Assessment of Toxicity After 68Ga-PSMA-PET-Imaging Based Treatment Planning and Dose-Escalation in Radiation Therapy in Prostate Cancer Patients

PURPOSE: Effective tumor control in prostate cancer demands elevated radiation doses given its low alpha/beta ratio. We investigated in primary and recurrent prostate cancer whether dose-escalated-radiation therapy (DE-RT) based on 68 Ga-PSMA-PET positive lesions yields increased toxicity. METHODS: We evaluated 90 patients (salvage DE-RT: 67 patients, DE-re-RT: 11 patients, denite DE-RT: 12 patients) who were treated based on a pre-treatment hybrid 68 Ga-PSMA. Findings from pre-treatment hybrid 68 Ga-PSMA-PET could result in an adaption of radiation planning. Common Toxicity Criteria of Adverse Effects (Version 4.03) were used to assess toxicity, this was done before the initiation and at the end of DE-RT, as well as at the rst and last follow-up (F/U) examination. We evaluated the change in toxicity for each interval for the collective as well as in a per-patient analysis. RESULTS: Findings in 68 Ga-PSMA-PET-imaging resulted in a change of TNM-stage in 61.1% and an adapted treatment concept in 71.1% of patients. When comparing overall toxicity before DE-RT and at the last follow-up, 5.9% treatment-related side effects (grade 1-3) occurred with 1.7% of them being severe (grade 3). In a patient-centered approach we examined the intra-individual changes in toxicity before and after DE-RT. At the last F/U, the majority out of 80 patients (range: 61.3-93.8%) stated unchanged toxicity rates compared to the toxicity examined at the initiation of RT. CONCLUSION: The rate of treatment-related toxicity (grade 1-2) due to DE-RT in our cohort is 4.2%. For grade 3 toxicity it is 1.7%, respectively. The overall level of toxicity is highest during and shortly after completion of DE-RT (+7.4%) and improves over time until the last reported F/U (+5.9%). Compared to historical data, the toxicity prole of DE-RT is not increased. Therefore it is possible to apply DE-RT with the aim to increase tumor control by precisely treating all involved areas according to PSMA-PET-imaging.


Introduction
Prostate cancer (PC) is a major medical and socio-economic challenge worldwide [1]. It is the second most commonly diagnosed tumor in men with over 1.2 million new cases per year (as of 2018) [2].
At the same time, PC is a well treatable disease with a 5-year survival rate of 91% and a lifetime mortality risk of 3.3% [2]. The possibility of an earlier diagnosis through steady improvements in tumor diagnostics and the continuous improvements in treatment options have resulted in a decline in mortality in recent decades [1]. Nevertheless, PC ranks fth concerning cancer-related death among men worldwide [3].
Radiation therapy (RT) for PC, in the primary setting as well as in a relapsed situation relies crucially on accurate clinical as well as imaging-based tumor staging. To differentiate between local, regional, or systemic disease, pre-treatment staging provides the basis for further treatment decisions and enables risk group-adapted treatment. To improve the outcome of PC patients after RT (e.g. progression-free survival, quality of life) individual diagnosis and tailored therapy planning is of utmost importance. Yet, conventional imaging yields low sensitivity for detection of small PC cancer deposits [4]. In recent years 68 Ga-PSMA-PET-imaging has proven to be more accurate for PC staging compared to conventional imaging including Choline and FACBC-PET-imaging [4][5][6][7][8][9].
Therefore 68 Ga-PSMA-PET-imaging has become the preferred imaging modality in recurrent PC for localization of disease extent and therapy planning of PC according to European guidelines [10][11][12][13]. In the planning process of de nitive RT it has not yet become a standard tool [14]. With a very low alpha/beta ratio between 0.47 and 4.14, PC tumor control demands elevated radiation doses in order to annihilate all present cancer cells [15]. However, dose escalation entails an increased risk for side effects. Since most treatments are of curative nature, this demands an effective but equally tolerable treatment minimizing the likelihood of severe acute or chronic side effects. Moderate hypofractionation is a possibility to combine both aims as previously clinically shown [15][16][17][18][19][20].
The present work is to evaluate whether 68 Ga-PSMA-PET-imaging based dose-escalated (DE)-RT to all visible lesions entails increased toxicity for the patient or if these concept-and target volume changes are feasible with acceptable side effect rates in patients treated with de nitive or salvage RT for PC.

