Multi-domain neurocognitive impairment following definitive intensity-modulated radiotherapy for nasopharyngeal cancer: A cross-sectional study.

INTRODUCTION: Neurocognitive impairment from inadvertent brain irradiation is common following intensity-modulated radiotherapy (IMRT) for nasopharyngeal carcinoma (NPC). This study aimed to determine the prevalence, pattern, and radiation dose-toxicity relationship of this late complication. MATERIALS AND METHODS: We undertook a cross-sectional study of 190 post-IMRT NPC survivors. Neurocognitive function was screened using the Montreal Cognitive Assessment-Hong Kong (HK-MoCA). Detailed assessments of eight distinct neurocognitive domains were conducted: intellectual capacity (WAIS-IV), attention span (Digit Span and Visual Spatial Span), visual memory (Visual Reproduction Span), verbal memory (Auditory Verbal Learning Test), processing speed (Color Trail Test), executive function (Stroop Test), motor dexterity (Grooved Pegboard Test) and language ability (Verbal Fluency Test). The mean percentiles and Z-scores were compared with normative population data. Associations between radiation dose and brain substructures were explored using multivariable logistic regression. RESULTS: The median post-IMRT interval was 7.0 years. The prevalence of impaired HK-MoCA was 25.3 % (48/190). Among the participants, 151 (79.4 %) exhibited impairments in at least one neurocognitive domain. The predominantly impaired domains included verbal memory (short-term: mean Z-score, -0.56, p < 0.001; long-term: mean Z-score, -0.70, p < 0.001), processing speed (basic: mean Z-score, -1.04, p < 0.001; advanced: mean Z-score, -0.38, p < 0.001), executive function (mean Z-score, -1.90, p < 0.001), and motor dexterity (dominant hand: mean Z-score, -0.97, p < 0.001). Radiation dose to the whole brain, hippocampus, and temporal lobe was associated with impairments in executive function, verbal memory, processing speed, and motor dexterity. CONCLUSIONS: Neurocognitive impairment is prevalent and profound in post-IMRT NPC survivors. Cognitive assessment and rehabilitation should be considered part of survivorship care.

Model Generalizability Investigation for GFCE-MRI Synthesis in NPC Radiotherapy Using Multi-Institutional Patient-Based Data Normalization.

Recently, deep learning has been demonstrated to be feasible in eliminating the use of gadoliniumbased contrast agents (GBCAs) through synthesizing gadolinium-free contrast-enhanced MRI (GFCE-MRI) from contrast-free MRI sequences, providing the community with an alternative to get rid of GBCAs-associated safety issues in patients. Nevertheless, generalizability assessment of the GFCE-MRI model has been largely challenged by the high inter-institutional heterogeneity of MRI data, on top of the scarcity of multi-institutional data itself. Although various data normalization methods have been adopted to address the heterogeneity issue, it has been limited to single-institutional investigation and there is no standard normalization approach presently. In this study, we aimed at investigating generalizability of GFCE-MRI model using data from seven institutions by manipulating heterogeneity of MRI data under five popular normalization approaches. Three state-of-the-art neural networks were applied to map from T1-weighted and T2-weighted MRI to contrast-enhanced MRI (CE-MRI) for GFCE-MRI synthesis in patients with nasopharyngeal carcinoma. MRI data from three institutions were used separately to generate three uni-institution models and jointly for a tri-institution model. The five normalization methods were applied to normalize the data of each model. MRI data from the remaining four institutions served as external cohorts for model generalizability assessment. Quality of GFCE-MRI was quantitatively evaluated against ground-truth CE-MRI using mean absolute error (MAE) and peak signal-to-noise ratio(PSNR). Results showed that performance of all uni-institution models remarkably dropped on the external cohorts. By contrast, model trained using multi-institutional data with Z-Score normalization yielded the best model generalizability improvement.

Explainable machine learning via intra-tumoral radiomics feature mapping for patient stratification in adjuvant chemotherapy for locoregionally advanced nasopharyngeal carcinoma.

PURPOSE: This study aimed to discover intra-tumor heterogeneity signature and validate its predictive value for adjuvant chemotherapy (ACT) following concurrent chemoradiotherapy (CCRT) in locoregionally advanced nasopharyngeal carcinoma (LA-NPC). MATERIALS AND METHODS: 397 LA-NPC patients were retrospectively enrolled. Pre-treatment contrast-enhanced T1-weighted (CET1-w) MR images, clinical variables, and follow-up were retrospectively collected. We identified single predictive radiomic feature from primary gross tumor volume (GTVnp) and defined predicted subvolume by calculating voxel-wised feature mapping and within GTVnp. We independently validate predictive value of identified feature and associated predicted subvolume. RESULTS: Only one radiomic feature, gldm_DependenceVariance in 3 mm-sigma LoG-filtered image, was discovered as a signature. In the high-risk group determined by the signature, patients received CCRT + ACT achieved 3-year disease free survival (DFS) rate of 90% versus 57% (HR, 0.20; 95%CI, 0.05-0.94; P = 0.007) for CCRT alone. The multivariate analysis showed patients receiving CCRT + ACT had a HR of 0.21 (95%CI: 0.06-0.68, P = 0.009) for DFS compared to those receiving CCRT alone. The predictive value can also be generalized to the subvolume with multivariate HR of 0.27 (P = 0.017) for DFS. CONCLUSION: The signature with its heterogeneity mapping could be a reliable and explainable ACT decision-making tool in clinical practice.

