Normal Tissue Complication Probability Models Research Articles

Modeling Variable Proton Relative Biological Effectiveness (RBE) Using Voxel-Level Image Density Change for Non-Small Cell Lung Cancer (NSCLC) Patients Treated with Passive Scattering Proton Therapy (PSPT) or Intensity Modulated Photon Therapy (IMRT)

<h3>Purpose/Objective(s)</h3> To test the hypothesis that radiation damage measured via lung image density change is a function of variable RBE for protons with a coefficient of determination (R²) greater than 0.8. <h3>Materials/Methods</h3> Data from 61 NSCLC patients previously treated on a prospective clinical trial with PSPT (N = 24) or IMRT (N = 37) to 74 Gy(RBE) (proton RBE of 1.1) were acquired. Patients with noticeable atelectasis or pleural effusion were excluded. For each patient, a ∼5-month post-treatment PETCT was obtained to assess response and was deformed to the planning exhale phase of the 4DCT (exCT) via a biomechanical model-based image registration algorithm. Subsequently, voxel-level image density change (IDC) within the normal ipsilateral lung was obtained by subtracting the exCT from the deformed PETCT, as a signal of biological damage. IMRT dose (D_x) was recalculated on the exCT using a commercial treatment planning system while the PSPT dose (D_p) and dose-averaged linear energy transfer (LET_d) were recalculated on the exCT using a commissioned track-repeating Monte-Carlo system. The fitted variable RBE models included a linear model (RBE = 1 + λ•LET_d) and a linear-quadratic (LQ) model (RBE = 1/(2D_p)•{√[(α/β)² + 4D_p(α/β)(p₀+(p₁•LET_d)/(α/β) + 4D_p²(p₂ + p₃•√(α/β)•LET_d)] - α/β}) with α/β = 3 Gy for lung tissue. <b>RBE was modeled against the ratio between D_x and D_p at given IDC levels.</b> We adapted Lyman-Kutcher-Burman normal tissue complication probability (NTCP) model to fit the dose-IDC relationship for each cohort by averaging the dose and IDC of voxels in 5-Gy dose intervals (adaptations: n = 1, effective dose = voxel dose, and complication probability = IDC normalized against global min and max IDC). LET_d corresponding to each D_p was approximated by averaging the LET_d of voxels that received D_p with a margin of 0.001Gy. All NTCP and RBE models were fitted using non-linear least squares with the Python SciPy package. <h3>Results</h3> For NTCP fitting, min and max IDC were 2.7 HU and 136.3 HU, respectively, and TD₅₀ and m were fitted to be 31.5 Gy and 0.51 for PSPT (R² = 0.99) and 44.4 Gy and 0.40 for IMRT (R² = 0.98). The table shows the IDC levels, the corresponding LET_d (SD), and the fitted D_x and D_p. The linear model resulted in λ = 0.154 µm/keV (R² = 0.27), and the LQ model resulted in p₀ = -0.72, p₁ = 10.02, p₂ = 1.51, and p₃ = -0.08 (R² = 0.999). <h3>Conclusion</h3> The study demonstrates that voxel-level radiation damage and image changes for proton therapy are functions of dose and LET and, thus, variable RBE.

Automated clinical decision support system with deep learning dose prediction and NTCP models to evaluate treatment complications in patients with esophageal cancer

Background and purposeThis study aims to investigate how accurate our deep learning (DL) dose prediction models for intensity modulated radiotherapy (IMRT) and pencil beam scanning (PBS) treatments, when chained with normal tissue complication probability (NTCP) models, are at identifying esophageal cancer patients who are at high risk of toxicity and should be switched to proton therapy (PT). Materials and methodsTwo U-Net were created, for photon (XT) and proton (PT) plans, respectively. To estimate the dose distribution for each patient, they were trained on a database of 40 uniformly planned patients using cross validation and a circulating test set. These models were combined with a NTCP model for postoperative pulmonary complications. The NTCP model used the mean lung dose, age, histology type, and body mass index as predicting variables. The treatment choice is then done by using a ΔNTCP threshold between XT and PT plans. Patients with ΔNTCP ≥ 10% were referred to PT. ResultsOur DL models succeed in predicting dose distributions with a mean error on the mean dose to the lungs (MLD) of 1.14 ± 0.93% for XT and 0.66 ± 0.48% for PT. The complete automated workflow (DL chained with NTCP) achieved 100% accuracy in patient referral. The average residual (ΔNTCP ground truth - ΔNTCP predicted) is 1.43 ± 1.49%. ConclusionThis study evaluates our DL dose prediction models in a broader patient referral context and demonstrates their ability to support clinical decisions.

Independent validation of a dysphagia dose response model for the selection of head and neck cancer patients to proton therapy.

