ABSTRACT
Background: Adaptive treatment strategies that can dynamically react to individual cancer progression can provide effective personalized care. Longitudinal multi-omics information, paired with an artificially intelligent clinical decision support system (AI-CDSS) can assist clinicians in determining optimal therapeutic options and treatment adaptations. However, AI-CDSS is not perfectly accurate, as such, clinicians' over/under reliance on AI may lead to unintended consequences, ultimately failing to develop optimal strategies. To investigate such collaborative decision-making process, we conducted a Human-AI interaction case study on response-adaptive radiotherapy (RT). Methods: We designed and conducted a two-phase study for two disease sites and two treatment modalities-adaptive RT for non-small cell lung cancer (NSCLC) and adaptive stereotactic body RT for hepatocellular carcinoma (HCC)-in which clinicians were asked to consider mid-treatment modification of the dose per fraction for a number of retrospective cancer patients without AI-support (Unassisted Phase) and with AI-assistance (AI-assisted Phase). The AI-CDSS graphically presented trade-offs in tumor control and the likelihood of toxicity to organs at risk, provided an optimal recommendation, and associated model uncertainties. In addition, we asked for clinicians' decision confidence level and trust level in individual AI recommendations and encouraged them to provide written remarks. We enrolled 13 evaluators (radiation oncology physicians and residents) from two medical institutions located in two different states, out of which, 4 evaluators volunteered in both NSCLC and HCC studies, resulting in a total of 17 completed evaluations (9 NSCLC, and 8 HCC). To limit the evaluation time to under an hour, we selected 8 treated patients for NSCLC and 9 for HCC, resulting in a total of 144 sets of evaluations (72 from NSCLC and 72 from HCC). Evaluation for each patient consisted of 8 required inputs and 2 optional remarks, resulting in up to a total of 1440 data points. Results: AI-assistance did not homogeneously influence all experts and clinical decisions. From NSCLC cohort, 41 (57%) decisions and from HCC cohort, 34 (47%) decisions were adjusted after AI assistance. Two evaluations (12%) from the NSCLC cohort had zero decision adjustments, while the remaining 15 (88%) evaluations resulted in at least two decision adjustments. Decision adjustment level positively correlated with dissimilarity in decision-making with AI [NSCLC: ρ = 0.53 ( p < 0.001); HCC: ρ = 0.60 ( p < 0.001)] indicating that evaluators adjusted their decision closer towards AI recommendation. Agreement with AI-recommendation positively correlated with AI Trust Level [NSCLC: ρ = 0.59 ( p < 0.001); HCC: ρ = 0.7 ( p < 0.001)] indicating that evaluators followed AI's recommendation if they agreed with that recommendation. The correlation between decision confidence changes and decision adjustment level showed an opposite trend [NSCLC: ρ = -0.24 ( p = 0.045), HCC: ρ = 0.28 ( p = 0.017)] reflecting the difference in behavior due to underlying differences in disease type and treatment modality. Decision confidence positively correlated with the closeness of decisions to the standard of care (NSCLC: 2 Gy/fx; HCC: 10 Gy/fx) indicating that evaluators were generally more confident in prescribing dose fractionations more similar to those used in standard clinical practice. Inter-evaluator agreement increased with AI-assistance indicating that AI-assistance can decrease inter-physician variability. The majority of decisions were adjusted to achieve higher tumor control in NSCLC and lower normal tissue complications in HCC. Analysis of evaluators' remarks indicated concerns for organs at risk and RT outcome estimates as important decision-making factors. Conclusions: Human-AI interaction depends on the complex interrelationship between expert's prior knowledge and preferences, patient's state, disease site, treatment modality, model transparency, and AI's learned behavior and biases. The collaborative decision-making process can be summarized as follows: (i) some clinicians may not believe in an AI system, completely disregarding its recommendation, (ii) some clinicians may believe in the AI system but will critically analyze its recommendations on a case-by-case basis; (iii) when a clinician finds that the AI recommendation indicates the possibility for better outcomes they will adjust their decisions accordingly; and (iv) When a clinician finds that the AI recommendation indicate a worse possible outcome they will disregard it and seek their own alternative approach.
ABSTRACT
ABSTRACT: In this article, as part of this special issue on biomarkers of early response, we review currently available reports regarding magnetic resonance imaging apparent diffusion coefficient (ADC) changes in hepatocellular carcinoma (HCC) in response to stereotactic body radiation therapy. We compare diffusion image acquisition, ADC analysis, methods for HCC response assessment, and statistical methods for prediction of local tumor progression by ADC metrics. We discuss the pros and cons of these studies. Following detailed analyses of existing investigations, we cannot conclude that ADC is established as an imaging biomarker for stereotactic body radiation therapy assessment in HCC.
