Identifying Early Suboptimal Hematological Response in Patients with AL Amyloidosis Treated with Bortezomib-Based Chemotherapy

Background: Achievement of rapid and deep hematological responses is critical in the treatment of AL amyloidosis, particularly for those patients with advanced cardiac involvement. The timing and depth of FLC reduction that defines suboptimal response and the need to switch to second-line therapy is uncertain. We aimed to determine the impact of hematological response after 2 months of bortezomib-based therapy on subsequent best hematological and organ responses and survival in patients with and without advanced cardiac disease.

Methods: The cohort was derived from three amyloidosis centres in Australia (Brisbane, Perth and Sydney). Patients were identified with a diagnosis of AL amyloidosis, symptomatic organ involvement and initial treatment with a bortezomib-based regimen. The impact of hematological response after 2 months of treatment on subsequent best hematological, cardiac and renal response was assessed using Fisher’s exact test. The impact on OS was assessed by landmark analysis using log-rank or Cox regression analysis. Organ involvement and response criteria were as per International Society of Amyloidosis recommendations.

Results: 190 patients with AL amyloidosis were identified: Median age was 66 yrs and 38% were male. 74% of cases were lambda restricted, the median dFLC was 170mg/L, the median bone marrow plasmacytosis was 11% and 7% had symptomatic myeloma. Mayo 2004 cardiac stage was: 1 (14%), 2 (47%), 3A (23%) and 3B (16%). 69% had renal involvement with a median eGFR of 70mls/min and median proteinuria of 1.9g/d , with 7% dialysis dependent at presentation. Liver involvement was present in 16% and peripheral neuropathy in 27%. Bortezomib regimens were: VCD (75%), VCD+daratumumab (13%), MDV (5%), VD (5%) and VRD (2%), with a median of 6 cycles delivered (range 1-16).

Median OS was 6.2 years which was adversely associated with more advanced cardiac disease: stage 1 (not reached), stage 2 (90 months), stage 3A (74 months) and stage 3B (24 months)(p=0.0001).

Hematological response after 2 months of bortezomib-based induction was: CR (17%), VGPR (39%), PR (19%) and less than PR (13%). 9% of patients died prior to the 2 month response assessment and 5 patients (2%) had missing data. Failure to achieve PR by 2 months was associated with a low chance of achieving deep responses, with only 16% going on to achieve VGPR or better (p<0.001). Similarly, failure to achieve PR by 2 months predicted a low likelihood of improving organ function with less chance of cardiac (6% vs 56%, p=0.001) and renal (30% vs 46%, p=0.097) responses. In a landmark analysis, failure to achieve PR after 2 months predicted worse OS (median 31 vs 88 months, p=0.001), a finding confirmed in multivariate analysis including cardiac stage and age.

Failure to achieve VGPR by 2 months was also associated with a lower likelihood of achieving subsequent deep hematological responses and organ responses but 38% still managed to achieve VGPR or better by the completion of therapy and 31% a cardiac response. When analysed among patients with Mayo Stage 3A or 3B cardiac disease (i.e. advanced cardiac involvement), however, failure to achieve VGPR by 2 months was associated with reduced likelihood of cardiac response (25% vs 61%, p<0.001) and worse overall survival (median OS 30 vs 72 months, p<0.001). In contrast, among patients with cardiac stage 1 or 2 disease, failure to achieve VGPR by 2 months was associated with a trend to reduced cardiac responses (35% vs 58%, p=0.083) but not overall survival (median 78 vs 96 months, p=0.224)

Conclusion: Failure to achieve hematological PR after 2 months of bortezomib-based therapy defines a high chance of therapeutic failure in AL amyloidosis and such patients should be switched to alternate salvage therapy. In patients with advanced cardiac stage 3A and 3B disease rapid, deep responses are required and failure to achieve VGPR after 2 cycles also predicts poor cardiac response and survival and defines a suboptimal response. Failure to achieve VGPR by 2 months in patients with cardiac stage 1 or 2 disease, however, was not associated with therapeutic futility.