E-Poster Presentation ESA-SRB-ANZBMS 2021

Bipolar Androgen Therapy induces heterogeneous changes in the abundance and localisation of the androgen receptor in advanced prostate cancer (#304)

Georgia Cuffe 1 , Ashlee Clark 1 , Michelle Richards 1 , Melbourne Urological Research Alliance (MURAL) 1 , Renea Taylor 2 3 4 , Gail Risbridger 1 3 4 , Anthony Joshua 5 , Mitchell Lawrence 1 3 4
  1. Department of Anatomy and Developmental Biology, Monash Biomedicine Discovery Institute Cancer Program, Monash University, Clayton, VIC, Australia
  2. Department of Physiology, Monash Biomedicine Discovery Institute Cancer Program, Monash University, Clayton, VIC, Australia
  3. Cancer Research Division, Peter MacCallum Cancer Centre, Melbourne, VIC, Australia
  4. Sir Peter MacCallum Department of Oncology, The University of Melbourne, Parkville, VIC, Australia
  5. The Kinghorn Cancer Centre, St Vincent's Hospital, Sydney, NSW, Australia

Background: Androgen deprivation therapy is the primary treatment for advanced prostate cancer. Despite sustained suppression of androgens, patients eventually develop castration-resistant prostate cancer and require further treatment. A new treatment is Bipolar Androgen Therapy (BAT), where patients rapidly cycle between castrate and supraphysiological testosterone levels to prevent tumours adapting to treatment. Clinical trials of BAT are promising, but patient responses vary.

Hypothesis & Aims: We hypothesised that changes in expression of the androgen receptor (AR) and truncated AR-variants (ARVs) were associated with response to BAT. To test this, we profiled AR and ARVs in patient-derived xenografts (PDXs) of prostate cancer before and after treatment with BAT.

Methods: We treated five, phenotypically diverse PDXs (1x BAT-sensitive, 4x BAT-resistant) with vehicle or BAT (fortnightly, intramuscular injections, 1mg testosterone cypionate). Tumours were harvested at 24-hours, 5 days, and endpoint (7-weeks or 1000mm3). Expression of full-length AR (AR-FL), ARVs, and AR target-genes were measured using quantitative PCR. AR abundance and localisation was quantified using immunohistochemistry against different forms of the AR.

Results: Untreated PDXs had heterogeneous patterns of AR-FL and ARV expression. Acute BAT treatment down-regulated expression of AR mRNA, except in tumours with structural rearrangements of the AR gene (i.e. 27.1, 27.2). BAT also decreased nuclear AR protein levels in a BAT-sensitive PDX (i.e. 201.1), but not in BAT-resistant PDXs. Changes in AR-FL persisted at 5 days and endpoint, despite fluctuations in testosterone levels over a cycle of BAT. Expression of ARV mRNA mirrored differences in protein levels. Expression of PSMA, an AR target-gene, was suppressed over time, regardless of BAT response.

Conclusions: In patient-derived models, reduced nuclear AR protein expression following BAT may define a subset of tumours sensitive to BAT. BAT combination therapies may be an alternative for tumours that are resistant to BAT alone, and are currently in clinical trials.