Subject(s)
Carcinoma, Hepatocellular , Liver Neoplasms , Humans , Carcinoma, Hepatocellular/diagnostic imaging , Carcinoma, Hepatocellular/radiotherapy , Liver Neoplasms/diagnostic imaging , Liver Neoplasms/radiotherapy , Diffusion Magnetic Resonance Imaging/methods , Magnetic Resonance Imaging , Biomarkers , Retrospective StudiesABSTRACT
BACKGROUND: Stereotactic body radiation therapy (SBRT) produces excellent local control for patients with hepatocellular carcinoma (HCC). However, the risk of toxicity for normal liver tissue is still a limiting factor. Normal tissue complication probability (NTCP) models have been proposed to estimate the toxicity with the assumption of uniform liver function distribution, which is not optimal. With more accurate regional liver functional imaging available for individual patient, we can improve the estimation and be more patient-specific. PURPOSE: To develop normal tissue complication probability (NTCP) models using pre-/during-treatment (RT) dynamic Gadoxetic Acid-enhanced (DGAE) MRI for adaptation of RT in a patient-specific manner in hepatocellular cancer (HCC) patients who receive SBRT. METHODS: 24 of 146 HCC patients who received SBRT underwent DGAE MRI. Physical doses were converted into EQD2 for analysis. Voxel-by-voxel quantification of the contrast uptake rate (k1) from DGAE-MRI was used to quantify liver function. A logistic dose-response model was used to estimate the fraction of liver functional loss, and NTCP was estimated using the cumulative functional reserve model for changes in Child-Pugh (C-P) scores. Model parameters were calculated using maximum-likelihood estimations. During-RT liver functional maps were predicted from dose distributions and pre-RT k1 maps with a conditional Wasserstein generative adversarial network (cWGAN). Imaging prediction quality was assessed using root-mean-square error (RMSE) and structural similarity (SSIM) metrics. The dose-response and NTCP were fit on both original and cWGAN predicted images and compared using a Wilcoxon signed-rank test. RESULTS: Logistic dose response models for changes in k1 yielded D50 of 35.2 (95% CI: 26.7-47.5) Gy and k of 0.62 (0.49-0.75) for the whole population. The high baseline ALBI (poor liver function) subgroup showed a significantly smaller D50 of 11.7 (CI: 9.06-15.4) Gy and larger k of 0.96 (CI: 0.74-1.22) compared to a low baseline ALBI (good liver function) subgroup of 54.8 (CI: 38.3-79.1) Gy and 0.59 (CI: 0.48-0.74), with p-values of < 0.001 and = 0.008, respectively, which indicates higher radiosensitivity for the worse baseline liver function cohort. Subset analyses were also performed for high/low baseline CP subgroups. The corresponding NTCP models showed good agreement for the fit parameters between cWGAN predicted and the ground-truth during-RT images with no statistical differences for low ALBI subgroup. CONCLUSIONS: NTCP models which incorporate voxel-wise functional information from DGAE-MRI k1 maps were successfully developed and feasibility was demonstrated in a small patient cohort. cWGAN predicted functional maps show promise for estimating localized patient-specific response to RT and warrant further validation in a larger patient cohort.