Background and purposeThe model based approach involves the use of normal tissue complication models for selection of head and neck cancer patients to proton therapy. Our goal was to validate the clinical utility of the related dysphagia model using an independent patient cohort.Materials and MethodsA dataset of 277 head and neck cancer (pharynx and larynx) patients treated with (chemo)radiotherapy between 2019 and 2021 was acquired. For the evaluation of the model discrimination we used statistical metrics such as the sensitivity, specificity and the area under the receiver operating characteristic curve. After the validation we evaluated if the dysphagia model can be improved using the closed testing procedure, the Brier and the Hosmer-Lemeshow score.ResultsThe performance of the original normal tissue complication probability model for dysphagia grade II-IV at 6 months was good (AUC = 0.80). According to the graphical calibration assessment, the original model showed underestimated dysphagia risk predictions. The closed testing procedure indicated that the model had to be updated and selected a revised model with new predictor coefficients as an optimal model. The revised model had also satisfactory discrimination (AUC = 0.83) with improved calibration.ConclusionThe validation of the normal tissue complication probability model for grade II-IV dysphagia was successful in our independent validation cohort. However, the closed testing procedure indicated that the model should be updated with new coefficients.

Concepts and methods for the dosimetry of radioembolisation of the liver with Y-90-loaded microspheres

This article aims at presenting in a didactic way, dosimetry concepts and methods that are relevant for radio-embolization of the liver with 90Y-microspheres. The application of the medical internal radiation dose formalism to radio-embolization is introduced. This formalism enables a simplified dosimetry, where the absorbed dose in a given tissue depends on only its mass and initial activity. This is applied in the single-compartment method, partition model, for the liver, tumour and lung dosimetry, and multi-compartment method, allowing identification of multiple tumours. Voxel-based dosimetry approaches are also discussed. This allows taking into account the non-uniform uptake within a compartment, which translates into a non-uniform dose distribution, represented as a dose–volume histogram. For this purpose, dose–kernel convolution allows propagating the energy deposition around voxel-sources in a computationally efficient manner. Alternatively, local-energy deposition is preferable when the spatial resolution is comparable or larger than the beta-particle path. Statistical tools may be relevant in establishing dose–effect relationships in a given population. These include tools such as the logistic regression or receiver operator characteristic analysis. Examples are given for illustration purpose. Moreover, tumour control probability modelling can be assessed through the linear-quadratic model of Lea and Catcheside and its counterpart, the normal-tissue complication probability model of Lyman, which is suitable to the parallel structure of the liver. The selectivity of microsphere administration allows tissue sparing, which can be considered with the concept of equivalent uniform dose, for which examples are also given. The implication of microscopic deposition of microspheres is also illustrated through a liver toxicity model, even though it is not clinically validated. Finally, we propose a reflection around the concept of therapeutic index (TI), which could help tailor treatment planning by determining the treatment safety through the evaluation of TI based on treatment-specific parameters.

Risk of Cerebrovascular Events Among Childhood and Adolescent Patients Receiving Cranial Radiation Therapy: A PENTEC Normal Tissue Outcomes Comprehensive Review

Radiation-induced cerebrovascular toxicity is a well-documented sequelae that can be both life-altering and potentially fatal. We performed a meta-analysis of the relevant literature to create practical models for predicting the risk of cerebral vasculopathy after cranial irradiation. A literature search was performed for studies reporting pediatric radiation therapy (RT) associated cerebral vasculopathy. When available, we used individual patient RT doses delivered to the Circle of Willis (CW) or optic chiasm (as a surrogate), as reported or digitized from original publications, to formulate a dose-response. A logistic fit and a Normal Tissue Complication Probability (NTCP) model was developed to predict future risk of cerebrovascular toxicity and stroke, respectively. This NTCP risk was assessed as a function of prescribed dose. The search identified 766 abstracts, 5 of which were used for modeling. We identified 101 of 3989 pediatric patients who experienced at least one cerebrovascular toxicity: transient ischemic attack, stroke, moyamoya, or arteriopathy. For a range of shorter follow-ups, as specified in the original publications (approximate attained ages of 17 years), our logistic fit model predicted the incidence of any cerebrovascular toxicity as a function of dose to the CW, or surrogate structure: 0.2% at 30 Gy, 1.3% at 45 Gy, and 4.4% at 54 Gy. At an attained age of 35 years, our NTCP model predicted a stroke incidence of 0.9% to 1.3%, 1.8% to 2.7%, and 2.8% to 4.1%, respectively at prescribed doses of 30 Gy, 45 Gy, and 54 Gy (compared with a baseline risk of 0.2%-0.3%). At an attained age of 45 years, the predicted incidence of stroke was 2.1% to 4.2%, 4.5% to 8.6%, and 6.7% to 13.0%, respectively at prescribed doses of 30 Gy, 45 Gy, and 54 Gy (compared with a baseline risk of 0.5%-1.0%). Risk of cerebrovascular toxicity continues to increase with longer follow-up. NTCP stroke predictions are very sensitive to model variables (baseline stroke risk and proportional stroke hazard), both of which found in the literature may be systematically erring on minimization of true risk. We hope this information will assist practitioners in counseling, screening, surveilling, and facilitating risk reduction of RT-related cerebrovascular late effects in this highly sensitive population.

