Graded Organ Response and Progression Criteria for Kidney Immunoglobulin Light Chain Amyloidosis

Eli Muchtar; Brendan Wisniowski; Susan Geyer; Giovanni Palladini; Paolo Milani; Giampaolo Merlini; Stefan Schönland; Kaya Veelken; Ute Hegenbart; Nelson Leung; Angela Dispenzieri; Shaji K Kumar; Efstathios Kastritis; Meletios A Dimopoulos; Michaela Liedtke; Patricia Ulloa; Vaishali Sanchorawala; Raphael Szalat; Katharine Dooley; Heather Landau; Erica Petrlik; Suzanne Lentzsch; Alexander Coltoff; Joan Bladé; M Teresa Cibeira; Oliver Cohen; Darren Foard; Jullian Gillmore; Helen Lachmann; Ashutosh Wechalekar; Morie A Gertz

doi:10.1001/jamaoncol.2024.2629

. 2024 Aug 1;10(10):1362–1369. doi: 10.1001/jamaoncol.2024.2629

Graded Organ Response and Progression Criteria for Kidney Immunoglobulin Light Chain Amyloidosis

Eli Muchtar ^1,^✉, Brendan Wisniowski ², Susan Geyer ³, Giovanni Palladini ^4,⁵, Paolo Milani ^4,⁵, Giampaolo Merlini ^4,⁵, Stefan Schönland ⁶, Kaya Veelken ⁶, Ute Hegenbart ⁶, Nelson Leung ¹, Angela Dispenzieri ¹, Shaji K Kumar ¹, Efstathios Kastritis ⁷, Meletios A Dimopoulos ⁷, Michaela Liedtke ⁸, Patricia Ulloa ⁸, Vaishali Sanchorawala ⁹, Raphael Szalat ⁹, Katharine Dooley ³, Heather Landau ¹⁰, Erica Petrlik ¹⁰, Suzanne Lentzsch ¹¹, Alexander Coltoff ¹¹, Joan Bladé ¹², M Teresa Cibeira ¹², Oliver Cohen ², Darren Foard ², Jullian Gillmore ², Helen Lachmann ², Ashutosh Wechalekar ², Morie A Gertz ¹

¹Division of Hematology, Mayo Clinic, Rochester, Minnesota

²National Amyloidosis Centre, University College London Medical School, Royal Free Hospital Campus, London, England

³Division of Clinical Trials and Biostatistics, Mayo Clinic, Rochester, Minnesota

⁴Department of Molecular Medicine, University of Pavia, Pavia, Italy

⁵Amyloidosis Research and Treatment Center, Fondazione IRCCS Policlinico San Matteo, and Department of Molecular Medicine, University of Pavia, Pavia, Italy

⁶Medical Department V, Amyloidosis Center, University of Heidelberg, Germany

⁷Department of Clinical Therapeutics, National and Kapodistrian University of Athens, Athens, Greece

⁸Stanford Amyloid Center, Stanford University School of Medicine, Stanford, California

⁹Section of Hematology and Oncology, Amyloidosis Center, Boston University School of Medicine, Boston Medical Center, Boston, Massachusetts

¹⁰Division of Hematologic Oncology, Memorial Sloan Kettering Cancer Center, New York, New York

¹¹Division of Hematology/Oncology, Columbia University Medical Center, New York, New York

¹²Department of Hematology, Institut d'Investigacions Biomèdiques August Pi i Sunyer, Hospital Clinic, Barcelona, Spain

Accepted for Publication: March 14, 2024.

Published Online: August 1, 2024. doi:10.1001/jamaoncol.2024.2629

^✉

Corresponding Author: Eli Muchtar, MD, Division of Hematology, Mayo Clinic, 200 First St, SW, Rochester, MN 55905 (muchtar.eli@mayo.edu).

Author Contributions: Drs Muchtar and Geyer had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Muchtar, Palladini, Dispenzieri, Kumar, Kastritis, Dimopoulos, Lentzsch, Gertz.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Muchtar, Geyer, Milani, Kumar, Dimopoulos, Szalat, Dooley, Wechalekar, Gertz.

Critical review of the manuscript for important intellectual content: Muchtar, Wisniowsk, Geyer, Palladini, Milani, Merlini, Schönland, Veelken, Hegenbart, Leung, Dispenzieri, Kumar, Kastritis, Dimopoulos, Leidtke, Ulloa, Sanchorawala, Szalat, Landau, Petrlik, Lentzsch, Coltoff, Bladé, Cibeira, Cohen, Foard, Gillmore, Lachmann, Wechalekar, Gertz.

Statistical analysis: Muchtar, Geyer, Leidtke, Dooley, Petrlik, Wechalekar.

Administrative, technical, or material support: Schönland, Veelken, Dispenzieri, Kastritis, Dimopoulos, Ulloa, Sanchorawala, Landau, Lentzsch, Coltoff, Foard.

Supervision: Wisniowski, Palladini, Merlini, Dispenzieri, Dimopoulos, Bladé, Wechalekar, Gertz.

