Persistence of Minimal Residual Disease By Multiparameter Flow Cytometry Can Hinder Recovery of Organ Damage in Patients with AL Amyloidosis Otherwise in Complete Response

Introduction. In multiple myeloma, Minimal Residual Disease (MRD) demonstrated by multiparameter flow cytometry (MFC) identifies subjects with significantly shorter survival among those who attain complete response (CR). Patients with AL amyloidosis generally have a lower clonal burden than subjects with multiple myeloma, and, despite a higher rate of early death due to advanced organ damage, if they respond to therapy they have a better long term outcome and are less likely to relapse. The role of MRD in AL amyloidosis has not been assessed so far. In the present proof-of-concept study, we assessed the MRD by MFC in patients with AL amyloidosis who attained CR.

Methods. Complete response was defined as per current criteria (negative serum and urine immunofixation and normal free light chain ratio). Immunofixation was performed with both a commercial semi-automated method (Hydragel, Sebia, Lisses, France) and our home-made high-resolution method in all cases and had to be negative by both techniques. Circulating free light chains were measured by the Freelite assay. For flow cytometry studies bone marrow samples were processed following the Euro Flow Bulk Lysis Standard Operating Protocol and stained with the EuroFlow-IMF MM MRD panel (Tube 1: CD138BV421 / CD27BV510 / CD38FITC / CD56PE / CD45PerCP-Cy5.5 / CD19PE-Cy7 / CD117APC / CD81APC-C750, and Tube 2: identical to Tube 1 except for CyKappaAPC / CyLambdaAPC-C750). At least 5x10⁶ events were measured using a FACSCanto II (BD Biosciences, San Jose, USA) instrument. Data were analyzed using the Infinicyt software (version 1.7; Cytognos SL, Salamanca, Spain). Patients were identified as having residual disease if a discreet population of clonal plasma cells comprising ≥ 50 events was identified (0.001% limit of detection). Patients exposed to different treatment types, to different numbers of treatment lines, and at different points in time after achievement of CR were tested, in order to assess the possible impact of these variables on MRD. Differences in variables between patients with and without MRD were tested for significance by mid p Fisher exact test.

Results. Seventeen patients were tested. Six subjects were found to have relapsed at the time of MRD assessment with monoclonal components detectable at immunofixation and/or abnormal FLC ratio. All of them also had detectable MRD. Eleven patients satisfied current criteria for CR. All of them had renal and 6 (54%) had cardiac involvement at diagnosis. Two and 3 lines of therapy were required to achieve CR in 3 and 1 subjects, respectively. Median time to CR was 8 months (range 3-23 months). Two patients underwent autologous stem cell transplant and 9 received bortezomib. Three patients (50%) had achieved cardiac response and 5 (45%) renal response at the time of attainment of CR. The median time from CR to MRD evaluation was 20 months (range 6-36 months). Flow cytometry identified MRD in 5 patients (45%). A median of 1089 (range 600-2500) corresponding to 0.03% (range 0.02-0.15%) plasma cells with abnormal phenotype were detected in patients with MRD. No differences in organ involvement, cardiac and renal stage, type of therapy (with 1 transplanted patient in each group), number of treatments, and organ response at the time of CR was found between patients with and without MRD. However, a further improvement of cardiac function compared to the time of CR attainment was observed in all 4 evaluable patients without MRD and in none of the 2 patients with MRD, with borderline statistical significance (P=0.067). Compared to the time of CR achievement, renal response was obtained in 5 subjects without MRD (83%) and in 1 (20%) with MRD (P=0.069). Overall, further improvement of cardiac or renal function after CR was significantly associated with absence of MRD (P=0.002). Interestingly, 1 patient with MRD had otherwise unexplained increase in proteinuria while still in CR, and anti-clone therapy was started based on MRD results.

Conclusion. This proof-of-concept study indicates that almost 50% of patients with AL amyloidosis satisfying current criteria for CR have MRD detectable by MFC. MRD makes further organ improvement less likely and can explain organ progression. A validation study in a larger sample is ongoing at our center. The possible impact of MRD should be considered in trials aiming at increasing organ response rate in patients achieving CR.