Radiomic feature repeatability and its impact on prognostic model generalizability: A multi-institutional study on nasopharyngeal carcinoma patients.

BACKGROUND AND PURPOSE: To investigate the radiomic feature (RF) repeatability via perturbation and its impact on cross-institutional prognostic model generalizability in Nasopharyngeal Carcinoma (NPC) patients. MATERIALS AND METHODS: 286 and 183 NPC patients from two institutions were included for model training and validation. Perturbations with random translations and rotations were applied to contrast-enhanced T1-weighted (CET1-w) MR images. RFs were extracted from primary tumor volume under a wide range of image filtering and discretization settings. RF repeatability was assessed by intraclass correlation coefficient (ICC), which was used to equally separate the RFs into low- and high-repeatable groups by the median value. After feature selection, multivariate Cox regression and Kaplan-Meier analysis were independently employed to develop and analyze prognostic models. Concordance index (C-index) and P-value from log-rank test were used to assess model performance. RESULTS: Most textural RFs from high-pass wavelet-filtered images were susceptible to image perturbations. It was more prominent when a smaller discretization bin number was used (e.g., 8, mean ICC = 0.69). Using high-repeatable RFs for model development yielded a significantly higher C-index (0.63) in the validation cohort than when only low-repeatable RFs were used (0.57, P = 0.024), suggesting higher model generalizability. Besides, significant risk stratification in the validation cohort was observed only when high-repeatable RFs were used (P < 0.001). CONCLUSION: Repeatability of RFs from high-pass wavelet-filtered CET1-w MR images of primary NPC tumor was poor, particularly when a smaller bin number was used. Exclusive use of high-repeatable RFs is suggested to safeguard model generalizability for wide-spreading clinical utilization.

Post-radiation primary hypothyroidism in patients with head and neck cancer: External validation of thyroid gland dose-volume constraints with long-term endocrine outcomes.

BACKGROUND: Post-radiation primary hypothyroidism is a common late complication in head and neck cancer (HNC) survivors. No radiation dose-volume constraint of the thyroid gland has been externally validated for predicting long-term thyroid function outcomes. MATERIALS AND METHODS: This external validation study evaluated the diagnostic properties of 22 radiation dose-volume constraints of the thyroid gland proposed in the literature. Radiation dosimetric data from 488 HNC patients who underwent neck irradiation from January 2013 to December 2015 at two tertiary oncology centers were reviewed. The diagnostic metrics of candidate constraints were computed by inverse probability of censoring weighting and compared using time-dependent receiver operating characteristic (ROC) curves with death designated as a competing event. Multivariable regression analyses were performed using the Fine-Gray sub-distribution hazard model. RESULTS: Over a median follow-up period of 6.8 years, 205 (42.0 %) patients developed post-radiation primary hypothyroidism. The thyroid volume spared from 60 Gy (VS60) had the largest area under ROC curve of 0.698 at 5 years after radiotherapy. Of all evaluated constraints, VS60 at a cutoff value of 10 cc had the highest F-score of 0.53. The 5-year hypothyroidism risks of patients with thyroid VS60 ≥ 10 cc and < 10 cc were 14.7 % and 38.2 %, respectively (p < 0.001). The adjusted sub-hazard ratio for post-radiation primary hypothyroidism for VS60 < 10 cc was 1.87 (95 % confidence interval, 1.22-2.87; p < 0.001). CONCLUSION: Thyroid VS60 is the best radiation dose-volume parameter to predict the long-term risk of primary hypothyroidism in patients with HNC who underwent neck irradiation. VS60 ≥ 10 cc is a robust constraint that limits the 5-year primary hypothyroidism risk to less than 15 % and should be routinely employed during radiotherapy optimization.

Improving radiomic model reliability using robust features from perturbations for head-and-neck carcinoma.