Methods
We evaluated 90 patients with histologically proven PC that were treated with DE-RT based on a 68 Ga -PSMA-PET/CT or 68 Ga -PSMA-PET/MRI between 2013 and 2016. 67 patients (74.4%) were treated with an intensi ed salvage RT concept after radical prostatectomy and 11 patients (12.2%) received an intensi ed second RT (DE-re-RT) after a previous resection and RT. Here patients received a RT of the pelvic lymph drainage including a boost to affected lymph nodes after a previous RT of the prostate bed as an annexation of the previous radiation volume. 12 patients (13.3%) received a de nite DE-RT. Patient characteristics are shown in Table 1. All patients were documented and followed prospectively within the institutional database. All gave written informed consent for the purpose of anonymized evaluation and publication of their data. All reported investigations were conducted in accordance with the Helsinki Declaration and with national regulations.
This analysis from the database was approved by the Ethics Committee of the Technical University of Munich (TUM) with the permit 121/17S.
Findings from pre-treatment hybrid 68 Ga-PSMA-PET could result in an adaption of radiation planning [11] compared to results of standard imaging and clinical risk factors (TNM classi cation, Gleason score, risk according to Roach [21] and D'Amico [22]).
All cases were discussed in consensus in an interdisciplinary tumor board consisting of at least one experienced radiation oncologist, radiologist, nuclear medicine physician and urologist.
In our department the radio-oncological planning target volume (PTV) is generated according to the RTOG consensus. The exact de nition of clinical target volume (CTV) and PTV in the context of de nite RT [14] and salvage RT [23] has been published. Depending on the additionally obtained results of the 68 Ga-PSMA-PET the RT concept and volume were adapted. For patients who received a salvage or a re-RT all PSMA-PET positive areas were treated with a dose escalation. For patients who received a de nite RT the prostate was treated with a uniform dose and dose-escalation was only applied to PSMA-PET-positive lymph nodes. The exact dosage and fractionation are shown in Table 2. All patients received an intensity-modulated radiotherapy (IMRT). RT was performed on one of the two types of linear accelerators (Clinac® radiation devices (Varian® Medical systems Incorporated, Palo Alto, CA, USA)) and the TomoTherapie® radiation device (Accuray® Incorporated, Sunnyvale, CA, USA). The treatment was performed with a reproducible comfortably lled bladder and empty rectum under daily image guidance.
We collected data for toxicity before the initiation and at the end of DE-RT, as well as at the rst and last follow-up (F/U) examination.
Generally, F/U visits were scheduled at 6 weeks post radiation, then time intervals were prolonged to three months and six months before continuing on an annual basis. In the case of unexpectedly increased side effects, shorter intervals were scheduled by the treating physician. Documentation and classi cation were done according to American National Cancer Institute-Common Terminology Criteria for Adverse Events (CTCAE), Version 4.03 [24], wherever applicable.
Missing data was obtained by contacting the outpatients' urologist whenever possible.
All toxicity that occurred within three months after completion of RT was classi ed as acute and all toxicity that occurred thereafter was classi ed as chronic.
We evaluated the change in toxicity for each interval between the initiation of DE-RT, the rst F/U visit and the last F/U visit. In order to calculate each value, the percentage of patients showing a side effect at the earlier visit was subtracted from the percentage of the patients showing the same side effect at the later visit. Furthermore, we conducted a per-patient-analysis of the intra-individual changes in toxicity over time using a log-rank test. All statistical analyses were performed descriptively with exploratory intention using proportions and means.  Table 3 shows ndings in 68 Ga-PSMA-PET-imaging.