Conflict of Interest Disclosures: Dr Muchtar reported grants from Protego outside the submitted work. Dr Palladini reported personal fees from Janssen, Alexion, AbbVie, Glaxo Smith Kline, Prothena, Regeneron, and Pfizer as well as grants from Pfizer outside the submitted work. Dr Milani reported advisory board service with Janssen and honoraria from Pfizer and Siemensa outside the submitted work. Dr Schönland reported grants and personal fees from Janssen, Prothena, and Sanofi during the conduct of the study as well as research support from Neurimmune and Janssen; personal fees from Telix, AstraZeneca, Sobi, Alexion, Pfizer; and grants from Celgene, Binding Site, Takeda, and Jazz outside the submitted work. Dr Hegenbart reported travel grants from Janssen, Prothena, and Pfizer ; honoraria from Pfizer, Prothena, Alnylam, and Akcea and grants from Janssen outside the submitted work. Dr Leung reported stocks in AbbVie stocks and Checkpoint Therapeutics outside the submitted work. Dr Dispenzieri reported personal fees from Janssen; nonfinancial support from Pfizer, Alnylam, AbbVie; grants from BMS and Takeda; and personal fees from Oncopeptides outside the submitted work. Dr Kumar reported research funding for their institution from AbbVie, Amgen, Allogene, Astra-Zeneca, BMS, Carsgen, GSK, Janssen, Novartis, Roche-Genentech, and Takeda; consulting/advisory board participation for AbbVie, Amgen, BMS, Janssen, Roche-Genentech, Takeda, Astra-Zeneca, Bluebird Bio, Secura Biotherapeutics, Trillium, Loxo Oncology, K36, Sanofi, ArcellX, Oncopeptides, Beigene, and Antengene. Dr Kastritis reported grants from Janssen, GSK, and Pfizer during the conduct of the study as well as travel support and registration expenses from Janssen outside the submitted work. Dr Dimopoulos reported personal fees from Amgen, Sanofi, Regeneron, Menarini, Takeda, GSK, BMS, Janssen, and Beigene outside the submitted work. Dr Leidtke reported personal fees from AbbVie, BMS, GSK, Takeda, Kite, and Janssen outside the submitted work. Dr Sanchorawala reported grants from Janssen, Prothena, Alexion, AbbVie, Regeneron, and Protego and personal fees from Janssen, Prothena, and AbbVie outside the submitted work. Dr Landau reported personal fees from Nexcella, Prothena, and AbbVie and grants from Prothena, Protego, Alexion, and Janssen outside the submitted work. Dr Petrlik reported employment with Bristol Myers Squibb outside the submitted work. Dr Lentzsch reported grants from Sanofi, and Zentalis; personal fees from Pfizer, Regeneron, Janssen, GSK, Sanofi, BMS, AstraZeneca, Karyopharm, Angitia, Alexion, Takeda, Caelum Biosience, Adaptive, Oncopeptide, PeerView, Clinical Care Option, RedMed, Aptitude, Bio Ascend, Med Scape, and Calgary University; spouse having stock options in Magenta and Poseida; committee/board service for the National Cancer Institute (NCI), SOHO, IMS, American Society of Clinical Oncology, Caelum Bioscience; editorial service for the Journal of Clinical Oncology and Blood Cancer Donor; and a patent with royalties paid from Caelum Bioscience outside the submitted work. Dr Coltoff reported personal fees from Sobi Pharmaceuticals and Blueprint Medicine outside the submitted work. Dr Bladé reported personal fees from Janssen, Celgene/BMS, and Sanofi outside the submitted work. Dr Gillmore reported personal fees from Alnylam, ATTRalus, Bridgebio, AstraZeneca, Ionis, Intellia, and Lycia outside the submitted work. Dr Gertz reported personal fees from Ionis/Akcea, Alnylym, Prothena, Sanofi, Janssen, Aptitude and Ashfield, Physicians Education Resource, AbbVie, Johnson & Johnson, Celgene, Research to Practice, Sorrento, and i3Health outside the submitted work. No other disclosures were reported.

Funding/Support: Research at Memorial Sloan Kettering Cancer Center was supported in part through the National Institutes of Health/NCI Cancer Center Support grant P30 CA008748.

Role of the Funder/Sponsor: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Data Sharing Statement: See Supplement 2.

^✉

Corresponding author.

PMCID: PMC11295065 PMID: 39088206

Key Points

Question

Are graded kidney response criteria in light chain amyloidosis associated with outcomes?

Findings

In this cohort study with 732 patients with kidney light chain amyloidosis, graded kidney response criteria based on the reduction in 24-hour proteinuria levels was associated with kidney and overall survival; kidney survival discrimination was seen as early as 6 months and overall survival as early as 12 months from therapy initiation. Deeper kidney responses were associated with a lower risk of progression to dialysis and death, which suggest that the proposed new system has superior performance compared with the binary response assessment.

Meaning

The results of this study suggest that graded kidney response criteria allow for better assessment of therapeutic interventions and set new goals for kidney response.

Abstract

Importance

Kidney light chain (AL) amyloidosis is associated with a risk of progression to kidney replacement therapy (KRT) and death. Several studies have shown that a greater reduction in proteinuria following successful anticlonal therapy is associated with improved outcomes.

Objective

To validate graded kidney response criteria and their association with kidney and overall survival (OS).

Design, Setting, and Participants

This retrospective, multicenter cohort was conducted at 10 referral centers for amyloidosis from 2010 to 2015 and included patients with kidney AL amyloidosis that was evaluable for kidney response and who achieved at least hematologic partial response within 12 months of diagnosis. The median follow-up was 69 (54-88) months. Data analysis was conducted in 2023.