Background: Using high robust radiomic features in modeling is recommended, yet its impact on radiomic model is unclear. This study evaluated the radiomic model's robustness and generalizability after screening out low-robust features before radiomic modeling. The results were validated with four datasets and two clinically relevant tasks. Materials and methods: A total of 1,419 head-and-neck cancer patients' computed tomography images, gross tumor volume segmentation, and clinically relevant outcomes (distant metastasis and local-regional recurrence) were collected from four publicly available datasets. The perturbation method was implemented to simulate images, and the radiomic feature robustness was quantified using intra-class correlation of coefficient (ICC). Three radiomic models were built using all features (ICC > 0), good-robust features (ICC > 0.75), and excellent-robust features (ICC > 0.95), respectively. A filter-based feature selection and Ridge classification method were used to construct the radiomic models. Model performance was assessed with both robustness and generalizability. The robustness of the model was evaluated by the ICC, and the generalizability of the model was quantified by the train-test difference of Area Under the Receiver Operating Characteristic Curve (AUC). Results: The average model robustness ICC improved significantly from 0.65 to 0.78 (P< 0.0001) using good-robust features and to 0.91 (P< 0.0001) using excellent-robust features. Model generalizability also showed a substantial increase, as a closer gap between training and testing AUC was observed where the mean train-test AUC difference was reduced from 0.21 to 0.18 (P< 0.001) in good-robust features and to 0.12 (P< 0.0001) in excellent-robust features. Furthermore, good-robust features yielded the best average AUC in the unseen datasets of 0.58 (P< 0.001) over four datasets and clinical outcomes. Conclusions: Including robust only features in radiomic modeling significantly improves model robustness and generalizability in unseen datasets. Yet, the robustness of radiomic model has to be verified despite building with robust radiomic features, and tightly restricted feature robustness may prevent the optimal model performance in the unseen dataset as it may lower the discrimination power of the model.

A Multi-Center Study of CT-Based Neck Nodal Radiomics for Predicting an Adaptive Radiotherapy Trigger of Ill-Fitted Thermoplastic Masks in Patients with Nasopharyngeal Carcinoma.

Significant lymph node shrinkage is common in patients with nasopharyngeal carcinoma (NPC) throughout radiotherapy (RT) treatment, causing ill-fitted thermoplastic masks (IfTMs). To deal with this, an ad hoc adaptive radiotherapy (ART) may be required to ensure accurate and safe radiation delivery and to maintain treatment efficacy. Presently, the entire procedure for evaluating an eligible ART candidate is time-consuming, resource-demanding, and highly inefficient. In the artificial intelligence paradigm, the pre-treatment identification of NPC patients at risk for IfTMs has become greatly demanding for achieving efficient ART eligibility screening, while no relevant studies have been reported. Hence, we aimed to investigate the capability of computed tomography (CT)-based neck nodal radiomics for predicting IfTM-triggered ART events in NPC patients via a multi-center setting. Contrast-enhanced CT and the clinical data of 124 and 58 NPC patients from Queen Elizabeth Hospital (QEH) and Queen Mary Hospital (QMH), respectively, were retrospectively analyzed. Radiomic (R), clinical (C), and combined (RC) models were developed using the ridge algorithm in the QEH cohort and evaluated in the QMH cohort using the median area under the receiver operating characteristics curve (AUC). Delong's test was employed for model comparison. Model performance was further assessed on 1000 replicates in both cohorts separately via bootstrapping. The R model yielded the highest "corrected" AUC of 0.784 (BCa 95%CI: 0.673-0.859) and 0.723 (BCa 95%CI: 0.534-0.859) in the QEH and QMH cohort following bootstrapping, respectively. Delong's test indicated that the R model performed significantly better than the C model in the QMH cohort (p < 0.0001), while demonstrating no significant difference compared to the RC model (p = 0.5773). To conclude, CT-based neck nodal radiomics was capable of predicting IfTM-triggered ART events in NPC patients in this multi-center study, outperforming the traditional clinical model. The findings of this study provide valuable insights for future study into developing an effective screening strategy for ART eligibility in NPC patients in the long run, ultimately alleviating the workload of clinical practitioners, streamlining ART procedural efficiency in clinics, and achieving personalized RT for NPC patients in the future.

Dose-volume predictors of post-radiation primary hypothyroidism in head and neck cancer: A systematic review.

BACKGROUND AND PURPOSE: This systematic review aims to identify radiation dose-volume predictors of primary hypothyroidism after radiotherapy in patients with head and neck cancer (HNC). MATERIALS AND METHODS: We performed a systematic literature search of Medline, EMBASE and Web of Science from database inception to July 1, 2021 for articles that discuss radiation dose-volume predictors of post-radiation primary hypothyroidism in patients with HNC. Data on the incidence, clinical risk factors and radiation dose-volume parameters were extracted. A meta-analysis was performed using the random-effects model to estimate the pooled odds ratio (OR) of thyroid volume as a predictor of the risk of post-radiation hypothyroidism, adjusted for thyroid radiation dosimetry. RESULTS: Our search identified 29 observational studies involving 4,530 patients. With median follow-up durations ranging from 1.0 to 5.3 years, the average crude incidence of post-radiation primary hypothyroidism was 41.4 % (range, 10 %-57 %). Multiple radiation dose-volume parameters were associated with post-radiation primary hypothyroidism, including the thyroid mean dose (Dmean), minimum dose, V25, V30, V35, V45, V50, V30-60, VS45 and VS60. Thyroid Dmean and V50 were the most frequently proposed dosimetric predictors. The pooled adjusted OR of thyroid volume on the risk of post-radiation primary hypothyroidism was 0.89 (95 % confidence interval, 0.85-0.93; p < 0.001) per 1 cc increment. CONCLUSION: Post-radiation primary hypothyroidism is a common late complication after radiotherapy for HNC. Minimizing inadvertent exposure of the thyroid gland to radiation is crucial to prevent this late complication. Radiation dose-volume constraints individualized for thyroid volume should be considered in HNC radiotherapy planning.

