Presentation Authors: Vasilis Stavrinides*, Francesco Giganti, Clement Orczyk, Shonit Punwani, Clare Allen, Alex Kirkham, Alex Freeman, Hayley Whitaker, Mark Emberton, Caroline M Moore, London, United Kingdom
Introduction: In MRI-based active surveillance (AS) for prostate cancer, the prognostic significance of visible disease at baseline is poorly defined. We present clinical outcomes from the UCLH MRI-based AS cohort (to our knowledge, the largest in the world) and investigate the association of baseline imaging with time to AS exit.
Methods: Follow up and clinical-histological outcomes were collected for 647 men in the UCLH AS database enrolled between August 2004 and December 2017 (Gleason 3+3 or low-volume 3+4 disease, PSA < 20 ng/mL and baseline mpMRI; mean age: 61.8 yrs; median PSA: 6.4 ng/mL). Starting time was defined as the date of the first MRI and all men included had at least 6 months of follow up. AS exit was defined as any treatment, transition to watchful waiting (WW) and/or pathological progression to Gleason 4+3 or higher on secondary biopsies. The Kaplan-Meier method and log rank test were used to compare time to AS exit in men with a baseline Likert score of 1-3 (no lesion) versus those with Likert 4-5 (lesion).
Results: Median follow up was 53 months (IQR 31.5-76). In total, 172 men were treated (44 radical prostatectomy, 83 focal, 12 hormones, 17 radiotherapy [12 combined with hormones], 3 brachytherapy). Fourteen patients transitioned to WW. Pathological progression on secondary biopsy was observed in 37 men in total; 15 progressed to 4+3 and received treatment, apart from 1 lost to follow up and 2 who transitioned to WW. Also, one patient was treated due to anxiety, 21 were lost to follow up and 36 were discharged for PSA monitoring in the community with specified PSA thresholds for re-referral. Five deaths were recorded, but only one was prostate cancer-related. There was a significant difference in time to AS exit between patients with a lesion and those without, which persisted regardless of Gleason grade at diagnosis (log rank test; p=0.0066 for Gleason 3+3, p=0.00069 for Gleason 3+4).
Conclusions: Men on AS with a Likert of 4-5 at baseline appear to have a significantly different clinical trajectory compared to those without a well-defined MRI lesion, regardless of cancer grade at diagnosis. This could reflect a divergence in the natural history of visible and non-visible disease, increased monitoring vigilance for men with a well-defined lesion, or both. Further research on these findings is under way.
Source of Funding: NIHR UCH/UCL Biomedical Research Centre, Movember Foundation.