Development and external validation of multivariate prediction models for erectile dysfunction in men with localized prostate cancer.

While the 10-year survival rate for localized prostate cancer patients is very good (>98%), side effects of treatment may limit quality of life significantly. Erectile dysfunction (ED) is a common burden associated with increasing age as well as prostate cancer treatment. Although many studies have investigated the factors affecting erectile dysfunction (ED) after prostate cancer treatment, only limited studies have investigated whether ED can be predicted before the start of treatment. The advent of machine learning (ML) based prediction tools in oncology offers a promising approach to improve the accuracy of prediction and quality of care. Predicting ED may help aid shared decision-making by making the advantages and disadvantages of certain treatments clear, so that a tailored treatment for an individual patient can be chosen. This study aimed to predict ED at 1-year and 2-year post-diagnosis based on patient demographics, clinical data and patient-reported outcomes (PROMs) measured at diagnosis. We used a subset of the ProZIB dataset collected by the Netherlands Comprehensive Cancer Organization (Integraal Kankercentrum Nederland; IKNL) that contained information on 964 localized prostate cancer cases from 69 Dutch hospitals for model training and external validation. Two models were generated using a logistic regression algorithm coupled with Recursive Feature Elimination (RFE). The first predicted ED 1 year post-diagnosis and required 10 pre-treatment variables; the second predicted ED 2 years post-diagnosis with 9 pre-treatment variables. The validation AUCs were 0.84 and 0.81 for 1 year and 2 years post-diagnosis respectively. To immediately allow patients and clinicians to use these models in the clinical decision-making process, nomograms were generated. In conclusion, we successfully developed and validated two models that predicted ED in patients with localized prostate cancer. These models will allow physicians and patients alike to make informed evidence-based decisions about the most suitable treatment with quality of life in mind.

Taking shared decision making for prostate cancer to the next level: Requirements for a Dutch treatment decision aid with personalized risks on side effects.

Background: Different curative treatment modalities need to be considered in case of localized prostate cancer, all comparable in terms of survival and recurrence though different in side effects. To better inform patients and support shared decision making, the development of a web-based patient decision aid including personalized risk information was proposed. This paper reports on requirements in terms of content of information, visualization of risk profiles, and use in practice. Methods: Based on a Dutch 10-step guide about the setup of a decision aid next to a practice guideline, an iterative and co-creative design process was followed. In collaboration with various groups of experts (health professionals, usability and linguistic experts, patients and the general public), research and development activities were continuously alternated. Results: Content requirements focused on presenting information only about conventional treatments and main side effects; based on risk group; and including clear explanations about personalized risks. Visual requirements involved presenting general and personalized risks separately; through bar charts or icon arrays; and along with numbers or words, and legends. Organizational requirements included integration into local clinical pathways; agreement about information input and output; and focus on patients' numeracy and graph literacy skills. Conclusions: The iterative and co-creative development process was challenging, though extremely valuable. The translation of requirements resulted in a decision aid about four conventional treatment options, including general or personalized risks for erection, urinary and intestinal problems that are communicated with icon arrays and numbers. Future implementation and validation studies need to inform about use and value in practice.

Development and validation of a patient decision aid for prostate Cancer therapy: from paternalistic towards participative shared decision making.

BACKGROUND: Patient decision aids (PDAs) can support the treatment decision making process and empower patients to take a proactive role in their treatment pathway while using a shared decision-making (SDM) approach making participatory medicine possible. The aim of this study was to develop a PDA for prostate cancer that is accurate and user-friendly. METHODS: We followed a user-centered design process consisting of five rounds of semi-structured interviews and usability surveys with topics such as informational/decisional needs of users and requirements for PDAs. Our user-base consisted of 8 urologists, 4 radiation oncologists, 2 oncology nurses, 8 general practitioners, 19 former prostate cancer patients, 4 usability experts and 11 healthy volunteers. RESULTS: Informational needs for patients centered on three key factors: treatment experience, post-treatment quality of life, and the impact of side effects. Patients and clinicians valued a PDA that presents balanced information on these factors through simple understandable language and visual aids. Usability questionnaires revealed that patients were more satisfied overall with the PDA than clinicians; however, both groups had concerns that the PDA might lengthen consultation times (42 and 41%, respectively). The PDA is accessible on http://beslissamen.nl/ . CONCLUSIONS: User-centered design provided valuable insights into PDA requirements but challenges in integrating diverse perspectives as clinicians focus on clinical outcomes while patients also consider quality of life. Nevertheless, it is crucial to involve a broad base of clinical users in order to better understand the decision-making process and to develop a PDA that is accurate, usable, and acceptable.

