Whereas the addition of thalidomide in TT2 improved CR rate and EFS, OS and, not widely appreciated, the 4-year continuous CR estimates were similar in the two study arms (see ) (Barlogie et al, 2006a
). With TT3, we observed that 92% of patients entering CR sustained such remission at 2 years, whereas with TT2 and TT1 relapses occurred earlier after the onset of CR (Barlogie et al, 1997
). This observation suggested that the addition of bortezomib had ushered in a fundamental shift in the durability of response and prompted a formal analysis of clinical outcomes of patients enrolled on TT3 and TT2, as presented here. Except for higher proportions of patients with advanced age, anaemia, B2M elevation and hypoalbuminaemia and thus higher ISS stage in TT3, the baseline characteristics of patients in the two protocols were similar (see ).
As a consequence of shortened induction and consolidation phases and an effort to complete the second autotransplantation within 2–3 months of the first high-dose therapy cycle, the progression through the protocol phases was indeed faster and the proportions of patients completing the intended therapies significantly higher in TT3 than in TT2 (see ). Thus, as intended, the proportion of patients suffering a relapse was significantly reduced for each of the protocol phases of TT3 vs. TT2 (see ). Importantly, despite similar times to and levels of CR in TT3 and TT2 plus thalidomide, CR duration was significantly superior in TT3 to TT2 regardless of study arm (see ). TT3 also affected superior EFS in comparison with both arms of TT2, whereas the follow-up is too short to comment on OS (see ). In a pair-mate analysis matching TT3 with TT2 patients on the four key adverse variables identified in TT2, TT3 was superior to TT2 in terms of CR duration and EFS, with a strong trend for OS (see ).
In the context of the recently reported GEP-defined risk groups (Shaughnessy et al, 2007
), it was remarkable to note superiority already for TT3 vs. TT2 in EFS in both low- and high-risk groups (see ), with a strong trend emerging for OS in the high-risk subgroup (see ). Remarkably, superior Kaplan–Meier plots were observed for CR duration for the low-risk subgroups of TT3 (see ), whereas no differences were observed between the two TT2 arms. These observations were supported by the results of multivariate analyses that included protocol assignment, showing superior EFS and CR duration with TT3 versus
both arms of TT2 (see ). In the context of GEP data, the GEP-defined high-risk category was the dominant baseline feature for both OS and EFS along with the presence of CA and LDH elevation, again with a significant contribution of TT3 for EFS and CR duration (see ). When examined in the context of the 7-GEP subgroups (Zhan et al, 2006
), patients with FGFR3/MMSET
disease benefited significantly from TT3 vs. TT2 in terms of all three endpoints examined (CR duration, EFS, OS) whereas, again, no difference existed between the TT2 arms (see ).
Finally, to address the issue of whether TT3’s superior performance was related to the addition of bortezomib or the greater protocol compliance in case of TT3, we re-examined OS and EFS in the subgroup of patients entering the maintenance phases of the protocols within 2 years from initiation of therapy. EFS was indeed superior with TT3 vs. TT2, with a trend observed in favour of TT3 for OS (see ), suggesting an important contribution of bortezomib to the improved outcome of patients receiving TT3.
We conclude from this historical comparison that, unlike with the effects of the addition of thalidomide in TT2, bortezomib in TT3 appeared to significantly prolong CR duration, currently limited to the low-risk subgroup, whereas the traditionally unfavourable FGFR3/MMSET subgroup, which did not benefit from the addition of thalidomide in TT2, experienced significant prolongations of all three clinical endpoints examined, including OS, with TT3.