Subject(s)
Carcinoma, Hepatocellular , Deep Learning , Liver Neoplasms , Radiosurgery , Humans , Carcinoma, Hepatocellular/diagnostic imaging , Carcinoma, Hepatocellular/radiotherapy , Liver Neoplasms/diagnostic imaging , Liver Neoplasms/radiotherapy , Probability , Radiotherapy DosageABSTRACT
Involvement of many variables, uncertainty in treatment response, and inter-patient heterogeneity challenge objective decision-making in dynamic treatment regime (DTR) in oncology. Advanced machine learning analytics in conjunction with information-rich dense multi-omics data have the ability to overcome such challenges. We have developed a comprehensive artificial intelligence (AI)-based optimal decision-making framework for assisting oncologists in DTR. In this work, we demonstrate the proposed framework to Knowledge Based Response-Adaptive Radiotherapy (KBR-ART) applications by developing an interactive software tool entitled Adaptive Radiotherapy Clinical Decision Support (ARCliDS). ARCliDS is composed of two main components: Artifcial RT Environment (ARTE) and Optimal Decision Maker (ODM). ARTE is designed as a Markov decision process and modeled via supervised learning. Given a patient's pre- and during-treatment information, ARTE can estimate treatment outcomes for a selected daily dosage value (radiation fraction size). ODM is formulated using reinforcement learning and is trained on ARTE. ODM can recommend optimal daily dosage adjustments to maximize the tumor local control probability and minimize the side effects. Graph Neural Networks (GNN) are applied to exploit the inter-feature relationships for improved modeling performance and a novel double GNN architecture is designed to avoid nonphysical treatment response. Datasets of size 117 and 292 were available from two clinical trials on adaptive RT in non-small cell lung cancer (NSCLC) patients and adaptive stereotactic body RT (SBRT) in hepatocellular carcinoma (HCC) patients, respectively. For training and validation, dense data with 297 features were available for 67 NSCLC patients and 110 features for 71 HCC patients. To increase the sample size for ODM training, we applied Generative Adversarial Networks to generate 10,000 synthetic patients. The ODM was trained on the synthetic patients and validated on the original dataset. We found that, Double GNN architecture was able to correct the nonphysical dose-response trend and improve ARCliDS recommendation. The average root mean squared difference (RMSD) between ARCliDS recommendation and reported clinical decisions using double GNNs were 0.61 [0.03] Gy/frac (mean [sem]) for adaptive RT in NSCLC patients and 2.96 [0.42] Gy/frac for adaptive SBRT HCC compared to the single GNN's RMSDs of 0.97 [0.12] Gy/frac and 4.75 [0.16] Gy/frac, respectively. Overall, For NSCLC and HCC, ARCliDS with double GNNs was able to reproduce 36% and 50% of the good clinical decisions (local control and no side effects) and improve 74% and 30% of the bad clinical decisions, respectively. In conclusion, ARCliDS is the first web-based software dedicated to assist KBR-ART with multi-omics data. ARCliDS can learn from the reported clinical decisions and facilitate AI-assisted clinical decision-making for improving the outcomes in DTR.
Subject(s)
Carcinoma, Hepatocellular , Carcinoma, Non-Small-Cell Lung , Decision Support Systems, Clinical , Liver Neoplasms , Lung Neoplasms , Humans , Carcinoma, Non-Small-Cell Lung/radiotherapy , Carcinoma, Non-Small-Cell Lung/pathology , Artificial Intelligence , Lung Neoplasms/pathology , Liver Neoplasms/radiotherapy , Radiotherapy DosageABSTRACT
Background: Imbalanced outcome is one of common characteristics of oncology datasets. Current machine learning approaches have limitation in learning from such datasets. Here, we propose to resolve this problem by utilizing a human-in-the-loop (HITL) approach, which we hypothesize will also lead to more accurate and explainable outcome prediction models. Methods: A total of 119 HCC patients with 163 tumors were used in the study. 81 patients with 104 tumors from the University of Michigan Hospital treated with SBRT were considered as a discovery dataset for radiation outcomes model building. The external testing dataset included 59 tumors from 38 patients with SBRT from Princess Margaret Hospital. In the discovery dataset, 100 tumors from 77 patients had local control (LC) (96% of 104 tumors) and 23 patients had at least one grade increment of ALBI (I-ALBI) during six-month follow up (28% of 81 patients). Each patient had a total of 110 features, where 15 or 20 features were identified by physicians as expert knowledge features (EKFs) for LC or I-ALBI prediction. We proposed a HITL based Bayesian network (HITL-BN) approach to enhance the capability of selecting important features from imbalanced data in terms of accuracy and explainability through humans' participation by integrating feature importance ranking and Markov blanket algorithms. A pure data-driven Bayesian network (PD-BN) method was applied to the same discovery dataset of HCC patients as a benchmark. Results: In the training and testing phases, the areas under receiver operating characteristic curves of the HITL-BN models for LC or I-ALBI prediction during SBRT are 0.85 (95% confidence interval: 0.75-0.95) or 0.89 (0.81-0.95) and 0.77 or 0.78, respectively. They significantly outperformed the during-treatment PD-BN model in predicting LC or I-ALBI based on the discovery cross-validation and testing datasets from the Delong tests. Conclusion: By allowing the human expert to be part of the model building process, the HITL-BN approach yielded significantly improved accuracy as well as better explainability when dealing with imbalanced outcomes in the prediction of post-SBRT treatment response of HCC patients when compared to the PD-BN method.
