Introduction

Prostate cancer is one of the predominant malignancies in men throughout the western world. Radiotherapy and prostatectomy are the main interventions applied in patients with a life expectancy long enough to justify possible treatment-related side effects and long-term sequelae. Generally, there is a broad consensus that both modalities offer a similar chance of a cure but due to the different profiles of side effects and impact on functional domains, they have their specific pros and cons; therefore, counselling of patients is much more time-consuming compared to other malignant diseases, where the superiority of a respective treatment option is unquestioned. One of the major drawbacks of external beam radiotherapy is the long time span needed to deliver a complete course of radiotherapy, usually amounting up to 2 months.

Mainly driven by a shortage of treatment facilities and/or long travelling distances in countries with healthcare systems providing fewer but larger therapeutic units, hypofractionation, i. e. shortening of overall treatment time by delivering larger doses per fraction up to a lower total dose, has attracted growing interest. In treating women with breast cancer, moderate hypofractionation (i. e. daily doses of approximately 2.7 Gy) is nowadays broadly accepted as an alternative to standard fractionation [1–3]. Prostate cancer as a relatively slowly growing malignancy shows a better prognosis than many other tumors, necessitating very long follow-up times to evaluate the safety profile of therapeutic modifications in terms of disease control as well as side effects; therefore, any adoption of new treatment concepts has to be scrutinized to a high degree. Especially in the last few years, a rapid increase in reports on hypofractionated radiotherapy for prostate cancer has been noted, which prompted the present overview, supplementing a recently published systematic review [4] and including guidance for daily practice.

Biology

Traditionally, external beam irradiation regimens have been developed over several decades and the mode of application, i. e. doses per fraction of 1.8–2.0 Gy given 5 times per week up to total doses exceeding 70 Gy have been shown to be safe, with severe side effects being very rare events. Most cancers and normal tissues behave differently when exposed to radiation. The linear-quadratic equation serves as a biomathematical model commonly applied to describe fractionation sensitivity of tissues and to calculate isoeffective doses for different doses per fraction. Tissue-specific α/β values derived from this model can be estimated from clinical and preclinical data. As almost every human organ is composed of different tissue types, the α/β values may be different for distinct endpoints evaluated within the same organ, emphasizing the need for cautious interpretation when testing new fractionation schemes.

Retrospective data derived from different modes of radiotherapy application and fractionation initially suggested very low α/β values for prostate cancer in the range of 1.5 Gy, i. e. lower than the α/β values of the surrounding dose-limiting normal tissues. These data led to the hypothesis that hypofractionation improves the therapeutic ratio for radiotherapy of prostate cancer. Based on this hypothesis randomized trials were initiated. The controversial discussion on fractionation sensitivity of prostate cancer is further complicated by the existence of a time factor [5]. Prostate cancer is often a slowly growing tumor which may predict a low α/β value (and hence a high fractionation sensitivity) and a negligible time factor (loss of effect by increase in overall treatment time); however, a large retrospective analysis on external beam radiotherapy for prostate cancer comprising 4839 patients demonstrated a significant and clinically relevant time factor of 0.24 Gy/day with a 95 % confidence interval of 0.03–0.44 Gy/day [6]. Of note, the effect of androgen deprivation on the time factor is unknown. Although the time factor for prostate cancer is not as pronounced as in other tumor types, it has important implications not only for trial design but also for clinical practice. As often with hypofractionation, the overall treatment time is shorter than for conventional fractionation, thus the suspected superiority of hypofractionation might not only be explained by fractionation sensitivity or in other words the α/β value for prostate cancer might be higher than initially suspected [7]. This is of high relevance for the design of biologically driven fractionation schedules. The lack of superiority of hypofractionation observed in the Fox Chase trial [8, 18] does not support the assumption of a very low α/β value for prostate cancer but suggests that the fractionation sensitivities of prostate cancer and the dose-limiting surrounding normal tissues overlap.

Given the contradictory and lacking evidence the German Society of Radiation Oncology (DEGRO) expert panel drew the following conclusions:

  • The assumption of an α/β value for prostate cancer of as low as 1.5 Gy might lead to an overestimation of the effects of hypofractionation.

  • Fractionation sensitivity of prostate cancer and surrounding tissues does not decisively differ; therefore, hypofractionated and accelerated study designs with a reduced total dose seem promising and may be possible without any deterioration of the therapeutic ratio.

  • The time factor may contribute in part to the efficacy of hypofractionation.

