A Pilot Study of the Combination of 5-Azacitidine and All-trans Retinoic Acid in Biochemically Recurrent Prostate Cancer.

Translational RelevanceDespite definitive local therapy, some men with prostate cancer develop biochemical recurrence (BCR-PCa) and progress to metastatic disease. Current standard-of-care, androgen receptor pathway inhibitors (ARPIs) with or without androgen deprivation therapy (ADT) carry substantial long-term morbidity. This trial tested a novel, non-hormonal approach of 5-azacitidine (AZA) plus all-trans retinoic acid (ATRA) to induce tumor cell dormancy via epigenetic reprogramming. The regimen was well tolerated, with manageable toxicities, and showed preliminary signals of delayed PSA progression and prolonged PSA doubling time in some patients, suggesting dormancy induction. One patient achieved durable disease stabilization with ATRA maintenance. These first-in-human findings indicate that epigenetic reprogramming may modulate dormancy in BCR-PCa, offering a potential strategy to delay or minimize ADT use and its toxicities. This approach highlights the translational potential of preventing or delaying overt metastases by activating dormancy pathways in early recurrent prostate cancer. PurposeBiochemical recurrence (BCR) after definitive local therapy remains a major clinical challenge in prostate cancer (PCa), with heterogeneous disease trajectories and few established strategies to delay further progression without prolonged androgen deprivation. This pilot study evaluated the combination of 5-azacitidine (AZA) and all-trans retinoic acid (ATRA) to induce tumor dormancy and delay clinical progression in patients with BCR. Experimental DesignIn a prospective, open-label, randomized, single-institution pilot trial, patients with BCR of PCa and no recent hormonal or definitive therapy received low-dose AZA and sequential ATRA. The co-primary endpoints were changes in prostate-specific antigen doubling time (PSADT) and time to next treatment (TTNT). Safety and biomarker analyses, including bone morphogenetic protein (BMP) signaling and dormancy marker NR2F1 in circulating tumor cells (CTCs), were evaluated to investigate treatment effects on minimal residual disease dormancy. ResultsFourteen patients were enrolled. Treatment resulted in an increase in median PSADT from 2.45 to 4.56 months. The median TTNT was 9.6 months, with 28.6% of patients experiencing TTNT over 12 months. No new safety signals were identified; adverse events were consistent with those expected for AZA and ATRA. Analysis of circulating BMP4 and BMP7 suggested that higher BMP4 levels may correlate with treatment response. Notably, all patients achieved testosterone recovery post-treatment, likely reflecting the avoidance of ongoing androgen deprivation. Across the cohort, treatment with AZA+ATRA led to a reduction in total CTC numbers and an apparent increase in the fraction of NR2F1-positive CTCs in responders, although the small cohort size limited statistical testing. ConclusionsThe combination of AZA and ATRA was feasible and prolonged PSA kinetics in a subset of patients with BCR of PCa, with a favorable safety profile. This epigenetic approach promoting tumor dormancy presents a potential strategy to defer progression and delay the need for continuous hormonal suppression. Larger studies are warranted to validate these findings and further explore biomarkers predictive of clinical benefit.