Reduced Early Mortality with Daratumumab-Based Frontline Therapy in Systemic AL Amyloidosis- Experience from the Columbia Amyloidosis Multidisciplinary Program

Introduction: Despite advances in treatmentof systemic light chain (AL) amyloidosis, early mortality (EM) has been substantial up until recently, with a 6-month EM rate of approximately 25% in the era preceding daratumumab (Muchtar et al. Blood. 2017).Daratumumab in combination with bortezomib, cyclophosphamide and dexamethasone (Dara-VCD) is the first FDA-approved therapy for systemic AL amyloidosis based on the ANDROMEDA trial with an unprecedented hematologic complete response (Heme-CR) rate ≈ 50%. However, data on hematologic and cardiac organ response rate and estimated early mortality (EM) in different stages since routine introduction of daratumumab-based frontline therapy are lacking.

Methods: We performed a retrospective cohort study of all consecutive patients treated with daratumumab-based combination therapy at Columbia Amyloidosis Multidisciplinary Program (CAMP). Hematologic and cardiac organ response rates are reported on an intention-to-treat (ITT) basis, with patients who died prior to response evaluation or those with missing data classified as non-responders. Time-to-event analyses were performed by the Kaplan-Meier method.

Results: Eighty-three consecutive patients treated with daratumumab-based frontline therapy were included in the analysis. The baseline characteristics of our cohort in comparison with Dara-VCD arm of ANDROMEDA is shown in Table 1. The median age at diagnosis was 68 years (range, 40-91), with 65% being males and 13 patients (16%) of Black/African-American race. The median dFLC (involved - uninvolved) was 21.9 mg/dl, with 61% of patients having baseline dFLC>18 mg/dl. Mayo 2004 stage IIIa/IIIb comprised 65% of our cohort compared to 37% of ANDROMEDA trial. Thirty-three patients (44%) had NYHA Class III/IV symptoms. The daratumumab-based frontline regimens used were Dara-VCD (n=78), Dara-VD (n=2), Dara-D (n=2), and Dara-Ixazomib-D (n=1). High-dose melphalan followed by autologous hematopoietic cell transplantation (HDM-AHCT) as a part of frontline therapy was performed in 11 patients (13.4%), with melphalan dose being 140 mg/m ² in 8, 200 mg/m2 in 2, and 100 mg/m ² in 1 patient.

Hematologic response was evaluable in 78 patients (i.e. dFLC>2 mg/dl at diagnosis). On an ITT basis, Heme-CR was achieved by 39 patients (50%), and very good partial response or better (Heme ≥VGPR) by 57 patients (73.1%). The median time to Heme-CR from treatment initiation was 2.7 months (range, 0.42-13.97). Among patients evaluable for cardiac organ response (i.e. NT-proBNP≥650 pg/ml; n=64), 35 (58.3%) achieved an organ response. The median time to initial cardiac response (>30% reduction in NT pro-BNP with absolute reduction of >300 pg/ml) was 2.95 months (range, 0.43-10.43). Figure 1 demonstrates comparable hematologic and numerically higher cardiac organ response rate in our cohort vs Dara-VCD arm of ANDROMEDA trial.

Next, we estimated the EM rate. The estimated EM at 6 months in our cohort was 7.6% (95% CI, 3.4-15.9). 6-month EM in patients with Mayo 2004 stage IIIb was 22.2% (95% CI, 9.5-43.6) vs 1.96% (95% CI, 0.27-12.65) in those with stage I-IIIa ( p<0.001) [Figure 2]. By Mayo 2012 staging, the estimated 6-month EM rate in patients with stage I, II, III, and IV disease was 0% (95% CI, 0-0), 0% (95% CI, 0-0), 9.6% (95% CI, 2.4-31.3), and 12.9% (95% CI, 4.9-29.8) respectively ( p=0.0425). Among patients who underwent upfront HDM-AHCT (n=11), there was no transplant-related mortality at 100 days. On evaluating the prognostic impact of individual factors in Mayo 2012 staging (NT-proBNP, Troponin, and dFLC) on OS, dFLC>18 mg/dl was no longer a significant adverse prognostic factor for survival in both univariate and multivariate analysis. Updated data will be presented at the meeting.

Conclusion: Despite a sicker population with more advanced cardiac involvement compared to ANDROMEDA trial, daratumumab-based frontline induction therapy led to similar Heme-CR rate and favorable cardiac organ response rate in our cohort. Additionally, estimated 6-month EM has decreased by approximately 3-fold compared to historical data, with dramatic reduction in Mayo 2004 stage I-IIIa and continued substantial EM in stage IIIb disease. Finally, as clone-directed therapy has significantly improved, light chain burden at baseline may not be a relevant prognostic factor in contemporary era and should be scrutinized in larger datasets to re-design the staging system.