Exposure

Four kidney response categories based on the reduction in pretreatment 24-hour urine protein (24-hour UP) levels: complete response (kidCR, 24-hour UP ≤200 mg), very good partial response (kidVGPR, >60% reduction in 24-hour UP), partial response (kidPR, 31%-60% reduction), and no response (kidNR, ≤30% reduction). Kidney response was assessed at landmark points (6, 12, and 24 months) and best kidney response.

Main Outcomes and Measures

Cumulative incidence of progression to KRT and OS.

Results

Seven-hundred and thirty-two patients (335 women [45.8%]) were included, with a median (IQR) age of 63 (55-69) years. The median (IQR) baseline 24-hour proteinuria and estimated glomerular filtration rate was 5.3 (2.8-8.5) g per 24 hours and 72 (48-92) mL/min/1.73m², respectively. In a competing-risk analysis, the 5-year cumulative incidence rates of progression to KRT decreased with deeper kidney responses as early as 6 months from therapy initiation (11%, 12%, 2.1%, and 0% for kidNR, kidPR, kidVGPR, and kidCR, respectively; P = .002) and were maintained at 12 months and 24 months and best kidney response. Patients able to achieve kidCR/kidVGPR by 24 months and at best response had significantly better OS compared with kidPR/kidNR. Kidney progression, defined as a 25% or greater decrease in estimated glomerular filtration rate, was associated with cumulative incidence of progression to KRT and OS.

Conclusions and Relevance

The results of this cohort study suggest that graded kidney response criteria offers clinically and prognostically meaningful information for treating patients with kidney AL amyloidosis. The response criteria potentially inform kidney survival based on the depth of reduction in 24-hour proteinuria levels and demonstrate an OS advantage for those able to achieve kidCR/kidVGPR compared with kidPR/kidNR. Taken together, achievement of at least kidVGPR by 12 months is needed to ultimately improve kidney and patient survival.

This cohort study examines graded kidney response criteria and their association with kidney and overall survival for patients with kidney light chain amyloidosis.

Introduction

Systemic light chain (AL) amyloidosis is a clonal plasma cell disorder associated with extracellular amyloid deposition that is followed by organ dysfunction and death.¹ Various organs can be affected, with kidney involvement seen in 50% to 70% of patients.^2,3,4 The principal clinical manifestations of kidney involvement are proteinuria, often in the nephrotic range, and kidney failure. End-stage kidney disease can develop in 30% to 40% of patients throughout the disease course.^5,6

The criteria for kidney response and progression in AL amyloidosis were formulated nearly 20 years ago.⁷ A decade later, a revised kidney response criterion lowered the threshold for kidney response from a 50% to 30% reduction in 24-hour proteinuria levels for better discrimination of kidney outcomes with the 30% threshold.⁶ Deeper kidney response (>75% reduction in proteinuria levels) has shown a trend toward superior survival compared with a smaller reduction,⁶ suggesting that depth of kidney response could be prognostic. Proposed graded kidney response criteria that stratified kidney response into 4 categories were associated with improved outcome prediction compared with the binary response system.⁸ We conducted this multicenter study to validate the prognostic value of graded kidney response criteria in AL amyloidosis.

Methods

The study was approved by the institutional review boards in the 10 participating centers, and informed consent was waived due to the retrospective nature of the study. Patients who received a diagnosis of AL amyloidosis between January 2010 and December 2015 were included if they (1) had kidney involvement and were evaluable for kidney response, which was defined as 24-hour proteinuria levels greater than 0.5 g per 24 hours, predominantly albumin; (2) achieved at least hematologic partial response within 12 months of diagnosis; and (3) received measurements of kidney function and 24-hour urine protein collection at least twice annually during the first 3 years and annually afterward. Patients were excluded if they were previously treated for associated hematologic disease (ie, multiple myeloma) or if they were receiving kidney replacement therapy (KRT) at diagnosis. Figure 1A lists the tested kidney response criteria, as previously proposed.⁸ Kidney response was assessed at fixed landmark points (6, 12, and 24 months from treatment initiation) and best kidney response reached at any time until death, last follow-up, or institution of subsequent chemotherapy beyond the first 12 months. Kidney stage was calculated based on the system proposed by Palladini et al,⁶ with alternative kidney staging as suggested by Kastritis et al.⁹

Figure 1. — A, Schematic representation of the definition of the graded kidney response criteria (in all response categories, a ≥25% decrease in estimated glomerular filtration rate from nadir must not occur). B, Percentage reduction in 24-hour proteinuria levels over time (median, IQR). C, Kidney response categories by landmark points and at best kidney response.

In the landmark and time-dependent covariate analyses, the best response observed by the landmark was used. When kidney response at a fixed point was missing, we imputed the kidney response as the best kidney response observed before that fixed point; if no kidney response was reported by that point (eg, by 6 months), we imputed kidney no response (kidNR). Kidney response was imputed in 165 (23%), 125 (18.3%), and 150 patients (23.6%) at 6, 12, and 24 months, respectively.

