Urethral and bladder dose–effect relations for late genitourinary toxicity following external beam radiotherapy for prostate cancer in the FLAME trial

Open AccessPublished:December 27, 2021DOI:https://doi.org/10.1016/j.radonc.2021.12.027

      Highlights

      • Dose to the bladder and urethra is related to GU toxicity.
      • Focal boost treatment plans should include a urethral dose constraint.
      • Urethral dose-effect relations for hypofractionated schemes should be analyzed.

      Abstract

      Purpose or objectives

      The FLAME trial (NCT01168479) showed that by adding a focal boost to conventional fractionated EBRT in the treatment of localized prostate cancer, the five-year biochemical disease-free survival increased, without significantly increasing toxicity. The aim of the present study was to investigate the association between radiation dose to the bladder and urethra and genitourinary (GU) toxicity grade ≥2 in the entire cohort.

      Material and methods

      The dose–effect relations of the urethra and bladder dose, separately, and GU toxicity grade ≥2 (CTCAE 3.0) up to five years after treatment were assessed. A mixed model analysis for repeated measurements was used, adjusting for age, diabetes mellitus, T-stage, baseline GU toxicity grade ≥1 and institute. Additionally, the association between the dose and separate GU toxicity subdomains were investigated.

      Results

      Dose-effect relations were observed for the dose (Gy) to the bladder D2 cm3 and urethra D0.1 cm3, with adjusted odds ratios of 1.14 (95% CI 1.12–1.16, p < 0.0001) and 1.12 (95% CI 1.11–1.14, p < 0.0001), respectively. Additionally, associations between the dose to the urethra and bladder and the subdomains urinary frequency, urinary retention and urinary incontinence were observed.

      Conclusion

      Further increasing the dose to the bladder and urethra will result in a significant increase in GU toxicity following EBRT. Focal boost treatment plans should incorporate a urethral dose-constraint. Further treatment optimization to increase the focal boost dose without increasing the dose to the urethra and other organs at risk should be a focus for future research, as we have shown that a focal boost is beneficial in the treatment of prostate cancer.

      Abbreviations:

      bDFS (biochemical disease-free survival), CT (Computed tomography), CTCAE (common toxicity criteria for adverse events), CTV (clinical target volume), DVH (dose-volume histograms), EBRT (external beam radiotherapy), FLAME (Focal Lesion Ablative Microboost in ProstatE cancer), GI (gastrointestinal), GTV (gross tumor volume), GU (genitourinary), IPSS (international prostate symptom score), IQR (interquartile range), Mp (multiparametric), MRI (magnetic resonance imaging), NKI (The Netherlands Cancer Institute), OAR (organs at risk), PTV (planning target volume), SAS (statistical analysis system), SD (standard deviation), SPSS (statistical package for social sciences), TSE (turbo spin echo), TURP (trans urethral resection of the prostate), UMCU (University Medical Center Utrecht), WHO (World Health organization)

      Keywords

      Whole-gland dose escalation up to 80 Gy to the entire prostate has shown to be effective regarding biochemical disease-free survival (bDFS) in the treatment of prostate cancer [
      • Kuban D.A.
      • Levy L.B.
      • Cheung M.R.
      • Lee A.K.
      • Choi S.
      • Frank S.
      • et al.
      Long-term failure patterns and survival in a randomized dose-escalation trial for prostate cancer. Who dies of disease?.
      ,
      • Zietman A.L.
      • Bae K.
      • Slater J.D.
      • Shipley W.U.
      • Efstathiou J.A.
      • Coen J.J.
      • et al.
      Randomized trial comparing conventional-dose with high-dose conformal radiation therapy in early-stage adenocarcinoma of the prostate: long-term results from proton radiation oncology group/american college of radiology 95–09.
      ,
      • Heemsbergen W.D.
      • Al-Mamgani A.
      • Slot A.
      • Dielwart M.F.H.
      • Lebesque J.V.
      Long-term results of the Dutch randomized prostate cancer trial: impact of dose-escalation on local, biochemical, clinical failure, and survival.
      ,
      • Beckendorf V.
      • Guerif S.
      • Le Prise E.
      • Cosset J.M.
      • Bougnoux A.
      • Chauvet B.
      • et al.
      70 Gy versus 80 Gy in localized prostate cancer: 5-year results of GETUG 06 randomized trial.
      ,
      • Dearnaley D.P.
      • Jovic G.
      • Syndikus I.
      • Khoo V.
      • Cowan R.A.
      • Graham J.D.
      • et al.
      Escalated-dose versus control-dose conformal radiotherapy for prostate cancer: long-term results from the MRC RT01 randomised controlled trial.
      ]. However, further increasing the dose to the entire prostate using external beam radiotherapy (EBRT) often results in higher genitourinary (GU) and gastrointestinal (GI) toxicity because of an increased dose to the surrounding organs at risk (OAR) [
      • Beckendorf V.
      • Guerif S.
      • Le Prise E.
      • Cosset J.M.
      • Bougnoux A.
      • Chauvet B.
      • et al.
      70 Gy versus 80 Gy in localized prostate cancer: 5-year results of GETUG 06 randomized trial.
      ,
      • Kuban D.A.
      • Tucker S.L.
      • Dong L.
      • Starkschall G.
      • Huang E.H.
      • Cheung M.R.
      • et al.
      Long-term results of the M. D. Anderson randomized dose-escalation trial for prostate cancer.
      ,
      • Dearnaley D.P.
      • Sydes M.R.
      • Graham J.D.
      • Aird E.G.
      • Bottomley D.
      • Cowan R.A.
      • et al.
      Escalated-dose versus standard-dose conformal radiotherapy in prostate cancer: first results from the MRC RT01 randomised controlled trial.
      ,
      • Al-Mamgani A.
      • van Putten W.L.J.
      • Heemsbergen W.D.
      • van Leenders G.J.L.H.
      • Slot A.
      • Dielwart M.F.H.
      • et al.
      Update of Dutch multicenter dose-escalation trial of radiotherapy for localized prostate cancer.
      ,
      • Rodda S.
      • Tyldesley S.
      • Morris W.J.
      • Keyes M.
      • Halperin R.
      • Pai H.
      • et al.
      ASCENDE-RT: an analysis of treatment-related morbidity for a randomized trial comparing a low-dose-rate brachytherapy boost with a dose-escalated external beam boost for high- and intermediate-risk prostate cancer.
      ]. Instead of further increasing the dose to the entire prostate, the phase 3 multicenter randomized controlled Focal Lesion Ablative Microboost in ProstatE cancer (FLAME) trial (NCT01168479) compared effectiveness and toxicity of EBRT with and without a simultaneous integrated focal boost up to 95 Gy to the macroscopic tumor(s) in intermediate- and high-risk localized prostate cancer patients [
      • Lips I.M.
      • van der Heide U.A.
      • Haustermans K.
      • van Lin E.N.
      • Pos F.
      • Franken S.PG.
      • et al.
      Single blind randomized phase III trial to investigate the benefit of a focal lesion ablative microboost in prostate cancer (FLAME-trial): study protocol for a randomized controlled trial.
      ]. The addition of a focal boost to conventionally fractionated EBRT significantly increased the five-year bDFS. The differences in cumulative GU toxicity rates were small and not statistically significant between the treatment arms of the FLAME trial [
      • Kerkmeijer L.G.W.
      • Groen V.H.
      • Pos F.J.
      • Haustermans K.
      • Monninkhof E.M.
      • Smeenk R.J.
      • et al.
      Focal boost to the intraprostatic tumor in external beam radiotherapy for patients with localized prostate cancer: results from the FLAME randomized phase III trial.
      ,
      • Monninkhof E.M.
      • van Loon J.W.L.
      • van Vulpen M.
      • Kerkmeijer L.G.W.
      • Pos F.J.
      • Haustermans K.
      • et al.
      Standard whole prostate gland radiotherapy with and without lesion boost in prostate cancer: Toxicity in the FLAME randomized controlled trial.
      ]. The development of GU toxicity is multifactorial and originates from clinical risk factors and irradiated volumes of the bladder and the urethra [
      • Viswanathan A.N.
      • Yorke E.D.
      • Marks L.B.
      • Eifel P.J.
      • Shipley W.U.
      Radiation dose-volume effects of the urinary bladder.
      ,
      • Olsson C.E.
      • Jackson A.
      • Deasy J.O.
      • Thor M.
      A systematic post-QUANTEC review of tolerance doses for late toxicity after prostate cancer radiotherapy.
      ,
      • Rosewall T.
      • Catton C.
      • Currie G.
      • Bayley A.
      • Chung P.
      • Wheat J.
      • et al.
      The relationship between external beam radiotherapy dose and chronic urinary dysfunction–a methodological critique.
      ]. Dosimetry studies addressing the association between GU toxicity and urethral dose parameters based on large whole-gland dose escalation trials are scarce because of the invisibility of the urethra on computed tomography (CT) scans, which were commonly used for the radiotherapy planning [
      • Cheung M.R.
      • Tucker S.L.
      • Dong L.
      • de Crevoisier R.
      • Lee A.K.
      • Frank S.
      • et al.
      Investigation of bladder dose and volume factors influencing late urinary toxicity after external beam radiotherapy for prostate cancer.
      ]. When MRI is used for contouring of the target volume and OAR, it is possible to delineate the urethra and carry out dose–effect relation analyses for the urethral dose in correlation with GU toxicity.
      By adding a focal boost dose to whole-gland EBRT, an inhomogeneous dose to the prostate was given. This inhomogeneous dose allows to differentiate between the dose to the bladder and urethra, Moreover, in the FLAME trial, we did not use a urethral dose-constraint in treatment planning. This resulted in a significant heterogeneity in the dose to the urethra, allowing us to perform a dose–effect analysis with a wide dose range for the urethra. The objective of this study was to perform a dose–effect relation analysis for the urethral and bladder dose parameters and GU toxicity grade ≥2 in patients with localized prostate cancer treated with EBRT in the FLAME trial in the study cohort, irrespective of randomization arm.

      Material and methods

       Study design and patient population

      In the FLAME trial, standard whole-gland EBRT was compared to an additional simultaneous integrated focal boost up to 95 Gy for localized prostate cancer. The University Medical Center Utrecht (UMCU), The Netherlands Cancer Institute (NKI), Radboudumc Nijmegen in The Netherlands and University Hospitals Leuven in Belgium were participating centers.
      Patients with intermediate- and high-risk prostate cancer according to the Ash criteria [
      • Ash D.
      • Flynn A.
      • Battermann J.
      • de Reijke T.
      • Lavagnini P.
      • Blank L.
      • et al.
      ESTRO/EAU/EORTC recommendations on permanent seed implantation for localized prostate cancer.
      ] were included. Patients were excluded if they had a WHO performance score >2, IPSS score ≥20, evidence of lymph node involvement or distant metastasis, history of prior pelvic irradiation, prostatectomy or trans urethral resection of the prostate (TURP) within three months prior to radiotherapy. Additionally, patients who could not undergo magnetic resonance imaging (MRI), or patients with an undefinable tumor on MRI were excluded.
      Approval was given by the medical ethics committee of the UMCU, The Netherlands (NL26038.041.08) and of the University Hospitals Leuven, Belgium (B322201110225). All included patients gave written informed consent.

       Radiotherapy dose and technique

      Patients were randomized between the standard arm (77 Gy in 35 fractions of 2.2 Gy to the whole prostate, during seven weeks) and the focal boost arm in a 1:1 ratio, with stratification per center. Patients in the focal boost arm received an additional simultaneous integrated boost to the macroscopic tumor up to 95 Gy, resulting in 35 fractions of 2.7 Gy. In order to reduce positioning errors, gold fiducial markers were implanted. Conventional linear accelerators were used to carry out either intensity-modulated radiotherapy or volumetric modulated arc therapy. For delineation of the target volumes and OAR, CT-scans and multiparametric (mp) MRI-scans with T2-weighted, diffusion-weighted and dynamic contrast-enhanced images were acquired. The boost dose to the gross tumor volume (GTV) in the focal boost arm varied between patients, as the dose-constraints to the OAR were prioritized over the focal boost dose. The planning target volume (PTV) was prescribed 77 Gy with a margin of 5–8 mm around the clinical target volume (CTV), depending on the participating center. The part of the PTV overlapping the rectum and bladder was prescribed 70 Gy. There was no margin around the GTV.
      The entire bladder was contoured, from the bladder neck to the bladder dome. Dose-constraints to the bladder were: V72Gy < 10% and D1 cm3 < 80 Gy. Bladder filling protocols differed per participating centre, patients were generally advised to have a comfortably filled bladder during planning CT, MRI and treatment. There was no dose-constraint for the urethra used for treatment planning. Therefore, to allow for the present dose–effect relation analysis, the urethra was contoured in all patients using the sagittal and axial images of the T2-weighted turbo spin echo (TSE) sequence. The prostatic urethra was delineated using a circle shape with 6 mm diameter. After delineation of the urethra, cumulative dose-volume histograms (DVHs) were calculated. Absolute dose parameters that represent the near maximum dose (bladder D2 cm3, urethra D0.1 cm3) were used, because these are considered to be most susceptible to the focal boost dose.

       Toxicity assessment

      Patients were reviewed weekly by the physician during treatment. After treatment, follow-up consisted of appointments with the physician at one month, six months, twelve months and yearly thereafter up to ten years. Treatment-related toxicity was scored using the Common Toxicity Criteria for Adverse Events (CTCAE) version 3.0. GU toxicity endpoints were: urinary frequency, retention, bladder spasm, incontinence, hematuria and dysuria. For the CTCAE 3.0, scores ranged from zero to grade five toxicity.

       Statistical analysis

      In this paper, we investigated the dose to the urethra and bladder, separately, in relation to GU toxicity grade ≥2, irrespective of the treatment arms. Generalized linear mixed effect models were used to assess the association between the dose parameters and GU toxicity over time. The cumulative incidence of toxicity was calculated as the number of patients that experienced any event of grade ≥2 GU toxicity at some point after radiotherapy.
      The different rates of acute GU toxicity (up to 90 days after start of treatment) and late GU toxicity (from 90 days after start of treatment up to five years) was accounted for with a separate intercept and a separate effect for time for acute and late toxicity. To account for multiple measurements per patient, we included a random effects intercept and a random effect for time. We accounted for the increasing dose during treatment up to seven weeks (35 fractions), and used the total planned dose thereafter. We adjusted the models for age, baseline GU toxicity grade ≥1, diabetes mellitus, T-stage and institute (fixed effects).
      The associations of the dose and urinary frequency, urinary retention and urinary incontinence were assessed without adjusting for potential confounders, because of the low number of toxicity events per subdomain. The endpoints hematuria and dysuria were considered to have too few events per measured time point and were not separately analyzed.
      Dose-toxicity curves were created based on the estimated probabilities of late GU toxicity and the planned dose to the bladder D2 cm3, and the D0.1 cm3 of the urethra. Probabilities for developing late GU toxicity were calculated based on the average patient. As a range around the average patient curve, we plotted the probability of developing late GU toxicity in an unfavorable risk group in which all potential risk factors for GU toxicity (higher T-stage (T3b or T4), diabetes mellitus and baseline GU toxicity grade ≥1) were present, and a favorable risk group in which none of the aforementioned risk factors were present, with a mean age of 71 years old.
      In addition, we analyzed the effect of acute GU toxicity on late GU toxicity in a generalized linear mixed effect model, adjusted for age, baseline GU toxicity grade ≥1, diabetes mellitus, T-stage and institute. Additional information on the statistical analyses were presented in the supplementary material Table S1.
      All statistical analyses were performed using SPSS version 25.0 (IBM Corp., Armonk, NY, USA) and SAS, version 9.4 (SAS Institute Inc., Cary, NC, USA).

      Results

      At time of the present analysis, all 571 patients had potentially reached five years of follow-up, with a median follow-up of 72 months (interquartile range (IQR) 58–86). Baseline characteristics are presented in Table 1. The mean age at time of randomization was 71 years (SD 6). For further analysis, 91 patients were excluded. Three patients were excluded because they received a different type of treatment than originally planned: one patient was treated with hormonal treatment alone because of metastatic disease on the planning CT/MRI scans, another patient received additional pelvic radiotherapy because of positive lymph nodes and the third patient underwent a cystoprostatectomy, as he was diagnosed with concomitant bladder carcinoma. Another ten patients without dosimetry data available and 78 patients who underwent a previous TURP were excluded.
      Table 1Patient and treatment characteristics of the FLAME trial participants at baseline.
      Total number of patients (n)571
      Age in years (mean, SD)716
      iPSA ng/mL (median, IQR)11.27.3–18.5
      Risk classification (EAU) (n, %)Low61%
      Intermediate8515%
      High48084%
      Center (n, %)UMC Utrecht32056%
      UZ Leuven9316%
      NKI10919%
      Radboudumc499%
      T stage (n, %)Missing20%
      T1c468%
      T2a5810%
      T2b377%
      T2c7613%
      T3a23742%
      T3b10218%
      T4132%
      N stage (n, %)Missing10%
      N045680%
      pN < 10 lymph nodes removed8114.%
      pN ≥10 lymph nodes removed336%
      M stage (n, %)Mx14325%
      M042875%
      Gleason (n, %)< 710318%
      727648%
      ≥ 819234%
      Cardiovascular disease (n, %)Missing30%
      No25545%
      Yes31355%
      Hormonal therapy (n, %)Missing51%
      No19033%
      Yes37666%
      Diabetes mellitus (n, %)Missing20%
      No50489%
      Yes6511%
      Baseline GU toxicity (n, %)Missing163%
      No35662%
      Grade 114726%
      Grade 2468%
      Grade 361%
      Grade 4NANA
      Grade 5NANA
      Abbreviations: SD = standard deviation, iPSA = initial prostate specific antigen, IQR = interquartile range, EAU = European Association of Urology, T stage = T refers to the size and extent of the main tumor, N stage = N refers to the number of nearby lymph nodes that have cancer, M stage = M refers to whether the cancer has metastasized, GU = genitourinary.
      For the present study we addressed the GU toxicity up to five years. The cumulative acute and late GU toxicity rates grade ≥2 are presented in Table 2. Cumulative acute toxicity grade ≥3 was seen in 3% (n = 18) of the patients, including urinary frequency (n = 9), urinary obstruction (n = 8), urinary incontinence (n = 1) and dysuria (n = 1). Cumulative late toxicity grade ≥3 was seen in 5% (n = 27) of the patients, including urinary frequency (n = 5), urinary obstruction (n = 11), urinary incontinence (n = 7), bladder spasm (n = 1), hematuria (n = 6) and dysuria (n = 2). Grade 4 GU toxicity occurred in one patient, who required a permanent urinary diversion due to severe incontinence, three years after treatment. Urethral strictures occurred in 18 patients (4%), requiring medical interventions including urethral dilatation, urethrotomy or daily intermittent self-catheterization. Over half of the strictures (13/18) occurred more than two years after treatment. In one patient a cystectomy including a partial prostatectomy was required four years after radiotherapy, because of urethral necrosis following a urethrotomy earlier that year. We did not determine the location of the urethral strictures.
      Table 2The association between urethra and bladder dose and cumulative GU toxicity: results of generalized linear mixed models with and without adjustment for potential confounding factors for GU toxicity. Results for late cumulative GU toxicity grade ≥2 and late separate GU toxicity grade ≥2 endpoints are shown.
      Total n = 480Overall GU toxicity

      grade ≥2
      Urinary frequency

      grade ≥2
      Urinary retention

      grade ≥2
      Urinary Incontinence

      grade ≥2
      Hematuria

      grade ≥2
      Dysuria

      grade ≥2
      Acute* cumulative toxicity47%

      (95% CI 42–51%, n = 225)
      37%

      (95% CI 33–42%, n = 179)
      15%

      (95% CI 12–19%, n = 74)
      3%

      (95% CI 2–5%, n = 13)
      0%

      (95% CI 0–1%, n = 1)
      5%

      (95% CI 3–7%, n = 23)
      Late* cumulative toxicity24%

      (95% CI 21–28%, n = 116)
      16%

      (95% CI 13–19%, n = 75)
      8%

      (95% CI 6–11%, n = 39)
      5%

      (95% CI 4–8%, n = 26)
      1%

      (95% CI 1–3%, n = 6)
      2%

      (95% CI 1–3%, n = 8)
      Unadjusted OR

      (95% CI)
      Adjusted OR

      (95% CI)
      Unadjusted OR

      (95% CI)
      Bladder D2cm31.15

      (1.13 1.17, p < 0.0001)
      1.14

      (1.12 1.16, p < 0.0001)
      1.17

      (1.15 1.20, p < 0.0001)
      1.17

      (1.13 1.22, p < 0.0001)
      1.12

      (1.05 1.20, p = 0.001)
      Urethra D0.1 cm31.13

      (1.11 1.15, p < 0.0001)
      1.12

      (1.11 1.14, p < 0.0001)
      1.15

      (1.13 1.17, p < 0.0001)
      1.15

      (1.11 1.18, p < 0.0001)
      1.11

      (1.05 1.18, p < 0.001)
      The ORs for the dose-parameters mean that when the planned dose to the urethra and bladder increases with 1 Gy, the odds of developing GU toxicity grade ≥2 increase with the corresponding given OR.
      *Acute toxicity was defined as toxicity from the start of treatment up to 90 days, late toxicity was defined as from 90 days after start of treatment up to five years after treatment.
      Abbreviations: GU = genitourinary, CI = confidence interval, OR = odds ratio.
      The median planned dose to the D2 cm3 of the bladder and the D0.1 cm3 of the urethra were 75 Gy (IQR (74–76) and 80 Gy (IQR 78–87), respectively, see Fig. 1 for the dose distributions per treatment arm. For the bladder D2cm3 we found a dose–effect relation with an unadjusted odds ratio of 1.15 (95% CI 1.13–1.17, p < 0.0001). This means that when the planned dose to the D2cm3 of the bladder increases with 1 Gy, the odds of developing GU toxicity grade ≥2 increases with 1.15. Adjusted for age, T-stage, diabetes mellitus, baseline GU toxicity grade ≥1 and institute, the odds ratio was 1.14 (95% CI 1.12–1.16, p < 0.0001). The odds ratios for the urethra D0.1 cm3 for developing GU toxicity grade ≥2 were 1.13 (95% CI 1.11–1.15, p < 0.0001) and (after adjustment for confounders) 1.12 (95% CI 1.11–1.14, p < 0.0001) per increase of 1 Gy (Table 2). These associations were visualized as dose–effect curves for the average patient, and for unfavorable and favorable risk groups based on patient characteristics in Fig. 2. When using a cut-off for the dose to the urethra of 80 Gy, cumulative toxicity rates were 22.3% (58/260) and 26.4% (58/220), respectively.
      Figure thumbnail gr1
      Fig. 1Stacked histogram of planned dose (Gy) to the A. bladder D2cm3 (standard arm and focal boost arm, median 75 Gy (IQR 74–76) and 75 Gy (IQR 74–77), respectively) and B. urethra D0.1 cm3 (standard arm and focal boost arm, median 78 Gy (IQR 78–79) and 86 Gy (IQR 82–91), respectively) per treatment arm. Abbreviations: IQR = interquartile range.
      Figure thumbnail gr2
      Fig. 2Dose-toxicity curves of the average and (un)favorable estimated cumulative GU toxicity grade ≥2, related to the planned dose based on the generalized linear mixed effects models adjusted for age, T-stage, diabetes mellitus, baseline GU toxicity grade ≥1. The risk groups are based on baseline patient characteristics potentially correlated with GU toxicity. In the unfavorable risk group, all aforementioned risk factors were present. In the favorable risk group, none of the risk factors were present. A. bladder D2cm3 (Gy) and B. urethra D0.1 cm3 (Gy). Abbreviations: GU = genitourinary, T-stage = T refers to the size and extent of the main tumor.
      The dose–effect relations of the GU toxicity subdomains are presented in Table 2. For the endpoints urinary retention and urinary incontinence the random effect for time was excluded, since the corresponding covariance parameter estimates were zero. An additional generalized linear mixed effect model showed that acute GU toxicity was associated with late GU toxicity with an adjusted odds ratio of 5.82 (95% CI 1.65–20.56, p = 0.006).

      Discussion

      Although cumulative toxicity showed no significant difference between treatment arms of the FLAME trial [
      • Kerkmeijer L.G.W.
      • Groen V.H.
      • Pos F.J.
      • Haustermans K.
      • Monninkhof E.M.
      • Smeenk R.J.
      • et al.
      Focal boost to the intraprostatic tumor in external beam radiotherapy for patients with localized prostate cancer: results from the FLAME randomized phase III trial.
      ], the large dose range to the bladder and urethra in the FLAME focal boost study allowed for the composition of dose–effect relations for the urethra and bladder, separately. By using a longitudinal repeated measures analysis, we found that an increased dose to the bladder and urethra will result in a significant increase in GU toxicity following EBRT. However, no clear threshold dose for the dose to the urethra and GU toxicity grade ≥2 could be observed. As we found a dose–effect relation for the urethra dose, it would be desirable to optimize the radiation plan also taking the urethra dose into account. We, therefore, propose that focal boost treatment plans should incorporate a urethral dose-constraint (pragmatically set at D0.1 cm3 ≤ 80 Gy as close to conventional whole prostate gland dose) in addition to the pre-existing bladder dose-constraints. A urethral constraint was incorporated in our subsequent hypo-FLAME study [
      • Draulans C.
      • van der Heide U.A.
      • Haustermans K.
      • Pos F.J.
      • van der Voort van Zyp J.
      • De Boer H.
      • et al.
      Primary endpoint analysis of the multicentre phase II hypo-FLAME trial for intermediate and high risk prostate cancer.
      ]. Genitourinary toxicity is multifactorial and depends on other factors than (urethral) dose only. However, when limiting the dose to the bladder and dose to the urethra, we expect to minimize treatment related toxicity.
      Whole-gland dose-escalation trials with doses up to 80 Gy, showed higher GU toxicity rates in the dose-escalation arms, showing a dose–effect relation mainly for the high/maximum doses to the bladder and/or urethra [
      • Cheung M.R.
      • Tucker S.L.
      • Dong L.
      • de Crevoisier R.
      • Lee A.K.
      • Frank S.
      • et al.
      Investigation of bladder dose and volume factors influencing late urinary toxicity after external beam radiotherapy for prostate cancer.
      ,
      • Heemsbergen W.D.
      • Al-Mamgani A.
      • Witte M.G.
      • van Herk M.
      • Pos F.J.
      • Lebesque J.V.
      Urinary obstruction in prostate cancer patients from the Dutch trial (68 Gy vs. 78 Gy): relationships with local dose, acute effects, and baseline characteristics.
      ,
      • Jolnerovski M.
      • Salleron J.
      • Beckendorf V.
      • Peiffert D.
      • Baumann A.-S.
      • Bernier V.
      • et al.
      Intensity-modulated radiation therapy from 70Gy to 80Gy in prostate cancer: six- year outcomes and predictors of late toxicity.
      ,
      • Mathieu R.
      • Arango J.D.
      • Beckendorf V.
      • Delobel J.B.
      • Messai T.
      • Chira C.
      • et al.
      Nomograms to predict late urinary toxicity after prostate cancer radiotherapy.
      ,
      • Barnett G.C.
      • De Meerleer G.
      • Gulliford S.L.
      • Sydes M.R.
      • Elliott R.M.
      • Dearnaley D.P.
      The impact of clinical factors on the development of late radiation toxicity: results from the Medical Research Council RT01 trial (ISRCTN47772397).
      ]. Frequent GU toxicity complaints are urethral strictures, urinary retention, incontinence and hematuria [
      • Viswanathan A.N.
      • Yorke E.D.
      • Marks L.B.
      • Eifel P.J.
      • Shipley W.U.
      Radiation dose-volume effects of the urinary bladder.
      ,
      • Heemsbergen W.D.
      • Al-Mamgani A.
      • Witte M.G.
      • van Herk M.
      • Pos F.J.
      • Lebesque J.V.
      Urinary obstruction in prostate cancer patients from the Dutch trial (68 Gy vs. 78 Gy): relationships with local dose, acute effects, and baseline characteristics.
      ,
      • Ghadjar P.
      • Zelefsky M.J.
      • Spratt D.E.
      • Munck af Rosenschöld P.
      • Oh J.H.
      • Hunt M.
      • et al.
      Impact of dose to the bladder trigone on long-term urinary function after high-dose intensity modulated radiation therapy for localized prostate cancer.
      ,
      • Mylona E.
      • Acosta O.
      • Lizee T.
      • Lafond C.
      • Crehange G.
      • Magné N.
      • et al.
      Voxel-based analysis for identification of urethrovesical subregions predicting urinary toxicity after prostate cancer radiation therapy.
      ,
      • McDonald A.M.
      • Baker C.B.
      • Popple R.A.
      • Cardan R.A.
      • Fiveash J.B.
      Increased radiation dose heterogeneity within the prostate predisposes to urethral strictures in patients receiving moderately hypofractionated prostate radiation therapy.
      ]. Previous studies showed that an increased dose to the bladder region or urethral surrogate structure receiving >75 Gy [
      • McDonald A.M.
      • Baker C.B.
      • Popple R.A.
      • Cardan R.A.
      • Fiveash J.B.
      Increased radiation dose heterogeneity within the prostate predisposes to urethral strictures in patients receiving moderately hypofractionated prostate radiation therapy.
      ] and >80 Gy [
      • Heemsbergen W.D.
      • Al-Mamgani A.
      • Witte M.G.
      • van Herk M.
      • Pos F.J.
      • Lebesque J.V.
      Urinary obstruction in prostate cancer patients from the Dutch trial (68 Gy vs. 78 Gy): relationships with local dose, acute effects, and baseline characteristics.
      ], respectively, results in an increase in urethral strictures and urinary obstruction. Mylona et al. also identified the dose to the urethra and bladder sub regions as predictors for various urinary symptoms [
      • Mylona E.
      • Acosta O.
      • Lizee T.
      • Lafond C.
      • Crehange G.
      • Magné N.
      • et al.
      Voxel-based analysis for identification of urethrovesical subregions predicting urinary toxicity after prostate cancer radiation therapy.
      ]. We found associations for urinary frequency, urinary retention, urinary incontinence and the bladder and urethral dose parameters. A correlation between acute GU toxicity and late GU toxicity was observed. Furthermore, a latency period was present, with GU toxicity occurring years after radiotherapy, including new onset urethral strictures occurring two years, up to seven years, after treatment. This should be taken into consideration and long follow-up is needed to account for (very) late onset genitourinary toxicity.
      Strengths of our study are the carefully considered longitudinal analysis that fits our data properly and increases power by using repeated measurements. Another strength is the use of risk groups for presenting the probability of GU toxicity. Notably, these risk groups represent differences in baseline risk of GU toxicity, which is not the same as prognostic modeling and should not be interpreted as such. The unique variation in focal boost dose used and the lack of a urethral dose-constraint in treatment planning, allowed us to carry out a urethral dose–effect relation analysis for a large dose range. Although the differences in cumulative GU toxicity between the study arms were small and not statistically significant [
      • Kerkmeijer L.G.W.
      • Groen V.H.
      • Pos F.J.
      • Haustermans K.
      • Monninkhof E.M.
      • Smeenk R.J.
      • et al.
      Focal boost to the intraprostatic tumor in external beam radiotherapy for patients with localized prostate cancer: results from the FLAME randomized phase III trial.
      ], in the present analysis, using the cohort irrespective of randomization, we did observe a dose–effect relation for the bladder and urethra. While this may seem contradictory, this is explained by the observation that in the focal boost arm of the trial, a wide range was found for the dose to the urethra, with some patients receiving a standard dose close to 77 Gy, while others received a much higher dose, depending on the location of the focal boost. The number of patients with a high dose to the urethra in the focal boost arm was too small to observe a significant difference in cumulative GU toxicity. When using cumulative toxicity, the highest toxicity grade at any moment is used to calculate the toxicity rates, not taking into account repeating events of toxicity. Yet, the finding of a significant dose–effect relation for the urethra in the study cohort irrespective of randomization arm, is explained by the longitudinal approach used in the present study. When using repeated measurements, more patients in the focal boost arm had recurring GU toxicity events compared to the standard treatment arm.
      Our study has a few limitations. First, the difficulty of delineating the urethra. The urethra was contoured using the T2-weighted sequence of a mpMRI scan [
      • Villeirs G.M.
      • L.Verstraete K.
      • De Neve W.J.
      • De Meerleer G.O.
      Magnetic resonance imaging anatomy of the prostate and periprostatic area: a guide for radiotherapists.
      ]. Even though the urethra is visualized better on MRI than CT, the delineation of the urethra even on MRI can be difficult and interobserver variation in contouring may exist [
      • Zakian K.L.
      • Wibmer A.
      • Vargas H.A.
      • Alberts E.
      • Kadbi M.
      • Mychalczak B.
      • et al.
      Comparison of motion-insensitive T2-weighted MRI pulse sequences for visualization of the prostatic urethra during MR simulation.
      ]. Second, we used center specific bladder filling protocols. As all participating centers aimed for a comfortably filled bladder, we do not expect this to have influenced our findings. Third, the considerable amount of missing toxicity data is a limiting factor. A generalized linear mixed effect model with random effects to determine the dose–effect relations was used to cope with this limitation. In addition to handling both single and recurring toxicity, these models provide unbiased estimations in the presence of missing data under the missing-at-random assumption [
      • Molenberghs G.
      • Verbeke G.
      Models for Discrete Longitudinal Data, Springer Series in Statistics.
      ]. Though we have no reason to doubt this assumption, missing-at-random is essentially unverifiable. Fourth, we decided to exclude patients who previously underwent a TURP, as the TURP cavity is not comparable to the anatomical structure of the urethra.
      In conclusion, dose–effect relations for both the dose to the bladder and urethra on GU toxicity were observed. For treatment planning of EBRT with a focal boost in the conventional fractionated FLAME scheme, we suggest a dose-constraint for the urethra D0.1 cm3 of ≤80 Gy, the optimal urethral dose-constraint for hypofractionated schemes is yet to be determined. Further treatment optimization to increase the focal boost without increasing the dose to the urethra, bladder and other OAR should be a focus for future research, as we have shown that a focal boost improves oncological outcomes in the treatment of prostate cancer.

      Registration number ClinicalTrials.gov

      NCT01168479.

      Funding source

      This study is supported by the Dutch Cancer Society (KWF, project 10088).

      Conflict of interest statement

      All authors declare having no conflict of interest related to the content of this manuscript.

      Acknowledgements

      This study is supported by the Dutch Cancer Society (KWF, project 10088).

      References

        • Kuban D.A.
        • Levy L.B.
        • Cheung M.R.
        • Lee A.K.
        • Choi S.
        • Frank S.
        • et al.
        Long-term failure patterns and survival in a randomized dose-escalation trial for prostate cancer. Who dies of disease?.
        Int J Radiat Oncol Biol Phys. 2011; 79: 1310-1317
        • Zietman A.L.
        • Bae K.
        • Slater J.D.
        • Shipley W.U.
        • Efstathiou J.A.
        • Coen J.J.
        • et al.
        Randomized trial comparing conventional-dose with high-dose conformal radiation therapy in early-stage adenocarcinoma of the prostate: long-term results from proton radiation oncology group/american college of radiology 95–09.
        J Clin Oncol. 2010; 28: 1106-1111
        • Heemsbergen W.D.
        • Al-Mamgani A.
        • Slot A.
        • Dielwart M.F.H.
        • Lebesque J.V.
        Long-term results of the Dutch randomized prostate cancer trial: impact of dose-escalation on local, biochemical, clinical failure, and survival.
        Radiother Oncol. 2014; 110: 104-109
        • Beckendorf V.
        • Guerif S.
        • Le Prise E.
        • Cosset J.M.
        • Bougnoux A.
        • Chauvet B.
        • et al.
        70 Gy versus 80 Gy in localized prostate cancer: 5-year results of GETUG 06 randomized trial.
        Int J Radiat Oncol Biol Phys. 2011; 80: 1056-1063
        • Dearnaley D.P.
        • Jovic G.
        • Syndikus I.
        • Khoo V.
        • Cowan R.A.
        • Graham J.D.
        • et al.
        Escalated-dose versus control-dose conformal radiotherapy for prostate cancer: long-term results from the MRC RT01 randomised controlled trial.
        Lancet Oncol. 2014; 15: 464-473
        • Kuban D.A.
        • Tucker S.L.
        • Dong L.
        • Starkschall G.
        • Huang E.H.
        • Cheung M.R.
        • et al.
        Long-term results of the M. D. Anderson randomized dose-escalation trial for prostate cancer.
        Int J Radiat Oncol Biol Phys. 2008; 70: 67-74
        • Dearnaley D.P.
        • Sydes M.R.
        • Graham J.D.
        • Aird E.G.
        • Bottomley D.
        • Cowan R.A.
        • et al.
        Escalated-dose versus standard-dose conformal radiotherapy in prostate cancer: first results from the MRC RT01 randomised controlled trial.
        Lancet Oncol. 2007; 8: 475-487
        • Al-Mamgani A.
        • van Putten W.L.J.
        • Heemsbergen W.D.
        • van Leenders G.J.L.H.
        • Slot A.
        • Dielwart M.F.H.
        • et al.
        Update of Dutch multicenter dose-escalation trial of radiotherapy for localized prostate cancer.
        Int J Radiat Oncol Biol Phys. 2008; 72: 980-988
        • Rodda S.
        • Tyldesley S.
        • Morris W.J.
        • Keyes M.
        • Halperin R.
        • Pai H.
        • et al.
        ASCENDE-RT: an analysis of treatment-related morbidity for a randomized trial comparing a low-dose-rate brachytherapy boost with a dose-escalated external beam boost for high- and intermediate-risk prostate cancer.
        Int J Radiat Oncol Biol Phys. 2017; 98: 286-295
        • Lips I.M.
        • van der Heide U.A.
        • Haustermans K.
        • van Lin E.N.
        • Pos F.
        • Franken S.PG.
        • et al.
        Single blind randomized phase III trial to investigate the benefit of a focal lesion ablative microboost in prostate cancer (FLAME-trial): study protocol for a randomized controlled trial.
        Trials. 2011; 12https://doi.org/10.1186/1745-6215-12-255
        • Kerkmeijer L.G.W.
        • Groen V.H.
        • Pos F.J.
        • Haustermans K.
        • Monninkhof E.M.
        • Smeenk R.J.
        • et al.
        Focal boost to the intraprostatic tumor in external beam radiotherapy for patients with localized prostate cancer: results from the FLAME randomized phase III trial.
        J Clin Oncol. 2021; 39: 787-796
        • Monninkhof E.M.
        • van Loon J.W.L.
        • van Vulpen M.
        • Kerkmeijer L.G.W.
        • Pos F.J.
        • Haustermans K.
        • et al.
        Standard whole prostate gland radiotherapy with and without lesion boost in prostate cancer: Toxicity in the FLAME randomized controlled trial.
        Radiother Oncol. 2018; 127: 74-80
        • Viswanathan A.N.
        • Yorke E.D.
        • Marks L.B.
        • Eifel P.J.
        • Shipley W.U.
        Radiation dose-volume effects of the urinary bladder.
        Int J Radiat Oncol Biol Phys. 2010; 76: S116-S122
        • Olsson C.E.
        • Jackson A.
        • Deasy J.O.
        • Thor M.
        A systematic post-QUANTEC review of tolerance doses for late toxicity after prostate cancer radiotherapy.
        Int J Radiat Oncol Biol Phys. 2018;
        • Rosewall T.
        • Catton C.
        • Currie G.
        • Bayley A.
        • Chung P.
        • Wheat J.
        • et al.
        The relationship between external beam radiotherapy dose and chronic urinary dysfunction–a methodological critique.
        Radiother Oncol. 2010; 97: 40-47
        • Cheung M.R.
        • Tucker S.L.
        • Dong L.
        • de Crevoisier R.
        • Lee A.K.
        • Frank S.
        • et al.
        Investigation of bladder dose and volume factors influencing late urinary toxicity after external beam radiotherapy for prostate cancer.
        Int J Radiat Oncol Biol Phys. 2007; 67: 1059-1065
        • Ash D.
        • Flynn A.
        • Battermann J.
        • de Reijke T.
        • Lavagnini P.
        • Blank L.
        • et al.
        ESTRO/EAU/EORTC recommendations on permanent seed implantation for localized prostate cancer.
        Radiother Oncol. 2000; 57: 315-321
        • Draulans C.
        • van der Heide U.A.
        • Haustermans K.
        • Pos F.J.
        • van der Voort van Zyp J.
        • De Boer H.
        • et al.
        Primary endpoint analysis of the multicentre phase II hypo-FLAME trial for intermediate and high risk prostate cancer.
        Radiother Oncol. 2020; 147: 92-98
        • Heemsbergen W.D.
        • Al-Mamgani A.
        • Witte M.G.
        • van Herk M.
        • Pos F.J.
        • Lebesque J.V.
        Urinary obstruction in prostate cancer patients from the Dutch trial (68 Gy vs. 78 Gy): relationships with local dose, acute effects, and baseline characteristics.
        Int J Radiat Oncol Biol Phys. 2010; 78: 19-25
        • Jolnerovski M.
        • Salleron J.
        • Beckendorf V.
        • Peiffert D.
        • Baumann A.-S.
        • Bernier V.
        • et al.
        Intensity-modulated radiation therapy from 70Gy to 80Gy in prostate cancer: six- year outcomes and predictors of late toxicity.
        Radiat Oncol (London, England). 2017; 12https://doi.org/10.1186/s13014-017-0839-3
        • Mathieu R.
        • Arango J.D.
        • Beckendorf V.
        • Delobel J.B.
        • Messai T.
        • Chira C.
        • et al.
        Nomograms to predict late urinary toxicity after prostate cancer radiotherapy.
        World J Urol. 2014; 32: 743-751
        • Barnett G.C.
        • De Meerleer G.
        • Gulliford S.L.
        • Sydes M.R.
        • Elliott R.M.
        • Dearnaley D.P.
        The impact of clinical factors on the development of late radiation toxicity: results from the Medical Research Council RT01 trial (ISRCTN47772397).
        Clin Oncolgy (Royal College of Radiologists (Great Britain)). 2011; 23: 613-624
        • Ghadjar P.
        • Zelefsky M.J.
        • Spratt D.E.
        • Munck af Rosenschöld P.
        • Oh J.H.
        • Hunt M.
        • et al.
        Impact of dose to the bladder trigone on long-term urinary function after high-dose intensity modulated radiation therapy for localized prostate cancer.
        Int J Radiat Oncol Biol Phys. 2014; 88: 339-344
        • Mylona E.
        • Acosta O.
        • Lizee T.
        • Lafond C.
        • Crehange G.
        • Magné N.
        • et al.
        Voxel-based analysis for identification of urethrovesical subregions predicting urinary toxicity after prostate cancer radiation therapy.
        Int J Radiat Oncol Biol Phys. 2019; 104: 343-354
        • McDonald A.M.
        • Baker C.B.
        • Popple R.A.
        • Cardan R.A.
        • Fiveash J.B.
        Increased radiation dose heterogeneity within the prostate predisposes to urethral strictures in patients receiving moderately hypofractionated prostate radiation therapy.
        Pract Radiat Oncol. 2015; 5: 338-342
        • Villeirs G.M.
        • L.Verstraete K.
        • De Neve W.J.
        • De Meerleer G.O.
        Magnetic resonance imaging anatomy of the prostate and periprostatic area: a guide for radiotherapists.
        Radiother Oncol. 2005; 76: 99-106
        • Zakian K.L.
        • Wibmer A.
        • Vargas H.A.
        • Alberts E.
        • Kadbi M.
        • Mychalczak B.
        • et al.
        Comparison of motion-insensitive T2-weighted MRI pulse sequences for visualization of the prostatic urethra during MR simulation.
        Pract Radiat Oncol. 2019; 9: e534-e540
        • Molenberghs G.
        • Verbeke G.
        Models for Discrete Longitudinal Data, Springer Series in Statistics.
        Springer Science & Business Media, 2005: 687