (89)	49/82 (60)	22/73 (30)	50/73 (68)	2/73 (3)	6/7 (86)
ADC	85	58/85 (68)	23/58 (40)	21/58 (36)	37/58 (64)	13/58 (22)	NR
All	256	213/256 (83)	122/203 (60)	54/194 (28)	148/203 (73)	16/213 (8)	24/32 (75)

Denominator reflects evaluable responders with reported data (not all enrolled patients).

MRD Neg, minimal residual disease negative; NR, not reported; PR, partial response.

Figure 1.

Hematologic response rates. (A) Hematologic response rates per treatment modality. (B) Hematologic response rates per individual study.

The depth of hematologic response was high across all therapeutic classes. Among evaluable patients, ≥VGPR was achieved in 69.1% of CAR-T recipients, 68.5% of BsAb recipients, and 64.3% of ADC recipients. CR was documented in 79.4% of evaluable CAR-T patients (50/63), 59.8% of BsAb patients (49/82), and 39.7% of ADC patients (23/58), highlighting the ability of these therapies to achieve deep hematologic remissions.

Individual agents within each therapeutic class showed varying efficacy profiles (supplemental Table 3). Among CAR-T products, cilta-cel showed an ORR of 90.3% (28/31 patients), with ide-cel achieving an ORR of 76.9% (10/13 patients). The NEXICART-2/NXC-201 and HBI0101 studies reported ORRs of 95.0% and 93.8%, respectively. Among BsAbs, teclistamab showed an ORR of 83.6% (46/55 patients), whereas elranatamab achieved 100% ORR in smaller cohorts (15/15 patients). Belantamab mafodotin demonstrated an ORR of 68.2% across all ADC-treated patients. The relationship between follow-up duration and ORR across individual studies is shown in supplemental Figure 1.

MRD and response kinetics

MRD assessment was reported in a subset of studies. Among evaluable CAR-T recipients, MRD negativity was achieved in 75% of patients (23/30 patients with available data). For BsAb-treated patients, MRD negativity was documented only in 86% (6/7 patients with available data). The median time to MRD negativity varied across studies, ranging from 10 to 30 days in CAR-T cohorts reporting this end point.

Response durability was encouraging in studies with longer follow-up. Among CAR-T recipients with at least 6 months of follow-up, 75% to 100% of responding patients maintained their hematologic response at 6 months and 80% to 100% maintained response at 12 months. Similar durability was observed among BsAb-treated patients, with 83% to 100% maintaining response at 6 months and comparable rates at 12 months among evaluable patients. Disease progression was relatively uncommon, occurring in 25% of CAR-T recipients, 16% of BsAb recipients, and 36% of ADC recipients during the reported follow-up periods.

Organ responses

Organ responses were documented across all therapeutic classes, although evaluability and reporting varied among studies (Table 3). Overall, 30.4% of evaluable patients (32/105) achieved renal responses across all therapeutic classes. Renal response rates were 29% for CAR-T recipients (10/34 evaluable patients), 35% for BsAb recipients (15/43 evaluable patients), and 25% for ADC recipients (7/28 evaluable patients).

Table 3.

Organ response rates by therapeutic class

Class	n	Renal involvement, n (%)	Renal response, n (%)	Cardiac involvement, n (%)	Cardiac response, n (%)	Other organ response, n (%)
CAR-T	89	34 (38)	10/34 (29)	48 (53)	19/47 (40)	6/8 (75)
BsAb	82	43 (59)	15/43 (35)	59 (81)	29/59 (49)	3/8 (38)
ADC	85	60 (71)	7/28 (25)	57 (67)	10/35 (29)	2/5 (40)
All	256	137 (54)	32/105 (30)	163 (64)	58/141 (41)	11/21 (52)

Denominator reflects evaluable responders with reported data (not all enrolled patients). Cardiac response is indicated by a ≥30% decrease in NT-proBNP (and >300 pg/mL) level. Renal response is indicated by a ≥30% decrease in 24-hour urine protein (minimum, 0.5 g/24 hours).

NT-proBNP, N-terminal pro-B-type natriuretic peptide.

Cardiac responses were observed in 41.1% of evaluable patients (58/141). CAR-T therapy was associated with cardiac responses in 40% of evaluable patients (19/47), BsAbs in 49% (29/59 patients), and ADCs in 29% (10/35 patients).

Other organ responses, including hepatic and neurologic improvements, were reported less frequently but demonstrated the potential for multiorgan benefit. Among patients with other organ involvement who were evaluable for organ response, 52.3% (11/21 patients) achieved improvement, with similar proportions across therapeutic classes.

