Comparison of the use of a clinically implemented deep learning segmentation model with the simulated study setting for breast cancer patients receiving radiotherapy.

BACKGROUND: Deep learning (DL) models for auto-segmentation in radiotherapy have been extensively studied in retrospective and pilot settings. However, these studies might not reflect the clinical setting. This study compares the use of a clinically implemented in-house trained DL segmentation model for breast cancer to a previously performed pilot study to assess possible differences in performance or acceptability. MATERIAL AND METHODS: Sixty patients with whole breast radiotherapy, with or without an indication for locoregional radiotherapy were included. Structures were qualitatively scored by radiotherapy technologists and radiation oncologists. Quantitative evaluation was performed using dice-similarity coefficient (DSC), 95th percentile of Hausdorff Distance (95%HD) and surface DSC (sDSC), and time needed for generating, checking, and correcting structures was measured. RESULTS: Ninety-three percent of all contours in clinic were scored as clinically acceptable or usable as a starting point, comparable to 92% achieved in the pilot study. Compared to the pilot study, no significant changes in time reduction were achieved for organs at risks (OARs). For target volumes, significantly more time was needed compared to the pilot study for patients including lymph node levels 1-4, although time reduction was still 33% compared to manual segmentation. Almost all contours have better DSC and 95%HD than inter-observer variations. Only CTVn4 scored worse for both metrics, and the thyroid had a higher 95%HD value. INTERPRETATION: The use of the DL model in clinical practice is comparable to the pilot study, showing high acceptability rates and time reduction.

Oncologic outcomes of de-escalating axillary treatment in clinically node-positive breast cancer patients treated with neoadjuvant systemic therapy - A two center cohort study.

BACKGROUND: The aim of the present study was to report the 5-year axillary recurrence-free interval (aRFI) in clinically node-positive breast cancer patients treated according to a de-escalating axillary treatment protocol after neoadjuvant systemic therapy (NST). METHODS: All patients diagnosed in two hospitals between October 2014 and March 2021 were identified retrospectively. Data on diagnostic workup, treatment and follow-up was collected. Adjuvant axillary treatment was considered based on the initial staging using 18F-FDG PET/CT and the results of axillary lymph node marking with a radioactive-iodine seed protocol or a targeted axillary dissection procedure. Follow-up was updated until 27th April 2024. Kaplan-Meier curves were calculated to report the 5-year aRFI with corresponding 95 % confident intervals (95%-CI). RESULTS: A total of 199 patients were included. Axillary pathological complete response was reported in 66 (33.2 %). Based on the treatment protocol and initial clinical staging, no adjuvant axillary treatment was indicated in 30 patients (15 %), while 139 (70 %) received axillary radiotherapy without performance of an axillary lymph node dissection (ALND). The remaining 30 patients (15 %) underwent an ALND with additional locoregional radiotherapy. A median follow-up of 62 months (30-106) showed that 4 (2 %) patients experienced an axillary recurrence after 7, 8, 36 and 36 months, respectively. In all 4 patients, synchronous distant metastases were diagnosed. The estimated 5-year aRFI was 97.8 % (95%-CI 95.6-99.9 %) CONCLUSION: Although longer follow-up should be awaited before final conclusions can be drawn regarding the oncological safety of this approach, the implementation of a de-escalating axillary treatment protocol appears to be safe since the estimated 5-year aRFI is 97.8 %.

Ovarian function recovery in breast cancer patients receiving adjuvant anastrozole treatment: updated results from the phase 3 DATA trial.