Consequences of Intermodality Registration Errors for Intramodality 3D Ultrasound IGRT.

Intramodality ultrasound image-guided radiotherapy systems compare daily ultrasound to reference ultrasound images. Nevertheless, because the actual treatment planning is based on a reference computed tomography image, and not on a reference ultrasound image, their accuracy depends partially on the correct intermodality registration of the reference ultrasound and computed tomography images for treatment planning. The error propagation in daily patient positioning due to potential registration errors at the planning stage was assessed in this work. Five different scenarios were simulated involving shifts or rotations of ultrasound or computed tomography images. The consequences of several workflow procedures were tested with a phantom setup. As long as the reference ultrasound and computed tomography images are made to match, the patient will be in the correct treatment position. In an example with a phantom measurement, the accuracy of the performed manual fusion was found to be ≤2 mm. In clinical practice, manual registration of patient images is expected to be more difficult. Uncorrected mismatches will lead to a systematically incorrect final patient position because there will be no indication that there was a misregistration between the computed tomography and reference ultrasound images. In the treatment room, the fusion with the computed tomography image will not be visible and based on the ultrasound images the patient position seems correct.

Critical assessment of intramodality 3D ultrasound imaging for prostate IGRT compared to fiducial markers.

PURPOSE: A quantitative 3D intramodality ultrasound (US) imaging system was verified for daily in-room prostate localization, and compared to prostate localization based on implanted fiducial markers (FMs). METHODS: Thirteen prostate patients underwent multiple US scans during treatment. A total of 376 US-scans and 817 matches were used to determine the intra- and interoperator variability. Additionally, eight other patients underwent daily prostate localization using both US and electronic portal imaging (EPI) with FMs resulting in 244 combined US-EPI scans. Scanning was performed with minimal probe pressure and a correction for the speed of sound aberration was performed. Uncertainties of both US and FM methods were assessed. User variability of the US method was assessed. RESULTS: The overall US user variability is 2.6 mm. The mean differences between US and FM are: 2.5 ± 4.0 mm (LR), 0.6 ± 4.9 mm (SI), and -2.3 ± 3.6 mm (AP). The intramodality character of this US system mitigates potential errors due to transducer pressure and speed of sound aberrations. CONCLUSIONS: The overall accuracy of US (3.0 mm) is comparable to our FM workflow (2.2 mm). Since neither US nor FM can be considered a gold standard no conclusions can be drawn on the superiority of either method. Because US imaging captures the prostate itself instead of surrogates no invasive procedure is required. It requires more effort to standardize US imaging than FM detection. Since US imaging does not involve a radiation burden, US prostate imaging offers an alternative for FM EPI positioning.

Active breathing control in combination with ultrasound imaging: a feasibility study of image guidance in stereotactic body radiation therapy of liver lesions.

PURPOSE: Accurate tumor positioning in stereotactic body radiation therapy (SBRT) of liver lesions is often hampered by motion and setup errors. We combined 3-dimensional ultrasound imaging (3DUS) and active breathing control (ABC) as an image guidance tool. METHODS AND MATERIALS: We tested 3DUS image guidance in the SBRT treatment of liver lesions for 11 patients with 88 treatment fractions. In 5 patients, 3DUS imaging was combined with ABC. The uncertainties of US scanning and US image segmentation in liver lesions were determined with and without ABC. RESULTS: In free breathing, the intraobserver variations were 1.4 mm in left-right (L-R), 1.6 mm in superior-inferior (S-I), and 1.3 mm anterior-posterior (A-P). and the interobserver variations were 1.6 mm (L-R), 2.8 mm (S-I), and 1.2 mm (A-P). The combined uncertainty of US scanning and matching (inter- and intraobserver) was 4 mm (1 SD). The combined uncertainty when ABC was used reduced by 1.7 mm in the S-I direction. For the L-R and A-P directions, no significant difference was observed. CONCLUSION: 3DUS imaging for IGRT of liver lesions is feasible, although using anatomic surrogates in the close vicinity of the lesion may be needed. ABC-based breath-hold in midventilation during 3DUS imaging can reduce the uncertainty of US-based 3D table shift correction.

