Simultaneous integrated boost with stereotactic radiotherapy for dominant intraprostatic lesion of localized prostate cancer: a dosimetric planning study

Dominant intraprostatic lesion (DIL) has been known as the most common local recurrence site of prostate cancer. We evaluated the feasibility of simultaneous integrated boost (SIB) to DIL with CyberKnife stereotactic body radiotherapy (CK-SBRT). We selected 15 patients with prostate cancer and visible DIL and compared 3 plans for each patient: 1) No boost plan of 35 Gy to prostate, 2) DIL_40 plan of SIB 40 Gy to DIL and 35 Gy to prostate, and 3) DIL_45 plan with 45 Gy to DIL and 35 Gy to the prostate in 5 fractions. All targets satisfied with the prescription coverage per protocol. However, some patients failed to meet the Dmax of the rectum in DIL_40 plans (n = 4), and DIL_45 plans (n = 6). Violations of bladder constraints occurred in four DIL_45 plans. Consequently, the DIL boost with SBRT was possible in 73% of patients with DIL_40 plans, and 60% of patients with DIL_45 plans without any violation of normal organ constraints. All patients who experienced constraint violations had DILs in posterior segments. DIL boost using CK-SBRT could be an option for localized prostate cancer patients. For patients who had DIL in posterior segments, a moderate dose escalation of 40 Gy seemed appropriate.


Results
showed the characteristics of patients and DIL. The median age of 15 patients was 66 years (range, 56-82), and the median initial PSA was 5.6 ng/mL (3-10.6). T stages were various with T2a (n = 10), T2b (n = 1), and T2c (n = 4) disease. Four patients presented with Gleason score (GS) 6, 10 patients had GS 7, and one patient diagnosed with GS 9. Four patients with T2c disease had 2 DILs, and one patient had a single tumor involving two segments. The locations of DIL were various, and more than half were posterior segments (12 out of 20 locations). Median DIL volume was 1.3 cc (range, 0.7-6.6), and median prostate volume was 46.2 cc (range, 18.9-63.4 cc).
The results of the planning study were shown in Table 3. All target volumes (DIL, prostate, and planning target volume [PTV]) could achieve the planning goal as we expected. The maximum dose points were all located in DIL. Figure 1 showed the dose-volume histogram of one of the patients. Dose to DIL was escalated ( Fig. 1.1), while irradiation to bladder, rectum, and urethra remained under constraints ( Fig. 1.2). There were also cases with violations in OAR constraints of DIL_40 and DIL_45 plans. In the aspect of D max of the rectum, 4 patients exceeded 35 Gy in DIL_40 plans (range, 36.4-38.4 Gy), and 2 more patients failed to meet constraints in DIL_45 plans (range, 37.7-43.2 Gy). Similarly, there were patients who demonstrated more than 1 cc of V 30.8 Gy of rectal volume in DIL_40 plans (n = 3; range, 1.69-5.09 cc) and DIL_45 plans (n = 5; range, 3.2-7.4 cc). Patients who resulted in rectal constraint violation were all had DIL in posterior segments.    www.nature.com/scientificreports/ Violations of bladder and urethra constraints only occurred in DIL_45 plans: 4 cases of exceeding D max of 38.5 Gy (range, 39.3-41.9 Gy), 2 cases of more than 1 cc of V 35.7 Gy (range, 2.1-6.3 cc), and 5 patients of maximal urethra dose over 42 Gy (range, 43.2-45.5 Gy). Likewise, the DIL of the patients was located in posterior segments. Consequently, the DIL boost with SBRT was possible in 73% of patients with DIL_40 plans, and 60% of patients with DIL_45 plans without any violation of normal organ constraints.