PURPOSE: Patients with chemotherapy-induced ovarian function failure (CIOFF) may experience ovarian function recovery (OFR). Earlier, we showed that OFR during treatment with anastrozole impacted the prognosis of hormone receptor-positive (HR+) breast cancer (BC) patients with CIOFF. Here, we present the long-term follow-up results. METHODS: Postmenopausal women with HR+ BC who were 45-57 years of age and received chemotherapy were identified from the phase 3 DATA study (NCT00301457) on the extended use of anastrozole. Eligible patients were categorised into two groups: patients with CIOFF and definitely postmenopausal patients. Patients with CIOFF were monitored for OFR. Disease-free survival (DFS), distant recurrence-free survival (DRFS), and overall survival (OS) were compared between patients with OFR and patients without OFR using multivariable Cox regression analyses, including OFR as a time-dependent covariate. BC-specific mortality (BCSM) was compared between groups using the Fine and Gray method. RESULTS: This study included 656 patients: 395 patients with CIOFF and 261 definitely postmenopausal patients. OFR occurred in 39 (12%) of 329 patients with CIOFF who were monitored for OFR. The median follow-up time was 13.3 years. Patients with OFR experienced a deterioration in DFS (hazard ratio (HR) = 1.54; 95% confidence interval (CI) 0.85-2.81), DRFS (HR = 1.51; 95% CI 0.73-3.11), OS (HR = 1.64; 95% CI 0.75-3.55), and BCSM (subdistribution HR = 1.98; 95% CI 0.84-4.63) when compared with patients without OFR. CONCLUSION: In patients with CIOFF, OFR during treatment with anastrozole was associated with a deterioration in BC outcomes. These findings underscore the importance of adequate ovarian function suppression in this subgroup of patients.

Breast conserving surgery with intraoperative electron beam radiation therapy for low-risk breast cancer: Five-year follow-up of 306 patients.

Recent studies have reported a higher than expected risk of ipsilateral breast tumor recurrence (IBTR) after breast conserving surgery (BCS) and a single dose of electron beam intra-operative radiotherapy (IORT). This finding was the rationale to perform a retrospective single center cohort study evaluating the oncologic results of consecutive patients treated with BCS and IORT. Women were eligible if they had clinical low-risk (N0, ≤2 cm unifocal, Bloom and Richardson grade 1-2), estrogen receptor-positive and human-epidermal-growth-factor-receptor-2-negative breast cancer. Prior to BCS, pN0 status was determined by sentinel lymph node biopsy. Data on oncologic follow-up were analyzed. Between 2012 and 2019, 306 consecutive patients were treated and analyzed, with a median age of 67 (50-86) years at diagnosis. Median follow-up was 60 (8-120) months. Five-year cumulative risk of IBTR was 13.4% (95% confidence interval [CI] 9.4-17.4). True in field recurrence was present in 3.9% of the patients. In 4.6% of the patients, the IBRT was classified as a local recurrence due to seeding of tumor cells in the cutis or subcutis most likely related to percutaneous biopsy. In 2.9% of the patients, the IBRT was a new outfield primary tumor. Three patients had a regional lymph node recurrence and two had distant metastases as first event. One breast cancer-related death was observed. Estimated 5-year overall survival was 89.8% (95% CI 86.0-93.6). In conclusion, although some of IBTR cases could have been prevented by adaptations in biopsy techniques and patient selection, BCS followed by IORT was associated with a substantial risk of IBTR.

European clinical practice guidelines for the definition, diagnosis, and treatment of oligometastatic esophagogastric cancer (OMEC-4).