Magnitude of speed of sound aberration corrections for ultrasound image guided radiotherapy for prostate and other anatomical sites.

PURPOSE: The purpose of this work is to assess the magnitude of speed of sound (SOS) aberrations in three-dimensional ultrasound (US) imaging systems in image guided radiotherapy. The discrepancy between the fixed SOS value of 1540 m∕s assumed by US systems in human soft tissues and its actual nonhomogeneous distribution in patients produces small but systematic errors of up to a few millimeters in the positions of scanned structures. METHODS: A correction, provided by a previously published density-based algorithm, was applied to a set of five prostate, five liver, and five breast cancer patients. The shifts of the centroids of target structures and the change in shape were evaluated. RESULTS: After the correction the prostate cases showed shifts up to 3.6 mm toward the US probe, which may explain largely the reported positioning discrepancies in the literature on US systems versus other imaging modalities. Liver cases showed the largest changes in volume of the organ, up to almost 9%, and shifts of the centroids up to more than 6 mm either away or toward the US probe. Breast images showed systematic small shifts of the centroids toward the US probe with a maximum magnitude of 1.3 mm. CONCLUSIONS: The applied correction in prostate and liver cancer patients shows positioning errors of several mm due to SOS aberration; the errors are smaller in breast cancer cases, but possibly becoming more important when breast tissue thickness increases.

In vivo dosimetry with a linear MOSFET array to evaluate the urethra dose during permanent implant brachytherapy using iodine-125.

PURPOSE: To develop a technique to monitor the dose rate in the urethra during permanent implant brachytherapy using a linear MOSFET array, with sufficient accuracy and without significantly extending the implantation time. METHODS AND MATERIALS: Phantom measurements were performed to determine the optimal conditions for clinical measurements. In vivo measurements were performed in 5 patients during the (125)I brachytherapy implant procedure. To evaluate if the urethra dose obtained in the operating room with the ultrasound transducer in the rectum and the patient in treatment position is a reference for the total accumulated dose; additional measurements were performed after the implantation procedure, in the recovery room. RESULTS: In vivo measurements during and after the implantation procedure agree very well, illustrating that the ultrasound transducer in the rectum and patient positioning do not influence the measured dose in the urethra. In vivo dose values obtained during the implantation are therefore representative for the total accumulated dose in the urethra. In 5 patients, the dose rates during and after the implantation were below the maximum dose rate of the urethra, using the planned seed distribution. CONCLUSION: In vivo dosimetry during the implantation, using a MOSFET array, is a feasible technique to evaluate the dose in the urethra during the implantation of (125)I seeds for prostate brachytherapy.

In vivo dosimetry using a linear Mosfet-array dosimeter to determine the urethra dose in 125I permanent prostate implants.

PURPOSE: In vivo dosimetry during brachytherapy of the prostate with (125)I seeds is challenging because of the high dose gradients and low photon energies involved. We present the results of a study using metal-oxide-semiconductor field-effect transistor (MOSFET) dosimeters to evaluate the dose in the urethra after a permanent prostate implantation procedure. METHODS AND MATERIALS: Phantom measurements were made to validate the measurement technique, determine the measurement accuracy, and define action levels for clinical measurements. Patient measurements were performed with a MOSFET array in the urinary catheter immediately after the implantation procedure. A CT scan was performed, and dose values, calculated by the treatment planning system, were compared to in vivo dose values measured with MOSFET dosimeters. RESULTS: Corrections for temperature dependence of the MOSFET array response and photon attenuation in the catheter on the in vivo dose values are necessary. The overall uncertainty in the measurement procedure, determined in a simulation experiment, is 8.0% (1 SD). In vivo dose values were obtained for 17 patients. In the high-dose region (> 100 Gy), calculated and measured dose values agreed within 1.7% +/- 10.7% (1 SD). In the low-dose region outside the prostate (< 100 Gy), larger deviations occurred. CONCLUSIONS: MOSFET detectors are suitable for in vivo dosimetry during (125)I brachytherapy of prostate cancer. An action level of +/- 16% (2 SD) for detection of errors in the implantation procedure is achievable after validation of the detector system and measurement conditions.

Total body irradiation, toward optimal individual delivery: dose evaluation with metal oxide field effect transistors, thermoluminescence detectors, and a treatment planning system.