Per-patient analysis
In a per-patient analysis we examined the intra-individual changes in toxicity over time. At the last F/U, the majority out of 80 patients (range: 61.3-93.8%) showed unchanged toxicity rates compared to the toxicity examined before the initiation of DE-RT.
The level of proctitis remained unchanged in 66 patients (73.3%). It increased by up to three degrees in 11 patients (13.8%) and decreased by two degrees of toxicity in three patients (3.8%).
The level of diarrhea remained unchanged in 68 patients (85.0%). It increased by a maximum of one degree in nine cases (11.3%). An improvement of up to three levels of toxicity was reported in 3 cases (3.8%) (Fig. 1).
Before the start of RT, 19 (21.1%) patients reported cystitis; 80% of patients showed no change concerning this toxicity at the time of the latest F/U. An increase of up to three degrees of severity occurred in seven cases (8.9%), simultaneously there was a decrease of up to two degrees of severity in 9 cases (11.3%) (Fig. 2). Stress incontinence (up to grade 2) was documented in 34 cases (37.8%) before RT. Up to the last F/U examination, there was no change in the degree of side effects in 60 cases (75%). For urinary urge and nocturia 17 (18.9%) respectively 59 (65.6%) patients showed at least grade 1 toxicity before the RT. Here unchanged rates were seen in 68.8% and 61.3% of the cases (shown in Fig. 2). At the time of the last F/U examination, unchanged erectile dysfunction was reported in 55 cases (68.8%). In eight patients (10.0%) the potency in the case of normal erectile function before radiotherapy was completely gone at the time of the last F/U examination (shown in Fig. 3).
Discussion conducted a 68 Ga-PSMA-PET planned DE-RT study with the aim of increasing tumor control while achieving acceptable side effects. In this paper the radiation induced toxicity is highlighted.
In our investigated group of 90 patients that received a 68 Ga-PSMA-PET planned DE-RT, acute side effects showed an inhomogeneous distribution with regard to the degree of manifestation of GI and UG toxicity. In 78.6 % of the patients no or only mild GI side effects up to grade 1 were found. Acute, above baseline, grade 3 UG toxicity was found in 6.7%. A high number of patients reporting erectile dysfunction (58 cases) can be explained by the inclusion of patients with previous radical prostatectomy in 42 cases and ongoing ADT in 48 cases. These were therefore most likely not treatment-related. Aside from erectile dysfunction no grade 3 UG toxicity is reported. 57.3% of the patients developed grade 2 toxicity and 42.7% of the patients presented with no or only mild grade 1 toxicity. Concerning chronic GI side effects only 2 cases of grade 3 proctitis were seen.
In order to answer the question if DE-RT concepts, as used in our study, come at an increased risk for patients concerning side effects, we looked at reported outcomes of conventionally planned and normo-fractionated RT studies. Abundant comparative information on toxicity rates for acute UG side effects was collected in the context of hypo-fractionated studies such as the Italian- [25], the CHHiP- [26] or the HYPRO-Trial [27][28][29]. Acute UG toxicity of at least grade 2 were observed in the conventionally fractionated arm in 40% − 58% of patients. This is comparable to the range of our numbers.
In a retrospective, two-institutional analysis 90 patients with biochemical recurrence, who received 68 Ga-PSMA-PET-CT before DE-RT were examined. Patients 1.8 Gy) to the prostate and affected lymph nodes (lymphatic drainage: 60 Gy) showing promising results: After a median F/U of two years, there was an excellent tolerance of the RT. Acute grade 1 or 2 GI toxicity was found in 48%/11% of the patients, and UG toxicity in 72%/24%. Late grade 1-3 GI toxicity was still present in 13%/2%/0% of the cases and UG toxicity in 28%/13%/4%. None of the toxicity correlated with dose escalation to PET positive lymph nodes [32].
Despite different numbers of patients, different dosing concepts and larger variety of treatment indications, our numbers are also comparable to other DE-RT studies and show even lower acute grade 1 and 2 toxicity (GI: 10.1%/18%, UG: 14.6%/18%) as well as mostly the same late grade 1-3 toxicity (GI: 10%/15%/2.5%, UG: 12.5%/15%/71.3%). Again, our UG grade 3 toxicity must be understood in the context of the already pre-existing erectile dysfunctions.
Examining the change of side effects, above baseline, before the initiation of RT to the rst and last F/U more closely, for the majority of the toxicity reported we found an initially quick deterioration, followed by a slower recovery to an overall comparably unchanged end result. This means that the increase of side effects attributable to the radiation itself is non-existent to mild.
On average, at the rst F/U 7.4% of the toxicity appeared for the rst time due to DE-RT. Most of it being grade 1 and 2 (6.6%). An amelioration over time occurred so that at the last F/U 5.9% more patients reported a side effect that was not present before RT. This was mainly caused by proctitis (9.9%), urinary urge symptoms (9.9%) and diarrhea (9.1%).
Cystitis, being one of the most common radiogenic side effects in pelvic irradiation, is a good example of illustrating worsening and improvement over time: At the rst F/U an increase of 16% was seen, but ultimately at the last F/U the rate had completely returned to the baseline.
As part of the ProtecT-study, the post-therapeutic GI and UG side effects and quality of life were assessed and compared six months, twelve months and every year after active surveillance, RPE or de nite RT among 1,643 patients with localized PC. The severity and recovery time differed in all three groups. After RT the sexual function recovered steadily from six months after RT and stabilized over the course of the six years; also, RT had little effect on urinary continence.
In contrast, bowel function, urinary retention and nocturia were worst six months after RT, while it was unchanged in the other groups. Over the course of time, however, most of the intestinal and micturition function stabilized after RT [33]. In contrast to the ProtecT study, our patients had multiple RT indications (postoperatively, re-salvage next to de nite), thus a different history and other symptoms at the beginning of the RT. Nevertheless, the side effects after our DE-RT behaved comparably to the conventionally fractionated RT in the ProtecT study. In particular, GI toxicity, nocturia and urinary urge showed the described rapid deterioration with slow recovery towards a stable result. With a shorter F/U period than the ProtecT study (6 years) up to now, a further improvement in our cohort could be expected.
The Cochrache analysis by Daly et al. supports these results. The review of three randomly controlled studies showed increased acute and late GI side effects, urinary strictures and incontinence after adjuvant RT compared to only RPE, but in the long run there was hardly any moderate to severe toxicity [34].
There are limitations to our study. We carried out a retrospective study with an inhomogeneous patient population. Our patients were in different stages of their illness, had different treatments beforehand and therefore different side effect pro les. This includes anti-hormonal therapy, which was taken by part of our patients throughout the timeframe of this study. On the other hand, this study represents a broad overview and a real-life-collective.
Also, to evaluate the superiority of 68 Ga-PSMA-PET-planned DE-RT, prospective studies would be favorable. A comparison with a conventionally treated control group would be desirable, but essentially impossible because of ethical and patient preference reasons.
Last, an analysis of biochemical progression free survival of this cohort is to follow.