Virtual Contrast-Enhanced Magnetic Resonance Images Synthesis for Patients With Nasopharyngeal Carcinoma Using Multimodality-Guided Synergistic Neural Network.

PURPOSE: To investigate a novel deep-learning network that synthesizes virtual contrast-enhanced T1-weighted (vceT1w) magnetic resonance images (MRI) from multimodality contrast-free MRI for patients with nasopharyngeal carcinoma (NPC). METHODS AND MATERIALS: This article presents a retrospective analysis of multiparametric MRI, with and without contrast enhancement by gadolinium-based contrast agents (GBCAs), obtained from 64 biopsy-proven cases of NPC treated at Hong Kong Queen Elizabeth Hospital. A multimodality-guided synergistic neural network (MMgSN-Net) was developed to leverage complementary information between contrast-free T1-weighted and T2-weighted MRI for vceT1w MRI synthesis. Thirty-five patients were randomly selected for model training, whereas 29 patients were selected for model testing. The synthetic images generated from MMgSN-Net were quantitatively evaluated against real GBCA-enhanced T1-weighted MRI using a series of statistical evaluating metrics, which include mean absolute error (MAE), mean squared error (MSE), structural similarity index (SSIM), and peak signal-to-noise ratio (PSNR). Qualitative visual assessment between the real and synthetic MRI was also performed. Effectiveness of our MMgSN-Net was compared with 3 state-of-the-art deep-learning networks, including U-Net, CycleGAN, and Hi-Net, both quantitatively and qualitatively. Furthermore, a Turing test was performed by 7 board-certified radiation oncologists from 4 hospitals for assessing authenticity of the synthesized vceT1w MRI against the real GBCA-enhanced T1-weighted MRI. RESULTS: Results from the quantitative evaluations demonstrated that our MMgSN-Net outperformed U-Net, CycleGAN and Hi-Net, yielding the top-ranked scores in averaged MAE (44.50 ± 13.01), MSE (9193.22 ± 5405.00), SSIM (0.887 ± 0.042), and PSNR (33.17 ± 2.14). Furthermore, the mean accuracy of the 7 readers in the Turing tests was determined to be 49.43%, equivalent to random guessing (ie, 50%) in distinguishing between real GBCA-enhanced T1-weighted and synthetic vceT1w MRI. Qualitative evaluation indicated that MMgSN-Net gave the best approximation to the ground-truth images, particularly in visualization of tumor-to-muscle interface and the intratumor texture information. CONCLUSIONS: Our MMgSN-Net was capable of synthesizing highly realistic vceT1w MRI that outperformed the 3 comparable state-of-the-art networks.

Quantitative Spatial Characterization of Lymph Node Tumor for N Stage Improvement of Nasopharyngeal Carcinoma Patients.

This study aims to investigate the feasibility of improving the prognosis stratification of the N staging system of Nasopharyngeal Carcinoma (NPC) from quantitative spatial characterizations of metastatic lymph node (LN) for NPC in a multi-institutional setting. A total of 194 and 284 NPC patients were included from two local hospitals as the discovery and validation cohort. Spatial relationships between LN and the surrounding organs were quantified by both distance and angle histograms, followed by principal component analysis. Independent prognostic factors were identified and combined with the N stage into a new prognostic index by univariate and multivariate Cox regressions on disease-free survival (DFS). The new three-class risk stratification based on the constructed prognostic index demonstrated superior cross-institutional performance in DFS. The hazard ratios of the high-risk to low-risk group were 9.07 (p < 0.001) and 4.02 (p < 0.001) on training and validation, respectively, compared with 5.19 (p < 0.001) and 1.82 (p = 0.171) of N3 to N1. Our spatial characterizations of lymph node tumor anatomy improved the existing N-stage in NPC prognosis. Our quantitative approach may facilitate the discovery of new anatomical characteristics to improve patient staging in other diseases.

External Validation of a Nomogram to Predict Survival and Benefit of Concurrent Chemoradiation for Stage II Nasopharyngeal Carcinoma.

