Baseline shift corrections towards the heart: External validation of the impact on survival in early-stage NSCLC patients.

PURPOSE: To externally validate Johnson-Hart et al. findings: the association of tumor baseline shifts towards the heart with overall survival (OS) in SBRT for NSCLC. Further analysis included investigating the presence of interfractional heart baseline shifts and the association of OS with heart dose change during treatment. METHODS: Data from 416 SBRT early-stage NSCLC patients was collected. Pearson's correlations (PCCs) between clinical variables and treatment-averaged tumor shifts towards/away from the heart were explored. Validation of published multivariable Cox model was performed. PCCs between heart and tumor baseline shifts were analyzed. Dose accumulation was performed following daily CBCT-to-pCT deformable registration. Maximum heart dose (D0) was computed for planned and accumulated doses. Differences in OS according to shifts towards/away from the heart or D0 increase/decrease were analyzed. Significant D0 differences between patients with D0 increase/decrease and different tumor locations were explored. RESULTS: Tumor shifts towards/away from the heart showed no significant association with OS (p = 0.91). Distance between PTV and heart correlated significantly (PCC = 0.18) with shifts to the heart. Cox model did not validate in our cohort. Heart presented baseline shifts positively correlated with tumor baseline shifts in all three directions (PCC ≥ 0.38; p < 0.001). Counterintuitively, patients experiencing increased D0 during treatment showed significantly better OS (p = 0.0077). Upper-lobe tumor patients with increased D0 had lower D0 than those with decreased D0 (right-upper-lobe p ≤ 0.018). CONCLUSIONS: In our SBRT cohort, the shifts towards the heart were not associated with worse OS. Moderate correlations were found between tumor and heart baseline shifts in each direction. Moreover, the distance between the PTV and the heart showed a significant correlation with shifts to the heart.

Estimating how contouring differences affect normal tissue complication probability modelling.

Background and purpose: Normal tissue complication probability (NTCP) models are developed from large retrospective datasets where automatic contouring is often used to contour the organs at risk. This study proposes a methodology to estimate how discrepancies between two sets of contours are reflected on NTCP model performance. We apply this methodology to heart contours within a dataset of non-small cell lung cancer (NSCLC) patients. Materials and methods: One of the contour sets is designated the ground truth and a dosimetric parameter derived from it is used to simulate outcomes via a predefined NTCP relationship. For each simulated outcome, the selected dosimetric parameters associated with each contour set are individually used to fit a toxicity model and their performance is compared. Our dataset comprised 605 stage IIA-IIIB NSCLC patients. Manual, deep learning, and atlas-based heart contours were available. Results: How contour differences were reflected in NTCP model performance depended on the slope of the predefined model, the dosimetric parameter utilized, and the size of the cohort. The impact of contour differences on NTCP model performance increased with steeper NTCP curves. In our dataset, parameters on the low range of the dose-volume histogram were more robust to contour differences. Conclusions: Our methodology can be used to estimate whether a given contouring model is fit for NTCP model development. For the heart in comparable datasets, average Dice should be at least as high as between our manual and deep learning contours for shallow NTCP relationships (88.5 ± 4.5 %) and higher for steep relationships.

Severe radiation-induced lymphopenia during concurrent chemoradiotherapy for stage III non-small cell lung cancer: external validation of two prediction models.

