Defining the index lesion for potential salvage partial or hemi-gland ablation after radiation therapy for localized prostate cancer.

Background: Salvage partial gland ablation (sPGA) has been proposed to treat some localized radiorecurrent prostate cancer. The role of prostate biopsy and magnetic resonance imaging (MRI) characteristics to identify patients eligible for sPGA is unknown.

Objective: To evaluate the ability of MRI and prostate biopsy characteristics to identify an index lesion suitable for sPGA and validate this selection using detailed tumor maps created from whole-mount slides from salvage radical prostatectomy (sRP) specimens.

Design, Setting, And Participants: Men who underwent sRP for recurrent prostate cancer following primary radiotherapy with external beam radiotherapy (EBRT) and/or brachytherapy between 2000 and 2014 at a single high-volume cancer center were eligible. Those with tumor maps, MRI and biopsy data were included in analysis.

Outcome Measurements And Statistical Analysis: Primary outcome was the ability of clinicopathologic and imaging criteria to identify patients who may be eligible for sPGA based on detailed tumor map from whole-mount sRP slides.

Results And Limitations: Of 216 men who underwent sRP following whole gland radiotherapy, tumor maps, MRI, and biopsy data were available for 77. Of these, 15 (19%) were determined to be eligible for sPGA based on biopsy-proven unilateral disease in contiguous sextant segments, a dominant lesion on MRI concordant with biopsy location or no focal region of interest, and no imaging evidence of extraprostatic disease. Review of tumor maps identified 6 additional men who would have met criteria for sPGA, resulting in sensitivity of 71% (95% C.I. 48%-89%) and specificity of 100% (lower bound of 95% C.I. 94%). None of the 15 men who met the criteria for sPGA on clinical data were identified incorrectly on tumor maps to require full gland surgery (upper bound of 95% C.I. 22%). Median tumor volume of the index lesion was 0.4 cc and recurrent cancer was noted in the apex, mid-gland, and base in 81%, 100%, and 29% of men.

Conclusions: In men with recurrent prostate cancer after radiotherapy, biopsy findings and MRI can be used to select index lesions potentially amenable for sPGA and can guide patient evaluation for inclusion in clinical trials of sPGA following radiation failure. Larger, prospective studies are required to evaluate both the role of MRI and clinical criteria in guiding focal salvage therapy and the effectiveness of this modality for radiorecurrent prostate cancer.