Exploiting ketone body metabolism as a therapeutic vulnerability in prostate cancer

Prostate cancer (PCa) progression remains a major clinical challenge with limited therapeutic options, and the complexity of metabolic pathways fueling advanced PCa rests poorly understood. We have identified a shift from glycolysis to ketone body (KB) metabolism in castration-resistant PCa (CRPC), with an upregulation of ketolytic/ketogenic enzymes. From these, ACAT1 is a potential therapeutic candidate, as its expression is elevated in aggressive PCa and associates with unfavorable clinical outcomes. Here, we investigated the contribution of the bone niche -the dominant PCa metastatic site- to KB metabolism during PCa progression and evaluated ACAT1 inhibition as a treatment strategy for advanced disease. Co-culture experiments between PCa cells and bone progenitors (MC3T3, Raw264.7) were used to evaluate bone-driven metabolic adaptations. To validate in vivo the effect of the bone niche, we implanted MDA PCa patient-derived xenograft (PDX) 183 either intrafemorally ( i.f. ) or subcutaneously ( s.c. ) in sham or castrated CB17 SCID mice. We assessed cell viability in PCa cells lines and in PDX-derived organoids (PDX.DO), treated with increasing concentrations of ACAT1 inhibitor, arecoline hydrobromide (AH). ACAT1 and phospho-ACAT1 levels were measured by Western blot and immunofluorescence. Intracellular KB and ATP content were quantified by luciferase-based assays. Lipid uptake and accumulation were quantified by flow cytometry using Bodipy probes. In vivo , mice bearing MDA PCa PDXs 183, 203 and 173 were treated with AH (50 mg/kg/day i.p., 21 days). Tumor volume, KB content, and ACAT1 modulation were evaluated. We observed transcriptional activation of lipid and KB metabolism in PC3 cells from PCa-bone co-cultures, and consistently, in i.f. PDXs compared to matched s.c. tumors, alongside a significant increase in ACAT1 expression levels. Castration dynamically modulated ACAT1 levels in MDA PCa 183 tumors growing i.f. , with initial downregulation followed by restoration over time, underscoring KB metabolism as a mechanism of resistance within the bone niche. Functional studies showed that AH treatment reduced cell viability in PC3, C42B, and 22Rv1 cells, as well as PDX.DO, decreased ACAT1 and phospho-ACAT1 expression, lowered KB levels, and altered lipid dynamics by increasing lipid accumulation, uptake, and decreasing ATP content (P<0.05). In vivo , AH treatment in MDA PCa 183 tumors led to a significant reduction in tumor volume and intratumoral KB content (P<0.05). Moreover, in the CRPC model MDA PCa 203 -derived from a longitudinal sample of the same patient as 183-, and in the aggressive hormone-naïve model MDA PCa 173, AH treatment significantly reduced both tumor volume and KB content (P<0.05). In conclusion, our findings reveal KB metabolism as a critical vulnerability in PCa progression, with ACAT1 emerging as a druggable target whose inhibition disrupts tumor growth and metabolic fitness.