Conclusions
For acute and chronic toxicity rates, as well as for change in toxicity analysis, we see the majority of patients reporting no or only mild toxicity. The rate of new appearing toxicity in our study population is below 5% for grade 1-2 and below 2% for grade 3.
In most cases, DE-RT could be performed without an increase in long-term toxicity and we postulate 68 Ga-PSMA-planned DE-RT to be tolerable with acceptable side effects.
In summary, we see great potential for 68 Ga-PSMA-PET planned DE radiation concepts. It is possible to use essentially high radiation doses for tumor control by precisely treating all involved areas and thereby potentially sparing uninvolved areas from unnecessary toxicity. Compared to existing literature, we see that radiation volumes with elevated radiation dose (boost) to affected lymph nodes or within the prostate region are not producing more acute and long-term toxicity throughout all examined gastrointestinal and urogenital parameters and could therefore be used safely on a more regular basis. Finally, and seen in increasingly more studies, PSMA-PET planned DE radiation concepts and volumes can potentially translate into improved local control and/or overall survival for our patients. This will be analyzed for this study cohort of patients in a study to follow.

Declarations ETHICAL APPROVAL AND CONSENT TO PARTICIPATE
This analysis from the database was approved by the Ethics Committee of the Technical University of Munich (TUM) with the permit 121/17S.

CONSENT FOR PUBLICATION
All authors agree for publication.

AVALABILITY OF SUPPORTING DATA
The data and material can be accessed by contacting the corresponding authors

COMPETING INTERESTS
On behalf of all authors, the corresponding author states that there is no con ict of interest.

FUNDING
Not applicable AUTHORS' CONTRIBUTIONS KS made substantial contributions to conception and design, analysis and interpretation of data, involved in drafting the manuscript, revised it critically for important intellectual content, gave nal approval of the version to be publish; agrees to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
SD made substantial contributions to conception and design, analysis and interpretation of data, involved in drafting the manuscript, revised it critically for important intellectual content, gave nal approval of the version to be publish; agrees to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
LP made substantial contributions to conception and design, acquisition of data and analysis; agrees to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
MMEV revised it critically for important intellectual content, gave nal approval of the version to be publish; agrees to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
ES revised it critically for important intellectual content, gave nal approval of the version to be publish; agrees to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
MD revised it critically for important intellectual content, gave nal approval of the version to be publish; agrees to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. ME revised it critically for important intellectual content, gave nal approval of the version to be publish; agrees to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
JEG revised it critically for important intellectual content, gave nal approval of the version to be publish; agrees to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
SEC revised it critically for important intellectual content, gave nal approval of the version to be publish; agrees to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.