KEY POINTS
- ARIEL is a prospective phase II imaging study in men with unfavorable-intermediate or high-risk localized prostate cancer receiving definitive RT plus ADT. The study uses serial Restriction Spectrum Imaging MRI before treatment, after neoadjuvant ADT and after RT.
- 97 patients were included in the preliminary analysis. Baseline MRI identified 96 PI-RADS lesions in 88 patients, while 79 patients underwent the planned post-ADT, pre-RT MRI.
- Neoadjuvant ADT substantially altered the target. Only 60% of previously visible lesions remained clearly visible, meaning 40% were no longer identifiable on MRI before RT.
- Even lesions that remained visible changed disproportionately. Median lesion volume decreased by 55.8% after ADT, compared with only 21.5% reduction in prostate volume. After ADT plus RT, lesion volume had decreased by 72.8% and prostate volume by 32.3%.
- The quantitative RSI restriction score also changed markedly: mean maximum RSIrs within visible lesions fell from 329 before treatment to 209 after ADT and 107 after RT, while conventional ADC changed much less consistently after ADT.
- These geometric changes have direct implications for focal boosting. Because apparent lesion disappearance may not represent eradication of viable tumor, defining the boost only from a post-ADT MRI could risk missing part of the original intraprostatic target.
- The trial's actual primary endpoint is still pending: whether changes in RSIrs predict biochemical recurrence within three years. The present report therefore establishes imaging and planning effects, not a validated response biomarker or oncologic benefit.
CLINICAL TAKEAWAY
For prostate RT with a focal intraprostatic boost, the pretreatment MRI should not be discarded after neoadjuvant ADT. ADT can make the lesion shrink or disappear much more rapidly than the gland itself, so pre- and post-ADT imaging may both be needed for accurate target reconstruction; the alternative strategy of starting ADT and RT concurrently remains a hypothesis for future study.