Patterns of Relapse after Upfront Therapy in AL Amyloidosis

Introduction. In the last few years a major international effort allowed establishing baseline staging systems and response criteria in AL amyloidosis. However, we still lack validated progression criteria. This is acutely relevant for reporting progression free survival in clinical trials and because novel agents are first tested in the relapsed/refractory setting. We studied the patterns of relapse and outcome of 259 consecutive patients with AL amyloidosis who attained hematologic response after upfront therapy.

Methods. The prospectively maintained database of the Pavia Amyloidosis Research and Treatment Center, including 1069 treatment-naïve patients with AL amyloidosis diagnosed between 2004 and 2015, was searched for patients who achieved at least partial response (PR) after upfront treatment, and did not require second-line therapy for at least 6 months. Relapse was defined as the initiation of second-line therapy.

Results. A total of 259 consecutive patients were identified. Their median age was 66 years (range 39-84 years). The heart was involved in 184 patients (71%) and the kidney in 179 (69%) of subjects. Cardiac stage was I in 29% of patients, II in 46%, IIIa in 18%, and IIIb in 7%. Renal stage was I in 48% of patients, II in 39%, and III in 13%. Upfront treatment was melphalan and dexamethasone (MDex) in 129 patients (50%), cyclophosphamide, bortezomib and dexamethasone (CyBorD) in 71 (27%), bortezomib plus MDex (BMDex) in 46 (18%), bortezomib and dexamethasone in 10 (4%) in 3 subjects (1%) with IgM-AL amyloidosis. Best hematologic response after upfront therapy was CR in 82 patients (32%), very good partial response (VGPR) in 134 (52%), and partial response (PR) in 43 (16%). Cardiac response was achieved in 38% of patients and renal response in 27%. All patients in whom treatment was discontinued in PR had also achieved organ response. After a median follow-up of living patients of 41 months, 92 patients (35%) needed second line therapy (relapsed). At time of relapse, dFLC (difference between amyloidogenic - involved - and uninvolved free light chains) increased in 86 patients (93%), reaching a median of 60 mg/L [interquartile (IQR) range 26-108 mg/L], corresponding to 41% (IQR 19-84%) of baseline value, with 44 patients (48%) whose dFLC at relapse remained below the threshold of "measurable disease" (50 mg/L) generally required for enrollment in clinical trials. Progression of NT-proBNP was observed in 20 patients, and was preceded by a dFLC increase (to a median of 50% of baseline value; IQR: 20-83%) in 20 (91%). A >50% increase in proteinuria was observed in 24 patients (26%), 2 of whom maintained VGPR with no increase in dFLC. In 11 patients (12%) estimated glomerular filtration rate decreased by >25%, with dFLC increase in all cases. Overall, 70 patients (76%) had signs of cardiac or renal progression at time of second line therapy initiation. Median time to second line therapy was 57 months. The variables significantly associated with prolonged time to relapse were baseline dFLC <60 mg/L (11% vs 32% relapsing at 3 years, P=0.004), quality of hematologic response (21%, 35%, and 54% relapsing at 3 years, among patients in CR, VGPR, and PR, respectively; CR vs. VGPR: P=0.003, VGPR vs. PR: P=0.028), and treatment with BMDex [27% relapsing at 3 years, compared with 38% with CyBorD (P=0.014), and 39% with MDex (P=0.023)]. A multivariate analysis stratified by achievement of CR showed that treatment with BMDex independently prolonged time to second line therapy (HR=0.28, P=0.016), while dFLC <60 at diagnosis retained borderline significance (HR=0.15, P=0.067). Thirty-two patients died, and median survival after relapse was 58 months. Progression of NT-proBNP was associated with shorter survival (median 17 vs. 62 months, P=0.001).

Conclusion. Low dFLC burden at diagnosis and treatment with the combination of bortezomib, melphalan and dexamethasone are associated with more durable responses. An increase in NT-proBNP should not be awaited to start treatment, because it is associated with poorer survival. However, cardiac progression is preceded by a dFLC increase in >90% of cases. Increases in dFLC (>20%) should trigger second-line therapy in patients with cardiac involvement.