Targeting the SWI/SNF Chromatin Remodeling Complex Subunit SMARCA4 in ALL

Background: B-cell-derived acute lymphoblastic leukemia (B-ALL) is a type of blood cancer characterized by the overproduction of immature lymphocytes and is the most common cancer in children. The current overall survival (OS) rates for children with B-ALL are nearly 90% when diagnosed with good- and intermediate-risk subtypes. However, OS rates for children diagnosed with high-risk subtypes, such as KMT2A-Rearranged (KMT2A-R) and Philadelphia-like (Ph-like), are significantly lower, highlighting the urgent need to identify critical leukemogenic pathways for targeted therapeutic interventions. SMARCA4 is part of the SWI/SNF chromatin-remodeling complex, which is crucial for regulating gene expression by altering chromatin structure. SMARCA4 is mutated in about 10% of non-small cell lung cancer patients, resulting in significantly lower OS. More recently, SMARCA4 has been identified as essential for early B cell development, identity, and growth. However, its importance in ALL has not been studied.

Results: We performed single-cell DNA sequencing experiments paired with meta-analyses of whole exome and whole genome sequencing data from 1717 patients with B-ALL. We found that SMARCA4 is only mutated in 0.35% of patients with B-ALL, suggesting that, unlike in non-small cell lung cancer, SMARCA4 mutations may not benefit ALL growth. Gene expression analysis of B cells at different developmental stages demonstrated that SMARCA4 is specifically upregulated during early B cell development (early pro-B to small pre-B stages), overlapping with the stages where ALL cells are arrested. Direct comparisons of human pre-B cells, B-ALL, CD34+, and T-ALL showed that pre-B cells and B-ALL cells have the highest SMARCA4 expression levels, indicating a potential dependency of B-ALL on SMARCA4. Analysis using the DepMAP portal confirmed that B-ALL cells are highly dependent on SMARCA4, confirming that a SMARCA4 inhibitor-based therapy may be beneficial for children with B-ALL. However, survival analysis of 207 children from the COG trial P9906 indicated no impact on patient outcomes from varying SMARCA4 expression levels. When we separated the patients based on their B-ALL subtype, we found that SMARCA4 was significantly highly expressed in children with Ph-like ALL than in those with KMT2A-R B-ALL, which we confirmed via Western blotting using Ph-like and KMT2A-R ALL cell lines. A survival analysis focused on these subtypes revealed that poor outcomes correlated with higher SMARCA4 expression in Ph-like ALL, while, surprisingly, high SMARCA4 expression levels correlated with good outcomes in KMT2A-R ALL. Based on these results, we hypothesized that Ph-like ALL, but not KMT2A-R B-ALL, may depend on high SMARCA4 expression levels. Validating our hypothesis, pharmacologic SMARCA4 inhibition using two recently developed inhibitors (BRM014 and FHD-286) showed that Ph-like B-ALL was significantly more sensitive to SMARCA4 inhibition compared to KMT2A-R ALL. KMT2A-R cell lines were highly resistant to both inhibitors, while Ph-like cells were highly sensitive, with IC50s for BRM014 at 10-50nM and 1nM for FHD-286.

Conclusion: Our results identify that SMARCA4 has a very low mutation frequency in B-ALL and is highly expressed in Ph-like B-ALL. Furthermore, we demonstrate a direct correlation between high SMARCA4 expression and poor patient outcomes, as well as a clear dependency of Ph-like B-ALL on SMARCA4. Based on our data, applying a SMARCA4 inhibitor-based therapy may specifically benefit patients with Ph-like B-ALL.