Clinical Significance of Measurable Residual Disease (MRD) in Light-Chain (AL) Amyloidosis

Background: Novel treatments targeting the plasma cell clone are increasing the rate of hematologic complete response (CR) and improving outcomes in AL amyloidosis. In multiple myeloma (MM), MRD negativity is considered the new CR and pivoted from a prognostic factor into a treatment endpoint and a surrogate for accelerated drug approval. By contrast, the role of MRD in AL amyloidosis remains uncertain and, accordingly, it is not routinely performed in clinical trials and practice.

Aim: Investigate the clinical significance of MRD assessment in patients with AL amyloidosis.

Methods: This multicenter study included 319 patients with AL amyloidosis in whom MRD assessment was performed at the time of suspected hematological CR. In total, 337 and 47 MRD studies were respectively performed in bone marrow (BM) and peripheral blood (PB) samples using EuroFlow next-generation flow cytometry. Time to next treatment (TTNT) and progression-free survival (PFS) were measured from the date of MRD assessment. A new line of treatment and/or death from any cause were considered events in PFS analysis. Median follow up of the series was 3 years.

Results

MRD negativity at one or more time points was observed in BM aspirates from 137 of the 319 patients (43%). Of the 47 patients with paired BM and PB samples, only 22 (47%) showed concordant MRD results (20 BM-/PB-, 2 BM+/PB+) and in 25 cases there was positive MRD in BM while negative in PB. The negative predictive value of PB-MRD considering BM-MRD as the reference was 44%, which is lower than that observed in MM (>60%).

The median percentage of detectable MRD in BM was 0.008%, and MRD levels ≥10^-4, ≥10^-5 to <10^-4 MRD, and ≥2x10^-6 to <10^-5 were respectively detected in 49%, 32% and 19% of MRD positive samples. There were no differences in TTNT and PFS among patients with persistent MRD at levels ≥10^-4, ≥10^-5 to <10^-4 and ≥2x10^-6 to <10^-5 (P ≥ .3). Accordingly, MRD negativity was defined based on the 2x10^-6 threshold.

Median TTNT and PFS were not reached (NR) in MRD negative patients, whereas the respective medians in MRD positive patients were 57 and 49 months (P < .001). The 3 years rates of TTNT and PFS were 88% and 81% in MRD negative vs 68% and 63% in MRD positive patients. Achieving MRD negativity was associated with 59% reduction in the risk of progression and/or death (hazard ratio [HR] for PFS: 0.4, 95% confidence interval [CI] 0.3 – 0.6, P < .001). Of note, the prognostic value of MRD response was reproduced in different centers (Athens, Pavia, Ostrava, PETHEMA).

We next investigated the impact of MRD response in risk subgroups according to the Mayo 2004 staging with the European modification of 2013 into stages 1, 2, 3a and 3b (28%, 38%, 27%, and 7% of cases, respectively) Median PFS of MRD negative vs positive patients with stage 1 was NR vs 61 months (P = .007), stage 2 was NR vs 37 months (P = .03) and combined stage 3a/3b was 65 vs 50 months (P = .15). The favorable outcome of patients staged 3a/3b is likely due to the selection criteria for MRD assessment (ie, hematological CR). A multivariate analysis of PFS showed that both staging at diagnosis (HR: 0.6, 95% CI 0.4 – 1.0, P = .05) and MRD status (HR: 0.4, 95% CI 0.3 – 0.7, P < .001) were independent prognostic factors. In addition, there were trends between MRD and organ response.

An autologous transplant was performed in 27% of patients. Bortezomib and daratumumab were administered to 92% and 32% of patients, respectively. MRD assessment was prognostic in different treatment scenarios. MRD negativity was associated with prolonged PFS in transplanted patients (median 72 vs 61 months, P = .07) and those who were not (median NR vs 44.5 months, P < .001), in patients treated with bortezomib (median NR vs 50 months, P < .001) and those who were not (median NR vs 48.5 months, P = .05), as well as in patients treated with daratumumab (median NR vs 37 months, P = .07) and those who were not (median NR vs 57 months, P < .001).

Conclusions: To our knowledge this is the largest MRD study in AL amyloidosis. It shows that MRD assessment helps redefining hematological CR through the identification of MRD positive patients who experience inferior TTNT and PFS. MRD assessment also helps redefining patients risk determined at diagnosis. Because the prognostic value of MRD was observed in different treatment scenarios, MRD negativity could become a therapeutic endpoint and should be routinely performed in AL amyloidosis who attain hematological CR.