Curing myeloma at last: defining criteria and providing the evidence

Abstract

Does the dogma that multiple myeloma is incurable still hold?. The genomic chaos and resulting resistance to apoptosis of myeloma, long considered an obstacle to cure, formed the basis of Total Therapy (TT) program. The TT approach uses all myeloma-active drugs upfront to target drug-resistant subclones during initial treatment to prevent later relapse. Long-term follow-up of 1202 patients (TT1: n = 231, median follow-up: 21 years; TT2: 668, median follow-up: 12 years; TT3a: n = 303, median follow-up: 9 years) permitted investigation of whether progression-free survival (PFS) and complete response (CR) duration were consistent with curability, ie observation of plateaus in Kaplan-Meier plots for PFS and CR duration. In the subset of 627 patients with plasma cell gene expression profiling data, cure plateaus were apparent at 5 years in the 14% with high-risk myeloma compared with 10 years in the remainder with low-risk disease. A parametric model based on PFS and CR duration supported an increase in curability: 10-year PFS and CR estimates increased from 8.8%/17.9% in TT1 to 15.5%/28.2% in TT2's control arm to 25.1%/35.6% in TT2's thalidomide arm and to 32.9%/48.8% in TT3a. Toward developing novel therapies, we recommend a concerted focus on patients with high-risk myeloma whose outcome has not been advanced.

Figure 1

Components of Total Therapy trials TT1, TT2, and TT3a.

Figure 2

Survival estimates by protocol. Overall survival (OS) and PFS curves (A-D) show progressive narrowing with transition from TT1 to TT2-Thal to TT2+Thal to TT3a. Progressive improvement between protocols in OS and PFS is apparent (E-F). (A) OS and PFS: TT1. (B) OS and PFS: TT2 - Thal. (C) OS and PFS: TT2 + Thal. (D) OS and PFS: TT3a. (E) OS by protocol. (F) PFS by protocol.

Figure 3

Response and progression estimates by protocol. The cumulative incidence of CR (A) translates into longer CR duration (CRD) (B). Of note, despite similar times of onset of CR in TT2+Thal and TT3a, CRD is far superior with TT3a, suggesting that a deeper level of CR was effectuated. With transition to later protocols, the slopes of time to progression (TTP) (C) and time to relapse (TTR) from CR (D) progressively flatten and reach lower plateaus. (A) Cumulative incidence of CR. (B) CRD by protocol. (C) TTP. (D) TTR from CR.

Figure 4

Progression estimates by protocol and GEP-70 risk. PFS and TTP are analyzed according to plasma cell GEP-based low- and high-risk subsets. Common to all endpoints examined, plateaus are reached earlier at ∼5 years in high-risk (B: PFS, D: TTP) compared with 10 years in the majority of 85% of patients with low-risk myeloma (A: PFS, C: TTP). (A) PFS by protocol: GEP-70 low-risk only. (B) PFS by protocol: GEP-70 high-risk only. (C) TTP: GEP-70 low-risk only. (D) TTP: GEP-70 high-risk only.

Figure 5

Response estimates by protocol and GEP-70 risk. Cumulative incidence of CR (CRI) and CRD are analyzed according to plasma cell GEP-based low- and high-risk subsets. Common to all endpoints examined, plateaus are reached earlier in high-risk (B: CRI, D: CRD) than patients with low-risk myeloma (A: CRI, C: CRD). (A) Cumulative incidence of CR: GEP-70 low-risk only. (B) Cumulative incidence of CR: GEP-70 high-risk only. (C) CRD by protocol: GEP-70 low-risk only. (D) CRD by protocol: GEP-70 high-risk only.

Figure 6

PFS by CR status at 2-year landmark by protocol. Patients considered here were alive and progression-free at the 2-year landmark. Plateaus are seen in patients achieving CR and also in those with lesser responses prior to the 2-year landmark (A: TT1, B: TT2-Thal, C: TT2+Thal). (A) PFS by response at landmark. Landmarked 2 years from registration: TT1. (B) PFS by response at landmark. Landmarked 2 years from registration: TT2 - Thal. (C) PFS by response at landmark. Landmarked 2 years from registration: TT2 + Thal.

Figure 7

GEP of bone marrow biopsy material. GEP of whole bone marrow biopsy material encompasses in both hematopoietic and bone marrow microenvironmental tissues. In the case of myeloma and precursor conditions, myeloma plasma cells coexist in the biopsy material with normal hematopoietic and stromal elements. Marked differences in the 37-gene score were noted between CMM, AMM, MGUS, CR, and normal donors (A). Myeloma patients qualifying for CR status and whose bone marrow biopsy becomes normal donor-like (“normalization”) enjoy superior CR duration than their counterparts not qualifying for normalization. CRD according to “normalization” of bone marrow biopsies of patients with MM was compared (CRD by normalization status for patients in CR using first GEP sample after 2 years of CR) (B).