A nomogram was recently published by Sun et al. to predict overall survival (OS) and the additional benefit of concurrent chemoradiation (CCRT) vs. radiotherapy (RT) alone, in stage II NPC treated with conventional RT. We aimed to assess the predictors of OS and to externally validate the nomogram in the IMRT era. We analyzed stage II NPC patients treated with definitive RT alone or CCRT between 2001 and 2011 under the territory-wide Hong Kong NPC Study Group 1301 study. Clinical parameters were studied using the Cox proportional hazards model to estimate OS. The nomogram by Sun et al. was applied with 1000 times bootstrap resampling to calculate the concordance index, and we compared the nomogram predicted and observed 5-year OS. There were 482 patients included. The 5-year OS was 89.0%. In the multivariable analysis, an age > 45 years was the only significant predictor of OS (HR, 1.98; 95%CI, 1.15-3.44). Other clinical parameters were insignificant, including the use of CCRT (HR, 0.99; 95%CI, 0.62-1.58). The nomogram yielded a concordance index of 0.55 (95% CI, 0.49-0.62) which lacked clinically meaningful discriminative power. The nomogram proposed by Sun et al. should be interpreted with caution when applied to stage II NPC patients in the IMRT era. The benefit of CCRT remained controversial.

EGFR mutation-guided use of afatinib, erlotinib and gefitinib for advanced non-small-cell lung cancer in Hong Kong - A cost-effectiveness analysis.

INTRODUCTION: Tyrosine kinase inhibitors (TKIs) therapy targets at epidermal growth factor receptor (EGFR) gene mutations in non-small-cell lung cancer (NSCLC). We aimed to compare the EGFR mutation-guided target therapy versus empirical chemotherapy for first-line treatment of advanced NSCLC in the public healthcare setting of Hong Kong. METHODS: A Markov model was designed to simulate outcomes of a hypothetical cohort of advanced (stage IIIB/IV) NSCLC adult patients with un-tested EGFR-sensitizing mutation status. Four treatment strategies were evaluated: Empirical first-line chemotherapy with cisplatin-pemetrexed (empirical chemotherapy group), and EGFR mutation-guided use of a TKI (afatinib, erlotinib, and gefitinib). Model outcome measures were direct medical cost, progression-free survival, overall survival, and quality-adjusted life-years (QALYs). Incremental cost per QALY gained (ICER) was estimated. Sensitivity analyses were performed to examine robustness of model results. RESULTS: Empirical chemotherapy and EGFR mutation-guided gefitinib gained lower QALYs at higher costs than the erlotinib group. Comparing with EGFR mutation-guided erlotinib, the afatinib strategy gained additional QALYs with ICER (540,633 USD/QALY). In 10,000 Monte Carlo simulations for probabilistic sensitivity analysis, EGFR mutation-guided afatinib, erlotinib, gefitinib and empirical chemotherapy were preferred strategy in 0%, 98%, 0% and 2% of time at willingness-to-pay (WTP) 47,812 USD/QALY (1x gross domestic product (GDP) per capita), and in 30%, 68%, 2% and 0% of time at WTP 143,436 USD/QALY (3x GDP per capita), respectively. CONCLUSIONS: EGFR mutation-guided erlotinib appears to be the cost-effective strategy from the perspective of Hong Kong public healthcare provider over a broad range of WTP.

Multi-Organ Omics-Based Prediction for Adaptive Radiation Therapy Eligibility in Nasopharyngeal Carcinoma Patients Undergoing Concurrent Chemoradiotherapy.

PURPOSE: To investigate the role of different multi-organ omics-based prediction models for pre-treatment prediction of Adaptive Radiotherapy (ART) eligibility in patients with nasopharyngeal carcinoma (NPC). METHODS AND MATERIALS: Pre-treatment contrast-enhanced computed tomographic and magnetic resonance images, radiotherapy dose and contour data of 135 NPC patients treated at Hong Kong Queen Elizabeth Hospital were retrospectively analyzed for extraction of multi-omics features, namely Radiomics (R), Morphology (M), Dosiomics (D), and Contouromics (C), from a total of eight organ structures. During model development, patient cohort was divided into a training set and a hold-out test set in a ratio of 7 to 3 via 20 iterations. Four single-omics models (R, M, D, C) and four multi-omics models (RD, RC, RM, RMDC) were developed on the training data using Ridge and Multi-Kernel Learning (MKL) algorithm, respectively, under 10-fold cross validation, and evaluated on hold-out test data using average area under the receiver-operator-characteristics curve (AUC). The best-performing single-omics model was first determined by comparing the AUC distribution across the 20 iterations among the four single-omics models using two-sided student t-test, which was then retrained using MKL algorithm for a fair comparison with the four multi-omics models. RESULTS: The R model significantly outperformed all other three single-omics models (all p-value<0.0001), achieving an average AUC of 0.942 (95%CI: 0.938-0.946) and 0.918 (95%CI: 0.903-0.933) in training and hold-out test set, respectively. When trained with MKL, the R model (R_MKL) yielded an increased AUC of 0.984 (95%CI: 0.981-0.988) and 0.927 (95%CI: 0.905-0.948) in training and hold-out test set respectively, while demonstrating no significant difference as compared to all studied multi-omics models in the hold-out test sets. Intriguingly, Radiomic features accounted for the majority of the final selected features, ranging from 64% to 94%, in all the studied multi-omics models. CONCLUSIONS: Among all the studied models, the Radiomic model was found to play a dominant role for ART eligibility in NPC patients, and Radiomic features accounted for the largest proportion of features in all the multi-omics models.