INTRODUCTION: The OligoMetastatic Esophagogastric Cancer (OMEC) project aims to provide clinical practice guidelines for the definition, diagnosis, and treatment of esophagogastric oligometastatic disease (OMD). METHODS: Guidelines were developed according to AGREE II and GRADE principles. Guidelines were based on a systematic review (OMEC-1), clinical case discussions (OMEC-2), and a Delphi consensus study (OMEC-3) by 49 European expert centers for esophagogastric cancer. OMEC identified patients for whom the term OMD is considered or could be considered. Disease-free interval (DFI) was defined as the time between primary tumor treatment and detection of OMD. RESULTS: Moderate to high quality of evidence was found (i.e. 1 randomized and 4 non-randomized phase II trials) resulting in moderate recommendations. OMD is considered in esophagogastric cancer patients with 1 organ with ≤ 3 metastases or 1 involved extra-regional lymph node station. In addition, OMD continues to be considered in patients with OMD without progression in number of metastases after systemic therapy. 18F-FDG PET/CT imaging is recommended for baseline staging and for restaging after systemic therapy when local treatment is considered. For patients with synchronous OMD or metachronous OMD and a DFI ≤ 2 years, recommended treatment consists of systemic therapy followed by restaging to assess suitability for local treatment. For patients with metachronous OMD and DFI > 2 years, upfront local treatment is additionally recommended. DISCUSSION: These multidisciplinary European clinical practice guidelines for the uniform definition, diagnosis and treatment of esophagogastric OMD can be used to standardize inclusion criteria in future clinical trials and to reduce variation in treatment.

Long-term trends in incidence, characteristics and prognosis of screen-detected and interval cancers in women participating in the Dutch breast cancer screening programme.

BACKGROUND: No studies are available in which changes over time in characteristics and prognosis of patients with interval breast cancers (ICs) and screen-detected breast cancers (SDCs) have been compared. The aim was to study these trends between 1995 and 2018. METHODS: All women with invasive SDCs (N = 4290) and ICs (N = 1352), diagnosed in a southern mammography screening region in the Netherlands, were included and followed until date of death or 31 December 2022. RESULTS: The 5-year overall survival rate of women with SDCs increased from 91.4% for those diagnosed in 1995-1999 to 95.0% for those diagnosed in 2013-2018 (P < 0.001), and from 74.8 to 91.6% (P < 0.001) in the same periods for those with ICs. A similar trend was observed for the 10-year survival rates. After adjustment for changes in tumour characteristics, the hazard ratio (HR) for overall survival was 0.47 (95% confidence interval (CI): 0.38-0.59) for women with SDCs diagnosed in the period 2013-2018, compared to the women diagnosed in the period 1995-1999. For the women with ICs this HR was 0.27 (95% CI: 0.19-0.40). CONCLUSION: The prognosis of women with ICs has improved rapidly since 1995 and is now almost similar to that of women with SDCs.

5-year adherence to adjuvant endocrine treatment in Dutch women with early stage breast cancer: A population-based database study (2006-2016).

BACKGROUND: Hormonal receptor (HR) positive breast tumors are common. Adjuvant hormonal therapy (AHT) with tamoxifen or Aromatase Inhibitors (AIs) is beneficial depending on the stage of the tumor. Despite the fact that AHT has been shown to improve survival and recurrence, Dutch adherence rates, which were mostly dependent on Tamoxifen prescriptions until 2006, plummeted from 80% after one year to 50% after five years. Nonadherence with AHT reduces its effectiveness. This research presents more recent adherence statistics (from 2006 to 2016), on a larger sample (7,996 vs 1,451), as well as factors that influence AHT adherence. In addition to tamoxifen data, AIs are now included. OBJECTIVE: As low use of adjuvant endocrine therapy is a potentially important and modifiable risk factor for poor outcome, it is important to monitor the rate as an indicator of women's burden of disease and the direction of adherence trends. METHODS: The Netherlands Cancer Registry (NCR) was used to find women with early-stage breast cancer who started AHT within a year of surgery and were linked to the PHARMO Database Network (n = 8,679). The Kaplan-Meier approach was used to measure AHT adherence five years after treatment was started, with a 60-day gap between refills as our primary outcome. Furthermore, the Medication Possession Rate (MPR) was determined using a cutoff of ≥80%. Analysis was performed on influential factors of adherence. RESULTS: The proportion of persistent women declined over time to reach 46.6% at the end of the fifth year and 53.3% of the women had a MPR ≥80% during the fifth year. Older and being diagnosed in 2006-2010 were associated with AHT adherence. CONCLUSION: Dutch 5-year AHT adherence appears to remain poor. Improving AHT adherence in HR+ breast cancer survivors is a critical medical need.