Background: Severe radiation-induced lymphopenia (RIL) in patients undergoing chemoradiotherapy (CRT) for non-small cell lung cancer (NSCLC) is associated with decreased immunotherapy efficacy and survival. At The Christie and MD Anderson Cancer Center (MDACC), prediction models for lymphopenia were developed in lung and esophageal cancer patients, respectively. The aim of this study was to externally validate both models in patients with stage III NSCLC. Methods: Patients who underwent concurrent CRT for stage III NSCLC in 2019-2021 were studied. Outcomes were grade ≥3 and grade 4 lymphopenia during CRT. The Christie model predictors for grade ≥3 lymphopenia included age, baseline lymphocyte count, radiotherapy duration, chemotherapy, mean heart and lung doses, and thoracic vertebrae V20Gy. MDACC predictors for grade 4 lymphopenia were age, baseline lymphocyte count, planning target volume (PTV), and BMI. The external performance of both models was assessed. Results: Among 100 patients, 78 patients (78%) developed grade ≥3 lymphopenia, with grade 4 lymphopenia in 17 (17%). For predicting grade ≥3 lymphopenia, the Christie and MDACC models yielded c-statistics of 0.77 and 0.79, respectively. For predicting grade 4 lymphopenia, c-statistics were 0.69 and 0.80, respectively. Calibration for the Christie and MDACC models demonstrated moderate and good agreement, respectively. Conclusion: The PTV-based MDACC prediction model for severe RIL demonstrated superior external performance in NSCLC patients compared to the dosimetry-based Christie model. As such, the MDACC model can aid in identifying patients at high risk for severe lymphopenia. However, to optimize radiotherapy planning, further improvement and external validation of dosimetry-based models is desired.

Bronchial Stenosis in Central Pulmonary Tumors Treated With Stereotactic Body Radiation Therapy.

PURPOSE: Stereotactic body radiation therapy (SBRT) in lung tumors has an excellent local control due to the high delivered dose. Proximity of the proximal bronchial tree (PBT) to the high dose area may result in pulmonary toxicity. Bronchial stenosis is an adverse event that can occur after high dose to the PBT. Literature on the risk of developing bronchial stenosis is limited. We therefore evaluated the risk of bronchial stenosis for tumors central to the PBT and correlated the dose to the bronchi. METHODS AND MATERIALS: Patients with a planning tumor volume (PTV) ≤2 cm from PBT receiving SBRT (8 × 7.5 Gy) between 2015 to 2019 were retrospectively reviewed. Main bronchi and lobar bronchi were manually delineated. Follow-up computed tomography scans were analyzed for bronchial stenosis and atelectasis. Bronchial stenosis was assessed using Common Terminology Criteria for Adverse Events Version 4.0 (CTCAEv4). Patient, tumor, dosimetric factors and survival were evaluated between patients with and without stenosis using uni- and multivariate and Kaplan-Meier analysis. RESULTS: Fifty-one patients were analyzed with a median age of 70 years and World Health Organization (WHO) performance status ≤1 in 92.2%. Median follow-up was 36 months (interquartile range [IQR], 19.6-45.4) and median overall survival 48 months (IQR 21.5-59.3). In 15 patients (29.4%) bronchial stenosis was observed on follow-up computed tomography scan. Grade 1 stenosis was seen in 21.6% (n = 11), grade 2 in 7.8% (n = 4). No grade ≥3 stenosis was observed. Median time to stenosis was 9.6 months (IQR 4.4-19.2). Patients who developed stenosis had significantly larger gross tumor volume with a median of 19 cm3(IQR 7.7-63.2) versus 5.2 cm3 (IQR 1.7-11.3, P <.01). Prognostic factors in multivariate analysis for stenosis were age (P = .03; odds ratio [OR] 1.1), baseline dyspnea (P = .02 OR 7.7), and the mean lobar bronchus dose (P = .01; OR 1.1). CONCLUSIONS: Low-grade (≤2) lobar bronchial stenosis is a complication in approximately one-third of patients after SBRT for lung tumors with a PTV ≤2 cm from PBT. Prognostic risk factors were age, baseline dyspnea and mean dose on a lobar bronchus.

Deep learning model for automatic contouring of cardiovascular substructures on radiotherapy planning CT images: Dosimetric validation and reader study based clinical acceptability testing.