PURPOSE: To predict the three-dimensional dose distribution of our total body irradiation technique, using a commercial treatment planning system (TPS). In vivo dosimetry, using metal oxide field effect transistors (MOSFETs) and thermoluminescence detectors (TLDs), was used to verify the calculated dose distributions. METHODS AND MATERIALS: A total body computed tomography scan was performed and loaded into our TPS, and a three-dimensional-dose distribution was generated. In vivo dosimetry was performed at five locations on the patient. Entrance and exit dose values were converted to midline doses using conversion factors, previously determined with phantom measurements. The TPS-predicted dose values were compared with the MOSFET and TLD in vivo dose values. RESULTS: The MOSFET and TLD dose values agreed within 3.0% and the MOSFET and TPS data within 0.5%. The convolution algorithm of the TPS, which is routinely applied in the clinic, overestimated the dose in the lung region. Using a superposition algorithm reduced the calculated lung dose by approximately 3%. The dose inhomogeneity, as predicted by the TPS, can be reduced using a simple intensity-modulated radiotherapy technique. CONCLUSIONS: The use of a TPS to calculate the dose distributions in individual patients during total body irradiation is strongly recommended. Using a TPS gives good insight of the over- and underdosage in a patient and the influence of patient positioning on dose homogeneity. MOSFETs are suitable for in vivo dosimetry purposes during total body irradiation, when using appropriate conversion factors. The MOSFET, TLD, and TPS results agreed within acceptable margins.

Clinical implementation of MOSFET detectors for dosimetry in electron beams.

BACKGROUND AND PURPOSE: To determine the factors converting the reading of a MOSFET detector placed on the patient's skin without additional build-up to the dose at the depth of dose maximum (D(max)) and investigate their feasibility for in vivo dose measurements in electron beams. MATERIALS AND METHODS: Factors were determined to relate the reading of a MOSFET detector to D(max) for 4 - 15 MeV electron beams in reference conditions. The influence of variation in field size, SSD, angle and field shape on the MOSFET reading, obtained without additional build-up, was evaluated using 4, 8 and 15 MeV beams and compared to ionisation chamber data at the depth of dose maximum (z(max)). Patient entrance in vivo measurements included 40 patients, mostly treated for breast tumours. The MOSFET reading, converted to D(max), was compared to the dose prescribed at this depth. RESULTS: The factors to convert MOSFET reading to D(max) vary between 1.33 and 1.20 for the 4 and 15 MeV beams, respectively. The SSD correction factor is approximately 8% for a change in SSD from 95 to 100 cm, and 2% for each 5-cm increment above 100 cm SSD. A correction for fields having sides smaller than 6 cm and for irregular field shape is also recommended. For fields up to 20 x 20 cm(2) and for oblique incidence up to 45 degrees, a correction is not necessary. Patient measurements demonstrated deviations from the prescribed dose with a mean difference of -0.7% and a standard deviation of 2.9%. CONCLUSION: Performing dose measurements with MOSFET detectors placed on the patient's skin without additional build-up is a well suited technique for routine dose verification in electron beams, when applying the appropriate conversion and correction factors.

Clinical dosimetry with MOSFET dosimeters to determine the dose along the field junction in a split beam technique.

BACKGROUND AND PURPOSE: We report on the accuracy of metal oxide-silicon semiconductor field effect transistor (MOSFET) detectors to measure dose distributions in the region of a field junction in a split beam technique, compared to ionisation chamber and photographic film. We present a study on five patients receiving loco-regional treatment for breast cancer. MATERIALS AND METHODS: The dose variation at the junction was measured with the patient dose verification system model TN-RD-50 (MOSFET system). Phantom measurements were performed to investigate the MOSFET accuracy in a half-field matching method and the influence of factors related to the accelerator. The aim of the patient measurements was to determine the effect of patient related factors on the dose at the junction over a period of 10 irradiation fractions. RESULTS: The MOSFET detector overestimates the dose in the junction area, therefore a correction factor was determined to correct the MOSFET results. Phantom measurements showed overdosages as well as underdosages, depending on the matching field direction (X or Y). Patient measurements showed dose values that deviated up to 133%+/-3% (2 S.D.). The average values of 10 irradiation sessions showed a continuous, almost linear, variation of dose value across the junction. CONCLUSIONS: (1) Significant dose deviations in the junction area average out over repeated treatments; (2) one individual measurement in the junction area should not lead to any action; (3) a linac QA procedure to check the dose at the junction line should simulate the clinical situation sufficiently; and (4) MOSFET dosimeters overestimate dose values in the penumbra region.