The prognostic and predictive effect of body mass index in hormone receptor-positive breast cancer.

BACKGROUND: Obesity has been associated with an adverse prognosis and reduced efficacy of endocrine therapy in patients with hormone receptor-positive (HR+) breast cancer (BC). This study determines the prognostic and predictive effect of body mass index (BMI) on the disease-free survival (DFS) of postmenopausal HR+ BC patients. METHODS: Patients were identified from the DATA study (NCT00301457), a randomized controlled trial evaluating the efficacy of 6 vs 3 years of anastrozole after 2 to 3 years of adjuvant tamoxifen in postmenopausal women with HR+ BC. Patients were classified as normal weight (BMI: 18.5-24.9 kg/m2), overweight (25.0-29.9 kg/m2), or obese (≥30.0 kg/m2). The primary endpoint was DFS, evaluated from randomization (prognostic analyses) or 3 years after randomization onwards (predictive analyses; aDFS) using multivariable Cox regression analyses. P-values were 2-sided. RESULTS: This study included 678 normal weight, 712 overweight, and 391 obese patients. After a median follow-up of 13.1 years, overweight and obesity were identified as negative prognostic factors for DFS (hazard ratio (HR) = 1.16; 95% confidence interval (CI) = 0.97 to 1.38 and HR = 1.26; 95% CI = 1.03 to 1.54, respectively). The adverse prognostic effect of BMI was observed in women aged younger than 60 years, but not in women aged 60 years or older (P-interaction = .009). The effect of extended anastrozole on aDFS was similar in normal weight (HR = 1.00; 95% CI = 0.74 to 1.35), overweight (HR = 0.74; 95% CI = 0.56 to 0.98), and obese patients (HR = 0.97; 95% CI = 0.69 to 1.36) (P-interaction = .24). CONCLUSION: In this study among 1781 HR+ BC patients, overweight and obesity were adverse prognostic factors for DFS. BMI did not impact the efficacy of extended anastrozole.

Evaluation of a clinically introduced deep learning model for radiotherapy treatment planning of breast cancer.

Deep learning (DL) models are increasingly studied to automate the process of radiotherapy treatment planning. This study evaluates the clinical use of such a model for whole breast radiotherapy. Treatment plans were automatically generated, after which planners were allowed to manually adapt them. Plans were evaluated based on clinical goals and DVH parameters. Thirty-seven of 50plans did fulfill all clinical goals without adjustments. Thirteen of these 37 plans were still adjusted but did not improve mean heart or lung dose. These results leave room for improvement of both the DL model as well as education on clinically relevant adjustments.

Clinical evaluation of a deep learning segmentation model including manual adjustments afterwards for locally advanced breast cancer.

Introduction: Deep learning (DL) models are increasingly developed for auto-segmentation in radiotherapy. Qualitative analysis is of great importance for clinical implementation, next to quantitative. This study evaluates a DL segmentation model for left- and right-sided locally advanced breast cancer both quantitatively and qualitatively. Methods: For each side a DL model was trained, including primary breast CTV (CTVp), lymph node levels 1-4, heart, lungs, humeral head, thyroid and esophagus. For evaluation, both automatic segmentation, including correction of contours when needed, and manual delineation was performed and both processes were timed. Quantitative scoring with dice-similarity coefficient (DSC), 95% Hausdorff Distance (95%HD) and surface DSC (sDSC) was used to compare both the automatic (not-corrected) and corrected contours with the manual contours. Qualitative scoring was performed by five radiotherapy technologists and five radiation oncologists using a 3-point Likert scale. Results: Time reduction was achieved using auto-segmentation in 95% of the cases, including correction. The time reduction (mean ± std) was 42.4% ± 26.5% and 58.5% ± 19.1% for OARs and CTVs, respectively, corresponding to an absolute mean reduction (hh:mm:ss) of 00:08:51 and 00:25:38. Good quantitative results were achieved before correction, e.g. mean DSC for the right-sided CTVp was 0.92 ± 0.06, whereas correction statistically significantly improved this contour by only 0.02 ± 0.05, respectively. In 92% of the cases, auto-contours were scored as clinically acceptable, with or without corrections. Conclusions: A DL segmentation model was trained and was shown to be a time-efficient way to generate clinically acceptable contours for locally advanced breast cancer.