Discussion
The present study compared three CK-SBRT plans in each patient with information on the locations, volumes, and numbers of the DIL. All patients who resulted in OAR constraint violations had posterior segment DIL. Using DIL_40 plans, 3 out of 4 cases for rectal D max violation showed the range of violations lower than 2 Gy. Two out of 3 cases for V 30.8 Gy > 1 cc were also exceeding range within 2 cc. We had to point out that the other study groups used more generous OAR constraint than the present study; rectal D max ranged from 36 to 38 Gy (The present study, 35 Gy), rectal V 32.3 -36 Gy < 1 cc (The present study, V 30.8 Gy < 1 cc), and bladder D max 41.8 Gy (The present study, 38.5 Gy), V 38 Gy < 1 cc or V 37 Gy < 10 cc (The present study, V 35.7 Gy < 1 cc) [15][16][17] . Kim et al. reported the dosimetric predictors of rectal tolerance observed in a phase 1-2 trial of dose-escalated SBRT with 45, 47.5, and 50 Gy in 5 fractions to the whole prostate 18 . Grade ≥ 3 late rectal toxicity was related to V 50 Gy > 3 cc, and > 35% circumference of rectal wall to 39 Gy. Grade ≥ 2 acute rectal complication was correlated with > 50% circumference to 24 Gy. In our study, dose irradiated to 50% of the rectal volume was below 18.5 Gy. For the DIL boost, we are going to establish revised OAR constraints with 2 Gy-relaxation for rectal D max, and V 30.8 Gy of rectum < 2 cc, which able 93% of the patients to be treated with DIL_40 plan. Since 70% of the prostate cancers arise in posterior/peripheral zone 19 , for the further dose escalation with DIL_45 plans, we consider methods for rectal sparing such as the use of spacer 20 .
The number and volume of DIL were not significant factors, and the location alone mattered. Four patients had two DILs. Two patients who one or more DIL(s) was/were located in anterior segments met all OAR constraints for all plans. On the other hand, the rest failed to satisfy one or more constraints. Volumes of the DIL were mostly below 3 cc in the present study, likewise in previous Aluwini et al. (mean DIL volume 1.2 cc, range 0.46-4.1) 13 . Only two patients had the DIL volumes of 3.7 and 6.6 cc, respectively, and all plans were satisfied with the prescription coverage and OAR constraints.
Similar to our study, Tree et al. compared the two methods of SBRT delivery, CK and VMAT. The dose scheme was 47.5 Gy to DIL while maintaining 36.25 Gy to the whole prostate in 5 fractions 15 . Both CK and VMAT planning produced clinically acceptable plans if the same PTV margins were applied. However, in case further margin for intra-fraction motion control was applied to the VMAT system, more violations of OAR constraints were observed in VMAT planning. They provided D50, D20, D10, D5 and hottest 1 cc of the rectum and the measures were comparable to those of our plans 15 .
We summarized the clinical outcomes of SBRT for prostate cancer in Table 4.8-13 Aluwini et al. 13 alone reported data on DIL boost in SBRT setting with a dose regimen of 38 Gy delivered to prostate and 44 Gy delivered to DIL in 4 fractions. With a median follow-up of 23 months, BCRFS was 100%, and ≥ grade 2 late GI toxicity was 3% and ≥ grade 2 late GU toxicity was 16%. Although 2-year outcomes were feasible, long-term outcomes are wanting. Also, we should be aware of a higher grade 3 GU toxicity rate of 6% compared to those of the other studies except for Hannan et al. 11 (0-2.4%). This might attribute to a higher dose per fraction (9.5 Gy) compared to the other series (mostly below 8 Gy per fraction). Hannan et al. 11 reported the results of dose-escalated SBRT (45-50 Gy in 5 fractions) different from the other studies usually utilized 33.5-40 Gy in 5 fractions. Five-year BCRFS was excellent especially in 47.5 and 50 Gy arms (100%), however, ≥ grade 3 late GI and GU toxicity rates were higher than those of the other studies (Table 4). It seems that selective dose-escalation for DIL in a five-fraction setting is a feasible option to increase the biochemical control without the increment of toxicities.
The main hurdle for DIL boost is common invisibility in imaging. Although multiparametric magnetic resonance images (mpMRI) has an advantage of excellent tissue contrast for identifying clinically prostate cancer 21 , tumors with small-volume or low Gleason score or certain histological architectures were less likely to be detected 22 . We attributed the low number of patients with a visible tumor (27.3%) in the present study to the reasons above. However, the rapid evolvement of imaging modalities has been performed. For example, positron emission tomography and computed tomography (PET-CT) using 11 C-labelled or 18 F-labelled choline are increasingly being used for primary and recurrent prostate cancer 23 . Simultaneous DIL boost contoured derived by choline PET-CT has also reported its feasibility 24 . More recently, PET-CT using prostate-specific membrane antigen (PSMA) labeled with 68 Ga was reported that had better contrast than that of choline PET-CT and could improve detection, localization of prostate cancer 25 . Go with the advances in imaging, the interest in DIL boost is expected to be increased.
The present study has a limitation of a dosimetric planning study with no clinical outcomes. However, a planning study must be preceded clinical study for the safety of the patients. Several on-going phase II studies are evaluating SIB to DIL in SBRT settings 26 , including hypo-FLAME study (NCT02853110) investigating the feasibility of 35 Gy in 5 fractions to the whole prostate with an additional boost to DIL up to 50 Gy. As the patient accrual was completed in 2018, it will take some time for maturing the data. Choosing a proper candidate is also an important issue, so we demonstrated that the location of the DIL was the main factor to do DIL boost with CK-SBRT, and to decide the dose regimen.
In conclusion, the DIL boost using CK-SBRT could be considered for localized prostate cancer patients. contouring. For DIL delineation, T2 and diffusion-weighted MRI images were fused with the planning CT using the Varian Eclipse treatment planning system. The location of the tumor positive biopsy site was also considered. No margin was put around the DIL. Whole prostate and proximal seminal vesicles were included in clinical target volume (CTV). PTV was a 2-3 mm expansion of CTV. Bladder and rectum were delineated according to the Radiation Therapy Oncology Group consensus guideline 28 . The urethra was defined as the outer contour of the Foley catheter.
planning. CK plans were generated using the ray-tracing dose calculation algorithm. For the MultiPlan planning system, sequential optimization planning was performed for each patient using a 6 MV flatteningfilter-free beam generated by the CK radiosurgery system (version 9.5

Data availability
The datasets analyzed during the current study are available from the corresponding author on reasonable request.