Safety and toxicity

The safety profiles of BCMA-targeted therapies among patients with AL amyloidosis were generally manageable, although distinct toxicity patterns emerged for each therapeutic class (Table 4; Figure 2).

Table 4.

Safety profile by therapeutic class

Adverse event	CAR-T, any/≥G3, n (%)	BsAb, any/≥G3, n (%)	ADC, any/≥G3, n (%)	All, any grade, n (%)	All, ≥G3, n (%)
CRS	66 (74)/8 (9)	40 (49)/1 (1)	0/0	106 (41)	9 (4)
Neurotoxicity	6 (7)/1 (1)	14 (17)/low	0/0	20 (8)	Min
Infections	40 (45)/12 (13)	14 (17)/16 (20)	5 (6)/5 (6)	59 (23)	33 (13)
Organ toxicity	7 (8)/5 (6)	17 (21)/8 (10)	71 (84)/25 (29)	95 (37)	38 (15)
Neutropenia	45 (51)/42 (47)	4 (5)/9 (11)	0/0	49 (19)	51 (20)
Anemia	23 (26)/12 (13)	0/2 (2)	6 (7)/1 (1)	29 (11)	15 (6)
Lymphopenia	17 (19)/17 (19)	1 (1)/1 (1)	0/0	18 (7)	18 (7)

Organ toxicity for ADCs primarily represents ocular toxicity (keratopathy). CRS graded per ASTCT consensus criteria.

ASTCT, American Society for Transplantation and Cellular Therapy; G3, grade 3; Min, minimal.

Figure 2.

Treatment-related toxicities by anti-BCMA therapeutic class.

CRS

CRS was the most common reported toxicity associated with CAR-T and BsAb therapies. Among CAR-T recipients, 74% (66/89 patients) experienced CRS of any grade, with grade ≥3 CRS occurring in only 9% (8/89 patients). Most CRS events were grades 1 to 2 and manageable with supportive care and tocilizumab. Among BsAb-treated patients, 49% (40/82 patients) developed CRS, with grade ≥3 CRS in only 1% (1/82 patients), showing a lower incidence than CAR-T therapy. As expected, CRS was not observed among ADC-treated patients.

Neurotoxicity

ICANS and other neurologic toxicities were infrequent. Among CAR-T recipients, 6.7% (6/89 patients) experienced ICANS or neurotoxicity events. BsAb therapy was associated with neurotoxicity in 17% of patients (14/82), primarily low-grade events. The single case of immune effector cell–associated hematotoxicity syndrome was reported in a cilta-cel recipient.

Infections

Infections were observed across all therapeutic classes. Among CAR-T recipients, 44.9% (40/89 patients) experienced infections of any grade, with grade ≥3 infections in 13.4% (12/89 patients). BsAb therapy was associated with infections in 17% of patients (14/82), with grade ≥3 infections in 19.5% (16/82 patients). ADC recipients had a lower infection rate, with 5.8% experiencing any-grade infections (5/85 patients) and 5.8% experiencing grade ≥3 events (5/85 patients).

Organ Toxicity

Organ toxicity patterns varied significantly by therapeutic class. CAR-T therapy was associated with organ toxicity in 7.8% of patients (7/89), with grade ≥3 events in 5.6% (5/89 patients). Events included acute kidney injury, heart failure, and transient arrhythmias.

BsAb therapy resulted in organ toxicity in 20.7% of patients (17/82), with grade ≥3 events in 9.8% (8/82 patients). The most distinctive toxicity profile was observed with ADCs, in which ocular toxicity (keratopathy, visual impairment, and blurred vision) occurred in 83.5% of patients (71/85), with grade ≥3 events in 29.4% (25/85 patients). Other organ toxicities with ADCs included cardiac events and pulmonary symptoms.

Cytopenias

Hematologic toxicities were common but generally manageable. Neutropenia of any grade occurred in 50.5% of CAR-T recipients (45/89 patients), with grade ≥3 in 47% (42/89 patients). Anemia was less frequent, affecting 25.8% (23/89 patients), with grade ≥3 events in 13.4% (12/89 patients). BsAb therapy was associated with lower rates of neutropenia (4.8% [4/82 patients with any grade]; 11% [9/82 patients with grade ≥3]), whereas ADCs showed minimal hematologic toxicity. Lymphopenia was notable in CAR-T recipients (19% [17/89 patients with grade ≥3]) but rare in other classes.

Survival outcomes

Survival data, although limited by heterogeneous follow-up duration and reporting, suggested favorable outcomes (Table 5). Among CAR-T recipients, 84.8% (70/82 patients with available data) were alive at last follow-up. BsAb-treated patients demonstrated 83.7% survival (69/82), and ADC recipients showed 89% survival (76/85 patients). However, these figures must be interpreted cautiously given the varying follow-up durations and potential for differential reporting of survival outcomes across studies.

