Cure fractions (CF) modeled from event-free survival and complete response duration plots in total therapy (TT) trials for newly diagnosed multiple myeloma (MM).

Abstract

8119 Background: We recently reported that survival outcomes in TT protocols were compatible with progressively increasing CF with the transition from TT1 initiated in 1989 to TT2 started in 1996 and TT3 begun in 2003 (Barlogie et al, ASH 2009). Methods: Mathematical models were applied to fit event-free survival (EFS) and CR duration (CRD) plots with a cure plateau and exponentially declining survival. Cox regression analysis was applied to determine odds ratio (OR) values of CF. Results: In TT1,CF- EFS was 9.4% (p<0.0001) and CF-CRD 14.6% among the 94 CR patients (p=0.0003), with higher values in the absence of cytogenetic abnormalities (CA) (CF-EFS: 11.7% v 3.5%; CF-CRD: 17.2% v 9.6%). In TT2, CF-EFS was 16.5% (p=0.0008) and CF-CRD was 27.6% among 351 attaining CR (p=0.001). CA affected CF-CRD (no CA: 27.8%, p=0.06; CA: 15.8%, p=0.06) but not CF-EFS. In the context of GEP, significant CF-EFS (14.3%, p=0.01) and CF-CRD (26.6%, p=0.03) were observed only in the high-risk setting. With regard to thalidomide, CF-EFS was superior when patients were randomized to thalidomide (23.2%, p=0.003 v 9.9%, p=0.121), which also applied to CF-CRD (31.4%, p=0.009 v 22.4%, p=0.07). In TT3, CF-EFS was 52.7% (p<0.0001) and CF-CRD among 188 CR patients 65.5% (p=0.02) which, when examined in context of GEP, only applied to low-risk myeloma (CF-EFS=61.1%, p<0.0001; CF-CRD=87.6%, p=0.0001). Multivariate analysis of protocols and baseline variables identified significant odds ratio (OR) values for cure in case of TT2 plus thalidomide (OR=0.296, p=0.03), TT3 (OR=0.079, p<0.0001), whereas cure-adverse variables were CA (OR=2.411, p=0.03) and albumin <3.5g/dL (OR=3.374, p=0.03). Conclusions: This is the first report on multivariate modeling for CF in MM, revealing major cure contributions from thalidomide in TT2 and added bortezomib in TT3 whereas, among the non-cured patients, CA and low albumin increased the risk of recurrence and death. No significant financial relationships to disclose.