Combined Genetic Lesions in TP53 and CDKN2A/CDKN2B Drive B Cell Receptor-Dependent/Costimulatory Signal-Independent Proliferation in Richter Syndrome

B cell receptor (BCR) signals play a critical role in the pathogenesis of chronic lymphocytic leukemia (CLL), but their role in regulating CLL cell proliferation has still not been firmly established. Unlike normal B cells, CLL cells do not proliferate in vitro upon engagement of the BCR, suggesting that CLL cell proliferation is regulated by other signals from the microenvironment, such as those provided by Toll-like receptor (TLR) ligands or T cells. However, the rapid reduction in the percentage of proliferating CLL cells in patients treated with a BTK or SYK inhibitor suggests that the BCR may be more directly involved, at least in a subset of cases.

To further address this issue, we investigated the expression of cell cycle regulatory proteins in human and Eμ-TCL1-derived murine CLL cells stimulated with immobilized anti-IgM or cognate antigen, respectively. In both cases, BCR stimulation induced the expression of the positive cell cycle regulators MYC, CCND1, CCND2 and CDK4, consistent with G1 cell cycle entry, but also induced the expression of the negative regulators CDKN1A, CDKN2A and CDKN2B, which block cell cycle progression. Since combined deficiency of these negative regulators occurs in approximately one quarter of Richter syndrome cases because of deletion of the CDKN2A/CDKN2B locus and inactivating genetic lesions of the CDKN1A regulator TP53, we introduced these defects by CRISPR/Cas9 in autoreactive murine Eμ-TCL1 leukemia cells and investigated the effects on leukemia behavior. Adoptive transfer experiments showed that combined targeting of TP53, CDKN2A and CDKN2B results in accelerated leukemia growth and morphological changes consistent with Richter's transformation, including more diffuse infiltration, larger and more pleomorphic cells, and a higher proliferation rate. Moreover, in vitro experiments showed that cells with combined TP53/CDKN2A/2B deficiency had acquired the capacity for spontaneous proliferation, in contrast to control, TP53- or CDKN2A/2B-targeted cells which died after a couple of weeks in culture. Nucleotide sequencing of the TP53/CDKN2A/2B-targeted cells showed complete disappearance of the wild type alleles, suggesting that biallelic loss of all three genes is required for spontaneous growth in vitro. Combined disruption of TP53, CDKN2A and CDKN2B in two other autoreactive Eμ-TCL1 leukemias yielded the same results.

To determine whether the spontaneous in vitro proliferation is dependent on BCR signals, the TP53/CDKN2A/2B-deficient cells were transfected with Cas9 ribonucleoprotein complexes targeting the IgM heavy chain constant region (IGHM) gene or were treated with the BCR inhibitors ibrutinib, idelalisib and fostamatinib. Disruption of the IGHM gene or treatment with a BCR inhibitor resulted in almost complete block of proliferation. In contrast, knockdown of the TLR-adaptor protein MyD88 had no effect. Considering that T cells were not present in the culture conditions, these experiments establish that proliferation of Eμ-TCL1 leukemia cells with biallelic TP53/CDKN2A/2B disruption is BCR-dependent but independent of costimulatory signals.

To validate these findings in a human setting, we performed experiments with two recently established Richter syndrome patient-derived xenografts, one with biallelic inactivation/deletion of TP53, CDKN2A and CDKN2B (RS9737), and one with wild type TP53, CDKN2A and CDKN2B (RS1316). BrdU incorporation experiments showed that only RS9737 cells proliferate in vitro following BCR stimulation, whereas both RS9737 and RS1316 cells proliferate following stimulation with CpG + IL-15 or CD40L + IL-4 + IL-21.

Finally, we tested the activity of combined treatment with a BCR inhibitor and the CDK4/6 inhibitor palbociclib against the murine and human TP53/CDKN2A/2B-deficient Richter syndrome models. Combined treatment showed synergistic activity in vitro and significantly prolonged mouse survival in vivo compared to single agent treatment (n = 10 mice/group, P<0.001).

In conclusion, these data provide evidence that BCR signals are directly involved in regulating CLL cell proliferation and suggest that frequently co-occurring genetic lesions in TP53 and CDKN2A/2B contribute to Richter transformation by allowing for BCR dependent/costimulatory signal independent proliferation, which can be therapeutically targeted with a BCR and CDK4/6 inhibitor combination.