Combination Benefit of Capivasertib and Venetoclax in Preclinical Models of Diffuse Large B-Cell Lymphoma

Diffuse large B-cell lymphoma (DLBCL) is the most common B-cell malignancy among adults and despite approximately 65% of patients with DLBCL being cured with RCHOP therapy, nonresponsive and relapsed patients have inadequate treatment options, highlighting the importance for innovative treatment regimens. Blockade of B-cell receptor (BCR) downstream signaling components with various targeted agents is emerging as a clinically tractable treatment strategy across multiple B-cell malignancies. Protein Kinase B (AKT) signaling downstream of the BCR complex has been shown to be a central node in germinal center B-cell (GCB) DLBCL and the potent, selective inhibitor of AKT1, AKT2, AKT3, capivasertib, currently being evaluated in multiple clinical trials by targeting AKT-driven solid cancers, has been shown to induce apoptosis in a subset of GCB-DLBCL cell lines and cause tumor stasis in xenograft mouse models (Erdman et al., 2017). Since the monotherapy capivasertib responses in GCB DLBCL models are partial and lack durability, we hypothesized a combination approach could deliver even greater therapeutic benefit. To identify optimal partners, we conducted a capivasertib centric in vitro combination screen with specific with BH3 family members across a panel of 15 DLBCL cell lines, which revealed a synergistically active combination with the BCL2 inhibitor, venetoclax which is currently being evaluated in DLBCL. The activity was specifically enhanced in cell lines of the GCB subtype, with 4 PTEN ^del and 2 PTEN ^wt cell line models showing combination benefit. To determine the ability of this combination to drive stronger and durable responses, we assessed capivasertib and ventoclax activity in xenograft mouse models using two GCB-DLBCL cell line lines, SUDHL4 (PTEN ^wt) and WSU-DLCL2 (PTEN ^del). Oral administration of either monotherapy capivasertib (130 mg/kg BID, 4-day on/3-day off) or venetoclax (100 mg/kg QD) provided partial tumor growth inhibition (capivasertib TGI = 74% in SUDHL4 and 29% in WSU-DLCL2, and venetoclax TGI = 46% in SUDHL4 and 0% in WSU-DLCL2), whereas the combination of capivasertib and venetoclax both on a 4-day on/3-day off schedule produced complete tumor regression (100% regression) in both xenograft GCB cell line models during the dosing period. Notably, in both xenograft models all mice (5/5 per model) remained tumor free for at least 30 days following dosing cessation demonstrating high durability of response for the combination. Additionally, this combination is currently being evaluated in clinically relevant GCB and non-GCB PDX mouse models. Taken together, our results provide preclinical evidence for the rational combination of AKT and BCL-2 blockade with capivasertib and venetoclax respectively in patients with relapsed/refractory GCB-DLBCL.