ABSTRACT
Advancements in data-driven technologies and the inclusion of information-rich multiomics features have significantly improved the performance of outcomes modeling in radiation oncology. For this current trend to be sustainable, challenges related to robust data modeling such as small sample size, low size to feature ratio, noisy data, as well as issues related to algorithmic modeling such as complexity, uncertainty, and interpretability, need to be mitigated if not resolved. Emerging computational technologies and new paradigms such as federated learning, human-in-the-loop, quantum computing, and novel interpretability methods show great potential in overcoming these challenges and bridging the gap towards precision outcome modeling in radiotherapy. Examples of these promising technologies will be presented and their potential role in improving outcome modeling will be discussed.
Subject(s)
Radiation Oncology , Humans , Radiation Oncology/methods , Computing Methodologies , Quantum Theory , Machine LearningABSTRACT
In the precision medicine era, there is a growing need for precision radiotherapy where the planned radiation dose needs to be optimally determined by considering a myriad of patient-specific information in order to ensure treatment efficacy. Existing artificial-intelligence (AI) methods can recommend radiation dose prescriptions within the scope of this available information. However, treating physicians may not fully entrust the AI's recommended prescriptions due to known limitations or at instances when the AI recommendation may go beyond physicians' current knowledge. This paper lays out a systematic method to integrate expert human knowledge with AI recommendations for optimizing clinical decision making. Towards this goal, Gaussian process (GP) models are integrated with deep neural networks (DNNs) to quantify the uncertainty of the treatment outcomes given by physicians and AI recommendations, respectively, which are further used as a guideline to educate clinical physicians and improve AI models performance. The proposed method is demonstrated in a comprehensive dataset where patient-specific information and treatment outcomes are prospectively collected during radiotherapy of 67 non-small cell lung cancer (NSCLC) patients and are retrospectively analyzed.
Subject(s)
Carcinoma, Non-Small-Cell Lung , Lung Neoplasms , Artificial Intelligence , Carcinoma, Non-Small-Cell Lung/radiotherapy , Clinical Decision-Making , Humans , Lung Neoplasms/radiotherapy , Retrospective StudiesABSTRACT
Objective.Modern radiotherapy stands to benefit from the ability to efficiently adapt plans during treatment in response to setup and geometric variations such as those caused by internal organ deformation or tumor shrinkage. A promising strategy is to develop a framework, which given an initial state defined by patient-attributes, can predict future states based on pre-learned patterns from a well-defined patient population.Approach.Here, we investigate the feasibility of predicting patient anatomical changes, defined as a joint state of volume and daily setup changes, across a fractionated treatment schedule using two approaches. The first is based on a new joint framework employing quantum mechanics in combination with deep recurrent neural networks, denoted QRNN. The second approach is developed based on a classical framework, which models patient changes as a Markov process, denoted MRNN. We evaluated the performance characteristics of these two approaches on a dataset of 125 head and neck cancer patients, which was supplemented by synthetic data generated using a generative adversarial network. Model performance was evaluated using area under the receiver operating characteristic curve (AUC) scores.Main results.The MRNN framework had slightly better performance than the QRNN framework, with MRNN (QRNN) validation AUC scores of 0.742±0.021 (0.675±0.036), 0.709±0.026 (0.656±0.021), 0.724±0.036 (0.652±0.044), and 0.698±0.016 (0.605±0.035) for system state vector sizes of 4, 6, 8, and 10, respectively. Of these, only the results from the two higher order states had statistically significant differences(p<0.05).A similar trend was also observed when the models were applied to an external testing dataset of 20 patients, yielding MRNN (QRNN) AUC scores of 0.707 (0.623), 0.687 (0.608), 0.723 (0.669), and 0.697 (0.609) for states vectors sizes of 4, 6, 8, and 10, respectively.Significance.These results suggest that both stochastic models have potential value in predicting patient changes during the course of adaptive radiotherapy.
Subject(s)
Deep Learning , Head and Neck Neoplasms , Head and Neck Neoplasms/diagnostic imaging , Head and Neck Neoplasms/radiotherapy , Humans , Neural Networks, Computer , ROC CurveABSTRACT
PURPOSE: A situational awareness Bayesian network (SA-BN) approach is developed to improve physicians' trust in the prediction of radiation outcomes and evaluate its performance for personalized adaptive radiotherapy (pART). METHODS: 118 non-small-cell lung cancer patients with their biophysical features were employed for discovery (n = 68) and validation (n = 50) of radiation outcomes prediction modeling. Patients' important characteristics identified by radiation experts to predict individual's tumor local control (LC) or radiation pneumonitis with grade ≥ 2 (RP2) were incorporated as expert knowledge (EK). Besides generating an EK-based naïve BN (EK-NBN), an SA-BN was developed by incorporating the EK features into pure data-driven BN (PD-BN) methods to improve the credibility of LC or / and RP2 prediction. After using area under the free-response receiver operating characteristics curve (AU-FROC) to assess the joint prediction of these outcomes, their prediction performances were compared with a regression approach based on the expert yielded estimates (EYE) penalty and its variants. RESULTS: In addition to improving the credibility of radiation outcomes prediction, the SA-BN approach outperformed the EYE penalty and its variants in terms of the joint prediction of LC and RP2. The value of AU-FROC improves from 0.70 (95% CI: 0.54-0.76) using EK-NBN, to 0.75 (0.65-0.82) using a variant of EYE penalty, to 0.83 (0.75-0.93) using PD-BN and 0.83 (0.77-0.90) using SA-BN; with similar trends in the validation cohort. CONCLUSIONS: The SA-BN approach can provide an accurate and credible human-machine interface to gain physicians' trust in clinical decision-making, which has the potential to be an important component of pART.
Subject(s)
Carcinoma, Non-Small-Cell Lung , Lung Neoplasms , Radiation Pneumonitis , Awareness , Bayes Theorem , HumansABSTRACT
INTRODUCTION: In some phase I trial settings, there is uncertainty in assessing whether a given patient meets the criteria for dose-limiting toxicity. METHODS: We present a design which accommodates dose-limiting toxicity outcomes that are assessed with uncertainty for some patients. Our approach could be utilized in many available phase I trial designs, but we focus on the continual reassessment method due to its popularity. We assume that for some patients, instead of the usual binary dose-limiting toxicity outcome, we observe a physician-assessed probability of dose-limiting toxicity specific to a given patient. Data augmentation is used to estimate the posterior probabilities of dose-limiting toxicity at each dose level based on both the fully observed and partially observed patient outcomes. A simulation study is used to assess the performance of the design relative to using the continual reassessment method on the true dose-limiting toxicity outcomes (available in simulation setting only) and relative to simple thresholding approaches. RESULTS: Among the designs utilizing the partially observed outcomes, our proposed design has the best overall performance in terms of probability of selecting correct maximum tolerated dose and number of patients treated at the maximum tolerated dose. CONCLUSION: Incorporating uncertainty in dose-limiting toxicity assessment can improve the performance of the continual reassessment method design.
Subject(s)
Bayes Theorem , Dose-Response Relationship, Drug , Drug-Related Side Effects and Adverse Reactions , Research Design , Clinical Trials, Phase I as Topic , Computer Simulation , Humans , Maximum Tolerated Dose , UncertaintyABSTRACT
PURPOSE: Dose to normal lung has commonly been linked with radiation-induced lung toxicity (RILT) risk, but incorporating functional lung metrics in treatment planning may help further optimize dose delivery and reduce RILT incidence. The purpose of this study was to investigate the impact of the dose delivered to functional lung regions by analyzing perfusion (Q), ventilation (V), and combined V/Q single-photon-emission computed tomography (SPECT) dose-function metrics with regard to RILT risk in patients with non-small cell lung cancer (NSCLC) patients who received radiation therapy (RT). METHODS AND MATERIALS: SPECT images acquired from 88 patients with locally advanced NSCLC before undergoing conventionally fractionated RT were retrospectively analyzed. Dose was converted to the nominal dose equivalent per 2 Gy fraction, and SPECT intensities were normalized. Regional lung segments were defined, and the average dose delivered to each lung region was quantified. Three functional categorizations were defined to represent low-, normal-, and high-functioning lungs. The percent of functional lung category receiving ≥20 Gy and mean functional intensity receiving ≥20 Gy (iV20) were calculated. RILT was defined as grade 2+ radiation pneumonitis and/or clinical radiation fibrosis. A logistic regression was used to evaluate the association between dose-function metrics and risk of RILT. RESULTS: By analyzing V/Q normalized intensities and functional distributions across the population, a wide range in functional capability (especially in the ipsilateral lung) was observed in patients with NSCLC before RT. Through multivariable regression models, global lung average dose to the lower lung was found to be significantly associated with RILT, and Q and V iV20 were correlated with RILT when using ipsilateral lung metrics. Through a receiver operating characteristic analysis, combined V/Q low-function receiving ≥20 Gy (low-functioning V/Q20) in the ipsilateral lung was found to be the best predictor (area under the curce: 0.79) of RILT risk. CONCLUSIONS: Irradiation of the inferior lung appears to be a locational sensitivity for RILT risk. The multivariable correlation between ipsilateral lung iV20 and RILT, as well as the association of low-functioning V/Q20 and RILT, suggest that irradiating low-functioning regions in the lung may lead to higher toxicity rates.
ABSTRACT
PURPOSE: Novel actuarial deep learning neural network (ADNN) architectures are proposed for joint prediction of radiation therapy outcomes-radiation pneumonitis (RP) and local control (LC)-in stage III non-small cell lung cancer (NSCLC) patients. Unlike normal tissue complication probability/tumor control probability models that use dosimetric information solely, our proposed models consider complex interactions among multiomics information including positron emission tomography (PET) radiomics, cytokines, and miRNAs. Additional time-to-event information is also used in the actuarial prediction. METHODS AND MATERIALS: Three architectures were investigated: ADNN-DVH considered dosimetric information only; ADNN-com integrated multiomics information; and ADNN-com-joint combined RP2 (RP grade ≥2) and LC prediction. In these architectures, differential dose-volume histograms (DVHs) were fed into 1D convolutional neural networks (CNN) for extracting reduced representations. Variational encoders were used to learn representations of imaging and biological data. Reduced representations were fed into Surv-Nets to predict time-to-event probabilities for RP2 and LC independently and jointly by incorporating time information into designated loss functions. RESULTS: Models were evaluated on 117 retrospective patients and were independently tested on 25 newly accrued patients prospectively. A multi-institutional RTOG0617 data set of 327 patients was used for external validation. ADNN-DVH yielded cross-validated c-indexes (95% confidence intervals) of 0.660 (0.630-0.690) for RP2 prediction and 0.727 (0.700-0.753) for LC prediction, outperforming a generalized Lyman model for RP2 (0.613 [0.583-0.643]) and a generalized log-logistic model for LC (0.569 [0.545-0.594]). The independent internal test and external validation yielded similar results. ADNN-com achieved an even better performance than ADNN-DVH on both cross-validation and independent internal test. Furthermore, ADNN-com-joint, which yielded performance similar to ADNN-com, realized joint prediction with c-indexes of 0.705 (0.676-0.734) for RP2 and 0.740 (0.714-0.765) for LC and achieved an area under a free-response receiving operator characteristic curve (AU-FROC) of 0.729 (0.697-0.773) for the joint prediction of RP2 and LC. CONCLUSION: Novel deep learning architectures that integrate multiomics information outperformed traditional normal tissue complication probability/tumor control probability models in actuarial prediction of RP2 and LC.
Subject(s)
Carcinoma, Non-Small-Cell Lung/radiotherapy , Computational Biology , Deep Learning , Lung Neoplasms/radiotherapy , Humans , Prognosis , Radiation Pneumonitis/diagnosis , Radiation Pneumonitis/etiology , Retrospective StudiesABSTRACT
INTRODUCTION: Radiation therapy for the management of intrahepatic malignancies can adversely affect liver function. Liver damage has been associated with increased levels of inflammatory cytokines, including tumor necrosis factor alpha (TNFα). We hypothesized that an inflammatory state, characterized by increased soluble TNFα receptor (sTNFR1), mediates sensitivity of the liver to radiation. MATERIALS/METHODS: Plasma samples collected during 3 trials of liver radiation for liver malignancies were assayed for sTNFR1 level via enzyme-linked immunosorbent assay (ELISA). Univariate and multivariate logistic regression and longitudinal models were used to characterize associations between liver toxicity (defined as a ≥2-point increase in Child-Pugh [CP] score within 6 months of radiation treatment) and sTNFR1 levels, ALBI score, biocorrected mean liver dose (MLD), age, and baseline laboratory values. RESULTS: Samples from 78 patients given liver stereotactic body radiation therapy [SBRT] (92%) or hypofractionated radiation were examined. There was a significant association between liver toxicity and sTNFR1 levels, and higher values were associated with increased toxicity over a range of mean liver doses. When ALBI score and biocorrected dose were included in the model with sTNFR1, baseline ALBI score and change in ALBI (ΔALBI) were significantly associated with toxicity, but sTNFR1 was not. Baseline aminotransferase levels also predicted toxicity but not independently of ALBI score. CONCLUSIONS: Elevated plasma sTNFR1 levels are associated with liver injury after liver radiation, suggesting that elevated inflammatory cytokine activity is a predictor of radiation-induced liver dysfunction. Future studies should determine whether administration of agents that decrease inflammation prior to treatment is warranted.
ABSTRACT
PURPOSE: To quantitatively evaluate published experiences with hepatic stereotactic body radiation therapy (SBRT), to determine local control rates after treatment of primary and metastatic liver tumors and to examine whether outcomes are affected by SBRT dosing regimen. METHODS AND MATERIALS: We identified published articles that reported local control rates after SBRT for primary or metastatic liver tumors. Biologically effective doses (BEDs) were calculated for each dosing regimen using the linear-quadratic equation. We excluded series in which a wide range of BEDs was used. Individual lesion data for local control were extracted from actuarial survival curves, and data were aggregated to form a single dataset. Actuarial local control curves were generated using the Kaplan-Meier method after grouping lesions by disease type and BED (<100 Gy10 vs >100 Gy10). Comparisons were made using log-rank testing. RESULTS: Thirteen articles met all inclusion criteria and formed the dataset for this analysis. The 1-, 2-, and 3-year actuarial local control rates after SBRT for primary liver tumors (n = 431) were 93%, 89%, and 86%, respectively. Lower 1- (90%), 2- (79%), and 3-year (76%) actuarial local control rates were observed for liver metastases (n = 290, log-rank P = .011). Among patients treated with SBRT for primary liver tumors, there was no evidence that local control is influenced by BED within the range of schedules used. For liver metastases, on the other hand, outcomes were significantly better for lesions treated with BEDs exceeding 100 Gy10 (3-year local control 93%) than for those treated with BEDs of ≤100 Gy10 (3-year local control 65%, P < .001). CONCLUSIONS: Stereotactic body radiation therapy for primary liver tumors provides high rates of durable local control, with no clear evidence for a dose-response relationship among commonly utilized schedules. Excellent local control rates are also seen after SBRT for liver metastases when BEDs of >100 Gy10 are utilized.
Subject(s)
Liver Neoplasms/radiotherapy , Liver Neoplasms/secondary , Radiosurgery/methods , Actuarial Analysis , Colorectal Neoplasms/pathology , Dose-Response Relationship, Radiation , Humans , Kaplan-Meier Estimate , Linear Models , Liver Neoplasms/mortality , Models, Biological , Models, Theoretical , Probability , Radiotherapy Dosage , Relative Biological Effectiveness , Treatment OutcomeABSTRACT
Stereotactic body radiation therapy (SBRT) has emerged as an effective, noninvasive treatment option for primary liver cancer and metastatic disease occurring in the liver. Although SBRT can be highly effective for establishing local control in hepatic malignancies, a tradeoff exists between tumor control and normal tissue complications. The objective of the present study was to review the normal tissue dose-volume effects for SBRT-induced liver and gastrointestinal toxicities and derive normal tissue complication probability models.
Subject(s)
Gastrointestinal Tract/radiation effects , Liver Neoplasms/radiotherapy , Liver/radiation effects , Organs at Risk/radiation effects , Radiosurgery/methods , Dose-Response Relationship, Radiation , Humans , Liver/pathology , Liver Neoplasms/diagnostic imaging , Liver Neoplasms/secondary , Models, Biological , Models, Statistical , Models, Theoretical , Organ Size , Radiation Injuries/etiology , Radiosurgery/adverse effects , Treatment OutcomeABSTRACT
PURPOSE: Previous reports of stereotactic body radiation therapy (SBRT) for hepatocellular carcinoma (HCC) suggest unacceptably high rates of toxicity in patients with Child-Turcotte-Pugh (CTP) B liver disease. We hypothesized that an individualized adaptive treatment approach based on midtreatment liver function would maintain good local control while limiting toxicity in this population. METHODS AND MATERIALS: Patients with CTP-B liver disease and HCC were treated on prospective trials of individualized adaptive SBRT between 2006 and 2018. Patients underwent pre- and midtreatment liver function assessments using indocyanine green. Treatment-related toxicity was defined as a ≥2-point increase in CTP score from pretreatment within 6 months of treatment. In addition, we performed analyses with a longitudinal model to assess changes in CTP score over 12 months after SBRT. RESULTS: Eighty patients with CTP-B (median tumor size, 2.5 cm) were treated: 37 patients were CTP-B-7, 28 were CTP-B-8, and 15 were CTP-B-9. The median treatment dose was 36 Gy in 3 fractions. One-year local control was 92%. In a multivariate model controlling for tumor size, treatment dose, and baseline CTP score, higher treatment dose was associated with improved freedom from local progression (hazard ratio: 0.97; 95% confidence interval, 0.94-1.00; P = .04). Eighteen patients (24%) had a ≥2-point increase in CTP score within 6 months of SBRT. In a longitudinal model assessing changes in CTP score over 12 months after SBRT, controlling for baseline CTP and tumor size, increasing mean liver dose was associated with larger increases in CTP score (P = .04). CONCLUSIONS: An individualized adaptive treatment approach allows for acceptable toxicity and effective local control in patients with HCC and CTP-B liver disease. Because increasing dose may increase both local control and toxicity, further work is needed to optimize treatment in patients with compromised liver function.
Subject(s)
Carcinoma, Hepatocellular/complications , Carcinoma, Hepatocellular/radiotherapy , Liver Cirrhosis/complications , Liver Neoplasms/complications , Liver Neoplasms/radiotherapy , Radiosurgery/adverse effects , Safety , Aged , Female , Humans , Liver Cirrhosis/pathology , Male , Middle Aged , Precision Medicine , Radiotherapy Planning, Computer-Assisted , Retrospective Studies , Survival AnalysisABSTRACT
Recent years have witnessed tremendous growth in the application of machine learning (ML) and deep learning (DL) techniques in medical physics. Embracing the current big data era, medical physicists equipped with these state-of-the-art tools should be able to solve pressing problems in modern radiation oncology. Here, a review of the basic aspects involved in ML/DL model building, including data processing, model training, and validation for medical physics applications is presented and discussed. Machine learning can be categorized based on the underlying task into supervised learning, unsupervised learning, or reinforcement learning; each of these categories has its own input/output dataset characteristics and aims to solve different classes of problems in medical physics ranging from automation of processes to predictive analytics. It is recognized that data size requirements may vary depending on the specific medical physics application and the nature of the algorithms applied. Data processing, which is a crucial step for model stability and precision, should be performed before training the model. Deep learning as a subset of ML is able to learn multilevel representations from raw input data, eliminating the necessity for hand crafted features in classical ML. It can be thought of as an extension of the classical linear models but with multilayer (deep) structures and nonlinear activation functions. The logic of going "deeper" is related to learning complex data structures and its realization has been aided by recent advancements in parallel computing architectures and the development of more robust optimization methods for efficient training of these algorithms. Model validation is an essential part of ML/DL model building. Without it, the model being developed cannot be easily trusted to generalize to unseen data. Whenever applying ML/DL, one should keep in mind, according to Amara's law, that humans may tend to overestimate the ability of a technology in the short term and underestimate its capability in the long term. To establish ML/DL role into standard clinical workflow, models considering balance between accuracy and interpretability should be developed. Machine learning/DL algorithms have potential in numerous radiation oncology applications, including automatizing mundane procedures, improving efficiency and safety of auto-contouring, treatment planning, quality assurance, motion management, and outcome predictions. Medical physicists have been at the frontiers of technology translation into medicine and they ought to be prepared to embrace the inevitable role of ML/DL in the practice of radiation oncology and lead its clinical implementation.
Subject(s)
Deep Learning , Physics , Electronic Data Processing , Image Processing, Computer-AssistedABSTRACT
PURPOSE: To study the dosimetric risk factors for radiation-induced proximal bronchial tree (PBT) toxicity in patients treated with radiation therapy for non-small cell lung cancer (NSCLC). METHODS AND MATERIALS: Patients with medically inoperable or unresectable NSCLC treated with conventionally fractionated 3-dimensional conformal radiation therapy (3DCRT) in prospective clinical trials were eligible for this study. Proximal bronchial tree (PBT) and PBT wall were contoured consistently per RTOG 1106 OAR-Atlas. The dose-volume histograms (DVHs) of physical prescription dose (DVHp) and biological effective dose (α/ß = 2.5; DVH2.5) were generated, respectively. The primary endpoint was PBT toxicities, defined by CTCAE 4.0 under the terminology of bronchial stricture/atelectasis. RESULTS: Of 100 patients enrolled, with a median follow-up of 64 months (95% confidence interval [CI], 50-78), 73% received 70 Gy or greater and 17% developed PBT toxicity (grade 1, 8%; grade 2, 6%; grade 3, 0%; and grade 4, 3%). The median time interval between RT initiation and onset of PBT toxicity was 8.4 months (95% CI, 4.7-44.1). The combined DVHs showed that no patient with a PBT maximum physical dose <65 Gy developed any PBT toxicity. Cox proportional hazards analysis and receiver operating characteristic analysis demonstrated that V75 of PBT was the most significant dosimetric parameter for both grade 1+ (P = .035) and grade 2+ (P = .037) PBT toxicities. The dosimetric thresholds for V75 of PBT were 6.8% and 11.9% for grade 1+ and grade 2+ PBT toxicity, respectively. CONCLUSIONS: V75 of PBT appeared be the most significant dosimetric parameter for PBT toxicity after conventionally fractionated thoracic 3DCRT. Constraining V75 of PBT can limit clinically significant PBT toxicity.