Application of hypoglossal nerve constraint in definitive radiotherapy for nasopharyngeal carcinoma: A dosimetric feasibility study.

PURPOSE: Radiation-induced hypoglossal nerve palsy is an infrequent but debilitating late complication after definitive radiotherapy for head and neck cancers. D1cc < 74 Gy (equivalent dose in 2 Gy fractions, EQD2) has been proposed as a potential dose constraint that limits 8-year palsy risk to < 5%. This study sets to perform detailed dosimetric assessments on the applicability of this novel dose constraint in advanced nasopharyngeal carcinoma (NPC). MATERIALS AND METHODS: This is a retrospective single-institution dosimetry study. NPC radiotherapy plans were identified from an institutional database, with an aim to select 10 eligible cases. Bilateral hypoglossal nerves were retrospectively contoured following a standard atlas. Cases with either one, or both, hypoglossal nerves D1cc exceeded 74 Gy EQD2 were included. Dosimetry of hypoglossal nerves, planning target volumes (PTV) and normal structures before and after application of the new hypoglossal nerve constraint were compared and analyzed. RESULTS: Ten NPC cases were replanned. All hypoglossal nerve contours overlapped with high-dose PTV, predominantly at regions of gross nodal diseases. D1cc in 15 out of 20 hypoglossal nerves exceeded 74G y EQD2 at initial plans. All nerves fulfilled the pre-specified constraint of 74Gy EQD2 after re-plan. Median hypoglossal nerve D1cc reduced from 74.8Gy (range, 74.1 to 77.4Gy) to 73.5Gy (range, 72.4 to 74.0Gy) (p < 0.001), corresponded to a projected reduction in 8-year palsy risk from 5%-14% to 3%-5%. PTV V100 was maintained above 95% in all cases. Dose increments in near-maximum (D2) and decrements in near-minimum (D98) were < 1 Gy. Safety dosimetric parameters of standard head and neck organs-at-risk showed no significant changes. CONCLUSIONS: Hypoglossal nerve D1cc < 74 Gy EQD2 is a dosimetrically feasible constraint in definitive radiotherapy for NPC. Tumor target coverage and normal organ dosimetry were not compromised with its usage. Its routine application should be considered in definitive radiotherapy for head and neck cancers.

Second primary cancer after intensity-modulated radiotherapy for nasopharyngeal carcinoma: A territory-wide study by HKNPCSG.

OBJECTIVES: Long-term risk of second primary cancer (SPC) after definitive intensity-modulated radiotherapy (IMRT) for nasopharyngeal carcinoma (NPC) remains unclear. This study aims to evaluate the risk, predictive factors and survival impact of SPC in a large territory-wide cohort of NPC survivors in an endemic region. MATERIALS AND METHODS: In this multicenter study, consecutive NPC patients (n = 3166) who underwent definitive IMRT in all six public oncology centers in Hong Kong between 2001 and 2010 were included. SPC risks were quantified by standardized incidence ratios (SIRs) and absolute excess risks (AERs) estimated from corresponding age-, sex-, and calendar year-specific population cancer incidence data from the Hong Kong Cancer Registry. Predictive factors and SPC-specific mortality were analyzed. RESULTS: Over a median follow-up period of 10.8 years, 290 cases of SPC were observed with a crude incidence of 9.2%. Cancer risk in NPC survivors was 90% higher than that in general population [SIR, 1.9; 95% confidence interval (CI), 1.7-2.2], with an AER of 52.1 (95% CI, 36.8-67.3) per 10,000 person-years at risk. Significant excess cancer risks were observed for oral cavity, sarcoma, oropharynx, paranasal sinus, salivary gland, thyroid, skin and lung. Advanced age, smoking, hepatitis B status, and re-irradiation were independent predictive factors. SPC accounted for 9.4% of all deaths among NPC survivors during the study period, and 10-year SPC-specific mortality was 3.4%. CONCLUSIONS: Second cancer risk after IMRT was substantial among NPC patients. SPC impairs long-term survival, and close surveillance is warranted as part of survivorship care.

Pseudo-CT generation from multi-parametric MRI using a novel multi-channel multi-path conditional generative adversarial network for nasopharyngeal carcinoma patients.

