ABRUPT 2.0 is a non-comparative, randomized phase II trial focusing on acute genitourinary (GU) toxicity as the primary endpoint in a similar patient population. Acute GU toxicity was selected because it has been shown to be predictive of late GU toxicity and represents a clinically meaningful early safety endpoint. Using PACE-C as a benchmark, the trial is designed to test the hypothesis that acute GU toxicity in each treatment arm will remain within predefined clinically acceptable limits. Patients are randomized in a 1:1 ratio to receive either 36.25 Gy SBRT in five fractions over two weeks or 24 Gy SDRT with urethra-sparing and a focal isotoxic GTV boost up to 27 Gy. Randomization is performed using a minimization algorithm balanced for NCCN risk group (intermediate-unfavorable vs selected high-risk) and incorporating a random element. Treatment allocation is not masked. All participants are treated at a single institution (Fondazione IRCCS San Gerardo dei Tintori and University of Milan-Bicocca).
All patients undergo non-contrast-enhanced CT and mpMRI simulation on the same day. Anatomic reproducibility is ensured through rectal micro-enema administration. In the SDRT arm, the bladder is filled with 150 mL of saline solution via a 16-F Foley catheter before simulation and treatment delivery. In the five-fraction SBRT arm, bladder preparation consists of drinking 500 mL of still water 30-45 minutes before simulation and each fraction. The use of a hyaluronic acid rectal spacer is encouraged and left at physician's discretion. Patients are positioned supine with arms placed over the chest. CT images and T2-weighted 3D MRI scans are fused for target and OAR delineation. The GTV was defined as the visible tumor lesion on mpMRI in collaboration with an experienced radiologist. The clinical target volume (CTV) encompasses the entire prostate gland and seminal vesicles. The PTV is generated by applying a 3-mm isotropic margin to the CTV. OARs include the urethra, bladder, bladder trigone, rectum, femoral heads, penile bulb, bowel, and neurovascular bundles. In the SDRT arm, a 3-mm isotropic margin is added to the urethra, bladder, and rectum to generate planning OAR volumes (PRV), and the minimum prescribed dose to the GTV and PTV is constrained by OAR dose limits. In the five-fraction SBRT arm, the treatment prescription consists of 36.25 Gy administered in five fractions over two weeks on alternate days, with a mandatory additional CTV dose target of 40 Gy. Both dose levels are prescribed to the isodose covering 95% of the target volume, with controlled dose heterogeneity within the target and a maximum PTV dose of 42 Gy to protect the urethra by limiting its estimated dose to ≤95Gy EQD2. No focal boost is delivered to the GTV in this arm. In both arms, treatment plans are optimized using a 10-MV flattening-filter-free single partial arc (140°-220°), allowing a substantial reduction in treatment time.
In order to preserve identical 3-mm PTV margins in both treatment arms, continuous intrafraction motion monitoring is systematically implemented. Two distinct non-ionizing real-time tracking systems are employed. In the SDRT arm, prostate motion is monitored using an electromagnetic transmitter embedded within a urethral catheter. In the five-fraction SBRT arm, real-time tracking is performed using a 4D transperineal ultrasound system. All treatments are delivered on a linear accelerator. Initial patient positioning is verified using cone-beam CT with soft-tissue matching. Beam delivery is interrupted and patient positioning corrected whenever prostate displacement exceeds 2.5 mm in any of the three spatial directions.
The sample size has been determined using a Simon two-stage minimax design, independently powered for each treatment group to test whether the incidence of acute G2+ toxicity is significantly different from 51% (null hypothesis), using a one-sided 0.05-level test with 90% power to detect a true toxicity rate of 25%. Each group includes 29 patients (stage 1: 25 patients, and stage 2: 4 patients). The trial continued to complete stage 2 as 9 or fewer patients among the first 25 will develop G2+ GU toxicity within 3 months after treatment. If 10 patients or fewer in a group will develop G2+ GU toxicity, this would allow rejecting the null hypothesis of a 51% toxicity rate, which is 1.5 times higher than the rate observed in the PACE-C trial. Under the two-stage design, interim analysis will be provided after 10 patients are treated in each group to ensure that acute toxicity rates are within acceptable limits before proceeding to full accrual. The trial is not powered to detect statistical differences between the two treatment groups; therefore, data will be analysed descriptively.
