BackgroundProteasome, a complex involved in the intracellular protein degradation, consists of multiple subunits, but only three subunits have enzymatic activity to cleave and degrade proteins, namely β1, β2 and β5. Carfilzomib (CFZ), a second-generation proteasome inhibitor (PI), can induce cell death by selective and irreversible inhibition of β5 subunit of proteasome. Preclinical data suggested that high-dose CFZ could co-inhibit predominantly β2 proteasome activity, followed by β1 inhibition (Besse et al, Cell Chem Biol. 2019). Over the past few years, CFZ has become a corner stone for multiple myeloma (MM) therapy. Currently, CFZ is approved by the FDA in different dosing schedules in combination with lenalidomide or daratumumab and dexamethasone. However, the optimal CFZ dosing is still a matter of debate, with the approved dosage ranging from 20to 70mg/m 2 in different regimens. In addition, if response can be recaptured by escalating CFZ dose in patients progressing from low-dose CFZ has yet to be determined. The aim of our current study was to analyse the profile of proteasome inhibition in the respective dose cohorts and to elucidate if high-dose CFZ could recapture response in patients resistant to low-dose CFZ.MethodsWe prospectively collected clinical data and peripheral blood mononuclear cells (PBMC) of 32 patients with relapsed/refractory (RR) MM before and 1-8 hours after CFZ administration. PBMC were lysed and labelled for the activity of individual proteasome subunits using activity based proteasome probes and the proteasome subunits were separated using SDS-PAGE. The activity of constitutive and immunoproteasome β1, β2 and β5 subunits was evaluated by densitometry analysis and combination of the activity of constitutive and immunoproteasome individual subunit was used for further analysis.ResultsOverall, six, nine, twelve and five patients received CFZ at a dose of 20, 27, 36 and 56 mg/m 2, respectively. As expected, the total activity of proteasome decreased with higher doses of CFZ. Significant inhibition (median inhibition > 50%) of β5 subunit was observed already at 20 mg/m 2 dose, while β2 subunit started to be co-inhibited only at a dose of ≥27 mg/m 2. Significant co-inhibition of β2 activity was seen at 36 mg/m 2 dose, at which also β1 subunit started to be co-inhibited. Finally, at 56 mg/m 2, the activity of all active subunits was inhibited with a median inhibition of > 50%, with the strongest inhibition of the β5 subunit, followed by β2 and then β1. When we compared the patient groups low-dose CFZ (20 or 27 mg/m 2) versus high-dose CFZ (36 or 56 mg/m 2), we observed a significant difference in β2 (P=0.002) and β5 (P=0.02) subunit inhibition between the both groups. In terms of total proteasome activity, high-dose CFZ demonstrated a significantly higher proteasome inhibition in comparison with patients receiving low-dose CFZ (P=0.01). In brief, our results suggested that high-dose CFZ, in contrast to low-dose CFZ, could obtain superior proteasome inhibition by co-inhibiting β2 subunit of proteasome complex.In light of this finding, we successfully treated six RRMM patients who were resistant to low-dose CFZ with CFZ dose escalation. All six patients were heavily pretreated with 3-12 lines of therapy including daratumumab, two PIs, two immunomodulatory drugs and autologous stem cell transplant. Additionally, one and two patients received prior treatment with B-cell maturation antigen targeted bi-specific antibody and chimeric antigen receptor modified T-cell, respectively. In the last line of treatment, these six patients showed progression during CFZ based regimens with low-dose CFZ, namely 20 or 27 mg/m 2. We therefore increased the CFZ dose to 36 or 56 mg/m 2 and the doses of agents other than CFZ in the combination regimens remained the same. High-dose CFZ dose recaptured response in all six patients with four and two patients that achieved partial remission and very good partial remission, respectively, and the progression free survival ranged from 1-13 months.ConclusionIn summary, high-dose CFZ, namely ≥ 36mg/m 2, showed more effective proteasome inhibition via blocking β5 and β2 subunits, while low-dose CFZ could not achieve a sufficient inhibition of β2 subunit. We provided the first in human evidence that high-dose CFZ could recapture response in RRMM patients resistant to low-dose CFZ by co-inhibiting the β2 subunit activity of proteasome complex. [Display omitted] DisclosuresEinsele: Janssen, Celgene/BMS, Amgen, GSK, Sanofi: Consultancy, Honoraria, Research Funding.