Novel Cytokine-Mediated Mechanism of Action Identified By Quantitative Seroproteomics in Multiple Myeloma Patients Treated with Tagraxofusp, a Novel CD123-Directed Targeted Therapy

Introduction Plasmacytoid dendritic cells (pDCs) express CD123/IL-3Rα and promote tumor growth and immunosuppression in multiple myeloma (MM) (Chauhan et al, Cancer Cell 2009, 16:309-323; Ray et al, Leukemia, 2018, 32:843-846). Tagraxofusp is a novel targeted therapy directed against CD123, and is FDA-approved for the treatment of patients with blastic plasmacytoid dendritic cell neoplasm [BPDCN]). Tagraxofusp can also trigger anti-MM activity by reducing the viability of immunologically defective and tumor-promoting pDCs in MM. Furthermore, tagraxofusp synergistically enhances the anti-MM activity of anti-MM agents bortezomib and pomalidomide. Our preclinical findings led to a recently completed phase 1/2 clinical trial of tagraxofusp with pomalidomide/dexamethasone in relapsed/refractory MM patients (NCT02661022). Results demonstrated preliminary safety and efficacy, with 5 of 9 heavily pretreated patients achieving durable partial response (PR) (ASH 2019). Here, we report the early results of our translational correlative studies using bone marrow (BM), peripheral blood (PB), and serum from the study cohort.

Methods Tagraxofusp is a bioengineered targeted therapy directed to CD123 comprised of human IL-3 fused to a truncated diphtheria toxin (DT) payload (Stemline Therapeutics, NY). pDCs and patient MM cells were purified from BM/PB samples after informed consent, and quantified using FACS, as described (Ray et al, Leukemia, 2018). A novel high throughput seroproteomics platform SOMAscan was used to analyze 1,310 protein analytes in serum samples from MM patients (n = 9). SOMAscan data were subjected to meta-analysis to generate heatmaps, followed by hierarchical cluster analysis. SOMAscan results were validated with ELISA using supernatants from MM patient pDCs cultured with or without tagraxofusp.

Results Analysis of BM/PB samples from MM patients receiving tagraxofusp therapy showed a distinct reduction in the frequency of viable pDCs [average 2% at screening vs 0.75% post-tagraxofusp; n = 6; p = 0.036]. Of note, pDCs isolated from tagraxofusp-treated patients showed decreased ability to trigger MM cell growth. SOMAscan analysis of patient serum before and after tagraxofusp therapy showed alterations in the levels of 100 proteins [Median Fold Change in expression: 0.39 to 4.5; n = 6; 3 each; p < 0.05]. Importantly, tagraxofusp treatment reduced pDC-related soluble proteins including IFN-α (fold change: 0.8, treated vs untreated; p < 0.05). Pathway analysis further show that treatment affected immune signaling. For example, tagraxofusp decreased the levels of immunosuppressive proteins, soluble CD40L and IL1R2 (0.071-fold and 0.088 fold vs untreated; p = 0.02 and p = 0.013, respectively), promoting immune response. Moreover, analysis of end of treatment samples showed decreased soluble C-reactive protein, affecting the complement cascade after treatment (0.53-fold, p = 0.0173) via the downregulation of several C-C motif soluble chemokines (p < 0.05). Our earlier study showed that pDC-MM interactions triggered secretion of IL-3, which in turn promotes both pDC survival and MM cell growth. Importantly, tagraxofusp in this trial decreased serum IL-3 levels (fold change 0.75, treated vs untreated; p < 0.05).

Conclusions In the present study, we validate the target specificity of tagraxofusp against MM pDCs in relapsed and refractory MM patients enrolled in a phase 1/2 clinical trial. A future clinical trial of tagraxofusp in combination with bortezomib and pomalidomide will examine the utility of tagraxofusp to improve outcome in patients with relapsed refractory MM.