  • Due to the described uncertainties in assuming fractionation sensitivities and the steep dose-response effects for tumor control and for normal tissue toxicity, fractionation concepts have to be tested in well-designed randomized trials, such as the CHHiP, HYPRO and the RTOG 0415 studies and should not be mathematically derived.

  • Only evidence-based fractionation schedules should be used outside of clinical trials.

  • Treatment interruptions leading to prolonged overall treatment times can have negative consequences and have to be adequately compensated.

Technology

The technological basis for external beam radiotherapy has continuously developed over the last 2 decades and a modern standard is now broadly available. A highly conformal radiotherapy, e. g. intensity modulated radiotherapy (IMRT), with daily verification of the prostate position by image-guided radiation therapy (IGRT) is the prerequisite for all hypofractionation concepts. The target volume concept in radiation therapy of prostate cancer forms the basis for understanding the reduction in normal tissue complication probability (NTCP) by IGRT. Due to histological multifocality of prostate cancer, the target of radiotherapy is the entire prostate gland and also a subclinical infiltration zone around the prostate needs to be considered carefully to improve the probability of tumor control [9].

As radiation therapy is fractionated in clinical practice, i. e. applied in small daily doses over several weeks a further safety margin, the planning target volume (PTV) needs to be defined around the clinical target volume (CTV). The PTV includes possible positioning errors of the CTV by potential inaccuracies in the daily set up, organ movement from day to day (interfractional organ movement) and during beam application (intrafractional organ movement). By definition, this safety margin within the PTV around the CTV contains only normal tissue. Another mandatory factor for hypofractionated radiotherapy of prostate cancer is a daily verification of target volumes immediately before irradiation. By visualizing the localization of the prostate, the positioning error and the interfractional organ movement can be widely corrected, thus reducing the PTV in the ideal case without intra-fractional movement to the CTV. Intrafractional movement can be minimized by fast beam application, e. g. rotational IMRT and/or using high dose rate flattening filter free (FFF) beam delivery. Institutional IGRT protocols may specifically account for intrafractional corrections [10]. As a result of these technical developments, the volume of normal tissue receiving high doses and thus the NTCP might be reduced. In the clinical routine, several IGRT methods have been established. Especially in the case of prostate cancer, IGRT procedures should allow 3‑dimensional imaging with soft tissue contrast or fiducial-based techniques (intraprostatic markers) should be applied.

Moderate hypofractionation

Defining standard fractionation with single doses of 1.8–2.0 Gy is easier than defining the dose commonly regarded as hypofractionation. For purposes of this review, the arbitrary definition of moderate hypofractionation with doses per fraction in the range of 2.2 up to 4.0 Gy and extreme hypofractionation with a single dose beyond 4.0 Gy has been chosen.

Data from eight randomized trials on moderate hypofractionation with appropriate sample size are available (table 1; [8, 11–25]). Single doses ranging from 2.4 Gy to 3.4 Gy and total doses from 52.5 Gy to 72.0 Gy were applied in the experimental arms. The largest study by far, the CHHiP trial with more than 3200 patients included was first partially published as a subset of 457 patients [17] and just recently 5‑year follow-up data became available [16]. Two other large studies, the RTOG 0415 and the HYPRO trial have also just been published in detail [11, 15]. The studies from Yeoh et al. [20, 26] and Lukka et al. [12] are of limited interest, as total doses applied in each of the study arms nowadays would be regarded as far below established standards of care. This is true for the techniques used in these studies as well. Some of these studies did use or at least permit hypofractionation in conjunction with simultaneous integrated boost techniques (e. g. CHHIP and HYPRO) and only very rarely examined the treatment of the pelvic lymph nodes with hypofractionated external beam radiotherapy [8, 18].

Table 1 Prospective randomized studies on moderate hypofractionated external beam radiotherapy for prostate cancer
Full size table

