Portal de Periódicos da CAPES

Reduced early mortality with Daratumumab‐based frontline therapy in AL amyloidosis: A retrospective cohort study

2023; Wiley; Volume: 99; Issue: 3 Linguagem: Inglês

10.1002/ajh.27179

1096-8652

Rajshekhar Chakraborty, Divaya Bhutani, Markus Y. Mapara, Ran Reshef, Matthew J. Maurer, Jai Radhakrishnan, Suzanne Lentzsch,

Chronic Lymphocytic Leukemia Research

In the past decade, significant advancements in therapies targeting plasma cell clone have brought about a transformative shift in the treatment landscape for immunoglobulin light chain (AL) amyloidosis. Despite these strides, early mortality (EM), secondary to advanced cardiac involvement, has remained a substantial challenge, with 6-month EM ~25% in the bortezomib-era.1 Currently, the frontline regimen for AL amyloidosis is a combination of the anti-CD38 monoclonal antibody daratumumab (Dara) with CyBorD (cyclophosphamide-bortezomib-dexamethasone), based on the ANDROMEDA trial.2 While the data on overall survival (OS) from this trial remains immature, the proportion of deaths was numerically lower in the Dara-CyBorD compared to CyBorD arm (17% vs. 24%).3 The objective of our retrospective cohort study was to assess EM rates at 6 and 12 months in patients receiving frontline daratumumab-based treatments, regardless of eligibility for HDM-AHCT or ANDROMEDA trial. Additionally, we aimed to compare the rates of hematologic and organ responses with those in the ANDROMEDA trial. We included all consecutive patients with systemic AL amyloidosis who were treated with daratumumab-based frontline therapy at Columbia University up until April 2023. The cut-off for our analysis was 7/31/2023. Survival analysis was done using the Kaplan–Meier method, with differences between curves tested for statistical significance by two-sided log-rank test. Variables associated with OS on univariate analysis at p < .1 were entered into a Cox proportional hazards regression model for multivariable analysis. Statistical analysis was done using JMP Pro 15.0 (SAS Institute Inc., Cary, NC). Detailed methods are summarized in Supplementary Appendix I. Eighty-four patients treated with daratumumab-based frontline therapies were included. The racial/ethnic makeup of our cohort was 71% White, 17% Blacks/African-Americans, 8.5% Asians, and 3.5% others/mixed race, with 4% of Hispanic ethnicity. The baseline clinical characteristics of our cohort in comparison with Dara-VCD arm of ANDROMEDA trial are summarized in Supplementary Appendix II. Importantly, a majority of our patients (66%) were Mayo 2004 stage IIIa or IIIb. Patients in our cohort had a high clone burden at diagnosis, as evidenced by 76% with ≥10% bone marrow plasma cells, and 61% with dFLC>18 mg/dL. The most common first-line regimen was Dara-VCD (n = 79), followed by Dara-VD (n = 2), Dara-Dex (n = 2), and Dara-Ixazomib-Dex (n = 1). The median follow-up from treatment initiation was 23.5 months (95% CI, 19.6–26.1). The Kaplan–Meier curve for OS for the entire cohort is shown in Figure 1. The median OS for our cohort is not reached and estimated 2-year OS is 79.4% (95% CI, 67.9–87.6). The cumulative incidence of all-cause mortality at 3, 6, and 12 months was 3.6% (95% CI, 1.2–10.5), 7.5% (95% CI, 3.4–15.7), and 11.7% (95% CI, 6.1–21.1) respectively. Kaplan–Meier curve for cumulative incidence of EM by stage at diagnosis (Mayo 2004 stage IIIb vs. I–IIIa) is shown in Figure 1. The 6-month EM rate in stage I–IIIa versus IIIb was 1.9% (95% CI, 0.3–12.2) versus 22.8% (95% CI, 9.7–44.6) respectively (p < .001). Within stages I–-IIIa, the 6-month EM rate in stages I, II, and IIIa were 0%, 6.25%, and 0% respectively. Kaplan–Meier curve for EM by Mayo 2012 stage at diagnosis is shown in Supplementary Appendix III. Among patients who underwent HDM-AHCT as part of frontline therapy (n = 12), there was no transplant-related mortality by day#100. Three out of 84 patients underwent OHT, with the timing being 1.4–4.0 months from the first dose of daratumumab-based frontline therapy. All three patients are alive at latest follow-up. 11/84 patients (13.4%) had an infection in the first 6 months that was either ≥grade 3 or led to treatment discontinuation. The most common site of infection was respiratory tract (n = 6). There were no infection-related deaths in the first 12 months. Among 9/84 patients who died within 12 months of treatment initiation, the cause of death was progressive cardiac failure (n = 6), sudden cardiac arrest (n = 1), massive ischemic stroke (n = 1), and unknown (n = 1). Next, we investigated hematologic and organ response rates in our cohort in comparison with ANDROMEDA trial. Among 81 patients that were evaluable for hematologic response, 39 achieved Heme-CR (48%) and 60 achieved a very good partial response (VGPR) or better (74%). Of note, one patient had missing hematologic response data and was coded as a non-responder. Among patients that achieved a heme-CR, the median time to CR from treatment initiation was 2.7 months (range, 0.43–13.97). The proportion of patients with iFLC<2 mg/dL and dFLC 0.5 g/24 h, 28 had available data on renal response. A renal response was achieved in 18/28 patients (64.3%). At 6 and 12-month time-points from treatment initiation, the renal response rates were 43% and 50% respectively. The hematologic and organ response rates in our cohort in comparison with that of ANDROMEDA trial are shown