Dose distribution prediction for head-and-neck cancer radiotherapy using a generative adversarial network: influence of input data.

Purpose: A three-dimensional deep generative adversarial network (GAN) was used to predict dose distributions for locally advanced head and neck cancer radiotherapy. Given the labor- and time-intensive nature of manual planning target volume (PTV) and organ-at-risk (OAR) segmentation, we investigated whether dose distributions could be predicted without the need for fully segmented datasets. Materials and methods: GANs were trained/validated/tested using 320/30/35 previously segmented CT datasets and treatment plans. The following input combinations were used to train and test the models: CT-scan only (C); CT+PTVboost/elective (CP); CT+PTVs+OARs+body structure (CPOB); PTVs+OARs+body structure (POB); PTVs+body structure (PB). Mean absolute errors (MAEs) for the predicted dose distribution and mean doses to individual OARs (individual salivary glands, individual swallowing structures) were analyzed. Results: For the five models listed, MAEs were 7.3 Gy, 3.5 Gy, 3.4 Gy, 3.4 Gy, and 3.5 Gy, respectively, without significant differences among CP-CPOB, CP-POB, CP-PB, among CPOB-POB. Dose volume histograms showed that all four models that included PTV contours predicted dose distributions that had a high level of agreement with clinical treatment plans. The best model CPOB and the worst model PB (except model C) predicted mean dose to within ±3 Gy of the clinical dose, for 82.6%/88.6%/82.9% and 71.4%/67.1%/72.2% of all OARs, parotid glands (PG), and submandibular glands (SMG), respectively. The R2 values (0.17/0.96/0.97/0.95/0.95) of OAR mean doses for each model also indicated that except for model C, the predictions correlated highly with the clinical dose distributions. Interestingly model C could reasonably predict the dose in eight patients, but on average, it performed inadequately. Conclusion: We demonstrated the influence of the CT scan, and PTV and OAR contours on dose prediction. Model CP was not statistically different from model CPOB and represents the minimum data statistically required to adequately predict the clinical dose distribution in a group of patients.

Focal 18 F-FDG uptake predicts progression of pre-invasive squamous bronchial lesions to invasive cancers.

INTRODUCTION: Pre-invasive squamous lesions of the central airways can progress into invasive lung cancers. Identifying these high-risk patients could enable detection of invasive lung cancers at an early stage. In this study, we investigated the value of 18 F-fluorodeoxyglucose (18 F-FDG) positron emission tomography (PET) scans in predicting progression in patients with pre-invasive squamous endobronchial lesions. METHODS: In this retrospective study, patients with pre-invasive endobronchial lesions, who underwent an 18 F-FDG PET scan at the VU University Medical Center Amsterdam, between January 2000 and December 2016, were included. Autofluorescence bronchoscopy (AFB) was used for tissue sampling and was repeated every 3 months. The minimum and median follow-up was 3 and 46.5 months. Study endpoints were the occurrence of biopsy proven invasive carcinoma, time-to-progression and overall survival (OS). RESULTS: A total number of 40 of 225 patients met the inclusion criteria of which 17 (42.5%) patients had a positive baseline 18 F-FDG PET scan. A total of 13 of 17 (76.5%) developed invasive lung carcinoma during follow-up, with a median time to progression of 5.0 months (range, 3.0-25.0). In 23 (57.5%) patients with a negative 18 F-FDG PET scan at baseline, 6 (26%) developed lung cancer, with a median time to progression of 34.0 months (range, 14.0-42.0 months, p < 0.002). With a median OS of 56.0 months (range, 9.0-60.0 months) versus 49.0 months (range, 6.0-60.0 months) (p = 0.876) for the 18 F-FDG PET positive and negative groups, respectively. CONCLUSIONS: Patients with pre-invasive endobronchial squamous lesions and a positive baseline 18 F-FDG PET scan were at high-risk for developing lung carcinoma, highlighting that this patient group requires early radical treatment.

The Impact of the Availability of Immunotherapy on Patterns of Care in Stage III NSCLC: A Dutch Multicenter Analysis.