PURPOSE: To develop and evaluate a novel method for pseudo-CT generation from multi-parametric MR images using multi-channel multi-path generative adversarial network (MCMP-GAN). METHODS: Pre- and post-contrast T1-weighted (T1-w), T2-weighted (T2-w) MRI, and treatment planning CT images of 32 nasopharyngeal carcinoma (NPC) patients were employed to train a pixel-to-pixel MCMP-GAN. The network was developed based on a 5-level Residual U-Net (ResU-Net) with the channel-based independent feature extraction network to generate pseudo-CT images from multi-parametric MR images. The discriminator with five convolutional layers was added to distinguish between the real CT and pseudo-CT images, improving the nonlinearity and prediction accuracy of the model. Eightfold cross validation was implemented to validate the proposed MCMP-GAN. The pseudo-CT images were evaluated against the corresponding planning CT images based on mean absolute error (MAE), peak signal-to-noise ratio (PSNR), Dice similarity coefficient (DSC), and Structural similarity index (SSIM). Similar comparisons were also performed against the multi-channel single-path GAN (MCSP-GAN), the single-channel single-path GAN (SCSP-GAN). RESULTS: It took approximately 20 h to train the MCMP-GAN model on a Quadro P6000, and less than 10 s to generate all pseudo-CT images for the subjects in the test set. The average head MAE between pseudo-CT and planning CT was 75.7 ± 14.6 Hounsfield Units (HU) for MCMP-GAN, significantly (P-values < 0.05) lower than that for MCSP-GAN (79.2 ± 13.0 HU) and SCSP-GAN (85.8 ± 14.3 HU). For bone only, the MCMP-GAN yielded a smaller mean MAE (194.6 ± 38.9 HU) than MCSP-GAN (203.7 ± 33.1 HU), SCSP-GAN (227.0 ± 36.7 HU). The average PSNR of MCMP-GAN (29.1 ± 1.6) was found to be higher than that of MCSP-GAN (28.8 ± 1.2) and SCSP-GAN (28.2 ± 1.3). In terms of metrics for image similarity, MCMP-GAN achieved the highest SSIM (0.92 ± 0.02) but did not show significantly improved bone DSC results in comparison with MCSP-GAN. CONCLUSIONS: We developed a novel multi-channel GAN approach for generating pseudo-CT from multi-parametric MR images. Our preliminary results in NPC patients showed that the MCMP-GAN method performed apparently superior to the U-Net-GAN and SCSP-GAN, and slightly better than MCSP-GAN.

Dose volume effects of re-irradiation for locally recurrent nasopharyngeal carcinoma.

BACKGROUND: This study analyzed the dose volume effects of re-irradiation for locally recurrent nasopharyngeal carcinoma (NPC) and attempts to determine the optimal dose for the best survival. METHODS: Ninety-one patients were studied. The local control, fatal complication, and overall survival were analyzed against the dose (in Equivalent Dose at 2 Gy/fractions) and recurrent gross tumor volume (GTV). RESULTS: The local control and fatal complication rate appear to increase with prescribed dose. The overall survival peaks at around 60 Gy10 . Local control decreases significantly with increasing GTV (P < .001) while overall survival shows similar trend (P = .06). No correlation was observed between the fatal complication rate and GTV volume. The dose response of local control appears to be stronger for smaller tumors. CONCLUSION: GTV volume plays a significant role in local control. A 60 Gy10 appears to be optimal for the best survival outcome; higher doses might be considered for small tumors.

Pretreatment Prediction of Adaptive Radiation Therapy Eligibility Using MRI-Based Radiomics for Advanced Nasopharyngeal Carcinoma Patients.

Background and purpose: Adaptive radiotherapy (ART) can compensate for the dosimetric impacts induced by anatomic and geometric variations in patients with nasopharyngeal carcinoma (NPC); Yet, the need for ART can only be assessed during the radiation treatment and the implementation of ART is resource intensive. Therefore, we aimed to determine tumoral biomarkers using pre-treatment MR images for predicting ART eligibility in NPC patients prior to the start of treatment. Methods: Seventy patients with biopsy-proven NPC (Stage II-IVB) in 2015 were enrolled into this retrospective study. Pre-treatment contrast-enhanced T1-w (CET1-w), T2-w MR images were processed and filtered using Laplacian of Gaussian (LoG) filter before radiomic features extraction. A total of 479 radiomics features, including the first-order (n = 90), shape (n = 14), and texture features (n = 375), were initially extracted from Gross-Tumor-Volume of primary tumor (GTVnp) using CET1-w, T2-w MR images. Patients were randomly divided into a training set (n = 51) and testing set (n = 19). The least absolute shrinkage and selection operator (LASSO) logistic regression model was applied for radiomic model construction in training set to select the most predictive features to predict patients who were replanned and assessed in the testing set. A double cross-validation approach of 100 resampled iterations with 3-fold nested cross-validation was employed in LASSO during model construction. The predictive performance of each model was evaluated using the area under the receiver operator characteristic (ROC) curve (AUC). Results: In the present cohort, 13 of 70 patients (18.6%) underwent ART. Average AUCs in training and testing sets were 0.962 (95%CI: 0.961-0.963) and 0.852 (95%CI: 0.847-0.857) with 8 selected features for CET1-w model; 0.895 (95%CI: 0.893-0.896) and 0.750 (95%CI: 0.745-0.755) with 6 selected features for T2-w model; and 0.984 (95%CI: 0.983-0.984) and 0.930 (95%CI: 0.928-0.933) with 6 selected features for joint T1-T2 model, respectively. In general, the joint T1-T2 model outperformed either CET1-w or T2-w model alone. Conclusions: Our study successfully showed promising capability of MRI-based radiomics features for pre-treatment identification of ART eligibility in NPC patients.

