Focal dose escalation for prostate cancer using Ga-HBED-CC PSMA PET/CT and MRI: a planning study based on histology reference.

Background: Focal radiation therapy has gained of interest in treatment of patients with primary prostate cancer (PCa). The question of how to define the intraprostatic boost volume is still open. Previous studies showed that multiparametric MRI (mpMRI) or PSMA PET alone could be used for boost volume definition. However, other studies proposed that the combined usage of both has the highest sensitivity in detection of intraprostatic lesions. The aim of this study was to demonstrate the feasibility and to evaluate the tumour control probability (TCP) and normal tissue complication probability (NTCP) of radiation therapy dose painting using Ga-HBED-CC PSMA PET/CT, mpMRI or the combination of both in primary PCa.

Methods: Ten patients underwent PSMA PET/CT and mpMRI followed by prostatectomy. Three gross tumour volumes (GTVs) were created based on PET (GTV-PET), mpMRI (GTV-MRI) and the union of both (GTV-union). Two plans were generated for each GTV. Plan95 consisted of whole-prostate IMRT to 77 Gy in 35 fractions and a simultaneous boost to 95 Gy (Plan95/Plan95/Plan95). Plan80 consisted of whole-prostate IMRT to 76 Gy in 38 fractions and a simultaneous boost to 80 Gy (Plan80/Plan80/Plan80). TCPs were calculated for GTV-histo (TCP-histo), which was delineated based on PCa distribution in co-registered histology slices. NTCPs were assessed for bladder and rectum.

Results: Dose constraints of published protocols were reached in every treatment plan. Mean TCP-histo were 99.7% (range: 97%-100%) and 75.5% (range: 33%-95%) for Plan95 and Plan80, respectively. Plan95 had significantly higher TCP-histo values than Plan95 (p = 0.008) and Plan95 (p = 0.008). Plan80 had significantly higher TCP-histo values than Plan80 (p = 0.012), but not than Plan80 (p = 0.472). Plan95 had significantly lower NTCP-rectum than Plan95 (p = 0.012). No significant differences in NTCP-rectum and NTCP-bladder were observed for all other plans (p > 0.05).

Conclusions: IMRT dose escalation on GTVs based on mpMRI, PSMA PET/CT and the combination of both was feasible. Boosting GTV-union resulted in significantly higher TCP-histo with no or minimal increase of NTCPs compared to the other plans.