P932: ON-DEMAND PLERIXAFOR WITH CYCLOPHOSPHAMIDE AND G-CSF FOR HEMATOPOIETIC STEM-CELL MOBILIZATION IN MULTIPLE MYELOMA PATIENTS: FINAL RESULTS OF THE MOZOBL06877 STUDY

Abstract

Background: High-dose melphalan and autologous stem-cell transplant (ASCT) are pillars for the upfront treatment of ASCT-eligible newly diagnosed multiple myeloma (ND)MM patients (pts) or as salvage strategy at relapse. Failure to collect an adequate number of hematopoietic stem cells (HSC) to receive ASCT (<2×106 CD34+ cells/kg) occurs in 5%-15% of MM pts undergoing HSC mobilization with granulocyte colony-stimulating factor (G-CSF) or G-CSF+cyclophosphamide (G-CSF/CY). Aims: We report the final results of the observational MOZOBL06877 study (NCT03406091; partially supported by Sanofi investigation funds) to prospectively assess the performance of HSC mobilization with G-CSF/CY plus on-demand CXCR4 inhibitor plerixafor (PLX) in NDMM pts treated with novel agents. Methods: NDMM pts undergoing HSC mobilization with CY (2-4 g/m2) and G-CSF (5-10 mcg/kg/day) were enrolled and observed up to 30 days after mobilization. According to its label, “on-demand” PLX was administered in pts with <20 CD34+ cells/ul after ≥4 days of G-CSF or in case <1×106 CD34+ cells/kg were collected on the first apheresis day. Pts gave written informed consent. The primary endpoint was the poor mobilizer rate, defined as the rate of pts collecting <2×106 CD34+ cells/kg or requiring PLX. Secondary endpoints were the identification of predictive factors for PLX use and safety during mobilization. Results: 301 NDMM pts were enrolled and analyzed; 72% received induction with bortezomib-thalidomide-dexamethasone, 9% a lenalidomide (Len)-based and 3% a daratumumab (Dara)-based regimen. Overall, 48 pts (16%) were poor mobilizers: 14 (5%) failed to yield ≥2×106/Kg CD34+, while 34 (11%) required rescue with PLX. Among pts yielding <2×106/Kg CD34+ cells, 4 received PLX, while 10 did not. Among pts who successfully collected ≥2×106/Kg CD34+ (n=287, 95%), 34 (12%) required PLX: 23 (68%) due to a CD34+/uL count ≤20 after ≥4 days of G-CSF and 11 (32%) to a collection <1×106/Kg after the first apheresis day. In pts mobilized with PLX, median number of CD34+×106/L cells increased from 17.5 (IQR 10.8-25.6) before PLX to 58.3 (IQR 34.2-100.2) after PLX. Median number of CD34+/Kg collected in pts who did not require PLX was 10.2×106 (IQR 8.3-13.2). In pts who required PLX, a median of 6.5×106 CD34+/Kg (IQR 4.6-9.6) was collected. An adequate HSC yield for ≥2 ASCTs (≥4×106/Kg) was obtained in 96% and 85% of pts not receiving or receiving PLX, respectively. Median number of apheresis days was 1 (IQR 1-2) in pts not requiring PLX and 2 (IQR 1-2) in pts requiring PLX. In a multivariate analysis (Table), factors predicting PLX use were advanced disease stage (R-ISS 3 vs 1-2, OR 5.5, P=0.01), bone marrow (BM) plasma cell (PC) infiltration at diagnosis (PC>60% vs ≤60%, OR 4.2, P<0.001), pre-mobilization absolute neutrophil count (ANC, < vs >2500/mmc, OR 2.8, P=0.01), and the use of Len-based (Yes vs No, OR 3.4, P=0.02) or Dara-based induction regimens (Yes vs No, OR 5.3; P=0.04). Grade 3 infections occurred in 3 pts (1%), with no grade 4-5 reported. Image:Summary/Conclusion: Among NDMM pts treated with novel-agent-based induction and undergoing HSC mobilization with CY/GCSF, “on-demand” PLX in case of low CD34+ cell count or insufficient HSC yield was a safe and effective rescue strategy, reducing mobilization failure from 16% to 5%. Predictive factors for PLX use were advanced disease stage, high BM plasmacytosis, low ANC values before mobilization and the use of Len/Dara during induction. Preemptive PLX in pts with ≥1 risk factors should be investigated to further reduce the risk of mobilization failure.