Comparison of the output of a deep learning segmentation model for locoregional breast cancer radiotherapy trained on 2 different datasets.

Introduction: The development of deep learning (DL) models for auto-segmentation is increasing and more models become commercially available. Mostly, commercial models are trained on external data. To study the effect of using a model trained on external data, compared to the same model trained on in-house collected data, the performance of these two DL models was evaluated. Methods: The evaluation was performed using in-house collected data of 30 breast cancer patients. Quantitative analysis was performed using Dice similarity coefficient (DSC), surface DSC (sDSC) and 95th percentile of Hausdorff Distance (95% HD). These values were compared with previously reported inter-observer variations (IOV). Results: For a number of structures, statistically significant differences were found between the two models. For organs at risk, mean values for DSC ranged from 0.63 to 0.98 and 0.71 to 0.96 for the in-house and external model, respectively. For target volumes, mean DSC values of 0.57 to 0.94 and 0.33 to 0.92 were found. The difference of 95% HD values ranged 0.08 to 3.23 mm between the two models, except for CTVn4 with 9.95 mm. For the external model, both DSC and 95% HD are outside the range of IOV for CTVn4, whereas this is the case for the DSC found for the thyroid of the in-house model. Conclusions: Statistically significant differences were found between both models, which were mostly within published inter-observer variations, showing clinical usefulness of both models. Our findings could encourage discussion and revision of existing guidelines, to further decrease inter-observer, but also inter-institute variability.

Hospital practice variation in the proportion of patients with esophagogastric cancer discussed during an expert multidisciplinary team meeting.

BACKGROUND: Multidisciplinary team meetings (MDTM) and especially MDTMs in which expert centres are involved (expert MDTMs) are a key element in adequate cancer care. However, variation among hospitals in the proportion of patients presented during an expert MDTM has been described. This study aims to investigate national practice variation in the proportion of patients with oesophageal or gastric cancer being discussed during an expert MDTM. METHODS: Patients diagnosed with oesophageal or gastric cancer in 2018-2019 were selected from the Netherlands Cancer Registry (n = 6,921). Multilevel logistic regression analyses were used to analyse the association between patient, and tumour characteristics, and the probability to be discussed in an expert MDTM. Variation was analysed according to the hospital and region of diagnosis for: all patients, patients with a potentially curable (cT1-4A cTX, any cN, cM0) or incurable tumour stage (cT4b and/or cM1). RESULTS: In total, 79% of patients were discussed during an expert MDTM, of whom 84% (n = 3,424) and 71% (n = 2,018) with potentially curable, or incurable oesophageal or gastric cancer, respectively. The proportion of patients discussed during an expert MDTM ranged from 54% to 98%, and 17% to 100% between hospitals for potentially curable and incurable patients, respectively (all p < 0.0001). Adjusted analyses showed significant hospital (all p < 0.0001), but no regional variation regarding the patients discussed during an expert MDTM. CONCLUSION: For patients with oesophageal or gastric cancer the probability of being discussed during an expert MDTM varies considerably according to the hospital of diagnosis.

Definition, diagnosis and treatment of oligometastatic oesophagogastric cancer: A Delphi consensus study in Europe.

