Role of Previous Thalidomide Administration on Final Outcome for 72 Patients with Relapsed Multiple Myeloma (MM) Treated with Bortezomib–Based Therapy

New drugs such as thalidomide, bortezomib and lenalidomide have expanded the therapeutic options for MM while improving outcome in both young and elderly patients. However, the best novel agents sequence in the therapeutic strategy for MM is still not definitely delineated as relapsed MM right after first line thalidomide therapy seems to be more resistant jeopardizing final outcome as per overall survival while still making questionable when to administer it, either at the beginning or later during the course of the disease. We analyzed 72 relapsed MM patients who were enrolled in two salvage study protocols which included bortezomib, dexamethasone and chemotherapy (Offidani et al, ASH 2007 and EHA 2008) and who had been treated with thalidomide first (18 patients) or subsequently (30 patients) or not treated at all with thalidomide (24 patients). We compared these three groups of patients in terms of response rate, post-relapse PFS and post-relapse OS with the aim to assess the role of previous administration of thalidomide on final outcome in this patient population. Median age for the 72 patients was 65 years (range 31–82); ISS stage 2–3 assessed in 51% of patients and unfavourable cytogenetics in 42%. Thirty four patients had been rescued in first relapse, 19 in second and 19 in third or subsequent relapse. Median disease history was 34 months (range 8–173). Forty four patients relapsed after high-dose therapy and autologous stem cell transplantation. The 48 patients were previously treated with thalidomide a median time of 8 months (range 4–48 months). VGPR or better response rate in the groups of patients treated with thalidomide in first line, second or subsequent line or never treated with thalidomide were 44%, 42% (p=0.795) and 79% (p=0.003; p=0.002), respectively.. Multivariate stepwise regression analysis selected only previous thalidomide treatment (OR=1.9; 95%CI=1.5–2.4; p=0.024) as factors affecting response whereas age, previous therapy lines, previous remission duration, previous transplant, previous disease history, sCRP, ISS stage and cytogenetics were not significantly associated to response. In the same groups post-relapse PFS was 9 months, 14 months (p=0.308) and not reached (p=0.018; p=0.055) while post-relapse 2 years OS was 51%, 50% (p=0.564) and 72% (p=0.074; p=0.135). Cox regression analysis showed that the presence of ISS 2–3 (p=0.010), previous thalidomide administration (p=0.052), and response < VGPR (p<0.0001) translated in significantly poorer post-relapse PFS and OS. Stepwise Cox regression analysis selected only response < VGPR as factor significantly associated to poor post-relapse OS (2yrs OS 26% vs 83%; p<0.0001; HR=7.0 95CI=2.6–18.9). In the group of 48 patients previously treated with thalidomide, time on thalidomide (cut-offs 4, 8, 12, 24 months) did not affect post-relapse PFS and OS. These data suggest that, even in relapse, response to salvage therapy is the most powerful predictor of PFS and OS. Unfortunately, previous thalidomide administration, particularly as front-line therapy, apart from time on treatment, significantly decreases VGPR and consequently post-relapse PFS. This lead to a final outcome in terms of OS not different, if worse, in those patients previously treated vs those never treated with thalidomide. These and other data (Barlogie et al NEJM 2006, Palumbo et al, JCO 2008), strongly questioned the use of front-line thalidomide as the best therapeutic strategy for patients with MM since alternatives are now possible.