Summary statistics were used to describe study cohort characteristics. The Pearson χ² test and the Kruskal-Wallis test were used to ascertain differences in nominal and continuous variables between groups, respectively. Cumulative incidence of progression to KRT was evaluated from therapy initiation to progression to KRT. Patients who were alive and had not progressed to KRT were censored. Death before progression to KRT was treated as a competing event to avoid informative censoring. We used cumulative incidence with competing risk analyses, competing risk regression models, and the methods of Fine and Gray to evaluate and compare the cumulative incidence functions between groups. Cause-specific hazard ratios (csHRs) were used to compare groups for the specific event of interest (ie, event of KRT vs event of competing risk of death). Overall survival (OS) was calculated from the time of therapy initiation until death of any cause; patients alive at last follow-up were censored at that date. Cumulative incidence of progression to KRT and OS were plotted using the Kaplan-Meier method, and distributions between groups compared using the log-rank test. Multivariable Cox proportional hazards models were used to assess and identify factors prognostic for cumulative incidence of KRT and OS. In these models, we adjusted for key factors, such as type of first-line treatment, cardiac stage, best hematologic response, age, and light chain burden. To evaluate the prognostic utility of kidney response at given points and avoid survivorship bias, landmark analyses for OS were used; patients who had an event or were lost to follow-up before that landmark point were excluded (eFigure 1 in Supplement 1). In addition, time-dependent covariate Cox regression models for OS with longitudinal kidney response assessment (fixed times and best kidney response) were used to robustly evaluate the association of kidney response over time with OS. Comparison of 4-level with 2-level kidney response classifications was performed using time-dependent receiver operating characteristic (ROC) and area under the curve (AUC) methods using timeROC, compareC, and risksetROC packages in R (R Foundation). P < .05 was considered significant. Statistical analyses were performed using SAS (version 9.4.1; SAS Institutes), R (version 4.2.2; R Foundation), and JMP (SAS Institute).

Results

The study cohort consisted of 732 patients. The participants’ baseline characteristics are summarized in eTable 1 in Supplement 1. The median (IQR) age was 63 (55-69) years. The median (IQR) 24-hour proteinuria and estimated glomerular filtration rate (eGFR) was 5.3 (2.8-8.5) g per 24 hours and 72 (48-92) mL/min/1.73m², respectively. Kidney stage I (proteinuria ≤5 g/24-hour and eGFR ≥50), was assigned to 252 patients (34%). A total of 585 patients (80%) received 1 line of therapy during the initial 12 months of their diagnosis, while 147 (20%) received 2 lines or more. Bortezomib-based therapy was administered for 439 patients (60%). Hematologic complete response was achieved for 323 patients (44.1%), followed by hematologic very good partial response (280 [38.3%]) and hematologic partial response (129 [17.6%]). KRT was initiated during follow-up for 104 patients (14.2%), and 206 patients (28.0%) died. The median follow-up was 69 months (IQR, 54-88 months).

Evolution of Kidney Response Over Time

A reduction in 24-hour proteinuria levels from baseline improved over time, with a median (IQR) percentage reduction of 34% (0%-63%), 50% (14%-77%), and 71% (36%-91%) by 6, 12, and 24 months, respectively. Response depth at each time point is depicted in Figure 1B. The proportion of kidNR gradually declined between 6 months to 12 and 24 months (58% vs 39.4% vs 27.8%, respectively). In contrast, the combined proportion of kidney very good partial response (kidVGPR) and kidney complete response (kidCR) more than doubled by these points (22.1%, 40.1% and 56.9%, respectively).

At best response, kidCR, kidVGPR, kidney partial response (kidPR), and kidNR were achieved in 191 (26.1%), 232 (31.7%), 106 (14.5%), and 203 patients (27.7%), respectively (Figure 1C). The median time to best kidney response was 16.0 months (IQR, 7.0-28.9) and was longer for kidCR (22.1 months; IQR, 9.9-39.6) compared with kidVGPR (16.2 months; IQR, 7.1-25.6) or kidPR (10.3 months; IQR, 5.5-17.9).

Association of Baseline Characteristics and Best Kidney Response

Patients who achieved kidCR had lower baseline 24-hour proteinuria levels and higher eGFR compared with patients who achieved kidVGPR, kidPR, or kidNR (eTable 2 in Supplement 1). Thus, a higher proportion of kidney stage I at diagnosis was observed among patients with kidCR compared with kidVGPR, kidPR, and kidNR (50% vs 30% vs 32% vs 27%, respectively; P < .001). Patients who were treated with autologous stem cell transplant were more likely to achieve kidCR than those treated with bortezomib-based, alkylator-based, or immunomodulatory drug–based therapy (36.8% vs 22.1% vs 20.7% vs 19.2%, respectively; P = .001), despite similar proportion of kidney stage I at diagnosis (37% vs 34% in the non–autologous stem cell transplant groups; P = .37). The proportion of hematologic complete response within each kidney response category increased with deeper kidney response (kidNR, 22.9%; kidPR, 32.4%; kidVGPR, 48.9%; kidCR, 67.5%; P < .001).

