Prolonged Follow-up Confirmed a Role for Upfront Tandem Auto-Allo Transplant in Multiple Myeloma Also in the Era of New Drugs

Introduction: Before the introduction of new drugs, we designed a trial where treatment of newly diagnosed multiple myeloma (MM) patients with double autografts or autograft followed by nonmyeloablative allograft was based on the presence or absence of HLA identical siblings (Bruno B et al. N Engl J Med 2007) We reported an update with special focus on long term outcomes.

Methods: From September 1998 to July 2004, 162 consecutive patients with newly diagnosed MM up to the age of 65 years and at least one sibling were enrolled at 5 Italian centers, and divided into 2 groups: donor (N=80) vs no donor (N=82). First-line treatments consisted of a cytoreductive autograft followed by a HLA identical sibling nonmyeloablative allograft or a second melphalan-based autograft (N=58 and N=46, respectively, completed the protocol).

Results: Median follow-up was 12.3 years (range, 7.7-15.3) from allograft, and 12.1 years (range, 10.5-15.4) from second autograft. The 5-year cumulative incidence of non-relapse mortality was 17.2% (95%CI: 7.4 to 27.1) in the allograft arm and 4.3% (95%CI:0 to 10.3) in the autograft arm. One of the main concern post allograft is the impact of chronic graft-versus-host disease (cGVHD): in our setting its 2-year cumulative incidence was 67.2% (95%CI: 54.9 to 79.5). We also evaluated the cumulative incidence of immunesuppression discontinuation in patients with cGVHD, considering both death and relapse as competing events: 26.8% of cGVHD patients (95%CI: 13 to 40.6) at 24 months and 39% (95%CI:23.6 to 54.4) at 60 months were alive and without therapy.

Median overall survival (OS) and progression-free survival (PFS) from second transplant were 137 and 43 months in the allograft arm and 46 and 18 months in the autograft one (p=0.006 and p=0.001, respectively).

In the allograft arm, 33 out of 58 patients relapsed at least once, and first salvage treatments included donor lymphocyte infusion (DLI, N=13), thalidomide (N=10) and bortezomib (N=8). Of note, 2 patients lost complete remission status but did not require further therapy. Nineteen out of 33 patients required a second post-transplant salvage treatment: 1 received chemotherapy, 1 DLI, 1 received debulking treatment with bortezomib followed by a second allograft, and 16 patients were treated with new drugs containing regimens. In the autograft arm, 36/46 patients relapsed and received salvage treatments consisting of: allograft (N=1), 3^rd autograft prepared with a new-drugs containing regimen (N=6), thalidomide (N=17), bortezomib (N=7), lenalidomide (N=1), chemotherapy alone (N=4). Among these, 19 required a third-line treatment: 1 received an allograft, and 18 a regimen containing new drugs. Median OS from 1^st relapse was 89.8 months (95%CI: 33.3 to n.r.) in the allograft arm vs 23.5 months (95%CI: 12.5 to 50.5) in the autograft (p=0.009).

Conclusions: Our update showed that more then a third of patients developing cGVHD were relapse-free and cGVHD-free at 5-years post-transplant and that the advantage in OS in the allograft arm is maintained also after relapse, suggesting a synergism between graft-vs-myeloma effect and new agents. Upfront allograft in MM remains a matter of debate, and it should be performed only within clinical trials. The main limit of the present study was the lack of novel agents as part of the pre-transplant approach, nevertheless our results suggested that allograft may have a role, and it might be considered in young patients with high-risk features such as del [13], t(4;14), del(17p), and t(14;16), who remain at poor prognosis even in the era of new drugs.