2023 - Del8p and TNFRSF10B loss are associated with a poor prognosis and resistance to fludarabine in chronic lymphocytic leukemia

Introduction. Chronic lymphocytic leukemia (CLL) is a heterogeneous disease, with a highly variable clinical course. Recurrent chromosomal abnormalities have a key role in CLL oncogenesis and drug resistance. We characterized a rare abnormality, the deletion of the short arm of chromosome 8 (del8p), associated with aggressive CLL. Additionally, we investigated the role of TNFRSF10 genes included in the minimal common deleted region.
Methods. We correlated patients' cytogenetic and clinical outcomes using standard statistical methods. We performed in vitro analyses in primary CLL cells and in TNFRSF10B CRISPR/Cas9 edited OSU-CLL cell lines using flow cytometry (annexin V and propidium iodide co-staining) for drug response assays, and ddPCR for gene expression assessment.
Results. Comparing the largest series to date of del8p CLL (n=57) to a non-del8p control cohort (n=155), del8p was significantly associated with poor prognostic factors: complex and highly complex karyotypes (33/46, 72% and 20/46, 44%, vs. 34/155 (22%) and 12/155 (8%) respectively, p < .001), del11q (27/52, 52% vs. 27/155, 17%; p < .001), del17p (20/55, 36% vs. 16/154, 10%; p < .001) and unmutated IGHV (35/48, 73%, vs. 70/151, 46%; p=.002). Del8p patients had a shorter time to first treatment (TTFT, median 0.25y vs 1.7y, p=.007), worse overall survival (OS) (10year OS: 30% vs. 82%; p < .0001), and a higher risk of Richter transformation (RT) (10y: 22% vs 5.5%, p=.034) (Figure 1). A multivariate analysis showed that large del8p and concomitant TP53 disruption had a negative impact on the OS. Interestingly, del8p patients treated with fludarabine-based regimens had a shorter time to next treatment (40 vs. 63 months; p=.002) and a shorter OS (10y OS: 46% vs. 82%; p < .001), including those with mutated IGHV (42 vs. 107 months; p=.015).
We found that the TNFRSF10B gene in 8p21.3 (coding for a pro-apoptotic receptor activated by TRAIL) had been lost in 51/56 (91%) cases. TNFRSF10B was haploinsufficient in del8p CLL: the TNFRSF10B expression was significantly lower in del8p CLL cells than in non-del8p CLL cells (2.4-fold; p < .001). We then demonstrated using CRISPR-edited cell lines that TNFRSF10B receptor was functional in inducing cell death in CLL cells. In primary CLL cells, TRAIL alone failed to induce significant cell death. We therefore hypothesized that exposure of cells to anti-CLL drugs might modulate TNFRSF10B expression, resulting in improved sensitivity to TRAIL. In non-del8p primary cells, TNFRSF10B expression was significantly greater after exposure to fludarabine (2.8-fold), but not after ibrutinib or venetoclax. This increase was not observed in del8p cells. We then performed a sequential approach to evaluate the cytotoxic effect of TRAIL with fludarabine. We observed that TRAIL had an additional cytotoxic effect in 13/15 (87%) of non-del8p samples. By contrast, TRAIL+fludarabine failed to elicit a synergistic cytotoxic effect in 6/7 del8p samples (Figure 2). We conclude that the lack of fludarabine-induced overexpression of TNFRSF10B abrogates the synergy between the purine analog and TRAIL. Finally, we hypothesized that p53 tumor suppressor protein was involved in fludarabine-induced TNFRSF10B expression. To check p53’s activation after fludarabine treatment we measured the expression of three p53 target genes: MDM2, BAX and CDKN1A. In non-del8p samples, fludarabine induced a greater expression of all three genes: MDM2, 12.5-fold; BAX, 4.3-fold; CDKN1A, 8.7-fold. In contrast, the fludarabine-induced expression was substantially lower in del8p samples (2.2-fold, 1.0-fold and 3.4-fold, respectively), irrespectively of the presence or the absence of del17p. Thus, del8p in CLL appeared to interfere with the p53-dependent gene expression induced by fludarabine.
Conclusions. In this large cohort, we found that del8p is associated with poor cytogenetic and molecular prognostic factors, as well as with short TTFT and OS, and higher risk of RT. We also demonstrated that TNFRSF10B has an important role in fludarabine resistance in del8p CLL: we showed that del8p in CLL prevents the increase in expression of the TNRSF10B receptor and thus abrogates the apoptogenic synergy between fludarabine and TRAIL. These in vitro results are in line with our in vivo data showing that del8p CLL patients treated with fludarabine-based regimens have shorter OS. In patients who could benefit from the classical fludarabine-cyclophosphamide-rituximab treatment but who harbor del8p, other treatments (such as ibrutinib and venetoclax) might be more effective than fludarabine.