The combination of STX-478 and lorlatinib for the treatment of PIK3CA mutant colorectal cancer

Background: Over 15% of colorectal cancer (CRC) patients are PIK3CA mutant, making it a critical pathway for drug development. Recently, we performed a high throughput drug screen with STX-478, a mutant-selective PIK3CA inhibitor, and identified lorlatinib, an ALK inhibitor, as a potential means to enhance therapeutic efficacy. Here we confirm this efficacy and evaluate potential mechanisms of action, including the potential impact of lorlatinib on MYC protein abundance and target gene expression. Methods: CRC cell lines SW48 and SW48PK (Horizon Discovery; SW48 with a PIK3CA H1047R mutation) were used in a cell viability assay. Immunoblotting was performed on SW48, SW48PK and patient derived cancer organoids (PDCOs) with STX-478, lorlatinib, and combo treatment. ImageJ was used to quantify these blots to evaluate PI3K protein expression. PIK3CA H1047R mutant PDCOs were treated, imaged, and the longest diameter of any organoid was measured. Glass's delta (GD) was used to determine effect size for each treatment group compared to control. Gene expressions for five MYC target genes were measured by quantitative PCR (qPCR). Results: In SW48 and SW48PK cells, there were no significant differences in cell viability following lorlatinib single-agent treatment. However, combo treatment with STX-478 and lorlatinib significantly reduced cell viability in both cell lines (lorlatinib 500 nm, STX-478 500 nm; SW48 p=0.001, SW48PK p<0.001) In SW48PK, immunoblot analysis demonstrated a marked decrease in phosphorylation levels of ribosomal protein S6 (RPS6) and 4EBP1 under combo treatment, indicating suppression of downstream PI3K signaling ( pRPS6: p=0.02, 4EBP1: p=0.03). There were no significant changes in protein expression of c-MYC and pAKT in any of the treatment groups. RT- qPCR analysis overall did not demonstrate a change in MYC target gene expression. The expression of LDHA, a key glycolytic gene upregulated under hypoxic conditions, was significantly decreased in SW48PK cells following treatment, indicating a potential reduction in glycolytic activity associated with the combined therapy (p=0.005). PIK3CA H1047R mutant PDCOs treated with the combo of STX-478 and lorlatinib had a significant reduction in growth compared to control (relative change in diameter control versus combo = 42.9% vs 12.3%; GD=1.37). Conclusion: The combination of STX-478 and lorlatinib was confirmed to have significant efficacy in PIK3CA mutant CRC models. This activity does not seem to be related to effects on MYC. Further studies look to determine the mechanism of action for the combo treatment, and confirm efficacy in vivo.