INTRODUCTION: Treatment patterns in stage III NSCLC can vary considerably between countries. The PACIFIC trial reported improvements in progression-free and overall survival with adjuvant durvalumab after concurrent chemoradiotherapy (CCRT). We studied treatment decision-making by three Dutch regional thoracic multidisciplinary tumor boards between 2015 and 2019, to identify changes in practice when adjuvant durvalumab became available. METHODS: Details of patients presenting with stage III NSCLC were retrospectively collected. Both CCRT and multimodality schemes incorporating planned surgery were defined as being radical-intent treatment (RIT). RESULTS: Of 855 eligible patients, most (95%) were discussed at a thoracic multidisciplinary tumor board, which recommended a RIT in 63% (n = 510). Only 52% (n = 424) of the patients finally received a RIT. Predictors for not recommending RIT were age greater than or equal to 70 years, WHO performance score greater than or equal to 2, Charlson comorbidity index greater than or equal to 2 (excluding age), forced expiratory volume in 1 second less than 80% of predicted value, N3 disease, and period of diagnosis. Between 2015 to 2017 and 2018 to 2019, the proportion of patients undergoing CCRT increased from 34% to 42% (p = 0.02) and use of sequential chemoradiotherapy declined (21%-16%, p = 0.05). Rates of early toxicity and 1-year mortality were comparable for both periods. After 2018, 57% of the patients who underwent CCRT (90 of 159) received adjuvant durvalumab. CONCLUSIONS: After publication of the PACIFIC trial, a significant increase was observed in the use of CCRT for patients with stage III NSCLC with rates of early toxicity and mortality being unchanged. Since 2018, 57% of the patients undergoing CCRT went on to receive adjuvant durvalumab. Nevertheless, approximately half of the patients were still considered unfit for a RIT.

Nuclear Imaging for Bone Metastases in Prostate Cancer: The Emergence of Modern Techniques Using Novel Radiotracers.

Accurate staging of prostate cancer (PCa) at initial diagnosis and at biochemical recurrence is important to determine prognosis and the optimal treatment strategy. To date, treatment of metastatic PCa has mostly been based on the results of conventional imaging with abdominopelvic computed tomography (CT) and bone scintigraphy. However, these investigations have limited sensitivity and specificity which impairs their ability to accurately identify and quantify the true extent of active disease. Modern imaging modalities, such as those based on the detection of radioactively labeled tracers with combined positron emission tomography/computed tomography (PET/CT) scanning have been developed specifically for the detection of PCa. Novel radiotracers include 18F-sodium fluoride (NaF), 11C-/18F-fluorocholine (FCH), 18F-fluordihydrotestosterone (FDHT), 68Gallium and 18F-radiolabeled prostate-specific membrane antigen (e.g., 68Ga-PSMA-11, 18F-DCFPyL). PET/CT with these tracers outperforms conventional imaging. As a result of this, although their impact on outcome needs to be better defined in appropriate clinical trials, techniques like prostate-specific membrane antigen (PSMA) PET/CT have been rapidly adopted into clinical practice for (re)staging PCa. This review focuses on nuclear imaging for PCa bone metastases, summarizing the literature on conventional imaging (focusing on CT and bone scintigraphy-magnetic resonance imaging is not addressed in this review), highlighting the prognostic importance of high and low volume metastatic disease which serves as a driver for the development of better imaging techniques, and finally discussing modern nuclear imaging with novel radiotracers.

Factors influencing multi-disciplinary tumor board recommendations in stage III non-small cell lung cancer.

OBJECTIVES: Treatment patterns in patients with stage III non-small cell lung cancer (NSCLC) vary considerably between countries, for reasons that are not well understood. We studied factors influencing treatment decision-making at thoracic multidisciplinary tumor boards (MDT's) and outcome for patients treated between 2015-2017, at a regional network comprising 5 hospitals. MATERIALS AND METHODS: Details of all patients, including comorbidities, with stage III NSCLC were collected in an ethics-approved database. Weekly MDT's were conducted. The preferred radical intent treatments (RIT) for suitable patients were assumed to be concurrent chemoradiotherapy and/or surgery and other therapies were non-radical intent treatments (n-RIT). RESULTS: Of 197 patients identified, 95 % were discussed at an MDT. RIT were recommended in 61 % of patients, but only 48 % finally received RIT. The estimated median OS was significantly better for patients undergoing RIT (28.3 months, CI-95 % 17.3-39.3), versus those who did not (11.2 months, CI-95 % 8.0-14.3). Patient age ≥70 years and a WHO-PS ≥2 were the most important predictors of not recommending RIT. Deaths due to progressive lung cancer within 2 years were observed in 36, 26 and 29 % of patients who received RIT, sequential chemoradiotherapy or radical radiotherapy. Corresponding comorbidity related deaths within 2 years were 3, 12 and 38 %. CONCLUSION: A large number of patients who underwent MDT review were considered too old or not fit for RIT. More effective and better tolerated systemic treatments are required for patients presenting with stage III NSCLC.

Chest wall pain and rib fracture after stereotactic radiotherapy for peripheral non-small cell lung cancer.

Stereotactic body radiotherapy is an emerging treatment option for peripheral non-small cell lung cancer in medically inoperable patients. With high dose per fraction radiotherapy, late side effects are of possible concern. In our initial cohort of 42 patients treated with 54 to 60 Gy in three fractions, nine patients have rib fracture. The median dose to rib fracture sites was 46 to 50 Gy, depending on the method of dose calculation. We describe a typical case of poststereotactic radiotherapy rib fracture and present dosimetric analysis of patients with rib fracture.