Toxicity

Whereas the median follow-up might already be sufficient to estimate outcome in terms of side effects, this is not true for the primary endpoint of disease control where longer follow-up is needed. Some caveats exist. In the HYPRO study, cumulative acute gastrointestinal (GI) toxicity grade 2 and worse was significantly increased (42 % vs. 31.2 %, p = 0.0015) in the hypofractionated arm, leading to the statement that hypofractionated radiotherapy was not non-inferior in terms of acute side effects [14]. In the gastrointestinal subitems evaluated, there was a marked difference with respect to increased stool frequency ≥6 times a day (15 % vs. 8 %, p = 0.0035) and in pain needing drugs (9 %vs. 5 %, p = 0.021). In genitourinary (GU) toxicity there was no difference in general but in the subitem increased frequency at night more than 7 times (grade 3) there again was a significant increase in the hypofractionation group (12 % vs. 7 %, p = 0.019). Late toxicity of the HYPRO study has been recently reported [13]. Grade 2 or worse GU toxicity at 3 years was increased (hazard ratio HR 1.16) from 39.0 % to 41.3 % and grade 2 or worse gastrointestinal toxicity at 3 years increased from 17.7 % to 21.9 % (HR 1.19). Especially cumulative grade 3 or worse late GU toxicity was significantly higher with an increase from 12.9 % to 19.0 % (p = 0.021). The subitems that were of special concern were nycturia ≥6 (1 % vs. 6 %, p = 0.0005), incontinence (14 % vs. 20 %, p = 0.04) and stool frequency ≥6 (3 % vs. 7 %, p = 0.034). Thus, the authors again had to state that with respect to late toxicity non-inferiority could not be shown. As there was no significant difference, neither in 5‑year relapse-free survival nor in treatment failure, the authors concluded that their hypofractionated radiotherapy regimen could not be regarded as the new standard of care [15].

With regard to acute side effects the results of the CHHiP trial pointed in the same direction as there was a statistically significant increase in acute grade 2 or more GI toxicity in the two hypofractionated arms of the trial (25 % vs. 38 % p < 0.0001) [16]. The 5‑year biochemical or clinical failure-free survival rates were in a relatively narrow range from 85.9 % for the 57 Gy in 19 fractions regimen up to 90.6 % for 60 Gy in 20 fractions and the conventionally fractionated radiotherapy regimen lying just in between with 88.3 %. Due to the predefined hazard ratios (HR) for the 57 Gy regimen, non-inferiority in comparison to standard fractionation could not be claimed, whereas for the 60 Gy regimen non-inferiority was documented. At 5 years there were no significant differences in grade 1, grade 2, grade 3 or worse bowel, bladder or sexual symptoms.

The Fox Chase and the HYPRO studies noted an increased risk for late GU toxicity in patients with impaired urinary function prior to the commencement of radiotherapy [15, 18, 27]. For other outcome domains of interest, such as sexual functioning and well-being [28, 29] there are not enough data available yet.

In the RTOG 0415 study, no differences in early GI or GU adverse events were observed. Late grade 2 and 3 GI adverse events were approximately 60 % more likely in men who were assigned to treatment with hypofractionated RT (HR, 1.55–1.59). Similarly, late grade 2 and 3 GU adverse events were more likely in men treated with hypofractionated radiotherapy (HR, 1.31 to 1.56). No differences in more severe events were observed [11].

Of note, both the HYPRO and RTOG studies might have used higher biologically effective doses in their experimental (hypofractionated) arms than in the control cohorts. Both the HYPRO and the RTOG trial were based on the assumption of α/β values of 1.5 Gy for prostate cancer. In the RTOG trial, even if assuming an α/β of 2 Gy for prostate cancer, the experimental arm resulted in an EQD2 (equivalent dose for a 2 Gy fraction) of 79 Gy vs. 70 Gy for the standard treatment. Likewise, in the HYPRO study, the respective groups were treated with an EQD2 of 87 Gy vs. 78 Gy to the prostate. Assuming an α/β value of 3.0 Gy for late rectal reactions, the EQD2 values would have been 6 Gy higher in the experimental arms of both trials and the observed differences in toxicity rates may be also attributable to biologically effective higher doses in the HF arms (see Fig. 1a–d), where the different fractionation schemes of the most relevant phase 3 studies are compared on basis of α/β values ranging from 1 Gy to 10 Gy.

Fig. 1
figure 1

Major randomized trials on moderate hypofractionation in prostate cancer. Biologically effective doses (BED) are depicted and compared for different alpha/beta values. The assumptions made in the study protocols on alpha/beta values for prostate cancer and organs at risk are shown in vertical lines. OTT overall treatment time, IMRT intensity modulated radiotherapy, SIB simultaneous integrated boost, 3D-CRT 3-dimensional conformal radiotherapy

Full size image

Hypofractionation in adjuvant and salvage treatment

Data on moderate hypofractionation in postoperative or salvage radiotherapy of prostate cancer are sparse and the few reports available are retrospective in nature. As there are already reports on unexpected high rates of up to 28 % grade 3 late GU toxicity [30, 31] it seems very wise to abstain from hypofractionation in the postoperative setting outside carefully designed clinical trials, keeping in mind that in this situation the target volume irradiated mainly consists of normal tissue.