BACKGROUND AND PURPOSE: Large radiotherapy (RT) planning imaging datasets with consistently contoured cardiovascular structures are essential for robust cardiac radiotoxicity research in thoracic cancers. This study aims to develop and validate a highly accurate automatic contouring model for the heart, cardiac chambers, and great vessels for RT planning computed tomography (CT) images that can be used for dose-volume parameter estimation. MATERIALS AND METHODS: A neural network model was trained using a dataset of 127 expertly contoured planning CT images from RT treatment of locally advanced non-small-cell lung cancer (NSCLC) patients. Evaluation of geometric accuracy and quality of dosimetric parameter estimation was performed on 50 independent scans with contrast and without contrast enhancement. The model was further evaluated regarding the clinical acceptability of the contours in 99 scans randomly sampled from the RTOG-0617 dataset by three experienced radiation oncologists. RESULTS: Median surface dice at 3 mm tolerance for all dedicated thoracic structures was 90% in the test set. Median absolute difference between mean dose computed with model contours and expert contours was 0.45 Gy averaged over all structures. The mean clinical acceptability rate by majority vote in the RTOG-0617 scans was 91%. CONCLUSION: This model can be used to contour the heart, cardiac chambers, and great vessels in large datasets of RT planning thoracic CT images accurately, quickly, and consistently. Additionally, the model can be used as a time-saving tool for contouring in clinic practice.

Delivered dose-effect analysis of radiation induced rib fractures after thoracic SBRT.

BACKGROUND AND PURPOSE: Anatomical changes during the stereotactic body radiation therapy (SBRT) of early stage non-small cell lung cancer (NSCLC) may cause the delivered dose to deviate from the planned dose. We investigate if normal tissue complication probability (NTCP) models based on the delivered dose predict radiation-induced rib fractures better than models based on the planned dose. MATERIAL AND METHODS: 437 NSCLC patients treated to a median dose of 3x18 Gy were included. Delivered dose was estimated by accumulating EQD2-corrected fraction doses after being deformed with daily CBCT-to-planning CT deformable image registration. Dosimetric parameters Dx (dose to a relative volume x) were extracted for each rib included in the CBCTs field-of-view. An NTCP model was constructed for both planned and delivered dose, optimizing the parameters TD50 (dose with 50% toxicity risk), m (steepness of the curve) and x, using maximum likelihood estimation. Best NTCP model was determined using Akaike weights (Aw). Differences between the models were tested for significance using the Vuong's test. RESULTS: Median time to fracture of 110 fractured ribs was 22.5 months. The maximum rib dose, D0, best predicted fractures for both planned and delivered dose. The average delivered D0 was significantly lower than planned (p < 0.001). NTCP model based on the delivered D0 was the best, with Aw = 0.95. The models were not significantly different. CONCLUSION: Delivered maximum dose to the ribs was significantly lower than planned. The NTCP model based on delivered dose improved predictions of radiation-induced rib fractures but did not reach statistical significance.

The prognostic value of volumetric changes of the primary tumor measured on Cone Beam-CT during radiotherapy for concurrent chemoradiation in NSCLC patients.

INTRODUCTION: The aim of this study was to identify subgroups of locally advanced NSCLC patients with a distinct treatment response during concurrent chemoradiotherapy (CCRT). Subsequently, we investigated the association of subgroup membership with treatment outcomes. METHODS: 394 NSCLC-patients treated with CCRT between 2007 and 2013 were included. Gross Tumor Volume (GTV) during treatment was determined and relative GTV-volume change from the planning-CT was subsequently calculated. Latent Class Mixed Modeling (LCMM) was used to identify subgroups with distinct volume changes during CCRT. The association of subgroup membership with overall survival (OS), progression free survival (PFS) and local regional control (LRC) was assessed using cox regression analyses. RESULTS: Three subgroups of GTV-volume change during treatment were identified, with each subsequent subgroup showing a more profound reduction of GTV during treatment. No associations between subgroup membership and OS, PFS nor LRC were observed. Nonetheless, baseline GTV (HR1.42; 95%CI 1.06-1.91) was significantly associated with OS. CONCLUSIONS: Three different subgroups of GTV-volume change during treatment were identified. Surprisingly, these subgroups did not differ in their risk of treatment outcomes. Only patients with a larger GTV at baseline had a significantly worse OS. Therefore, risk stratification at baseline might already be accurate in identifying the best treatment strategy for most patients.