Table 5.

Response durability and survival outcomes

Class	n	Response, 6 mo, %	Response, 12 mo, %	Progression, n (%)
CAR-T	89	75-100	80-100	9 (10)
BsAb	82	83-100	83-100	8 (10)
ADC	85	66-100	9-100	16 (19)
All	256	Variable	Variable	33 (12.8)

Response maintenance rates represent ranges observed across studies with available follow-up data.

Discussion

This comprehensive pooled analysis of 256 patients showed that BCMA-targeted therapies achieve high hematologic response rates with manageable safety profiles among patients with heavily pretreated AL amyloidosis. The pooled ORR of 83.2% across all therapeutic classes, with CAR-T and BsAbs achieving response rates exceeding 89%, represents an excellent response for patients with relapsed or refractory disease who have limited treatment options and poor prognosis.

The depth of hematologic response observed with BCMA-directed therapies is particularly noteworthy. Approximately 70% of CAR-T and BsAb recipients achieved ≥VGPR, with 75% and 86% of evaluable CAR-T and BsAb patients achieving MRD negativity. These deep responses are critical in AL amyloidosis, in which the depth and rapidity of hematologic response correlate with organ response and survival.^36,37 The ability to achieve CR and MRD negativity may provide the best opportunity for organ recovery and prolonged disease control in this aggressive disease.

Organ responses, observed in 30% to 40% of evaluable patients for cardiac and renal involvement, represent clinically meaningful outcomes. The cardiac improvement in 41% of evaluable patients is particularly encouraging given the prognostic importance of cardiac involvement in AL amyloidosis.³⁸ However, organ response assessment was limited by variable reporting across studies, heterogeneous definitions, and the challenge of distinguishing true organ improvement from stabilization or natural fluctuations in biomarkers. Future studies should employ standardized organ response assessments with longer follow-up to better define the impact of BCMA-directed therapies on organ function and survival.

The comparative efficacy among BCMA-targeted therapeutic classes requires careful interpretation. Although CAR-T therapy was associated with the highest ORR (92.1%) and the deepest responses, direct comparisons are limited by differences in patient selection, disease characteristics, and study designs. CAR-T recipients may have been more carefully selected for favorable performance status and organ function to tolerate the manufacturing period and potential toxicities. In addition, the smaller sample sizes and single-center nature of many CAR-T studies may introduce selection bias. BsAbs offer the advantage of off-the-shelf availability, potentially enabling the treatment of patients who are sicker and cannot wait for CAR-T manufacturing or who may not survive bridging therapy. The lower ORR observed with ADCs (68.2%) may reflect the inclusion of less selected patients or intrinsic differences in mechanism of action, although the ocular toxicity profile presents a significant adverse event.

Safety profiles were generally manageable but differed substantially among therapeutic classes. CRS, the most common toxicity with CAR-T and BsAbs, was predominantly grade 1 to 2 and manageable with established supportive care algorithms.³⁹ The low incidence of grade ≥3 CRS (9% for CAR-T and 1.2% for BsAbs) is reassuring, particularly given the concerns about cardiovascular stress among patients with cardiac amyloidosis. The lower CRS rates compared with MM may reflect the plasma cell burden differences between these diseases or more cautious dosing and step-up strategies employed in AL amyloidosis trials.

Infections represented a notable concern, occurring in 44.9% of CAR-T recipients and 17% of BsAb recipients, with grade ≥3 events in 13.4% and 19.5%, respectively. These rates highlight the importance of infection prophylaxis, immune reconstitution monitoring, and prompt management of infectious complications in this immunocompromised population. The relatively lower infection rates with ADCs may reflect the absence of T-cell engagement and associated lymphodepletion.

The high incidence of ocular toxicity with belantamab mafodotin (83.5% with any grade and 29.4% with grade ≥3) represents a significant limitation of this agent. Keratopathy and visual impairment led to treatment discontinuation among many patients and raise concerns about the long-term quality of life. Although ocular toxicity is generally reversible with dose interruption or discontinuation, the impact on patient acceptance and treatment adherence cannot be underestimated.

The durability of responses observed in studies with longer follow-up is encouraging, with 75% to 100% of responders maintaining hematologic responses at 6 to 12 months. However, longer-term data are needed to assess the true duration of benefit and survival outcomes. Furthermore, the management of patients who progress after BCMA-directed therapy remains an important unresolved question because treatment options become increasingly limited and the mechanisms of resistance may affect subsequent therapies.

