Session: 651. Myeloma - Biology and Pathophysiology, excluding Therapy: Poster II
Methods: Imaging parameters included standard metastatic skeletal survey (MBS) delineating the number of osteolytic lesions, STIR-derived MRI-based foal lesion (FL) number (limited to the axial skeleton bone marrow of head, spine, and pelvis). In the absence of FL, diffuse hyper-intense marrow (DHIM) involvement was also annotated1. As for PET scanning, FL number, SUVmax and EMD parameters were captured. PET studies were repeated on day 7 after initiation of the DT-PACE portion of the first induction cycle, prior to first transplant, and prior to maintenance initiation. Imaging data were complete for 429 patients at baseline including 394 with additional information on gene expression profiling (GEP)-derived high-risk. Day-7 PET data were available on 308 and 277 patients, respectively. Cox regression modeling was employed for overall survival (OS), progression free survival (PFS) and complete response duration (CRD).
Results: Patient characteristics were fairly comparable among the TT3A and TT3B patients except higher proportions of patients with high-risk features (high beta-2-microglobulin, low albumin, high GEP-derived centrosome index) in TT3B. The number of MBS-OL (>2) adversely affected OS (p=0.0001) and PFS (p=0.005) in TT3A (p=0.0001), PFS (p=0.03) in TT3B with a trend apparent for OS (p=0.06). For CRD, trends were apparent for shorter CRD in the presence of OL >2 in TT3A. In case of baseline PET, results were quite consistent for the 3 endpoints examined and between protocols applied. Thus, patients with 0 or 1-3 FL had equally favorable OS, PFS and, less pronounced in TT3B, CRD. Those with FL >3 fared poorly. Adverse consequences of FL-associated SUVmax >3.9 are apparent for OS, PFS and CRD in TT3A, whereas with TT3B only trends were apparent. Among follow-up imaging, Day-7 PET was the earliest measure. As is apparent in Figure 1, the presence of more than 3 FL imparted shorter OS and PFS in both TT3A and TT3B, with comparable outcomes noted for patients presenting with 0 or 1-3 FL at that time. Similar trends were apparent for CRD with both TT3A and TT3B. Cox regression modeling that considered additional information on day-7 PET results revealed an independent negative impact of day-7 FL>3 in the absence and presence of GEP data for both OS and PFS. Importantly, such information made baseline PET-FL irrelevant for OS and PFS, with MBS-OL >2 remaining in the model for OS, even when GEP data were also analyzed. The other surviving baseline variables included GEP-70-defined high risk for all 3 endpoints examined including CRD, where no imaging parameter emerged as significant. In the case of day-7 PET-SUV max, prognostic significance only applied to OS and PFS in TT3A.
Conclusions: The data presented herein, demonstrate that failure of FDG suppression as early as post-induction chemotherapy carries adverse outcome and can serve as prognostic indicators of long-term survival. We have thus established that early PET follow-up scanning can identify further risk discrimination that can be exploited toward therapy change.
Disclosures: No relevant conflicts of interest to declare.
See more of: Oral and Poster Abstracts
*signifies non-member of ASH