Session Item

Androgen deprivation therapy (ADT) is a common neoadjuvant treatment for prostate cancer (PCa) treated with primary radiation therapy (RT), particularly for patients with high-risk disease. Recurrence of PCa after RT is not uncommon, and if unmanaged, can result in disease progression. Quantitative imaging using magnetic resonance imaging (MRI) has shown potential as a non-invasive method for monitoring treatment response and may provide a means for early detection of local recurrence. However, the utility of quantitative MRI (qMRI) parameters to monitor response to RT in patients receiving ADT remains unclear. In this prospective study, we aimed to further understand the potential of qMRI as a biomarker of response to RT through the characterisation of post-RT response kinetics in patients treated with and without ADT.

Twenty-one patients were recruited to the Sequential Imaging Biofocussed Radiotherapy clinical trial (UTN U1111-1221-9589) between May and December 2019. Thirteen patients received neoadjuvant ADT prior to RT (2 – 4 months), with median duration of 22 months (range 4 – 24 months). Multiparametric MRI was acquired pre-RT and at 6-, 12-, and 18- months post-RT. ADT patients were also imaged prior to commencing ADT. ADC, D, f, HS, R2*, T1, Ktrans, kep, ve maps were generated from diffusion weighted (DWI) and dynamic contrast enhanced imaging (DCE) at each time point and deformably registered to the baseline (first) scan. An inhouse developed deep learning model was used to segment the tumour on a combination of T2w, DWI and ADC maps. Mean qMRI parameter values in the tumour were calculated at each imaging timepoint. Statistical significance of the treatment-related changes was assessed with rANOVA and post hoc two-tailed t-test.

All qMRI parameters, except f, had significant post-RT changes (Fig1 and 2). Both cohorts had similar trends in post-RT changes of DWI-derived parameters and T1. The magnitude of post-RT changes in ADT patients were enhanced in DWI-derived parameters and diminished for T1 compared to non-ADT patients. Although both cohorts had significant changes in DCE-derived parameters, the onset of changes for ADT patients occurred prior to RT. R2* changed significantly after RT in hormone naïve patients. While a significant change in R2* occurred after the initiation of ADT, the post-RT and pre-treatment values were comparable.

All qMRI parameters, except f, demonstrated potential as biomarkers of treatment response. However, the kinetics and magnitude of qMRI parameter change after RT varied between patients treated with and without ADT. Although the effect of RT cannot be distinguished from ADT in DCE-derived parameters, a lack of change in these parameters could be associated with poorly or non-responding tumours with risk of recurrence. As such, in developing predictive models of treatment response, the effects of ADT need to be considered.