Definition of an Normal Tissue Complication Probability Model for the Inner Ear in Definitive Radiochemotherapy of Nasopharynx Carcinoma.

Simple SummaryDefinitive radiochemotherapy is the treatment of choice for locally advanced nasopharyngeal carcinoma. Due to the vicinity of the nasopharynx to the inner ear and the use of ototoxic platinum-based chemotherapy, there is a risk for irreversible damage to the auditory system. To avoid or minimize these critical side effects, radiation exposure to each inner ear must be balanced between target volume coverage and toxicity. However, normal tissue complication probability (NTCP) models of the inner ear validated by clinical data are rare. In this retrospective study of 46 patients, an NTCP model and a cutoff dose logistic regression model (CDLR) were created. There is a sigmoidal relation between radiation dose and incidence of inner ear toxicities. Dose constraints for the inner ear of <44 Gy (Dmean) or <58 Gy (Dmax) are suggested to limit the probability of inner ear toxicity < 25%.Background: Definitive radiochemotherapy is the treatment of choice for locally advanced nasopharyngeal carcinoma. Due to the vicinity of the nasopharynx to the inner ear and the use of ototoxic platinum-based chemotherapy, there is a risk for irreversible damage to the auditory system. To avoid or minimize these critical side effects, radiation exposure to each inner ear must be balanced between target volume coverage and toxicity. However, normal tissue complication probability (NTCP) models of the inner ear validated by clinical data are rare. Patients and Methods: This retrospective study investigates the inner ear toxicity of 46 patients who received radio(chemo-)therapy for nasopharyngeal carcinoma at our institution from 2004 to 2021 according to CTCAE criteria. For each inner ear, the mean and maximum dose in Gray was evaluated and correlated with clinical toxicity data. Based on the data, an NTCP model and a cutoff dose logistic regression model (CDLR) were created. Results: In 11 patients () hearing impairment and/or tinnitus was observed as a possible therapy-associated toxicity. was between 15– Gy, whereas was between 30– Gy. There was a dose-dependent, sigmoidal relation between inner ear dose and toxicity. A of Gy and Gy was associated with inner ear damage in and of patients, respectively. The maximum curve slope (m) was found at and is . The Dmax values showed a and complication probability at Gy and Gy, respectively, and a maximum slope of the sigmoid curve at with . Conclusion: There is a sigmoidal relation between radiation dose and incidence of inner ear toxicities. Dose constraints for the inner ear of Gy or Gy are suggested to limit the probability of inner ear toxicity .

Redefine the Role of Proton Beam Therapy for the Locally-Advanced Non-Small Cell Lung Cancer Assisting the Reduction of Acute Hematologic Toxicity.

PurposeTo investigate the potential clinical benefit of utilizing intensity-modulated proton therapy (IMPT) to reduce acute hematologic toxicity for locally advanced non-small cell lung cancer (LA-NSCLC) patients and explore the feasibility of a model-based patient selection approach via the normal tissue complication probability (NTCP).MethodsTwenty patients with LA-NSCLC were retrospectively selected. Volumetric modulated arc photon therapy (VMAT) and IMPT plans were generated with a prescription dose of 60 Gy in 30 fractions. A wide range of cases with varied tumor size, location, stations of metastatic lymph nodes were selected to represent the general cancer group. Contouring and treatment planning followed RTOG-1308 protocol. Doses to thoracic vertebral bodies (TVB) and other organ at risks were compared. Risk of grade ≥ 3 acute hematologic toxicity (HT3+) were calculated based on the NTCP model, and patients with a reduction on NTCP of HT3+ from VMAT to IMPT (△NTCP_HT3+) ≥ 10% were considered to ‘significantly benefit from proton therapy.’ResultsCompared to VMAT, IMPT significantly reduced the dose to the TVB, the lung, the heart, the esophagus and the spinal cord. Tumor distance to TVB was significantly associated with △NTCP _HT3+ ≥ 10%. For the patients with tumor distance ≤ 0.7 cm to TVB, the absolute reduction of dose (mean, V30 and V40) to TVB was significantly lower than that in patients with tumor distance > 0.7 cm.ConclusionIMPT decreased the probability of HT3+ compared to VMAT by reducing the dose to the TVB in LA-NSCLC patients. Patients with tumor distance to TVB less than 0.7 cm are likely to benefit most from proton over photon therapy.