BACKGROUND: Local treatment improves the outcomes for oligometastatic disease (OMD, i.e. an intermediate state between locoregional and widespread disseminated disease). However, consensus about the definition, diagnosis and treatment of oligometastatic oesophagogastric cancer is lacking. The aim of this study was to develop a multidisciplinary European consensus statement on the definition, diagnosis and treatment of oligometastatic oesophagogastric cancer. METHODS: In total, 65 specialists in the multidisciplinary treatment for oesophagogastric cancer from 49 expert centres across 16 European countries were requested to participate in this Delphi study. The consensus finding process consisted of a starting meeting, 2 online Delphi questionnaire rounds and an online consensus meeting. Input for Delphi questionnaires consisted of (1) a systematic review on definitions of oligometastatic oesophagogastric cancer and (2) a discussion of real-life clinical cases by multidisciplinary teams. Experts were asked to score each statement on a 5-point Likert scale. The agreement was scored to be either absent/poor (<50%), fair (50%-75%) or consensus (≥75%). RESULTS: A total of 48 experts participated in the starting meeting, both Delphi rounds, and the consensus meeting (overall response rate: 71%). OMD was considered in patients with metastatic oesophagogastric cancer limited to 1 organ with ≤3 metastases or 1 extra-regional lymph node station (consensus). In addition, OMD was considered in patients without progression at restaging after systemic therapy (consensus). For patients with synchronous or metachronous OMD with a disease-free interval ≤2 years, systemic therapy followed by restaging to consider local treatment was considered as treatment (consensus). For metachronous OMD with a disease-free interval >2 years, either upfront local treatment or systemic treatment followed by restaging was considered as treatment (fair agreement). CONCLUSION: The OMEC project has resulted in a multidisciplinary European consensus statement for the definition, diagnosis and treatment of oligometastatic oesophagogastric adenocarcinoma and squamous cell cancer. This can be used to standardise inclusion criteria for future clinical trials.

Extended adjuvant aromatase inhibition after sequential endocrine therapy in postmenopausal women with breast cancer: follow-up analysis of the randomised phase 3 DATA trial.

Background: The DATA study evaluated the use of two different durations of anastrozole in patients with hormone receptor-positive breast cancer who were disease-free after 2-3 years of tamoxifen. We hereby present the follow-up analysis, which was performed after all patients reached a minimum follow-up of 10 years beyond treatment divergence. Methods: The open-label, randomised, phase 3 DATA study was performed in 79 hospitals in the Netherlands (ClinicalTrials.gov, number NCT00301457). Postmenopausal women with hormone receptor-positive breast cancer who were disease-free after 2-3 years of adjuvant tamoxifen treatment were assigned to either 3 or 6 years of anastrozole (1 mg orally once a day). Randomisation (1:1) was stratified by hormone receptor status, nodal status, HER2 status, and prior tamoxifen duration. The primary outcome was adapted disease-free survival, defined as disease-free survival from 3 years after randomisation onwards. Adapted overall survival was assessed as a secondary outcome. Analyses were performed according to the intention-to-treat design. Findings: Between June 28, 2006, and August 10, 2009, 1912 patients were randomly assigned to 3 years (n = 955) or 6 years (n = 957) of anastrozole. Of these, 1660 patients were eligible and disease-free at 3 years after randomisation. The 10-year adapted disease-free survival was 69.2% (95% CI 55.8-72.3) in the 6-year group (n = 827) and 66.0% (95% CI 62.5-69.2) in the 3-year group (n = 833) (hazard ratio (HR) 0.86; 95% CI 0.72-1.01; p = 0.073). The 10-year adapted overall survival was 80.9% (95% CI 77.9-83.5) in the 6-year group and 79.2% (95% CI 76.2-81.9) in the 3-year group (HR 0.93; 95% CI 0.75-1.16; p = 0.53). Interpretation: Extended aromatase inhibition beyond 5 years of sequential endocrine therapy did not improve the adapted disease-free survival and adapted overall survival of postmenopausal women with hormone receptor-positive breast cancer. Funding: AstraZeneca.

