The clinical trials landscape in immunoglobulin light chain amyloidosis: a systematic review

TO THE EDITOR:

Immunoglobulin light chain (AL) amyloidosis is a rare hematologic disorder characterized by end-organ damage from deposition of fibrillar aggregates of kinetically unstable light chains.¹ The key to successful management of AL amyloidosis is rapid and profound reduction of amyloidogenic free light chains (FLCs) using plasma cell clone–directed therapies.² The treatment armamentarium in AL amyloidosis is therefore mostly borrowed from that of multiple myeloma. However, although drug development in myeloma has proceeded rapidly with >15 new therapies approved since 2010, just 1 therapy has received FDA-accelerated approval specifically for AL amyloidosis: anti-CD38 monoclonal antibody daratumumab (2021).³ Potential reasons for this gap include the rarity of AL amyloidosis and its historically poor prognosis, especially patients with cardiac amyloid with baseline N-terminal pro-brain natriuretic peptide (NT-proBNP) >8500 pg/mL,^4,5 who often have early mortality or cardiac adverse events that might deter pharmaceutical companies from investing in this space.

Furthermore, the lack of contemporary criteria to standardize assessment of hematologic progression complicates trial design in AL amyloidosis, especially in the relapsed/refractory setting. The 2005 International Society of Amyloidosis criteria for hematologic progression requires the involved FLC to be at least 10 mg/dL to define hematologic progression in most patients.⁶ However, a recent survey of global amyloidosis experts revealed their discomfort to wait until such an involved FLC elevation, likely due to the risk of organ progression.⁷ Another challenge is that clinical trials often exclude patients with advanced cardiac and renal dysfunction. We performed a systematic review to define the current landscape of clinical trials in AL amyloidosis and highlight unmet needs in this population.

A search on ClinicalTrials.gov was performed on September 2023, using the term “amyloidosis” under the field, condition/disease. We included all trials in AL amyloidosis testing a pharmaceutical agent and that are currently ongoing, defined as having an enrollment status of “not yet recruiting,” “recruiting,” “active, not recruiting,” and “enrolling by invitation.” We excluded studies that had “suspended,” “withdrawn,” or “unknown” status. Data on key elements of the study design (inclusion/exclusion criteria, intervention, end points, and sponsor) were extracted by 2 independent reviewers (R.R. and V. Shah), and discrepancies were resolved by a third reviewer (R.C.).

Our search generated 280 unique entries, among which 32 trials that tested a pharmaceutical agent in systemic AL amyloidosis were included for analysis. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flowchart is shown in supplemental Appendix I. The characteristics of included trials are summarized in Table 1. Notably, the majority of trials were in patients with previously treated AL amyloidosis (17/32 [53%]), followed by the newly diagnosed setting (12/32 [38%]). Just 9 of 32 trials (28.1%) tested novel agents for AL amyloidosis rather than those already approved for myeloma. Most trials (23/32 [71.9%]) were multicenter, and just a minority were industry sponsored (8/32 [25%]). Among the 5 phase-3 trials, 3 were industry–sponsored registrational trials. The mean estimated sample size for industry-sponsored vs non–industry-sponsored trials was 143 (±28.8) vs 59 (±16.6) patients, respectively (P = .0174). The mean sample size of multicenter trials was significantly higher than that of single-center trials (95 ± 18 vs 42 ± 29, respectively; P = .025). Industry-sponsored trials had a substantially higher likelihood of being in the newly diagnosed setting (62.5%) than non–industry-sponsored trials (29.2%; P = .146). Of 32 trials, 31 reported the cutoff in difference between involved and uninvolved FLC (dFLC) used for trial inclusion. The most common cutoff was 5 mg/dL (16/31 [52%]), followed 2 mg/dL (8/31 [26%]). Of 2 trials that had no dFLC cutoff, 1 evaluated the different durations of daratumumab maintenance (NCT05898646), and 1 tested siltuximab to reduce symptom-burden after autotransplant (NCT03315026). Serum M-spike was allowed as measurable disease in 10 trials, with the most common cutoff being 0.5 g/dL (n = 9). Among 27 trials with available data on the upper limit of NT pro-BNP cutoff for exclusion, the most common cutoff was 8500 pg/mL (17/27 [63%]), followed by 1800 (2/27 [7%]), 5000 (2/27 [7%]), and 7500 pg/mL (1/27 [4%]). Of 5 trials that did not exclude patients based on NT-proBNP, 4 were specifically targeted to patients with stage IIIb disease, and 1 was a pragmatic trial that included all comers. Notably, 23 trials excluded patients with renal function below a specified eGFR cutoff: most commonly 40 mL per minute per 1.73 m² (6 trials), 30 mL (6 trials), and 20 mL (6 trials). Only 2 trials with an eGFR cutoff tested experimental drugs that are excreted renally (lenalidomide [NCT03252600] and pomalidomide [NCT04270175]).

