Prognostic value of central gland volume on MRI for biochemical recurrence after prostate radiotherapy.

PURPOSE

This study evaluates pretreatment prostate magnetic resonance imaging (MRI) metrics and clinical characteristics in predicting biochemical recurrence (BCR) after prostate radiotherapy (RT).

METHODS

In this retrospective single institution study, we identified men in our prostate cancer database who underwent MRI within 6 months prior to completing definitive RT from May 2011 to February 2023. Central gland volume and peripheral zone volume were measured by a radiologist using manual segmentation, along with Prostate Imaging-Reporting and Data System (PI-RADS) score. The primary objective was to determine the association of central gland volume with biochemical recurrence per Phoenix criteria. Multivariable and inverse probability weighted (IPW) Cox proportional hazards regression models were constructed.

RESULTS

A total of 373 men were included, with a median follow-up of 28 months. Thirteen (3.5%) were low risk, 97 (26%) favorable intermediate risk, 201 (53.9%) unfavorable intermediate risk, and 62 (16.6%) high risk. Fifty-four (14.5%) patients received conventionally fractionated RT, 105 (28.2%) moderately hypofractionated RT, 121 (32.4%) high-dose rate brachytherapy, and 93 (24.9%) stereotactic body RT. The 3- and 5-year rates of BCR were 7.8% and 18.3%, respectively. Higher central gland volume (per 5 cc) was associated with decreased risk of BCR (hazard ratio [HR]: 0.69, 95% confidence interval [CI]: 0.50-0.94, p = 0.02) on the multivariable Cox model and IPW model (HR: 0.75, 95% CI: 0.65-0.87, p < 0.001). No significant association was seen with peripheral zone volume, PI-RADS score, or RT modality.

CONCLUSION

Increased central gland volume on pretreatment prostate MRI is independently associated with a lower risk of biochemical recurrence after definitive radiation for prostate cancer. Central gland volume may improve patient selection and oncologic risk stratification prior to offering RT.