Corrigendum to "Harmonization of breast cancer radiotherapy treatment planning in the Netherlands" [Tech. Innov. Patient Support Radiat. Oncol. 19 (2021) 26-32].

[This corrects the article DOI: 10.1016/j.tipsro.2021.06.004.].

Trends and variation in the use of radiotherapy in non-metastatic prostate cancer: A 12-year nationwide overview from the Netherlands.

PURPOSE: This population-based study describes nationwide trends and variation in the use of primary radiotherapy for non-metastatic prostate cancer in The Netherlands in 2008-2019. METHODS: Prostate cancer patients were selected from the Netherlands Cancer Registry (N = 103,059). Treatment trends were studied over time by prognostic risk groups. Multilevel analyses were applied to identify variables associated with external beam radiotherapy (EBRT) and brachy-monotherapy versus no active treatment in low-risk disease, and EBRT versus radical prostatectomy in intermediate and high-risk disease. RESULTS: EBRT use remained stable (5-6%) in low-risk prostate cancer and increased from 21% to 32% in intermediate-risk, 37% to 45% in high-risk localized and 50% to 57% in high-risk locally advanced disease. Brachy-monotherapy decreased from 19% to 6% and from 15% to 10% in low and intermediate-risk disease, respectively, coinciding an increase of no active treatment from 55% to 73% in low-risk disease. Use of EBRT or brachy-monotherapy versus no active treatment in low-risk disease differed by region, T-stage and patient characteristics. Hospital characteristics were not associated with treatment in low-risk disease, except for availability of brachy-monotherapy in 2008-2013. Age, number of comorbidities, travel time for EBRT, prognostic risk group, and hospital characteristics were associated with EBRT versus prostatectomy in intermediate and high-risk disease. CONCLUSION: Intermediate/high-risk PCa was increasingly managed with EBRT, while brachy-monotherapy in low/intermediate-risk PCa decreased. In low-risk PCa, the no active treatment-approach increased. Variation in treatment suggests treatment decision related to patient/disease characteristics. In intermediate/high-risk disease, variation seems furthermore related to the treatment modalities available in the diagnosing hospitals.

Clinical evaluation of two AI models for automated breast cancer plan generation.

BACKGROUND: Artificial intelligence (AI) shows great potential to streamline the treatment planning process. However, its clinical adoption is slow due to the limited number of clinical evaluation studies and because often, the translation of the predicted dose distribution to a deliverable plan is lacking. This study evaluates two different, deliverable AI plans in terms of their clinical acceptability based on quantitative parameters and qualitative evaluation by four radiation oncologists. METHODS: For 20 left-sided node-negative breast cancer patients, treated with a prescribed dose of 40.05 Gy, using tangential beam intensity modulated radiotherapy, two model-based treatment plans were evaluated against the corresponding manual plan. The two models used were an in-house developed U-net model and a vendor-developed contextual atlas regression forest model (cARF). Radiation oncologists evaluated the clinical acceptability of each blinded plan and ranked plans according to preference. Furthermore, a comparison with the manual plan was made based on dose volume histogram parameters, clinical evaluation criteria and preparation time. RESULTS: The U-net model resulted in a higher average and maximum dose to the PTV (median difference 0.37 Gy and 0.47 Gy respectively) and a slightly higher mean heart dose (MHD) (0.01 Gy). The cARF model led to higher average and maximum doses to the PTV (0.30 and 0.39 Gy respectively) and a slightly higher MHD (0.02 Gy) and mean lung dose (MLD, 0.04 Gy). The maximum MHD/MLD difference was ≤ 0.5 Gy for both AI plans. Regardless of these dose differences, 90-95% of the AI plans were considered clinically acceptable versus 90% of the manual plans. Preferences varied between the radiation oncologists. Plan preparation time was comparable between the U-net model and the manual plan (287 s vs 253 s) while the cARF model took longer (471 s). When only considering user interaction, plan generation time was 121 s for the cARF model and 137 s for the U-net model. CONCLUSIONS: Two AI models were used to generate deliverable plans for breast cancer patients, in a time-efficient manner, requiring minimal user interaction. Although the AI plans resulted in slightly higher doses overall, radiation oncologists considered 90-95% of the AI plans clinically acceptable.