The most common primary end points were hematologic response rate (13/32 [41%]) and safety-related end points (13/32 [41%]); overall survival (OS) was a primary end point in just 4 of 32 trials (13%). Industry-sponsored trials had a higher incidence of having safety as primary end point than non–industry-sponsored trials (75% vs 29.2%, respectively; P = .022). Health-related quality of life (HRQoL) was measured as a prespecified end point in just 12 of 32 trials (38%). The majority of interventions were fixed-duration (26/31 [84%]); industry-sponsored trials were significantly more likely to treat until progression than non–industry-sponsored trials (50% vs 4.4%; P = .0045).

We demonstrate substantial heterogeneity in the eligibility criteria and definitions of measurable disease used in current AL amyloidosis clinical trials. Two-thirds of trials continue to exclude patients with NT-proBNP >8500 pg/mL, despite an improved prognosis for these patients in the daratumumab era.^8,9 Because early mortality in the era of daratumumab-bortezomib-cyclophosphamide-dexamethasone frontline therapy predominantly occurs in patients with stage IIIb disease (ie, baseline NT-proBNP >8500 pg/mL),¹⁰ it will be more difficult for emerging therapies to demonstrate an OS benefit in a reasonable time frame without including this high-risk group. Additionally, trials in relapsed/refractory setting should not exclude patients with stage IIIb disease at diagnosis because the steep drop in survival slope is limited to the first year after diagnosis. Despite the availability of validated patient-reported outcomes (PRO) instruments in AL amyloidosis and a well-defined HRQoL trajectory,^11-15 only one-third of trials measured a PRO end point.

We also observed discordance in the dFLC cutoffs used for measurable disease. Notably, the current hematologic response criteria in AL amyloidosis defines very good partial response as dFLC <4 mg/dL and requires a baseline dFLC ≥5 mg/dL to be considered response evaulable.¹⁶ However, several seminal studies have now demonstrated the strong prognostic impact of achieving dFLC <1 mg/dL at the end of treatment, highlighting that even small amounts of persistent clonal light chain can lead to ongoing organ damage.^17-20 Additionally, several groups had proposed a new response category named “low-dFLC partial response (PR)” for patients with baseline dFLC of 2 to 5 mg/dL, defined as a posttreatment dFLC <1 mg/dL without achieving a CR.^21-23 Because hematologic response evaluation is now feasible for all patients with dFLC ≥2 mg/dL (due to creation of the new response category low-dFLC PR), regulatory authorities should encourage inclusion of these patients in clinical trials. Furthermore, novel assays to check light chain burden such as mass spectrometry-based FLC assays²⁴ should be evaluated in clinical trials. Although significant advances have been made in clone-directed therapy, in large part owing to successful development in myeloma, therapies targeting light chain cytotoxicity and deposited amyloid fibrils in vital organs remain an unmet need and should be tested in RCTs. Furthermore, because patients with AL are frailer than those with myeloma, trials should be designed with the shortest possible treatment duration needed to achieve the desired efficacy. Table 2 highlights some suggestions in clinical trial design in AL amyloidosis, along with areas of unmet need.

In conclusion, our study highlights the urgent need to harmonize definitions of measurable disease and hematologic progression criteria in AL amyloidosis clinical trials, greater inclusion of patients with advanced organ involvement, and increased use of PRO end points.

Contribution: R.R. wrote the first draft of the manuscript; D.S. and R.R. participated in manuscript writing; R.R., C.L., and V. Shah abstracted the data; R.C. designed the research, edited the manuscript, and approved the final version; and V. Sanchoralwala, S.L., and E.R.S.C. provided critical input and approved the final draft of the manuscript.

Conflict-of-interest disclosure: E.R.S.C. receives research funding from Arnold Ventures. R.C. reports consulting/advisory board fees from Janssen, Sanofi, and Adaptive Biotech. The remaining authors declare no competing financial interests.

Correspondence: Rajshekhar Chakraborty, Columbia University Irving Medical Center, 161 Fort Washington Ave, Herbert Erving Pavillion, New York, NY 10032; email: rc3360@cumc.columbia.edu.