Ramucirumab plus erlotinib in patients with untreated, EGFR-mutated, advanced non-small-cell lung cancer (RELAY): a randomised, double-blind, placebo-controlled, phase 3 trial.

BACKGROUND: Dual blockade of the EGFR and VEGF pathways in EGFR-mutated metastatic non-small-cell lung cancer (NSCLC) is supported by preclinical and clinical data, yet the approach is not widely implemented. RELAY assessed erlotinib, an EGFR tyrosine kinase inhibitor (TKI) standard of care, plus ramucirumab, a human IgG1 VEGFR2 antagonist, or placebo in patients with untreated EGFR-mutated metastatic NSCLC. METHODS: This is a worldwide, double-blind, phase 3 trial done in 100 hospitals, clinics, and medical centres in 13 countries. Eligible patients were aged 18 years or older (20 years or older in Japan and Taiwan) at the time of study entry, had stage IV NSCLC, with an EGFR exon 19 deletion (ex19del) or exon 21 substitution (Leu858Arg) mutation, an Eastern Cooperative Oncology Group performance status of 0 or 1, and no CNS metastases. We randomly assigned eligible patients in a 1:1 ratio to receive oral erlotinib (150 mg/day) plus either intravenous ramucirumab (10 mg/kg) or matching placebo once every 2 weeks. Randomisation was done by an interactive web response system with a computer-generated sequence and stratified by sex, geographical region, EGFR mutation type, and EGFR testing method. The primary endpoint was investigator-assessed progression-free survival in the intention-to-treat population. Safety was assessed in all patients who received at least one dose of study treatment. This trial is registered at ClinicalTrials.gov, NCT02411448, and is ongoing for long-term survival follow-up. FINDINGS: Between Jan 28, 2016, and Feb 1, 2018, 449 eligible patients were enrolled and randomly assigned to treatment with ramucirumab plus erlotinib (n=224) or placebo plus erlotinib (n=225). Median duration of follow-up was 20·7 months (IQR 15·8-27·2). At the time of primary analysis, progression-free survival was significantly longer in the ramucirumab plus erlotinib group (19·4 months [95% CI 15·4-21·6]) than in the placebo plus erlotinib group (12·4 months [11·0-13·5]), with a stratified hazard ratio of 0·59 (95% CI 0·46-0·76; p<0·0001). Grade 3-4 treatment-emergent adverse events were reported in 159 (72%) of 221 patients in the ramucirumab plus erlotinib group versus 121 (54%) of 225 in the placebo plus erlotinib group. The most common grade 3-4 treatment-emergent adverse events in the ramucirumab plus erlotinib group were hypertension (52 [24%]; grade 3 only) and dermatitis acneiform (33 [15%]), and in the placebo plus erlotinib group were dermatitis acneiform (20 [9%]) and increased alanine aminotransferase (17 [8%]). Treatment-emergent serious adverse events were reported in 65 (29%) of 221 patients in the ramucirumab plus erlotinib group and 47 (21%) of 225 in the placebo plus erlotinib group. The most common serious adverse events of any grade in the ramucirumab plus erlotinib group were pneumonia (seven [3%]) and cellulitis and pneumothorax (four [2%], each); the most common in the placebo plus erlotinib group were pyrexia (four [2%]) and pneumothorax (three [1%]). One on-study treatment-related death due to an adverse event occurred (haemothorax after a thoracic drainage procedure for a pleural empyema) in the ramucirumab plus erlotinib group. INTERPRETATION: Ramucirumab plus erlotinib demonstrated superior progression-free survival compared with placebo plus erlotinib in patients with untreated EGFR-mutated metastatic NSCLC. Safety was consistent with the safety profiles of the individual compounds in advanced lung cancer. The RELAY regimen is a viable new treatment option for the initial treatment of EGFR-mutated metastatic NSCLC. FUNDING: Eli Lilly.

Patterns of care and treatment outcomes for local recurrence of NPC after definite IMRT-A study by the HKNPCSG.

BACKGROUND: This study evaluates the contemporary care for patients with locally recurrent nasopharyngeal carcinoma after failure of the primary course of intensity modulated radiotherapy. METHODS: Eligible patients were identified through the Hong Kong Cancer Registry database. Patterns of care and treatment outcomes were analyzed. RESULTS: Two hundred seventy-two patients with locally recurrent tumors were identified. Of them, 30.9% received surgery, whereas 35.7% received re-irradiation (re-RT). The 5-year overall survival (OS) for the whole group was 30.2%. Old age and advanced rT classification were adverse prognostic factors, whereas surgery (mainly in resectable recurrence) was associated with favorable survival outcome. The 5-year OS rates for patients who received surgery and re-RT were 56.3% and 21.8%, respectively. CONCLUSIONS: Early detection of resectable recurrence is of paramount importance as surgery for resectable tumors offers the potential to achieve excellent outcomes. Re-RT could be considered in selected patients with unresectable disease and favorable prognostic features.