Neoadjuvant Chemoradiotherapy Versus Upfront Surgery for Resectable and Borderline Resectable Pancreatic Cancer: Long-Term Results of the Dutch Randomized PREOPANC Trial.

PURPOSE: The benefit of neoadjuvant chemoradiotherapy in resectable and borderline resectable pancreatic cancer remains controversial. Initial results of the PREOPANC trial failed to demonstrate a statistically significant overall survival (OS) benefit. The long-term results are reported. METHODS: In this multicenter, phase III trial, patients with resectable and borderline resectable pancreatic cancer were randomly assigned (1:1) to neoadjuvant chemoradiotherapy or upfront surgery in 16 Dutch centers. Neoadjuvant chemoradiotherapy consisted of three cycles of gemcitabine combined with 36 Gy radiotherapy in 15 fractions during the second cycle. After restaging, patients underwent surgery followed by four cycles of adjuvant gemcitabine. Patients in the upfront surgery group underwent surgery followed by six cycles of adjuvant gemcitabine. The primary outcome was OS by intention-to-treat. No safety data were collected beyond the initial report of the trial. RESULTS: Between April 24, 2013, and July 25, 2017, 246 eligible patients were randomly assigned to neoadjuvant chemoradiotherapy (n = 119) and upfront surgery (n = 127). At a median follow-up of 59 months, the OS was better in the neoadjuvant chemoradiotherapy group than in the upfront surgery group (hazard ratio, 0.73; 95% CI, 0.56 to 0.96; P = .025). Although the difference in median survival was only 1.4 months (15.7 months v 14.3 months), the 5-year OS rate was 20.5% (95% CI, 14.2 to 29.8) with neoadjuvant chemoradiotherapy and 6.5% (95% CI, 3.1 to 13.7) with upfront surgery. The effect of neoadjuvant chemoradiotherapy was consistent across the prespecified subgroups, including resectable and borderline resectable pancreatic cancer. CONCLUSION: Neoadjuvant gemcitabine-based chemoradiotherapy followed by surgery and adjuvant gemcitabine improves OS compared with upfront surgery and adjuvant gemcitabine in resectable and borderline resectable pancreatic cancer.

Artificial intelligence based treatment planning of radiotherapy for locally advanced breast cancer.

BACKGROUND AND PURPOSE: Treatment planning of radiotherapy for locally advanced breast cancer patients can be a time consuming process. Artificial intelligence based treatment planning could be used as a tool to speed up this process and maintain plan quality consistency. The purpose of this study was to create treatment plans for locally advanced breast cancer patients using a Convolutional Neural Network (CNN). MATERIALS AND METHODS: Data of 60 patients treated for left-sided breast cancer was used with a training, validation and test split of 36/12/12, respectively. The in-house built CNN model was a hierarchically densely connected U-net (HD U-net). The inputs for the HD U-net were 2D distance maps of the relevant regions of interest. Dose predictions, generated by the HD U-net, were used for a mimicking algorithm in order to create clinically deliverable plans. RESULTS: Dose predictions were generated by the HD U-net and mimicked using a commercial treatment planning system. The predicted plans fulfilling all clinical goals while showing small (≤0.5 Gy) statistically significant differences (p < 0.05) in the doses compared to the manual plans. The mimicked plans show statistically significant differences in the average doses for the heart and lung of ≤0.5 Gy and a reduced D2% of all PTVs. In total, ten of the twelve mimicked plans were clinically acceptable. CONCLUSIONS: We created a CNN model which can generate clinically acceptable plans for left-sided locally advanced breast cancer patients. This model shows great potential to speed up the treatment planning process while maintaining consistent plan quality.