This analysis has several limitations that warrant acknowledgment. First, the inclusion of case series, single-center experiences, and early-phase trials introduces substantial heterogeneity among patient populations, treatment protocols, response assessment, and follow-up duration. Second, the relatively small sample sizes for individual agents and the descriptive nature of this pooled analysis preclude formal comparative effectiveness assessments or multivariable analyses. Third, survival data were incompletely reported and limited by variable follow-up duration, preventing robust survival analyses. Fourth, organ response assessment was inconsistently reported across studies, and the criteria used varied, thereby limiting definitive conclusions about organ benefit. Fifth, cardiac staging by the Mayo criteria was not uniformly reported across all included studies, precluding a standardized assessment of baseline cardiac risk. Sixth, granular supportive care data, including the use of IV immunoglobulin prophylaxis, duration of hypogammaglobulinemia, prophylactic tocilizumab use, steroid use for CRS management, and baseline plasma cell burden as a correlate of CRS severity, were not consistently reported and could not be systematically analyzed. Seventh, the median time to hematologic response was infrequently reported and could not be pooled. Finally, publication bias may exist, with positive results more likely to be reported than negative experiences.

Despite these limitations, this pooled analysis provides the most comprehensive overview to date of BCMA-targeted therapy outcomes in AL amyloidosis. The results support continued investigation and the clinical use of these approaches in carefully selected patients with relapsed or refractory disease. Several key questions remain for future research: (1) What are the optimal patient selection criteria, particularly regarding cardiac function and performance status? (2) What is the ideal timing for BCMA-directed therapy in the treatment algorithm? (3) How do outcomes compare in prospective head-to-head studies? (4) What are the long-term outcomes, including overall survival and quality of life? (5) How should disease progression after BCMA-directed therapy be managed? (6) Can biomarkers predict response or toxicity risk? (7) What are the optimal supportive care strategies to minimize toxicity and maximize efficacy? Ongoing and planned clinical trials will help address many of these questions. Several phase 2 studies of CAR-T therapies in AL amyloidosis are actively enrolling, and larger registry studies are collecting real-world data on outcomes and safety. The development of next-generation BCMA-targeted therapies with improved efficacy and safety profiles, including novel CAR constructs, next-generation BsAbs, and optimized ADCs, holds promise for further advancing the treatment landscape.

In conclusion, this pooled analysis shows that BCMA-targeted therapies achieve high hematologic response rates with deep, durable responses and meaningful organ improvements in patients with heavily pretreated AL amyloidosis. CAR-T and BsAbs show particularly promising efficacy, with MRD negativity achievable in most responding patients. Safety profiles are generally manageable, with CRS being the most common toxicity but predominantly low-grade and treatable. These findings support the continued investigation and judicious clinical application of BCMA-directed approaches in this challenging patient population. Ongoing studies with longer follow-up, standardized assessments, and comparative designs will further refine our understanding of the role of these transformative therapies in AL amyloidosis management.

Conflict-of-interest disclosure: A.M.K. reports consultancy with Genzyme (Sanofi); and research funding from Johnson & Johnson and Sanofi. A.A. reports consultancy with Bristol Myers Squibb (BMS), Sanofi, Pfizer, Johnson & Johnson, and Karyopharm; and research funding from Johnson & Johnson, AbbVie, Adaptive Biotechnologies, Regeneron, and K36 Therapeutics. C.S. reports honoraria from MJH Life Sciences, OncLive, and Pfizer; serves on an advisory board for Johnson & Johnson; and reports consulting fees from Johnson & Johnson. F.v.R. reports consulting or advisory role from Johnson & Johnson, EUSA/Recordati, BMS, GSK, Castleman Disease Collaborative Network, Sanofi, Regeneron, and Pfizer. L.D.A. Jr reports consultancy with Pfizer, Amgen, BMS, Janssen, BeiGene, Sanofi, Karyopharm, Celgene, Prothena, Cellectar, GSK, and AbbVie; and honoraria from Amgen, BMS, Janssen, BeiGene, Sanofi, Karyopharm, Celgene, Prothena, Cellectar, GSK, and AbbVie. V.S. reports research support from Celgene, Millennium-Takeda, Janssen, Prothena, Sorrento, Karyopharm, Oncopeptides, Caelum, and Alexion; consultancy with Pfizer, Janssen, Attralus, GateBio, AbbVie, and BridgeBio; and advisory board participation with Proclara, Caelum, AbbVie, Janssen, Regeneron, Protego, PharmaTrace, Telix, Prothena, AstraZeneca, Nexcella, Alexion, and GSK. S.A.H. reports receiving research funding from Alexion, Janssen, and Pfizer; and consulting fees from Janssen and Legend Biotech. The remaining authors declare no competing financial interests.