Risk-adapted moderate hypofractionation of prostate cancer

The prognosis of prostate cancer is highly dependent on its risk profile as developed by D’Amico et al. [1] and adopted by the RTOG and AUA. LR patients have a life expectancy similar to healthy individuals, no matter if they are actively treated or actively surveilled [2]. In contrast, HR patients are more prone to develop systemic and/or locoregional progression and are more likely to die from prostate cancer [3, 4].

However, these significant differences in outcome are not necessarily reflected in dose recommendations, although the dose–response relationship is clearly intact even above 80 Gy as shown by numerous dose escalation trials [5] and epidemiologic surveys [6]. NCCN guidelines recommend normofractionated as well as moderately hypofractionated schedules ranging from EQD2 72 to 84.2 Gy irrespective of risk status. German S3 guidelines recommend doses from 74–80 Gy in the setting of normofractionation for all risk groups; for hypofractionation, no explicit dose recommendations are given. This may contribute to a situation where patients of all risk groups receive rather the same radiation schedule in a ‘one size fits all’ manner.

We therefore aimed to tailor radiation treatment to the risk status by using three different radiation regimen which differ in applied dose, fractionation, use of rectum spacers, inclusion of pLN and prescription of ADT.

Hypofractionated treatment regimen were prescribed with doses of EQD2 78 Gy for LR, and 81 Gy for IR and HR patients. In the latter, pLN were treated with 50 Gy with inclusion of the common iliac nodes.

In order to mitigate the expected higher rectal toxicity in IR and HR patients, rectal spacers were implanted via perineal injection into the retroprostatic space behind the Denovillier fascia. Rectal spacers are effective in reducing dose to the anterior rectal wall [7] and have been shown to reduce gastrointestinal toxicity in randomized prospective trials [8, 9].

Here we report on acute toxicity, quality of life and biochemical control after tailored hypofractionated treatment of low, intermediate and high risk prostate cancer patients.

We have compared toxicity profiles of three different hypofractionation schedules which were designed under the assumption that HR prostate cancer requires escalated treatment compared to LR prostate cancer which holds a high risk of overtreatment. Treatment differed in applied dose, use of rectal spacers, use of antihormonal medication and inclusion of pLN.

We could show that all treatment regimen were in total tolerated well in terms of acute toxicity, scored by both, CTC and direct rectoscopic inspection of the rectal mucosa. However, the HR regimen featured slightly higher toxicity in nearly all assessed domains as well as quality of life. The difference was statistically significant in GI CTC compared to IR regimen. This higher GU and GI toxicity can mainly be attributed to the inclusion of pLN, which inevitably causes higher doses to the small bowel, rectum and bladder. Of note, in our series all toxicity domains reverted to near baseline levels after 6 months. To our mind, this slightly higher acute toxicity is outbalanced by the benefit of treating possible occult nodal disease in HR patients. Since the absolute toxicity was very low and only transient, no severe late toxicities are to be expected as consequential late effects [14], but this will be subject of longer follow-up. For scoring of mucosal reactions, the VRS was used which was designed to classify acute and late mucosal reactions and was validated with the EORTC classification of late radiation reactions. Using endoscopic scoring of rectal toxicities allowed the detection of clinically inapparent mucosal reactions.

The benefit of pLN irradiation is heavily debated [15] since the scarce and somewhat outdated randomized data do not provide clear evidence for an improvement in either overall or cancer-specific survival [16‐18]. However, these trials have in part been criticized for using inappropriate doses and insufficient field sizes. In addition, Spratt et al. recently showed that the common iliac lymph nodes are a site of frequent locoregional failure, and that these subvolumes are not typically covered when adhering to RTOG contouring guidelines [19]. These shortcomings might in part explain why no clear benefit for pLN irradiation has yet been demonstrated in prospective trials. However, experiences from other tumor entities suggest that pLN irradiation reduces regional failures and might improve overall survival.

The advent of PSMA PET CT provided a superior specificity and sensitivity in detection of lymph node metastases [20], thus, building a possible rationale to spare pelvic irradiation in patients who are node negative in PSMA PET-CT. This concept has to be evaluated in prospective trials.

For LR and IR prostate cancer, moderate hypofractionation has been shown to be non-inferior to normofractionated treatment in several prospective randomized trials [21‐24] and is now recommended in the primary setting by NCCN guidelines [25], amongst others. For HR patients, the benefit of hypofractionated radiotherapy is less clear. However, results from three large meta-analyses suggests that the low α/β ratio is an intrinsic property of all prostate cancer cells irrespective of their Gleason score or grading [26]. The same is true for dose escalation in the setting of long-term androgen deprivation. While there is no clear consensus that the dose–effect relationship is intact at doses above 80 Gy in the presence of ADT, several analyses point well in that direction [27, 28]. We have therefore opted to treat patients of all risk groups using moderate hypofractionation schedules for the prostate and normofractionation to the pLN in HR patients.

Although the applied dose to the prostate in IR and HR patients (EQD2 of 81 Gy assuming an α/β ratio of 1.5 Gy) was slightly higher than in the above mentioned hypofractionation trials, the observed toxicity compares favorably with a very low incidence of G2 GI and GU toxicity. We could not detect any differences in toxicity between the 20 × 3 Gy and the 21 × 3 Gy arm. The low rectal toxicity of the latter might in part be attributed to the rectal spacer.

In terms of oncologic outcome, follow-up is too short to draw any conclusions, but the bPFS of 99% at 12 months lies within the expected range.

To our knowledge, this is the first comparison of risk adapted tailored hypofractionated treatment regimen with or without inclusion of the pLN which provides insight on added combined toxicity in the context of a modern and risk-adapted hypofractionated regimen.

In HR patients, slightly higher transient acute reactions should be outweighed by possible oncological benefits.