Current trials in moderate hypofractionation

As already mentioned, there are several large scale trials that have already completed accrual and published results and hopefully in the near future more details especially subgroup analyses will be presented. The collaborate OCOG/TROG PROFIT Prostate Fractionated Irradiation Trial (NCT00304759) testing 78 Gy/39 fractions in 8 weeks vs. 60 Gy/20 fractions in 4 weeks (i. e. the same regimen as the 20 fractions schedule of the CHHiP trial) for intermediate risk prostate cancer has completed its accrual of more than 1200 men and first results were reported at the American Society of Clinical Oncology Annual Meeting 2016 [32]. Clinicaltrials.gov. lists the Canadian trial on Hypofractionated Dose Escalation Radiotherapy for High Risk Adenocarcinoma of the Prostate 76 Gy/38 fractions vs. 68 Gy/25 fractions (NCT01444820) and the closed MD Anderson phase III study (NCT00667888), comparing 75.6 Gy/42 fractions vs. 72 Gy/30 fractions with no further details. Some insight into the outcomes of extreme vs. moderate hypofractionation may be derived from the HEAT trial, comparing 70.2 Gy in 26 fractions vs. 36.25 Gy in 5 fractions (NCT01794403).

Summary on moderate hypofractionation

There is a growing body of evidence that modern moderately hypofractionated regimens are safe and non-inferior to conventional fractionation in terms of clinical and biochemical recurrence-free survival and late toxicity. Special precaution is needed for particular subgroups of patients with a substantially higher risk of acute GI toxicity and late GU toxicity who apparently are not eligible candidates for hypofractionation. An important caveat in the assessment of hypofractionation in prostate cancer remains the fact that although supportive evidence is growing, the present review and recommendations are limited by the fact of still immature data; therefore, the DEGRO expert panel strongly recommends:

  1. 1.

    To restrict the use of moderate hypofractionation to high-end techniques including IGRT and IMRT in carefully selected patients and to adhere to published phase 3 protocols with documented safety and efficacy. The CHHiP regimen with 60 Gy in 20 fractions over 4 weeks or the RTOG regimen with 70 Gy in 28 fractions over 6 weeks at present seem to be the first choices, although it has to be kept in mind that in the CHHIP trial equivalency was shown only in comparison to 74.0 Gy in standard fractionation, which might be regarded a rather low dose for the patients treated, who predominantly had intermediate risk disease.

  2. 2.

    Meticulous follow-up and documentation of outcome and late toxicity are mandatory.

  3. 3.

    Hypofractionated radiotherapy of the pelvic lymphatic vessels is experimental and should not be carried out except in clinical trials.

  4. 4.

    Hypofractionated radiation therapy in postoperative and in salvage situations is experimental and should not be carried out except in clinical trials.

Extreme hypofractionation

Extreme hypofractionation with single doses of >4–10 Gy up to total doses of 35–50 Gy, is most often applied with stereotactic body radiotherapy (SBRT) techniques. Throughout the last 5 years, a growing number of phase I/II studies as well as retrospective analyses of extreme hypofractionation schedules have been published, with median follow-up periods of roughly 6 years at maximum [33–35]. Four to five fractions of single doses between 7 Gy and 10 Gy have been used, with estimated biologically equivalent doses of up to an EQD2 of 164 Gy [36] applied in only 1–2 weeks; therefore, these regimens are also frequently termed stereotactic ablative radiotherapy (SABR). These very high doses necessitate small PTV margins and utmost precision with respect to both interfraction and intrafraction motion of the prostate. This can be achieved with live tracking image guidance strategies based on fiducials or on electromagnetic beacon transponder technologies. In most of these reports dedicated robotic radiosurgery units were used; alternatively, IGRT-IMRT on specially equipped linacs was performed.

Clinical evidence

In the vast majority these studies included only low-risk and intermediate-risk patients. For this selections early cancer outcomes have been excellent, with unanimously reported bNED rates of > 90 % up to 100 % in short and mid-term follow-up. In the largest single institution study with a 72-month median follow-up, Katz and Kang. [35] treated 515 patients to 35–36.25 Gy in 5 fractions. Low-risk patients showed a freedom from biochemical failure rate of 95.8 % at 7 years. Results for intermediate and high-risk patients were 89.3 % and 68.5 %, respectively. Loblaw et al. [37] treated only low-risk prostate cancer patients and had a similar follow-up of 55 months. The authors noted a 5-year biochemical control of 98 %, also confirmed in posttreatment biopsies in 85 % of the patients. Out of these, only 4 % were classified positive for residual cancer 3 years after treatment. These excellent results are comparable to a high-dose IMRT series obtained in a phase 3 trial for low-risk disease [38].