in Supplementary Appendix IV. The median time to first cardiac response and best cardiac response was 3.0 months (range, 0.4–10.0) and 13.3 months (range, 3.7–35.3) respectively. The median time to first renal response and best renal response was 4.1 months (range, 0.3–13.9 months) and 15.9 months (range, 2.9–51.1) respectively. Next, we investigated predictors of survival in the context of daratumumab-based frontline therapy. We picked the three prognostic factors that are included in the Mayo 2012 staging system (dFLC, NT-proBNP, and troponin). On the multivariate Cox proportional hazards model, using NT-proBNP (per 1000 pg/mL increase), hs-TnT (per 10 ng/L increase), and dFLC (per 10 mg/dL increase) as continuous variables, NT-proBNP and hs-TnT were independent predictors of OS, but not dFLC (Supplementary Appendix V). In summary, there are three salient findings of our study. First, 6-month EM in systemic AL amyloidosis is low at 7.5% in our cohort in the era of daratumumab-based frontline therapy, with substantial EM only limited to patients with stage IIIb disease at diagnosis. Second, hematologic and organ response rates were comparable to that in the ANDROMEDA trial despite more advanced organ involvement and a higher clone burden in our study. Third, higher baseline dFLC was not a predictor of OS in our dataset, likely reflecting high hematologic response rates of daratumumab-based combination therapies. The key drivers of EM in AL amyloidosis are delayed diagnosis that leads to advanced cardiac dysfunction at presentation and the toxic effect of circulating amyloidogenic light chain on cardiac tissue.4 A large study from the Mayo Clinic in the pre-daratumumab era showed >2 organ-involvement, Mayo 2004 stage III disease, and dFLC≥18 mg/dL to be the predictors of 6-month EM.1 However, not only has the heme-CR rate tripled with Dara-VCD compared to VCD, the time to heme-CR has substantially decreased to ~2 months.2 Owing to the S-shaped kinetics of fibril formation in amyloidosis, circulating precursor proteins lead to accelerated deposition of toxic amyloid aggregates in cardiac tissue, which necessitates a rapid reduction of the light chain to halt fibrillogenesis.4 Accordingly, the deep hematologic remissions with Dara-VCD have translated into early and deep cardiac organ responses. Compared to a historical median time to first cardiac response of 9 months,5 the median time to first cardiac response in our cohort was approximately 3 months. These results are also in line with ANDROMEDA trial, which showed an earlier decline in NT-proBNP in Dara-VCD compared to VCD arm.2 Importantly, higher dFLC was not a significant predictor of survival in our cohort. In a secondary analysis of ANDROMEDA trial, the heme-CR rate in Dara-VCD arm was similar in patients with dFLC<18 vs. dFLC≥18 mg/dL (63% vs. 55%, respectively), as was the cardiac response rate at 6 months in these two subgroups defined by dFLC (63% vs. 55%, respectively).6 However, in the VCD arm, heme-CR rate in patients with dFLC≥18 mg/dL was almost half of those with dFLC<18 (13% vs. 25%, respectively). These results call into question the external validity of Mayo 2012 staging system in the era of daratumumab-based frontline therapy. Large multicenter studies with mature follow-up are needed to accurately identify contemporary prognostic factors for survival and re-design the staging system. Our study has important implications for trial design in AL amyloidosis. Given a low incidence of mortality up until 2 years in all except stage IIIb patients, it will be difficult for clinical trials testing new agents on the backbone of Dara-VCD to demonstrate an early OS benefit unless patients with advanced cardiac involvement such as stage IIIb are included in the trial. The limitations of this study are a single center cohort with potential referral bias and a small sample size with a relatively short median follow-up of 2 years. In conclusion, daratumumab-based frontline therapy represents a major advance in the field of AL amyloidosis and may alter the natural history due to rapid and profound hematologic responses. While improved compared to historical data, patients with stage IIIb disease experience substantial EM and should be preferentially included in trials testing anti-amyloid strategies. Rajshekhar Chakraborty: Consulting/Advisory Board-Janssen, Sanofi Pasteur, and Adaptive Biotech. Suzanne Lentzsch: Chief Scientific Advisor and Shareholder of Caelum Biosciences; Consulting/Advisory Board-Janssen, Cellularity, Oncopeptide. M.M. is a consultant for Ossium Health Inc and a recipient of sponsored research agreements. The remaining authors do not have any relevant financial conflicts of interest. De-identified individual patient-level data can be made available to qualified investigators upon request to the corresponding author (subject to ethics committee approval). Supplementary Appendix I. Methods. Supplementary Appendix II. Baseline clinical and demographic characteristics of the study cohort (n = 84). Supplementary Appendix III: Kaplan–Meier curve for early mortality by Mayo 2012 stage at diagnosis. Supplementary Appendix IV: Hematologic and organ response rates in our cohort in comparison with that of ANDROMEDA trial. Supplementary Appendix V. Univariate and multivariate analysis to investigate predictors of survival. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.