Association of Fixed-Time Response Assessment With Kidney and OS

The cumulative incidence of progression to KRT was significantly associated with the depth of kidney response as early as 6 months. Depth of kidney response by 6 months was inversely associated with cumulative incidence of progression to KRT (5-year, 0%, 2.1%, 12%, and 11% for kidCR, kidVGPR, kidPR, and kidNR, respectively; P = .002; Figure 2A). This discriminatory association was maintained in the 12-month and 24-month landmarks (Figure 2, B and C). For example, the 5-year cumulative incidence of progression to KRT was 0%, 2.0%, 9.9%, and 11% for kidCR, kidVGPR, kidPR, and kidNR achieved by 24 months, respectively (P < .001). In the csHR models (comparing the risk of KRT with the competing risk of death), achieving kidCR or kidVGPR had lower risk of progression to KRT compared with kidPR at all points (csHR of 0.21 [95% CI, 0.08-0.56], 0.31 [95% CI, 0.15-0.66], and 0.24 [95% CI, 0.11-0.51], at 6, 12, and 24 months, respectively; P < .003). No patients who achieved a kidCR at the 6-month or 12-month points had a progression to KRT during follow-up.

Figure 2. — Cumulative incidence of progression to KRT as stratified by the kidney response by the 6-month (A), 12-month (B), and 24-month (C) landmarks.

Any kidney response by 6 months did not significantly discriminate OS (Figure 3A). However, at 12 months, the OS distributions associated with patients achieving a kidPR, kidVGPR, or kidCR were significantly better than those who achieved kidNR (5-year estimated OS for kidNR, 72.3% vs 84.7% [kidPR], 86.2% [kidVGPR], and 85.6% [kidCR]; P < .001) (Figure 3B). At 24-month, kidNR, and kidPR had significantly worse survival compared to patients who achieved kidVGPR or kidCR by this landmark (estimated 5-year OS of 75.2% [kidNR], 79.7% [kidPR], 91.4% [kidVGPR], and 91.6% [kidCR]; P < .001) (Figure 3C). A sensitivity analysis for OS by the 4-level kidney response to include only patients with available kidney response data at landmarks (ie, excluding imputed data at landmarks) is presented in eTable 3 in Supplement 1 and supports the previously described findings.

Association of Best Kidney Response With Kidney Survival and OS

The 5-year cumulative incidence of progression to KRT was inversely associated with best kidney response depth (0% vs 3.1% vs 18% vs 33% for kidCR, kidVGPR, kidPR, or kidNR, respectively; P < .001; eFigure 2A in Supplement 1). The association of best kidney response with the cumulative incidence of progression to KRT by kidney stage is depicted in eFigures 2B and C in Supplement 1. A deeper kidney response was associated with a longer OS (5-year OS of 90%, 85%, 66%, and 54% for kidCR, kidVGPR, kidPR, and kidNR, respectively; P < .001; eFigure 2D in Supplement 1), with survival discrimination maintained mostly between those collectively achieving kidCR/VGPR and patients achieving kidPR or kidNR. At the 1-year landmark, the respective 5-year OS rates were 91%, 87%, 70%, and 61% (P < .001; eFigure 2E in Supplement 1). The association of graded kidney response criteria with OS as stratified by kidney stage is depicted in eFigure 3 in Supplement 1. Utilizing all kidney response data over time, we observed that in the time-dependent covariate competing-risk regression model, the depth of kidney response was significantly associated with the cumulative incidence of progression to KRT after adjusting for established risk factors and the competing risk of death. In this multivariable model, the adjusted csHR showed minimal risk of progression to KRT in those who achieved a kidCR compared with kidNR (csHR, 0.03; 95% CI, 0.004-0.200; P < .001; eTable 4 in Supplement 1). The csHR of kidVGPR and kidPR compared with kidNR was 0.19 (95% CI, 0.10-0.37; P < .001) and 0.41 (95% CI, 0.20-0.83; P = .01), respectively. The graded kidney response criteria were tested in univariate analysis and time-dependent covariate Cox models for OS (Table). Along with age, cardiac stage, and best hematologic response, kidCR and kidVGPR showed superior independent OS compared with those with kidPR and kidNR. Alternative time-dependent covariate Cox models for OS in which kidPR is the reference kidney response are provided in eTables 5 and 6 in Supplement 1. Multivariate analyses for OS at the landmarks are presented in eTables 7 to 9 in Supplement 1 and suggest an OS advantage for kidCR/kidVGPR vs kidPR/kidNR at the 24-month point after adjusting for other risk factors.

Table. Univariate Analysis and Multivariable Model for Overall Survival Using Time-Varying Longitudinal Analysis.