Longitudinal study of irradiation-induced brain functional network alterations in patients with nasopharyngeal carcinoma

BackgroundTo investigate radiotherapy (RT)-related brain network changes in patients with nasopharyngeal carcinoma (NPC) over time and develop least absolute shrinkage and selection operator (LASSO)-based multivariable normal tissue complication probability (NTCP) models to predict RT-related brain network changes. Methods36 NPC patients were followed up at four timepoints: baseline, within 3 months (acute), 6 months (subacute), and 12 months (delayed) post-RT. 15 comparable healthy controls (HCs) were finally included and followed up in parallel. Functional neuroimaging data, dose–volume parameters of bilateral temporal lobes and Montreal Cognitive Assessment (MoCA) were acquired. Graph theoretical analysis and mixed-design analysis of variance were performed to investigate how the brain global and nodal changes were affected by RT. Multivariate logistic regression NTCP models were developed. LASSO with nested cross-validation strategy was used to select features. The relationships between network changes and MoCA changes were also examined. ResultsSignificant changes were detected in nodal efficiency (NE) in NPC patients but not in HCs over time. Altered NE was distributed in the bilateral frontal, temporal lobes and the right insula, which showed a “decrease-increase/recovery” pattern over time. Among all models, the model for predicting NE changes of STG.R showed a relatively good performance (area under the receiver operating curve: 0.68), and D20cc and V20 to right temporal lobe outperformed in this model. ConclusionOur findings indicate that RT-induced brain injury begin at the acute period and follow a recovery over time. Furthermore, our study presents prediction models for brain dysfunction based on the dosimetric and clinical parameters.

Evaluation of the Radiobiological Models Predicting the Radiation-Induced Hypothyroidism in the Partially Irradiated Thyroid Gland of Patients with Breast Cancer

Background: Radiation-induced hypothyroidism (RHT) is one of the side effects that might have an impact on the quality of life of patients with breast cancer treated with radiotherapy. Objectives: The aim of the current study was to evaluate the performances of the Lyman-Kutcher-Burman (LKB) and Log-Logistic models in the prediction of hypothyroidism (HT) as well as the estimation of the model parameters for the incidence of RHT among patients with breast cancer. Methods: Fifty-two patients treated with radiation therapy (RT) for breast cancer were prospectively evaluated. Patients' serum samples [tri-iodothyronine, thyroxine, thyroid-stimulating hormone (TSH), free triiodothyronine, and free thyroxine] were measured before RT and also at a regular time interval until 1 year after the completion of RT. For each patient, dose-volume histograms (DVHs) of the thyroid gland were derived from their treatment planning dataset. Patients whose TSH levels were higher than normal with a decrease in FT4 levels were considered as cases with RHT. The LKB and Log-Logistic radiobiological models were evaluated by comparing them with the resultant follow-up data. The parameters for radiobiological models have been deduced by fitting the models to the follow-up data. The models were fitted in a Bayesian setting and compared according to the widely applicable information criterion (WAIC). Results: Twenty-one (40%) patients developed RHT at a follow-up of 1 year after the end of radiation treatment. The fitted values of D50 for the LKB and Log-Logistic models were 37.71 and 25.50 Gy, respectively for the partially irradiated thyroid of patients with breast cancer. The mean time to the incidence of RHT was obtained at 6.7 months in the studied group. Conclusions: A volumetric effect was found for the thyroid gland in the implemented normal tissue complication probability models. Compared to the follow-up data, the Log-Logistic model was ranked as the best model for predicting the rate of RHT in patients with breast cancer.

Predictive performance of different NTCP techniques for radiation-induced esophagitis in NSCLC patients receiving proton radiotherapy

This study aimed to compare the predictive performance of different modeling methods in developing normal tissue complication probability (NTCP) models for predicting radiation-induced esophagitis (RE) in non–small cell lung cancer (NSCLC) patients receiving proton radiotherapy. The dataset was composed of 328 NSCLC patients receiving passive-scattering proton therapy and 41.6% of the patients experienced ≥ grade 2 RE. Five modeling methods were used to build NTCP models: standard Lyman–Kutcher–Burman (sLKB), generalized LKB (gLKB), multivariable logistic regression using two variable selection procedures-stepwise forward selection (Stepwise-MLR), and least absolute shrinkage and selection operator (LASSO-MLR), and support vector machines (SVM). Predictive performance was internally validated by a bootstrap approach for each modeling method. The overall performance, discriminative ability, and calibration were assessed using the Negelkerke R2, area under the receiver operator curve (AUC), and Hosmer–Lemeshow test, respectively. The LASSO-MLR model showed the best discriminative ability with an AUC value of 0.799 (95% confidence interval (CI): 0.763–0.854), and the best overall performance with a Negelkerke R2 value of 0.332 (95% CI: 0.266–0.486). Both of the optimism-corrected Negelkerke R2 values of the SVM and sLKB models were 0.301. The optimism-corrected AUC of the gLKB model (0.796) was higher than that of the SVM model (0.784). The sLKB model had the smallest optimism in the model variation and discriminative ability. In the context of classification and probability estimation for predicting the NTCP for radiation-induced esophagitis, the MLR model developed with LASSO provided the best predictive results. The simplest LKB modeling had similar or even better predictive performance than the most complex SVM modeling, and it was least likely to overfit the training data. The advanced machine learning approach might have limited applicability in clinical settings with a relatively small amount of data.

Assessment of gene expressions from squamous cell carcinoma of the head and neck to predict radiochemotherapy-related xerostomia and dysphagia

Purpose We tested the hypothesis that gene expressions from biopsies of locally advanced head and neck squamous cell carcinoma (HNSCC) patients can supplement dose-volume parameters to predict dysphagia and xerostomia following primary radiochemotherapy (RCTx). Material and methods A panel of 178 genes previously related to radiochemosensitivity of HNSCC was considered for nanoString analysis based on tumour biopsies of 90 patients with locally advanced HNSCC treated by primary RCTx. Dose-volume parameters were extracted from the parotid, submandibular glands, oral cavity, larynx, buccal mucosa, and lips. Normal tissue complication probability (NTCP) models were developed for acute, late, and for the improvement of xerostomia grade ≥2 and dysphagia grade ≥3 using a cross-validation-based least absolute shrinkage and selection operator (LASSO) approach combined with stepwise logistic regression for feature selection. The final signatures were included in a logistic regression model with optimism correction. Performance was assessed by the area under the receiver operating characteristic curve (AUC). Results NTCP models for acute and late xerostomia and the improvement of dysphagia resulted in optimism-corrected AUC values of 0.84, 0.76, and 0.70, respectively. The minimum dose to the contralateral parotid was selected for both acute and late xerostomia and the minimum dose to the larynx was selected for dysphagia improvement. For the xerostomia endpoints, the following gene expressions were selected: RPA2 (cellular response to DNA damage), TCF3 (salivary gland cells development), GBE1 (glycogen storage and regulation), and MAPK3 (regulation of cellular processes). No gene expression features were selected for the prediction of dysphagia. Conclusion This hypothesis-generating study showed the potential of improving NTCP models using gene expression data for HNSCC patients. The presented models require independent validation before potential application in clinical practice.

Comparison of Estimated Late Toxicities between IMPT and IMRT Based on Multivariable NTCP Models for High-Risk Prostate Cancers Treated with Pelvic Nodal Radiation.

PurposeTo compare the late gastrointestinal (GI) and genitourinary toxicities (GU) estimated using multivariable normal tissue complication probability (NTCP) models, between pencil-beam scanning proton beam therapy (PBT) and helical tomotherapy (HT) in patients of high-risk prostate cancers requiring pelvic nodal irradiation (PNI) using moderately hypofractionated regimen.Materials and MethodsTwelve consecutive patients treated with PBT at our center were replanned with HT using the same planning goals. Six late GI and GU toxicity domains (stool frequency, rectal bleeding, fecal incontinence, dysuria, urinary incontinence, and hematuria) were estimated based on the published multivariable NTCP models. The ΔNTCP (difference in absolute NTCP between HT and PBT plans) for each of the toxicity domains was calculated. A one-sample Kolmogorov-Smirnov test was used to analyze distribution of data, and either a paired t test or a Wilcoxon matched-pair signed rank test was used to test statistical significance.ResultsProton beam therapy and HT plans achieved adequate target coverage. Proton beam therapy plans led to significantly better sparing of bladder, rectum, and bowel bag especially in the intermediate range of 15 to 40 Gy, whereas doses to penile bulb and femoral heads were higher with PBT plans. The average ΔNTCP for grade (G)2 rectal bleeding, fecal incontinence, stool frequency, dysuria, urinary incontinence, and G1 hematuria was 12.17%, 1.67%, 2%, 5.83%, 2.42%, and 3.91%, respectively, favoring PBT plans. The average cumulative ΔNTCP for GI and GU toxicities (ΣΔNTCP) was 16.58% and 11.41%, respectively, favoring PBT. Using a model-based selection threshold of any G2 ΔNTCP >10%, 67% (8 patients) would be eligible for PBT.ConclusionProton beam therapy plans led to superior sparing of organs at risk compared with HT, which translated to lower NTCP for late moderate GI and GU toxicities in patients of prostate cancer treated with PNI. For two-thirds of our patients, the difference in estimated absolute NTCP values between PBT and HT crossed the accepted threshold for minimal clinically important difference.

A Patient Selection Approach Based on NTCP Models and DVH Parameters for Definitive Proton Therapy in Locally Advanced Sinonasal Cancer Patients.

Simple SummaryThe role of proton therapy as a radiation treatment option for locally advanced sinonasal cancer patients has increased in the last years, showing promising results in terms of clinical outcomes. Definition of strategies to identify patients who would benefit the most from proton therapy in terms of reduced toxicity is highly desirable, due to limited availability and higher costs of this treatment option. The novelty of our in silico study relies on assessing the suitability of a mixed dose volume histograms and normal tissue complication probability model-based approach, aiming at providing, to the scientific community, a patient selection criteria possibly leading the therapeutic choice between proton therapy and advanced photon techniques.(1) Background: In this work, we aim to provide selection criteria based on normal tissue complication probability (NTCP) models and additional explanatory dose-volume histogram parameters suitable for identifying locally advanced sinonasal cancer patients with orbital invasion benefitting from proton therapy. (2) Methods: Twenty-two patients were enrolled, and two advanced radiation techniques were compared: intensity modulated proton therapy (IMPT) and photon volumetric modulated arc therapy (VMAT). Plans were optimized with a simultaneous integrated boost modality: 70 and 56 Gy(RBE) in 35 fractions were prescribed to the high risk/low risk CTV. Several endpoints were investigated, classified for their severity and used as discriminating paradigms. In particular, when NTCP models were already available, a first selection criterion based on the delta-NTCP was adopted. Additionally, an overall analysis in terms of DVH parameters was performed. Furthermore, a second selection criterion based on a weighted sum of the NTCP and DVH was adopted. (3) Results: Four patients out of 22 (18.2%) were suitable for IMPT due to NTCP > 3% for at least one severe toxicity, 4 (18.2%) due to NTCP > 20% for at least three concurrent intermediate toxicities and 16 (72.7%) due to the mixed sum of NTCP and DVH criterion. Since, for some cases, both criteria were contemporary fulfilled, globally 17/22 patients (77.3%) would benefit from IMPT. (4) Conclusions: For this rare clinical scenario, the use of a strategy including DVH parameters and NTCPs when comparing VMAT and IMPT is feasible. We showed that patients affected by sinonasal cancer could profit from IMPT compared to VMAT in terms of optical and central nervous system organs at risk sparing.

A study on predicting cases that would benefit from proton beam therapy in primary liver tumors of less than or equal to 5 cm based on the estimated incidence of hepatic toxicity

BackgroundFor small primary liver tumors, favorable outcomes have been reported with both of proton beam therapy (PBT) and X-ray therapy (XRT). However, no clear criteria have been proposed in the cases for which and when of PBT or XRT has to be used. The aim of this study is to investigate cases that would benefit from PBT based on the predicted rate of hepatic toxicity. Materials and methodsEligible patients were those who underwent PBT for primary liver tumors with a maximum diameter of ≤ 5 cm and Child-Pugh grade A (n = 40). To compare the PBT-plan, the treatment plan using volumetric modulated arc therapy was generated as the XRT-plan. The rate of predicted hepatic toxicity was estimated using five normal tissue complication probability (NTCP) models with three different endpoints. The differences in NTCP values (ΔNTCP) were calculated to determine the relative advantage of PBT. Factors predicting benefits of PBT were analyzed by logistic regression analysis. ResultsFrom the dose-volume histogram comparisons, an advantage of PBT was found in sparing of the normal liver receiving low doses. The factors predicting the benefit of PBT differed depending on the selected NTCP model. From the five models, the total tumor diameter (sum of the target tumors), location (hepatic hilum vs other), and number of tumors (1 vs 2) were significant factors. ConclusionsFrom the radiation-related hepatic toxicity, factors were identified to predict benefits of PBT in primary liver tumors with Child-Pugh grade A, with the maximum tumor diameter of ≤ 5 cm.

Predicting acute severe toxicity for head and neck squamous cell carcinomas by combining dosimetry with a radiosensitivity biomarker: a pilot study.

Radiotherapy (RT) against head and neck squamous cell carcinomas (HNSCC) may lead to severe toxicity in 30-40% of patients. The normal tissue complication probability (NTCP) models, based on dosimetric data refined the normal tissue dose/volume tolerance guidelines. In parallel, the radiation-induced nucleoshuttling (RIANS) of the Ataxia-Telangiectasia Mutated protein (pATM) is a predictive approach of individual intrinsic radiosensitivity. Here, we combined NTCP with RADIODTECT©, a blood assay derived from the RIANS model, to predict RT toxicity in HNSCC patients. RADIODTECT© cutoff values (i.e. 57.8 ng/mL for grade⩾2 toxicity and 46 ng/mL for grade⩾3 toxicity) have been previously assessed. Validation was performed on a prospective cohort of 36 HNSCC patients treated with postoperative RT. Toxicity was graded with the Common Terminology Criteria for Adverse Events (CTCAE) scale and two criteria were considered: grade⩾2 oral mucositis (OM2), grade⩾3 mucositis (OM3) and grade⩾2 dysphagia (DY2), grade⩾3 dysphagia (DY3). pATM quantification was assessed in lymphocytes of HNSCC patients. The discrimination power of the pATM assay was evaluated through the Area Under the Receiver Operator Characteristics Curve (AUC-ROC). Two previously described NTCP models were considered, including the dose to the oral cavity and the mean dose to the parotid glands (OM2 and OM3) and the dose to the oral cavity, to the larynx and the volume of pharyngeal constrictor muscles (DY2 and DY3). Combining NTCP models with RADIODTECT© blood test improved the AUC-ROC. Considering the prediction of mucositis, AUC-ROCNTCP+RADIODTECT©=0.80 was for OM2, and AUC-ROCNTCP+RADIODTECT©=0.78 for OM3. Considering the prediction of acute dysphagia, AUC-ROCNTCP+RADIODTECT©=0.71 for DY2 and for DY3. Combining NTCP models with a radiosensitivity biomarker might significantly improve the prediction of toxicities for HNSCC patients.

Extending the relative seriality formalism for interpretable deep learning of normal tissue complication probability models

We formally demonstrate that the relative seriality (RS) model of normal tissue complication probability (NTCP) can be recast as a simple neural network with one convolutional and one pooling layer. This approach enables us to systematically construct deep relative seriality networks (DRSNs), a new class of mechanistic generalizations of the RS model with radiobiologically interpretable parameters amenable to deep learning. To demonstrate the utility of this formulation, we analyze a simplified example of xerostomia due to irradiation of the parotid gland during alpha radiopharmaceutical therapy. Using a combination of analytical calculations and numerical simulations, we show for both the RS and DRSN cases that the ability of the neural network to generalize without overfitting is tied to ‘stiff’ and ‘sloppy’ directions in the parameter space of the mechanistic model. These results serve as proof-of-concept for radiobiologically interpretable deep learning of NTCP, while simultaneously yielding insight into how such techniques can robustly generalize beyond the training set despite uncertainty in individual parameters.

Radiation Induced Lymphopenia Is Associated With the Effective Dose to the Circulating Immune Cells in Breast Cancer.

BackgroundLymphopenia is a known significant factor for treatment outcome in cancer patients, with underlying risk factor poorly understood in breast cancer. We hypothesize that the effective dose to the circulating immune cells (EDIC) which was related with lymphopenia in lung cancer will also have significant effect for radiation induced lymphopenia (RIL) in patients with breast cancer.Material and MethodsPatients treated with adjuvant radiotherapy (RT) and with complete blood tests within one week from RT end/start (post/preRT) were eligible in this study. Radiation dosimetric factors were collected retrospectively, and EDIC for each patient was calculated based on the doses to lung, heart and total body according to the model description, as previously reported. RIL was defined by the CTCAE5.0 based on postRT peripheral lymphocyte count (PLC). Linear regression was first used to test the correlation between EDIC with post/preRT PLC ratio and postRT PLC, using all these as continuous variables. Normal tissue complication probability (NTCP) was used to develop models that predict the CTCAE graded RIL from EDIC.ResultsA total of 735 patients were eligible. The mean post/preRT PLC ratio was 0.66 (95% CI: 0.64-0.68) and mean EDIC of breast cancer was 1.70Gy (95% CI: 1.64-1.75). Both post/preRT PLC ratio and postRT PLC were significantly correlated with EDIC (P<0.001), with R2 of 0.246. For patients with normal preRT PLC, the post/preRT PLC ratio was better associated with EDIC, and postRT PLC was expressed as PLCpreRT × (0.89 – 0.16 × EDIC). For patients with preRT lymphopenia, postRT PLC was better associated with EDIC and it was 1.1 – 0.17 × EDIC. Using binned EDIC as the dose variable, the bootstrap validated NTCPs fit the data nicely with R2 of 0.93, 0.96, and 0.94 for grade-1, grade-2, and grade-3 RIL, respectively. The corresponding EDIC to induce 50% of grade-1, grade-2 and grade-3 RIL was 1.2, 2.1 and 3.7 Gy, respectively.ConclusionEDIC is a significant factor for RIL in patients with breast cancer, and may be used to compute the risk of lymphopenia in each individual patient with the use of the conventional NTCP modeling. External validation is needed before the EDIC can be used to guide RT plan.

Impact of sarcopenia on acute radiation-induced toxicity in head and neck cancer patients

Background and purposeSarcopenia is related to late radiation-induced toxicities and worse survival in head and neck cancer (HNC) patients. This study tested the hypothesis that sarcopenia improves the performance of current normal tissue complication probability (NTCP) models of radiation-induced acute toxicity in HNC patients. Material/methodsThis was a retrospective analysis in a prospective cohort of HNC patients treated from January 2007 to December 2018 with (chemo)radiotherapy. Planning CT scans were used for evaluating skeletal muscle mass. Characteristics of sarcopenic and non-sarcopenic patients were compared. The impact of sarcopenia was analysed by adding sarcopenia to the linear predictors of current NTCP models predicting physician- and patient-rated acute toxicities. ResultsThe cut-off values of sarcopenia in the study population (n = 977) were established at skeletal muscle index < 42.0 cm2/m2 (men) and < 31.2 cm2/m2 (women), corresponding to the lowest sex-specific quartile. Compared to non-sarcopenic patients, sarcopenic patients were more frequently smokers (61% vs. 48%, p < 0.001), had more often advanced stage of disease (stage III-IV, p = 0.004), higher age (67 vs. 63 years, p < 0.001) and experienced more pretreatment complaints, such as dysphagia (grade ≥ 2, p < 0.001). Sarcopenia remained statistically significant, next to the linear predictor, only for physician-rated grade ≥ 3 dysphagia (week 3–6 during RT, p < 0.01). However, sarcopenia did not improve the performance of these NTCP models (p > 0.99). ConclusionSarcopenia in HNC patients was an independent prognostic factor for radiation-induced physician-rated acute grade ≥ 3 dysphagia, which might be explained by its impact on swallowing muscles. However, addition of sarcopenia did not improve the NTCP model performance.

Stricter Postoperative Oropharyngeal Cancer Radiation Therapy Normal Tissue Dose Constraints Are Feasible

PurposeAlthough dose de-escalation is one proposed strategy to mitigate long-term toxicity in human papillomavirus associated oropharyngeal cancer, applying more stringent normal tissue constraints may be a complementary approach to further reduce toxicity. Our study demonstrates that in a postoperative setting, improving upon nationally accepted constraints is achievable and leads to reductions in normal tissue complication probabilities (NTCP) without compromising disease control. Methods and MaterialsWe identified 92 patients at our institution between 2015 and 2019 with p16+ oropharyngeal cancer who were treated with adjuvant volumetric modulated arc therapy. We included patients treated to postoperative doses and standard volumes (including bilateral neck). Doses delivered to organs at risk were compared with recommended dose constraints from a recent cooperative group head and neck cancer trial of radiation therapy to 60 Gy. We applied validated and published NTCP models for dysphagia, dysgeusia, esophagitis, oral mucositis, and xerostomia relevant to oropharyngeal cancer. ResultsAchievable and delivered mean doses to most normal head and neck tissues were well below national recommended constraints. This translates to notable absolute NTCP reductions for salivary flow (10% improvement in contralateral parotid, 35% improvement in submandibular gland), grade ≥ 2 esophagitis (23% improvement), grade ≥ 3 mucositis (17% improvement), dysgeusia (10% improvement), and dysphagia (8% improvement). Locoregional control at a median follow-up of 26.3 months was 96.7%, with only 3 patients experiencing locoregional recurrence (1 local, 2 regional). ConclusionsModern radiation therapy planning techniques allow for improved normal tissue sparing compared with currently established dose constraints without compromising disease control. These improvements may lead to reduced toxicity in a patient population expected to have favorable long-term outcomes. Stricter constraints can be easily achieved and should be used in conjunction with other evolving efforts to mitigate toxicity.

Dosimetry in the Prediction and Reduction of Acute Skin Toxicity from the Use of Brass Bolus in Postmastectomy Chest Wall Radiotherapy Treatment

Purpose: Brass bolus has been shown to be an acceptable substitute for tissue-equivalent bolus (TEB), and may be advantageous for surface dose delivery to the often irregularly shaped chest wall in postmastectomy radiotherapy (PMRT) [1]. The purpose of this study was to analyse the in vivo TLD skin dose measurements, and use normal tissue complication probability (NTCP) and multivariate logistic regression models for the prediction of acute skin toxicity following PMRT.