Biomarker Proteasome Levels Predict Response to Combination Therapy with Carfilzomib, Lenalidomide, and Dexamethasone in Newly Diagnosed Multiple Myeloma Patients

Introduction

Carfilzomib is an effective irreversible proteasome inhibitor, exerting strong anti-myeloma effects and deep minimal residual disease negative responses. Updated results of a phase II clinical and correlative study combining carfilzomib, lenalidomide, and dexamethasone (CRd) in newly diagnosed multiple myeloma patients (NCT01402284) demonstrates a best response rate of very good partial response rate or higher (>VGPR) of 89%. As part of the prospective clinical correlative trial, we developed a novel flow cytometric assay for measuring proteasome and aggresome levels in bone marrow plasma cells pre- and post-carfilzomib single drug exposure in patient samples. The measurements were correlated with clinical treatment outcomes after CRd therapy.

Methods

Bone marrow aspirates were collected at baseline and C1D2 (single agent cfz exposed prior to len/dex administration). Permeabilized CD138 and CD38 positive plasma cells were tested using anti-19S proteasome or B-4 proteasome subunit antibodies subunit antibodies (Abcam, Cambridge, UK). In parallel, cells were labeled with ProteoStat Aggresome Detection Reagent (Enzo Life Sciences, Farmingdale, NY). Multicolor flow cytometric acquisition and analysis was performed using BD FACS CANTO and DIVA software. Data was expressed as mean fluorescence intensity (MFI) ratio using isotype-matched controls. Patients’ best responses to therapy, minimal residual disease status, and progression free survival data were correlated to the acquired proteasome and aggresome data. Statistical analysis was performed using DataPrism software. Statistical significance was considered as p-value <0.05.

Results

Twenty-six out of 45 patient pre-treatment bone marrow samples were available for analysis of 19S proteasome, B4 proteasome, or aggresome levels in plasma cells: 16 complete response/stringent (CR/sCR) (1 CR, 15 sCR) and 10 non-CR (7 VGPR, and 3 partial response) groups. The total percent of plasma cells was not statistically significantly different between two groups. The median plasma cell 19S MFI ratio was higher in the CR group: 20.7 months (range: 5.21-84.3) compared to the non-CR group: 6.4 months (range: 0.4-12.4) (p<0.01, Mann-Whitney). Median plasma cell B4 or aggresome MFI ratios did not vary significantly prior to treatment between CR and non-CR groups (p=NS, Mann-Whitney). After single agent carfilzomib treatment (C1D2), plasma cell 19S proteasome MFI ratio in the CR group decreased in 8 out of 10 samples compared to baseline (p=0.06, Wilcoxon signed rank test) and the mean percentage decline was 42.3%. A baseline 19S MFI ratio > 10 was associated with obtaining CR/sCR compared to 19S MFI ratio < 10 (p=0.02, Fisher’s exact test), and trended towards achieving MRD negative status (p=0.05, Fisher’s exact test). PFS and duration of response improved significantly in patients with higher baseline 19S MFI ratio > 10 compared to baseline 19S MFI ratio <10, median NR vs. 19 months (p=0.04, two-tailed log-rank) and NR vs. 16.3 months (p=0.04, two-tailed log-rank), respectively. However, median time to reach a CR/sCR did not vary between 19S MFI ratio > 10 vs. ratio <10 (p=NS, two-tailed log-rank).

Conclusion

In patients receiving combination therapy with CRd, higher pre-treatment 19S proteasome levels correlate with deeper clinical responses to treatment (ie, CR and sCR). Also higher pre-treatment proteasome levels were predictive of improved duration of response and PFS. In further support of these observations, decreased 19S proteasome levels after single drug carfilzomib exposure were associated with a higher likelihood of reaching CR/sCR. Our findings suggest that quantification of 19S proteasome levels in plasma cells, prior to initiation of carfilzomib-based therapy, may be a clinically useful biomarker to predict clinical outcomes in multiple myeloma patients.