Covariate	Univariate analysis		Multivariate analysis with 4-level kidney response classification		Multivariate analysis with 2-level kidney response classification
Covariate	HR (95% CI)	P value	HR (95% CI)	P value	HR (95% CI)	P value
Best kidney response
Any response (vs NR)	0.47 (0.35-0.64)	<.001	NA		0.62 (0.45-0.87)	.01
PR (vs NR)	0.69 (0.45-1.07)	.09	0.88 (0.55-1.42)	.61	NA
VGPR (vs NR)	0.38 (0.25-0.56)	<.001	0.51 (0.33-0.78)	.002
CR (vs NR)	0.47(0.31-0.71)	<.001	0.58 (0.36-0.94)	.03
Age ≥65 y (vs <65)	2.60 (1.95-3.46)	<.001	2.06 (1.48-2.86)	<.001	2.08 (1.50-2.87)	<.001
Cardiac stage
II (vs I)	4.40 (2.57-7.52)	<.001	3.88 (2.31-6.54)	<.001	3.85 (2.28-6.48)	<.001
IIIa (vs I)	7.00 (4.06-12.1)	<.001	5.58 (3.21-9.71)	<.001	5.70 (3.27-9.93)	<.001
IIIb (vs I)	14.1 (7.52-26.3)	<.001	12.01 (6.11-23.61)	<.001	11.82 (6.02-23.21)	<.001
Light chain burden
dFLC ≥180 mg/L (vs <180)	1.67 (1.27-2.19)	<.001	1.05 (0.75-1.46)	.79	1.03 (0.74-1.43)	.86
Best hematologic response
VGPR (vs PR)	0.55 (0.40-0.75)	<.001	0.58 (0.39-0.86)	.01	0.59 (0.39-0.87)	.01
CR (vs PR)	0.21 (0.14-0.30)	<.001	0.28 (0.18-0.44)	<.001	0.27 (0.17-0.42)	<.001
Type of first-line treatment
Bortezomib-based	1 [Reference]	NA	1 [Reference]	NA	1 [Reference]	NA
ASCT	0.34 (0.23-0.50)	<.001	0.67 (0.44-1.03)	.07	0.69 (0.45-1.05)	.09
Alkylator-based	1.69 (1.12-2.54)	.01	1.57 (1.03-2.39)	.04	1.56 (1.02-2.39)	.04
IMiD-based	1.25 (0.69-2.25)	.47	1.18 (0.56-2.48)	.67	1.16 (0.55-2.43)	.70

Open in a new tab

Abbreviations: ASCT, autologous stem cell transplantation; CR, complete response; dFLC, difference between involved to uninvolved light chain; HR, hazard ratio; IMiD, immunomodulatory drug; NA, not applicable; NR, no response; PR, partial response; VGPR, very good partial response.

Comparison of 2-Level and 4-Level Response Criteria on Cumulative Incidence of Progression to KRT and OS

Two-level kidney response classification (with a 30% reduction cutpoint) was associated with cumulative incidence of progression to KRT at all landmark models and best kidney response (eFigure 4 in Supplement 1). Using time-dependent AUCs, the 4-level kidney response performed better than the 2-level response in predicting cumulative incidence of progression to KRT at all points. The performance gap between the 4-level and 2-level response classifications gradually widened between the points (eFigure 6 in Supplement 1). A 2-level kidney response was able to discriminate OS by 12 months, which persisted by 24 months and at best kidney response (eFigure 5 in Supplement 1) and retained prognostic independence in the multivariable analysis (Table). The 4-level kidney response criteria did not significantly discriminate OS better than the 2-level response in the 6-month and 12-month landmark models (eFigure 6A-B in Supplement 1), but at 24 months and best response had superior survival discrimination compared with the 2-level response (eFigure 6C-D in Supplement 1).

Association of Kidney Progression With Risk of KRT and OS

eGFR progression, defined as a 25% or greater decrease in eGFR from nadir, was documented in 209 patients (28.6%). The median eGFR value at eGFR progression was 33 mL/min/1.73 m2 (IQR, 21-45), a median reduction of 27 from the baseline value (IQR, 17-41) (alternatively, a median reduction of 44% from baseline value [IQR, 32%-57%]). The median time from therapy initiation to eGFR progression was 13.5 months (IQR, 5.0-35.0). In a competing risk analysis, eGFR progression was associated with an increased risk of progression to KRT compared with no eGFR progression at all points (eFigure 7 in Supplement 1). eGFR progression occurring by 6 months had the highest cumulative incidence of progression to KRT compared to no eGFR progression (5-year cumulative incidence of KRT 42% vs 7.6%, P < .001). eGFR progression was associated with a significantly shorter OS compared with no eGFR progression (5-year OS, 65% vs 80%; P < .001; eFigure 8 in Supplement 1). The association of proteinuria progression with risk of KRT and OS is provided in Supplement 1. Proteinuria progression was not significantly associated with the outcome of interest.

Discussion

The results of this cohort study suggest that graded kidney response criteria were prognostic for kidney and OS and should replace the binary response criteria. These new response criteria highlight the importance of achieving deep kidney response (≥kidVGPR) to maximize patient outcomes. With the recognition of the importance of maximizing the reduction in proteinuria levels, a more careful assessment of kidney response is now feasible and can be used to identify new ways to assess therapies.

During the past 2 decades, staging and response assessment of kidney AL amyloidosis changed.^6,7,8,9 The present study comes at a time of improved therapeutic options,^{10,11,12,13,14,15} creating the need to refine and update disease risk assessment, including organ response depth, following therapy. The concept of kidney staging was important in solidifying kidney survival as a clinical end point in kidney amyloidosis and establishing a simple and robust method to assess the risk of dilaysis.^6,9 In this study, baseline kidney stage was associated with the depth of kidney response following treatment. This emphasizes the need for early diagnosis to improve outcomes. An increased probability of complete organ recovery in patients with milder organ dysfunction is seen also in patients with cardiac involvement.¹⁶

The graded kidney response criteria provide clinicians with tools to reassess the need to introduce or modify therapy to enhance the likelihood of organ and patient survival. The kidney response criteria suggested by Palladini et al⁶ and Kastritis et al⁹ found that responses as early as 6 months could differentiate the risk of end-stage kidney disease. This study validates this finding. We recommend the achievement of at least kidPR by 6 months and at least kidVGPR by 12 and 24 months to reduce the risk of progression to KRT and enhance outcomes. Failure to meet these end points should prompt consideration of alternate therapy for AL, especially if other causes of kidney dysfunction have been excluded and residual clonal disease is evident.¹⁷ The time-based kidney response recommendations may allow a timeline for clinical trial design, defining early therapeutic failures qualifying for salvage therapies.

