Diffuse Large B-Cell Lymphoma (DLBCL) is the most frequent B-cell Non-Hodgkin Lymphoma (B-NHL). It derives from germinal-center (GC) B cells, and has been classified into GC B cell like (GCB) and activated B cell like (ABC) subtypes, which differ in their cell of origin, genetic alterations and responses to therapy, with ABC having inferior prognosis in response to the current immuno-chemotherapeutic regimen (R-CHOP). This standard first-line treatment is successful in ~70% of cases based on 5-year overall survival, but the remaining ~30% of patients remains incurable. While recently developed drugs, including the Bruton Tyrosine Kinase (BTK) inhibitor Ibrutinib and PI3K-delta inhibitors, are providing novel targeted strategies, their impact on DLBCL is still limited to a fraction of cases, mostly in the ABC subtype. In order to identify drugs that can be repositioned for selective efficacy against DLBCL subtypes (GCB- and ABC-DLBCL) compared to non-GC derived lymphomas, we screened a focused set including 2,160 FDA-approved drugs and other targeted drugs in advanced clinical testing on a panel of 9 cell lines consisting of 6 DLBCL (3 ABC and 3 GCB) and 3 Mantle Cell Lymphoma (MCL) lines. This screen identified the multi-kinase inhibitor Dasatinib as the most DLBCL-specific agent. Dose-response experiments on an extended panel of 32 DLBCL lines showed that Dasatinib has broader activity than Ibrutinib, since it is active against all ABC-DLBCL Ibrutinib-sensitive lines, as well as an additional set of 9 ABC- and GCB-DLBCL lines. Notably, Dasatinib overcomes Ibrutinib-resistance caused by BTK C481S mutation, as shown by the fact that Dasatinib is active against 3 Ibrutinib-sensitive ABC-DLBCL lines rendered Ibrutinib-resistant by transduction of BTK C481S mutant. In these cell lines, Dasatinib, analogous to Ibrutinib, is unable to suppress BTK-C481S auto-phosphorylation at Tyrosine 223, indicating that it may act independently of BTK. To understand which kinase or combination of kinases needs to be suppressed for Dasatinib activity, we employed Dasatinib-resistant, gatekeeper mutants of each BCR-associated Src-Family Kinase (LYN, FYN and BLK), which can all be targeted by Dasatinib, alone or in combination. Our results showed that FYN suppression is essential for Dasatinib activity in both DLBCL subtypes. Integrated genomic and bioinformatics approaches to explore genetic determinants of resistance to Dasatinib revealed PTEN disruption as the most significant predictor. PTEN is inactivated in the majority of resistant lines by both genetic and epigenetic mechanisms, resulting in the inability by Dasatinib to restrict PI3K-mediated AKT activation. These data were further validated by showing that 3 Dasatinib-sensitive DLBCL lines were made resistant by transduction of a constitutively active PIK3CA mutant (PIK3CA-H1047R). Notably, suppression of AKT activation by mTORC2 inhibitors can revert Dasatinib resistance, suggesting that combinatorial strategies involving Dasatinib and drugs targeting the PI3K/AKT/mTOR axis may be effective in PTEN-negative DLBCLs. The above results were validated in vivo using DLBCL xenograft mouse models. Overall, these results provide a preclinical framework for Dasatinib-based targeted therapies for DLBCL.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.