Successful retreatment of CLL with venetoclax following clonal evolution during 37 months of tirabrutinib and idelalisib

ABSTRACT

The BCL2-specific inhibitor, venetoclax, has demonstrated remarkable clinical activity in the treatment of chronic lymphocytic leukemia (CLL), either alone or in combination with CD20 antibodies. Nevertheless, patients who fail to attain a complete remission relapse, and require further therapy. Data on retreatment with venetoclax at disease progression are currently limited. Here, we report patterns of clonal evolution in an R/R CLL patient that has demonstrated successful retreatment. A 57 year-old lady with chemotherapy- refractory (FCR, RCHOP, high dose methyl prednisolone) TP53 mutant CLL was treated for 21 months with single-agent venetoclax in 2014 (NCT01889186). She attained an MRD positive CR with the resolution of massive lymphadenopathy and with only low-level (0.01%) disease in the bone marrow. However, she subsequently progressed rapidly with a lymphocyte doubling time of only 4 weeks and was treated with tirabrutinib and idelalisib in combination (NCT02968563) from December 2015 for 37 months before progressing December 2019. She was retreated with venetoclax and rituximab but died of COVID-19-induced respiratory failure in March 2020. To study the clonal evolution underlying these events, in vitro drug sensitivity assays and whole exome sequencing (WES) were used to study peripheral blood mononuclear (PBMC) and bone marrow samples. WES of sample 1 showed multiple mutations in CLL driver genes SF3B1 R625C, KMT2C R4434Q, and TP53 R110L at VAFs of 37, 17, 35%, respectively. Mutations in other genes associated with CLL included FANCA L217F (47%) and SPEN P3402S (46%). At disease progression (sample 2), following venetoclax, there was the loss of detectable (WES at 100× coverage) TP53 R110L (with loss of 17p deletion on interphase FISH and analysis of copy number) but maintenance of SF3B1 R625C (44%), KMT2C R4434Q 30%), FANCA L217F (47%), and SPEN P3402S (55%). These data, therefore, suggest the TP53 mutant subclone was largely lost during therapy. No other mutations were identified as possible resistance mediators. There were no detectable BCL2 mutations. In vitro drug sensitivity testing to venetoclax showed an EC50 of 228nM (CLL EC50 usually 3-5 nM). The patient was then treated with the BTK inhibitor tirabrutinib in combination with idelalisib, with an excellent clinical response. After 10 months (sample 3, during the lymphocytosis induced by BTKi/PI3Kdi) SF3B1, KMT2C, FANCA, and SPEN mutations were detected at VAFs of 26, 30, 54, and 56%, respectively. At this point the TP53 R110L mutation was detected again at a VAF of 4%, indicating that stopping venetoclax allowed the clone to re-emerge. At this time, there were no detectable BTK or PLCG2 mutations. The patient then responded for a further 37 months before disease progression. At progression (sample 4), SF3B1, KMT2C, FANCA, and SPEN mutations were still detected in the peripheral blood at VAFs of 43, 31, 48, and 50%, respectively. The VAF of the TP53 R110L mutation had increased to 33%. Additionally, a BTK mutation (T474I) was identified with a VAF of 16%. Identical results were obtained using a bone marrow sample. Now, however, in vitro analysis demonstrated a high degree of sensitivity to venetoclax (EC50 0.72 nM). The patient was, therefore, retreated with venetoclax and rituximab. At the point of re-treatment, VAFs were maintained, with the emergence of a new subclonal NOTCH1 G1001D mutation at a VAF of 3%. The patient, unfortunately, died 4 months after commencing therapy due to COVID-19 associated pneumonitis. A full disease reassessment was not made but the patient's blood count had normalized, with rapid clearance of CLL cells from the peripheral blood, recovery of normal hematological indices, resolution of splenomegaly, and partial resolution of lymphadenopathy on CT scan. These data, therefore, suggest that re-treatment with venetoclax in CLL can be successful. Regaining sensitivity to venetoclax may largely depend on shifting clonal dynamics. The molecular basis of venetoclax resistance in this case is currently being investigated. A so in this particular case, it appears that the TP53 mutant subclone was more sensitive to BCL2 inhibition than TP53 wild-type subclone(s), and was largely eliminated by initial venetoclax treatment, contrasting with recently published data suggesting resistance of TP53 mutant hematological malignancies to BCL2 inhibition due to increased thresholds for BAX/BAK activation (Thijssen et al., 2021).