A deeper kidney response was associated with improved kidney survival. This aligned with other kidney diseases for which proteinuria reduction was associated with better outcomes.¹⁸ The OS advantage with a deeper kidney response is a novel observation in kidney AL amyloidosis, since patients with kidney AL amyloidosis have a lower risk of death compared with patients with dominant heart involvement.^2,19 It is possible that deeper kidney response indicates better control of the underlying plasma cell clone, which in turn is associated with reduced amyloid deposition outside the kidney (ie, heart) and better survival. However, a lower risk of progression to KRT with deeper kidney response may carry its own independent survival advantage.

At the 12-month landmark, the 5-year OS estimates were slightly higher for kidVGPR over kidCR; however, 95% CIs of these estimates were broadly overlapping. The median time to kidCR was longer than the time to kidVGPR (22.1 vs 16.2 months). Therefore, many patients who eventually achieve kidCR did not reach it early. Thus, the discriminatory power of kidCR at earlier landmarks is lower due to the slower response time, and some patients who eventually reached kidCR are reflected in lower response categories at earlier landmark analyses. While it seems possible to combine kidCR and kidVGPR into 1 response category given the relatively minor changes in outcomes between these 2 response categories, we suggest that these 2 kidney response categories be maintained. The concept of complete organ recovery is important in treating patients with amyloidosis, and this cannot be distinguished if the 2 response categories are combined. Further, we did observe differences in the magnitudes of the association for those able to achieve kidCR vs kidVGPR, especially in looking at kidney survival.

Limitations

This study had several limitations. It was a retrospective study with potential bias in patient selection. In addition, we could not provide landmark kidney response assessment for all patients related to incomplete data, although we used imputation to minimize data loss. Lastly, we did not assess the association of hematologic progression with the previously described criteria due to limited resources and differences in hematologic progression thresholds, which are associated with variation in the decision to reinstitute therapy.

Conclusions

The results of this cohort study confirmed the importance of graded kidney response criteria in AL amyloidosis. These criteria predicted kidney and OS based on the depth of reduction in 24-hour proteinuria levels. These criteria outperformed the binary kidney response system and allowed for better assessment of treatment effectiveness. Applying these criteria at fixed points provided kidney response time–based targets to maximize outcomes. The goal of therapy in kidney AL amyloidosis should be a kidPR by 6 months and kidVGPR by 12 and 24 months from treatment initiation. Failure to meet these end points should prompt consideration of alternate therapy. With improvement in the management of AL amyloidosis, earlier landmarks to achieve these organ response goals may be emerging.

Supplement 1.

eTable 1. Baseline characteristics of the study cohort

eTable 2. Comparison of the four renal response groups based on baseline parameters and response evaluation

eTable 3. Sensitivity analysis of OS by renal response categories at landmarks in the full data set (with imputed responses) compared to using exclusively renal responses data available at landmarks

eTable 4. Adjusted time-dependent competing risk regression of RRT with competing risk of death

eTable 5. Univariate analysis and multivariable model for overall survival using time-varying longitudinal analysis (renPR as the reference renal response; renCR, renVGPR separated)

eTable 6. Univariate analysis and multivariable model for overall survival using time-varying longitudinal analysis (renPR as the reference renal response; renCR, renVGPR combined)

eTable7. 6‐month landmark univariate and multivariate analysis for overall survival

eTable 8. 12‐month landmark univariate and multivariate analysis for overall survival

eTable 9. 24‐month landmark univariate and multivariate analysis for overall survival

eTable 10. The interaction effect between time-dependent eGFR and proteinuria progressions on overall survival

eFigure 1. Flow diagram for landmark analysis cohorts

eFigure 2. Cumulative incidence of progression to renal replacement therapy (RRT) and overall survival (OS) by best renal response

eFigure 3. Overall survival by best renal response stratified by renal stage

eFigure 4. Incidence proportion of progression to renal replacement therapy using two-level renal response classification

eFigure 5. Overall survival outcome based on 2-level renal response classification

eFigure 6. Adjusted dynamic receiver operator characteristic curve and corresponding AUC comparing two-level and four-level renal response for cumulative incidence of progression to renal replacement therapy for: (A) 6-month landmark, (B) 12-month landmark (C) 24-month landmark

eFigure 7. Adjusted dynamic receiver operator characteristic curve and corresponding AUC comparing two-level and four-level renal response for overall survival at: (A) 6-month landmark, (B) 12-month landmark (C) 24-month landmark

eFigure 8. Incidence proportion of progression to renal replacement therapy based on eGFR progression status at: A) 6-month landmark, (B) 12-month landmark (C) 24-month landmark

eFigure 9. Overall survival by eGFR progression status

jamaoncol-e242629-s001.pdf^{(2.4MB, pdf)}

Supplement 2.

