First line treatment of systemic non IgM AL amyloidosis is well defined in the recommendations of the French National Reference Center for AL amyloidosis. Melphalan or cyclophosphamide at standard dose with dexamethasone in combination with bortezomib are currently recommended. The objective is to rapidly obtain an hematological response, modulating treatment according to severity of heart involvement and hematologic response. At relapse there are no such precise guidelines. If the initial treatment has been effective, it may be chosen in case of late relapse occurring after at least one year. Combination of other drugs effective in plasma cell dyscrasias might be proposed, but studies on treatment in the relapse setting are scarce and a consensus remains to be established.

Here we report a multicenter retrospective study assessing treatment at relapse and evaluating their impact on hematologic response, progression-free survival (PFS), overall survival, organ responses and toxicity.

We included 84 patients who received, from 2006 to 2014, as first line treatment, a non-intensive chemotherapy in 15 French hospital centers. At diagnosis, 47 (56 %), 51 (61%) and 16 (19%) patients respectively presented with heart, kidney, and neurological involvement. Twenty three patients (27%) had a severe heart disease with a Mayo Clinic stage III. The median follow-up of the cohort was 58 months (9-135). The hematological response was assessed at 3 months based on free light chain measurement, according to the international consensus. At relapse, 29 patients (34,5%) received a bortezomib, cyclophosphamide and dexamethasone combination (VCD) resulting in 79% of very good partial response (VGPR) or better and 7% of partial response (PR), 18 patients (21%) had an organ response, and with a median follow up of 70,5 months (9,03-135) the estimated overall survival at 3 years was 79,4%. Twenty six patients (31%) experienced grade ≥ 3 toxicities that were peripheral neuropathy in 5 patients. Seventeen patients (21%) received a combination of lenalidomide and dexamethasone (RD), resulting in 17% of VGPR or better and 35% of PR, 12 % of organ response with an estimated overall survival at 3 years of 94% with a median follow up of 46,5 months (20,99-122,97). Four patients (23%) had a grade ≥ 3 toxicity, mainly hematologic. Five patients (6%) received a combination of bortezomib with lenalidomide and dexamethasone (VRD) resulting in VGPR or better in 4 patients and PR in 1 patients, 4 patients had an organ response with all patients being alive at 3 years. Grade ≥ 3 toxicity was observed in 1 patient. Regarding these 3 combinations, the median PFS were respectively 22 months for VCD, 8 months for RD (p = 0, 0012) and 17.3 months for VRD. At relapse 13 other therapeutic combinations were used. Other IMIDs (thalidomide or pomalidomide), melphalan, bendamustine or daratumumab were given to 4, 8, 6 and 5 patients respectively resulting in 1, 4, 3 and 1 of VGPR or better hematologic response. Bendamustine containing regimens were associated in grade ≥ 3 toxicity in 5 patients (83%). Sixty-seven patients (79%) were still alive at 3 years.

In conclusion, there is currently no consensus on the best treatment for AL amyloidosis patients in relapse. This study shows the different regimens used in France and their effectiveness in relapse. As usual in this disease the survival of relapsing patients is good with almost 80% of patients alive at 3 years. Bortezomib remains an important molecule in relapse. Lenalidomide alone with dexamethasone seems to be less effective to obtain a high rate of hematologic response. VRD has been used in few patients, we found no cumulative toxicity with the combination of proteasome inhibitor and lenalidomide with a very interesting response rate, PFS, OS and organ responses. VRD should be tested more systematically in relapsing or refractory patients. Finally, the role of daratumumab and new proteasome inhibitors remains to be defined.


Harel:janssen: Consultancy; takeda: Consultancy; amgen: Consultancy.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution