A Method for the Prediction of Late Organ-at-Risk Toxicity After Radiotherapy of the Prostate Using Equivalent Uniform Dose

Feature Importance Analysis of a Deep Learning Model for Predicting Late Bladder Toxicity Occurrence in Uterine Cervical Cancer Patients

(1) In this study, we developed a deep learning (DL) model that can be used to predict late bladder toxicity. (2) We collected data obtained from 281 uterine cervical cancer patients who underwent definitive radiation therapy. The DL model was trained using 16 features, including patient, tumor, treatment, and dose parameters, and its performance was compared with that of a multivariable logistic regression model using the following metrics: accuracy, prediction, recall, F1-score, and area under the receiver operating characteristic curve (AUROC). In addition, permutation feature importance was calculated to interpret the DL model for each feature, and the lightweight DL model was designed to focus on the top five important features. (3) The DL model outperformed the multivariable logistic regression model on our dataset. It achieved an F1-score of 0.76 and an AUROC of 0.81, while the corresponding values for the multivariable logistic regression were 0.14 and 0.43, respectively. The DL model identified the doses for the most exposed 2 cc volume of the bladder (BD_2cc) as the most important feature, followed by BD_5cc and the ICRU bladder point. In the case of the lightweight DL model, the F-score and AUROC were 0.90 and 0.91, respectively. (4) The DL models exhibited superior performance in predicting late bladder toxicity compared with the statistical method. Through the interpretation of the model, it further emphasized its potential for improving patient outcomes and minimizing treatment-related complications with a high level of reliability.

Normal tissue complication probability (NTCP) models of acute urinary toxicity (AUT) following carbon ion radiotherapy (CIRT) for prostate cancer

PURPOSE

To estimate the Lyman Kutcher Burman (LKB) and multivariate NTCP models predicting the AUT of prostate cancer treated with CIRT.

MATERIALS AND METHODS

A cohort of 154 prostate adenocarcinoma patients were retrospectively analyzed. The AUT levels were graded according to CTCAE 4.03. Based on dosimetric parameters and/or clinical factors, a set of variables with best-fit values determined in the two models was validated by the area under the receiver operating characteristic curve (AUC) and used to correlate the predicted and observed NTCP rates for both levels and related endpoints.

RESULT

59 (38.3%) patients experienced AUT. For LKB model, the equivalent uniform doses (EUDs) were calculated to be 62.0 GyE (following V_61.5 > 1.7%) and 61.2 GyE (following maximum dose > 63.0 GyE) with predicted NTCP rates of 37.0% (AUC: 0.71) and 15.6% (AUC: 0.65) for AUT G1&2 and G2 of bladder. While for the multivariate model, the predicted NTCP rates was 37.1% (AUC: 0.70) and 20.2% (AUC: 0.64) for AUT G1&2 and G2, associated with V₆₁ and V₆₅, respectively. Nocturia was associated with bladder volume and maximum dose for G1&2, with patient's age and maximum bladder dose for G2. Other predictable endpoints were associated with V_≥61. The predicted NTCPs agree with the observed complication rates for bladder and its wall.

CONCLUSIONS

The LKB model successfully predicted the NTCP rates of both AUT levels and urgency urination. The multivariate model predicted well on both levels and nocturia. Decreasing high bladder dose volume may reduce the incidence of AUT.

Understanding Urinary Toxicity after Radiotherapy for Prostate Cancer: First Steps Forward

One of the most relevant achievements of Professor Gianni Bonadonna was the implementation of the methodology of controlled clinical trials in medical oncology. It is valid for all cancer types, oncological disciplines and clinical endpoints, both survival and toxicity. This narrative review reports on the status of the current knowledge of the radiation-induced urinary syndrome after external-beam radiotherapy for prostate cancer. In recent years, the syndrome has been the object of large-scale prospective observational trials specifically devoted to investigating the association of patient and treatment features with acute/late urinary toxicity. The first results of these trials allow initial attempts at predictive modeling, which can serve as a basis for the optimization of patient selection and treatment planning.

Dosimetric and radiobiological comparison of Cyberknife and Tomotherapy in stereotactic body radiotherapy for localized prostate cancer

BACKGROUND AND PURPOSE

As recent studies have suggested relatively low α/β for prostate cancer, the interest in hypofractionated stereotactic body radiotherapy (SBRT) for prostate cancer is rising. The aim of this study is to compare dosimetric results of Cyberknife (CK) with Tomotherapy (HT) in SBRT for localized prostate cancer. Furthermore, the radiobiologic consequences of heterogeneous dose distribution are also analyzed.

MATERIAL AND METHOD

A total of 12 cases of localized prostate cancer previously treated with SBRT were collected. Treatments had been planned and delivered using CK. Then HT plans were generated for comparison afterwards. The prescribed dose was 37.5Gy in 5 fractions. Dosimetric indices for target volumes and organs at risk (OAR) were compared. For radiobiological evaluation, generalized equivalent uniform dose (gEUD) and normal tissue complication probability (NTCP) were calculated and compared.

RESULT

Both CK and HT achieved target coverage while meeting OAR constraints adequately. HT plans resulted in better dose homogeneity (Homogeneity index: 1.04±0.01 vs. 1.21±0.01; p = 0.0022), target coverage (97.74±0.86% vs. 96.56±1.17%; p = 0.0076) and conformity (new vonformity index: 1.16±0.05 vs. 1.21±0.04; p = 0.0096). HT was shown to predict lower late rectal toxicity as compared to CK. Integral dose to body was also significantly lower in HT plans (46.59±6.44 Gy'L vs 57.05±11.68 Gy'L; p = 0.0029).

CONCLUSION

Based on physical dosimetry and radiobiologic considerations, HT may have advantages over CK, specifically in rectal sparing which could translate into clinical benefit of decreased late toxicities.

The benefit of using bladder sub-volume equivalent uniform dose constraints in prostate intensity-modulated radiotherapy planning

Background

To assess the benefits of bladder wall sub-volume equivalent uniform dose (EUD) constraints in prostate cancer intensity-modulated radiotherapy (IMRT) planning.

Methods

Two IMRT plans, with and without EUD constraints on the bladder wall, were generated using beams that deliver 80 Gy to the prostate and 46 Gy to the seminal vesicles and were compared in 53 prostate cancer patients. The bladder wall was defined as the volume between the external manually delineated wall and a contraction of 7 mm apart from it. The bladder wall was then separated into two parts: the internal-bladder wall (bla-in) represented by the portion of the bladder wall that intersected with the planning target volume (PTV) plus 5 mm extension; the external-bladder wall (bla-ex) represented by the remaining part of the bladder wall. In the IMRT plan with EUD constraints, the values of “a” parameter for the EUD models were 10.0 for bla-in and 2.3 for bla-ex. The plans with and without EUD constraints were compared in terms of dose–volume histograms, 5-year bladder and rectum normal tissue complication probability values, as well as tumor control probability (TCP) values.

Results

The use of bladder sub-volume EUD constraints decreased both the doses to the bladder wall (V₇₀: 22.76% vs 19.65%, D_mean: 39.82 Gy vs 35.45 Gy) and the 5-year bladder complication probabilities (≥LENT/SOMA Grade 2: 20.35% vs 17.96%; bladder bleeding: 10.63% vs 8.64%). The doses to the rectum wall and the rectum complication probabilities were also slightly decreased by the EUD constraints compared to physical constraints only. The minimal dose and the V_76Gy of PTV_prostate were, however, slightly decreased by EUD optimization, nevertheless without significant difference in TCP values between the two plans, and the PTV parameters finally respected the Groupe d’Etude des Tumeurs Uro-Génitales recommendations.

Conclusion

Separating the bladder wall into two parts with appropriate EUD optimization may reduce bladder toxicity in prostate IMRT. Combining biological constraints with physical constraints in the organs at risk at the inverse planning step of IMRT may improve the dose distribution.

Predicting toxicity in radiotherapy for prostate cancer

This comprehensive review addresses most organs at risk involved in planning optimization for prostate cancer. It can be considered an update of a previous educational review that was published in 2009 (Fiorino et al., 2009). The literature was reviewed based on PubMed and MEDLINE database searches (from January 2009 up to September 2015), including papers in press; for each section/subsection, key title words were used and possibly combined with other more general key-words (such as radiotherapy, dose-volume effects, NTCP, DVH, and predictive model). Publications generally dealing with toxicity without any association with dose-volume effects or correlations with clinical risk factors were disregarded, being outside the aim of the review. A focus was on external beam radiotherapy, including post-prostatectomy, with conventional fractionation or moderate hypofractionation (<4Gy/fraction); extreme hypofractionation is the topic of another paper in this special issue. Gastrointestinal and urinary toxicity are the most investigated endpoints, with quantitative data published in the last 5years suggesting both a dose-response relationship and the existence of a number of clinical/patient related risk factors acting as dose-response modifiers. Some results on erectile dysfunction, bowel toxicity and hematological toxicity are also presented.

Temporal patterns of late bowel and bladder radiotherapy toxicity in a randomised controlled trial assessing duration of neo-adjuvant hormones in prostate cancer

BACKGROUND

To assess the temporal patterns of late gastrointestinal (GI) and genitourinary (GU) radiotherapy toxicity and resolution rates in a randomised controlled trial (All-Ireland Cooperative Oncology Research Group 97-01) assessing duration of neo-adjuvant (NA) hormone therapy for localised prostate cancer.

MATERIAL AND METHODS

Node negative patients with > 1 of: PSA > 20 ng/mL, Gleason score ≥ 7, and stage T3 or more, were included. Follow-up, including toxicity assessment, was three-monthly in the early stages and yearly thereafter.

RESULTS

Median follow-up from the end of RT was 6.8 years. In the interval between 90 days following the end of RT and the last toxicity assessment, GI and GU toxicity (any grade) was found in 50% and 51% of 240 and 241 patients, respectively. For those who did develop toxicity, the median time from end of RT until the first development of any grade GI or GU toxicity was 1.2 years and 1.6 years, respectively, whilst median time to final resolution was 1.6 years and 2.2 years, respectively. Grade 2 (G2) or greater GI and GU toxicity occurred in 29 (12.1%) and 40 (16.6%) patients, respectively. The proportion with unresolved G2 + GI and GU toxicity was 89% and 79%, respectively, in year 1, 69% and 65% in year 2, 59% and 52% in year 3 and 27% and 32% in year 5.

CONCLUSION

Long-term toxicities continue to occur many years after NA hormone therapy and RT. The rate of occurrence does not appear to reduce within the time frame during which our patients were followed. The percentage of patients suffering from G2 + toxicity at any time is however low. Resolution of these toxicities continues for the duration of the follow-up.

Novel parameter predicting grade 2 rectal bleeding after iodine-125 prostate brachytherapy combined with external beam radiation therapy

PURPOSE

To propose a novel parameter predicting rectal bleeding on the basis of generalized equivalent uniform doses (gEUD) after (125)I prostate brachytherapy combined with external beam radiation therapy and to assess the predictive value of this parameter.

METHODS AND MATERIALS

To account for differences among radiation treatment modalities and fractionation schedules, rectal dose-volume histograms (DVHs) of 369 patients with localized prostate cancer undergoing combined therapy retrieved from corresponding treatment planning systems were converted to equivalent dose-based DVHs. The gEUDs for the rectum were calculated from these converted DVHs. The total gEUD (gEUDsum) was determined by a summation of the brachytherapy and external-beam radiation therapy components.

RESULTS

Thirty-eight patients (10.3%) developed grade 2+ rectal bleeding. The grade 2+ rectal bleeding rate increased as the gEUDsum increased: 2.0% (2 of 102 patients) for <70 Gy, 10.3% (15 of 145 patients) for 70-80 Gy, 15.8% (12 of 76 patients) for 80-90 Gy, and 19.6% (9 of 46 patients) for >90 Gy (P=.002). Multivariate analysis identified age (P=.024) and gEUDsum (P=.000) as risk factors for grade 2+ rectal bleeding.

CONCLUSIONS

Our results demonstrate gEUD to be a potential predictive factor for grade 2+ late rectal bleeding after combined therapy for prostate cancer.