An overview of trials comprising of at least 50 patients and results following extreme hypofractionation is provided in table 2 [33–35, 37, 39–51]. Additional evidence is provided by the Registry for Prostate Cancer Radiosurgery (RPCR) with an analysis of the largest cohort so far treated by extreme hypofractionated SBRT [43]. Between 2010 and 2013, almost 1750 men from 45 participating sites were enrolled, with the majority (86 %) receiving SBRT as monotherapy. At 2‑year follow-up, biochemical disease-free survival amounted to 92 %, which is in line with the few prospective series. The data presented by Freeman et al. [43] are to some extent limited by incomplete data entry into the database and thus an inherent risk of underreporting of outcomes that were undesired. To what extent these registry data include patients already reported on in other series remains another open question.

Table 2 Major prospective studies on extreme hypofractionated external beam radiotherapy for prostate cancer
Full size table

Toxicity

Moderate to high-grade acute toxicity from extreme hypofractionation ranges between 10 % and 20 %, with urinary symptoms more common than those related to the bowels and rectum. Late grade 2 toxicity rates from these individual institutional experiences vary significantly (1–31 %), with grade 4 toxicity occasionally reported [36, 37, 49]. While urinary incontinence is uncommon in conventional fractionation, in one study it was reported to be as high as 10 % in previously continent men 3 years post-SBRT [52]. Many of the toxicity rates published are crude rates, not taking into account patients lost to follow-up and thus actuarial rates may be in fact higher than reported [4]. Earlier studies have noted particularly high urinary toxicity rates, with late grade 2 rates of up to 30–40 % [50, 53, 54], pointing at the necessity of high quality levels in performing marker-based, image-guided SBRT. The hypothesis of an α/β ratio for urethra and bladder that is lower than commonly assumed builds a caveat towards unexpectedly increased late toxicity. In a large retrospective case control study of Medicare claims, 1335 extreme HF-SBRT patients were matched to 2670 CF-IMRT patients and 2 years posttreatment, higher late GU toxicity events were noted in the hypofractionated group (43.9 % vs 36.3 %; p = 0.001) [53]. The increase in GU toxicity was due to claims indicative of urethritis, urinary incontinence and/or obstruction. Treatment on non-consecutive days may reduce late toxicity. One study found lower rates of grade 1–2 urinary (17 % vs 56 %; p = 0.007) and rectal (5 % vs 44 %; p = 0.001) toxicity with a regimen of every other day versus daily dosing [33].

A comprehensive overview of the late GI/GU toxicity rates as a function of prescription dose is provided by Koontz et al. comprising the majority of extreme HF study results published at that time [4]. Several of the studies have reported grade 4 GI toxicities (i.e. colostomy) at high total doses. In 1 study patients were treated with 5 fractions of 7 Gy and 2 studies used 9–10 Gy per fraction. An update of a dose-escalation study performed by Boike et al. [36] noted a 2-year actuarial rate of 8 % high-grade GI toxicity in 61 patients receiving 50 Gy in 5 fractions [55] and 5 patients required a diverting colostomy at a median time of 9.5 months. The volume of rectum receiving 50 Gy was highly significant on multivariate analysis. Katz et al. [35] noted decreased grade 2 GU toxicity in patients receiving 35 Gy rather than 36.25 Gy in 5 fractions (5.7 % vs 10.6 %) and overall grade 2 or higher toxicity was significantly higher after 36.25 Gy (p = 0.05). Another study noted increased high-grade rectal toxicity with increasing dose, recommending <50 % of the rectal volume receiving 4.8 Gy per fraction [55]. In a series of 204 patients reported by Rana et al., HF-SBRT with 35–36.25 Gy resulted in an acute increase in irritative urinary symptoms that peaked within the first month posttreatment. Irritative voiding symptoms returned to baseline in the majority of patients by 3 months post-SBRT and were actually improved from baseline at 3 years post-SBRT [56]. In the RPCR cohort, no grade 3 late urinary toxicity was reported. One patient developed grade 3 gastrointestinal toxicity (rectal bleeding). Erectile function was preserved in 80 % of men <70 years old [43].