Data sharing statement

jamaoncol-e242629-s002.pdf^{(14.3KB, pdf)}

References

1.Gertz MA, Dispenzieri A. Systemic amyloidosis recognition, prognosis, and therapy: a systematic review. JAMA. 2020;324(1):79-89. doi: 10.1001/jama.2020.5493 [DOI] [PubMed] [Google Scholar]
2.Muchtar E, Gertz MA, Kyle RA, et al. A modern primer on light chain amyloidosis in 592 patients with mass spectrometry-verified typing. Mayo Clin Proc. 2019;94(3):472-483. doi: 10.1016/j.mayocp.2018.08.006 [DOI] [PubMed] [Google Scholar]
3.Palladini G, Merlini G. What is new in diagnosis and management of light chain amyloidosis? Blood. 2016;128(2):159-168. doi: 10.1182/blood-2016-01-629790 [DOI] [PubMed] [Google Scholar]
4.Schulman A, Connors LH, Weinberg J, et al. Patient outcomes in light chain (AL) amyloidosis: the clock is ticking from symptoms to diagnosis. Eur J Haematol. 2020;105(4):495-501. doi: 10.1111/ejh.13472 [DOI] [PubMed] [Google Scholar]
5.Gertz MA, Leung N, Lacy MQ, et al. Clinical outcome of immunoglobulin light chain amyloidosis affecting the kidney. Nephrol Dial Transplant. 2009;24(10):3132-3137. doi: 10.1093/ndt/gfp201 [DOI] [PubMed] [Google Scholar]
6.Palladini G, Hegenbart U, Milani P, et al. A staging system for renal outcome and early markers of renal response to chemotherapy in AL amyloidosis. Blood. 2014;124(15):2325-2332. doi: 10.1182/blood-2014-04-570010 [DOI] [PubMed] [Google Scholar]
7.Gertz MA, Comenzo R, Falk RH, et al. Definition of organ involvement and treatment response in immunoglobulin light chain amyloidosis (AL): a consensus opinion from the 10th International Symposium on Amyloid and Amyloidosis, Tours, France, 18-22 April 2004. Am J Hematol. 2005;79(4):319-328. doi: 10.1002/ajh.20381 [DOI] [PubMed] [Google Scholar]
8.Muchtar E, Dispenzieri A, Leung N, et al. Depth of organ response in AL amyloidosis is associated with improved survival: grading the organ response criteria. Leukemia. 2018;32(10):2240-2249. doi: 10.1038/s41375-018-0060-x [DOI] [PubMed] [Google Scholar]
9.Kastritis E, Gavriatopoulou M, Roussou M, et al. Renal outcomes in patients with AL amyloidosis: Prognostic factors, renal response and the impact of therapy. Am J Hematol. 2017;92(7):632-639. doi: 10.1002/ajh.24738 [DOI] [PubMed] [Google Scholar]
10.Dispenzieri A, Kastritis E, Wechalekar AD, et al. A randomized phase 3 study of ixazomib-dexamethasone versus physician’s choice in relapsed or refractory AL amyloidosis. Leukemia. 2022;36(1):225-235. doi: 10.1038/s41375-021-01317-y [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Kastritis E, Leleu X, Arnulf B, et al. Bortezomib, melphalan, and dexamethasone for light-chain amyloidosis. J Clin Oncol. 2020;38(28):3252-3260. doi: 10.1200/JCO.20.01285 [DOI] [PubMed] [Google Scholar]
12.Kastritis E, Palladini G, Minnema MC, et al. ; ANDROMEDA Trial Investigators . Daratumumab-based treatment for immunoglobulin light-chain amyloidosis. N Engl J Med. 2021;385(1):46-58. doi: 10.1056/NEJMoa2028631 [DOI] [PubMed] [Google Scholar]
13.Merlini G, Dispenzieri A, Sanchorawala V, et al. Systemic immunoglobulin light chain amyloidosis. Nat Rev Dis Primers. 2018;4(1):38. doi: 10.1038/s41572-018-0034-3 [DOI] [PubMed] [Google Scholar]
14.Muchtar E, Gertz MA, Kumar SK, et al. Improved outcomes for newly diagnosed AL amyloidosis between 2000 and 2014: cracking the glass ceiling of early death. Blood. 2017;129(15):2111-2119. doi: 10.1182/blood-2016-11-751628 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Palladini G, Milani P, Merlini G. Management of AL amyloidosis in 2020. Blood. 2020;136(23):2620-2627. doi: 10.1182/blood.2020006913 [DOI] [PubMed] [Google Scholar]
16.Muchtar E, Dispenzieri A, Wisniowski B, et al. Graded cardiac response criteria for patients with systemic light chain amyloidosis. J Clin Oncol. 2023;41(7):1393-1403. doi: 10.1200/JCO.22.00643 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Palladini G, Merlini G. When should treatment of AL amyloidosis start at relapse? Early, to prevent organ progression. Blood Adv. 2019;3(2):212-215. doi: 10.1182/bloodadvances.2018021253 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.de Zeeuw D, Remuzzi G, Parving HH, et al. Proteinuria, a target for renoprotection in patients with type 2 diabetic nephropathy: lessons from RENAAL. Kidney Int. 2004;65(6):2309-2320. doi: 10.1111/j.1523-1755.2004.00653.x [DOI] [PubMed] [Google Scholar]
19.Sidana S, Tandon N, Gertz MA, et al. Clinical features, laboratory characteristics and outcomes of patients with renal versus cardiac light chain amyloidosis. Br J Haematol. 2019;185(4):701-707. doi: 10.1111/bjh.15832 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials