The use of bortezomib (B), a first-in-class proteasome inhibitor, administered with or without dexamethasone (D) for treatment of systemic AL amyloidosis, was reported in two small preliminary series to result in high rates of hematologic (HR) and organ (OR) response. We report here the efficacy and toxicity of bortezomib among 94 patients with AL amyloidosis attending three European amyloidosis centers. Consensus criteria were used for assessment of amyloidotic organ involvement and treatment response. Ninety-four patients treated with B±D were identified in the three centers’ databases. Median age at start of treatment was 62 years. Eighty-one percent had received prior chemotherapy (median 2 lines, range 1–6), including high dose melphalan in 12% and a thalidomide combination in 55%; 69% had clonal disease that had been completely refractory. The median number of amyloidotic organs was 2 (range 1–5); the heart was involved in 73% (62% had symptomatic CHF) and the kidneys in 75% of cases. Median serum creatinine was 1.3 mg/dL (range 0.4–8.9), and 12% were dialysis-dependent. The median serum concentration of amyloidogenic free light chains (FLC) was 183 mg/L (range 9–9464); serum NT-proBNP was >332 ng/L and >1266 ng/L in 81% and 64% patients respectively. B was administered according to standard schedule in 80% and weekly in 20%. Median dose of D per cycle was 80 mg; 10% of patients received B without D. Median number of treatment cycles was 4 (range 1–8). HR was achieved in 71% of patients, including CR in 25%; 67% of responding patients did so by the 3rd month of treatment. HR was not related to prior treatment response although in untreated patients CR rates were higher (47% vs. 20%, p=0.074). In multivariate analysis HR was independently associated with age <65 years (HR 4.225, 95% CI 1.4–13, p=0.012) and biweekly administration of B (HR 5.63, 95% CI 1.3–24, p=0.02). HR rates with B+D were 17% higher than with B alone (p=0.074). A response of at least one amyloidotic organ was observed among 30% of patients: heart in 22%, renal in 17% and liver in 19%. OR was strongly associated with HR: 27 (43%) patients who had HR vs. 1 (3%) who did not met the criteria of HR, had response in at least one affected organ. Hematologic CR was associated with higher OR rates (64% vs. 32% for PR, p=0.001). In multivariate analysis higher rates of OR were associated with HR (HR 18.8, 95% CI 2.36–142, p=0.006); HR was strongly associated with cardiac response (46% vs.0%, p<0.001) and reduction 30% of NT-proBNP (84% vs. 27%, p<0.001). Estimated 1 and 2-year survival rates were 78% and 71% respectively with median follow-up of 12 months (range 0.5–48) from commencement of B. In multivariate analysis improved survival was independently associated with HR (HR: 0.15, 95% CI 0.041–0.551, p=0.004) and reduction of NT-proBNP by 30% (HR 0.10, 95% CI 0.025–0.424, p=0.001). Twelve month survival for patients who achieved both HR and an NT-proBNP response was 97%, compared to 83% for those who achieved either and only 24% for patients who achieved neither. On intent to treat basis, 50% of patients have progressed (hematologic in 31% and/or organ progression in 30%). Median time to any progression is 13.3 months (95% CI 9.1–17.5). Common toxicities included peripheral sensory neuropathy (40%, G2 in 30%), neuropathic pain (14%, G2 in 9%), hypotension/orthostasis (36%, G3 in 13%), and peripheral edema (33%, G3 in 12%). Grade 3 or 4 toxicity occurred in a total of 29% of patients; five (5%) deaths occurred within the first two months of treatment. These findings extend our preliminary observations and confirm that bortezomib-based treatment has substantial efficacy and manageable toxicity in patients with relapsed and refractory AL amyloidosis. Biweekly administration of bortezomib and the addition of dexamethasone may be related with higher rates of hematologic response. Reduction of NT-proBNP by >30% following B treatment was associated with survival benefit.

Disclosures: Dimopoulos:Ortho-Biotech: Honoraria. Off Label Use: Bortezomib in AL Amyloidosis.

Author notes

Corresponding author

Sign in via your Institution