Current trials of extreme hypofractionation

Currently, three phase 3 trials of extreme hypofractionation are active [4]: the Scandinavian Hypofractionated Radiotherapy of Intermediate Risk Localized Prostate Cancer trial (HYPO-RT-PC; ISRCTN45905321) randomizes intermediate-risk men to 42.7 Gy in 7 fractions versus 78 Gy in 39 fractions. The Prostate Advances in Comparative Evidence (PACE) trial, active in multiple European centers, will randomize 1036 men to (1) robotic surgery or prostate SBRT or (2) moderate versus extreme hypofractionation with 5‑year biochemical progression-free survival as the primary end point (NCT01584258). A similar approach is tested by the RTOG in a phase II randomized multicenter trial to assess quality of life outcomes, acute and late toxicity of two different extreme hypofractionated regimens: 36.25 Gy in 5 fractions of 7.25 Gy twice a week and 51.6 Gy in 12 daily fractions of 4.3 Gy (RTOG 0938, NCT01434290). Finally, the Proton Cooperative Group is randomizing 192 patients in a phase 3 study of 79.2 Gy in 44 fractions or 38 Gy in 5 fractions, using a primary end point of 5‑year freedom from failure (NCT01230866). All these studies are still recruiting, and results are pending.

Summary

In selected non-randomized cohorts, clinical outcome following extreme hypofractionation regimens for low-risk prostate cancer shows good short and mid-term biochemical control up to 5 years, well comparable with current conventional high-dose fractionation. The American Society for Radiation Oncology recently released their policy on SBRT stating that while longer outcome is still necessary, it is regarded suitable to offer SBRT to selected low and intermediate-risk prostate cancer patients; however, in the light of the reports of higher grade urinary and rectal toxicity and in the absence of long-term experience derived from randomized controlled trials, the DEGRO expert panel strongly discourages its use outside prospective clinical protocols.

Hypofractionated radiotherapy with protons and ions

Prerequisites

Particle therapy of prostate cancer is currently not routine procedure. As there are potential benefits due to the physical differences in dose distribution compared to photon therapy, there is an increasing interest. Physically, it is possible to focus the particles in the tumor while minimizing the dose to the surrounding healthy tissue; therefore, the beneficial dose distribution might reduce the risks of long-term toxicity (assuming similar fractionation and dose concepts). Also, a reduction of radiation-associated secondary cancers is postulated due to the reduced dose in surrounding tissues; however, in clinical routine the same requirements are necessary as for photon therapy: in particular, a daily verification of target volumes immediately before irradiation (image guided proton therapy, IGPT) is essential. As the range of protons is very sensitive to the density of irradiated tissue, the set-up alignment focuses on bony structures in the beam; therefore, in contrast to photon therapy, the interfractional movement of the prostate cannot be easily corrected. This can only be compensated partly by careful preparation of the patient (e. g. rectal balloon, fiducials).

From the radiobiological point of view, there might be (after application of correction factors) probably little difference between radiotherapy for prostate cancer with protons or photons. In the hypofractionated setting this might be an issue, especially in Germany with respect to approval and regulations [57].

Clinical data

The information on carbon ion therapy for prostate cancer is very limited. Many of the reports originate from centers with much experience in particle therapy, mainly from Japan but are retrospective in nature and only very few cover a number of patients that will allow valid conclusions to be drawn [58, 59]. Prospective studies are usually of even smaller size [60]. Data on proton therapy are more robust, and proton therapy in standard fractionation is in routine use in many centers for definitive treatment of prostate cancer and detailed information is available [38, 61]. In a small phase 3 study, Vargas et al. [62] compared standard fractionated proton therapy with 79.2 GyE in 44 treatments with an extremely hypofractionated schedule of 38 GyE in 5 fractions. There were no major differences at an interim analysis although the American Urological Association Symptom Index at 12 months did show a significant deterioration (p = 0.04) in the hypofractionated arm. A median follow-up of only 18 months and a total number of patients of 82 limit the possibility of further interpretation. Finally, the Heidelberg University ion facility recently finished a pilot study to compare hypofractionated therapy with carbon ions to protons [63].

In summary, the DEGRO expert panel recommends that hypofractionation with heavy ions should not be used outside clinical trials.

In absence of long-term experience derived from randomized controlled trials, the DEGRO expert panel recommends the use of hypofractionated proton therapy in prostate cancer in prospective clinical protocols.