Safety and efficacy of reduced dose and margins to involved lymph node metastases in locally advanced NSCLC patients.

BACKGROUND AND PURPOSE: (Chemo)Radiotherapy for locally advanced non-small lung cancer (LA-NSCLC) causes severe dysphagia due to the radiation dose to the mediastinal lymphadenopathy. Reducing the dose to the mediastinum and the margins to the planning target volume (PTV) might reduce severe toxicity rates. The results of both adaptations in LA-NSCLC patients receiving (chemo)radiotherapy were analysed. MATERIALS AND METHODS: 308 LA-NSCLC patients were included in an observational study. Both cohorts received hypofractionated RT (24 × 2.75 Gy) of 70 Gy (EQD210) to the primary tumour. The reference-cohort (N = 170) received the same dose of 70 Gy (EQD210) to the involved lymph nodes, while the reduction-cohort (N = 138) received 24 × 2.42 Gy, biologically equivalent to 60 Gy (EQD210). Furthermore, the patient-specific PTV-margins for both the primary tumour and lymph nodes were reduced by 2-3 mm in the reduction-cohort after implementing a carina based correction strategy. The effects on toxicity, regional failure and overall survival (OS) were assessed. RESULTS: The acute grade 3 (G3) dysphagia and G3 pulmonary toxicity decreased significantly from 12.9% to 3.6% and 4.1% versus 0%, respectively. The regional failures were comparable: 5.9% versus 4.3% (p = 0.546). The median OS was significantly different: 26 months (reference-cohort) versus 35 months (reduction-cohort). After correction for confounders, the association between the reduction-cohort and OS remained significant (HR 0.63 versus HR 0.70). CONCLUSION: A reduction in PTV-margins and dose from 70 Gy to 60 Gy to the involved lymph nodes in LA-NSCLC patients receiving (chemo)radiotherapy did not result in an increase in regional failures. Moreover, significantly lower acute toxicities and an improved OS were observed in the reduction-cohort.

Quantification of Esophageal Tumor Motion and Investigation of Different Image-Guided Correction Strategies.

PURPOSE: To accurately quantify esophageal tumor position variability and to optimize image guided correction strategies. MATERIAL AND METHODS: Esophageal cancer patients receiving chemoradiotherapy (41.4-50.4 Gy in 23-28 fractions combined with carboplatin plus paclitaxel) were included in a prospective cohort study (NCT02139488). Gold fiducial markers were inserted into the esophageal tumors during diagnostic endoscopic ultrasound. Four-dimensional (4D) planning computed tomography (CT) and daily 4D cone beam (CB) CT scans were acquired. Each CBCT was registered to the planning CT using different regions of interest (bone; 3D), and carina, diaphragm, clinical target volume (CTV), and fiducial markers (4D) for alignment and using the fiducial markers as the true tumor position. Subsequently, a planning target volume (PTV) margin accounting for residual uncertainties, including the average respiratory motion, was calculated for each of these registrations. RESULTS: Fifty-six patients with tumors located in the proximal (n = 1), mid (n = 7), or distal esophagus (n = 25) or at the gastroesophageal junction (n = 23) were included. The average peak-to-peak respiratory tumor motion was 0.20, 0.92, and 0.34 cm on the planning CT in left-right (LR), cranial-caudal (CC), and anterior-posterior (AP) directions, respectively. The required PTV margin with average motion amplitude, depending on the correction strategy used for image guidance, ranged from 0.8 cm to 1.0 cm, 1.1 cm to 1.6 cm, and 0.7 cm to 0.9 cm in LR, CC, and AP direction, respectively. A registration based on the CTV resulted in the smallest PTV margins (0.8, 1.1, and 0.7 cm in LR, CC, and AP direction, respectively). For bone registration the calculated PTV margins were 1.0, 1.3, and 0.7 cm in LR, CC, and AP directions, respectively. The registration based on the diaphragm increased PTV margins. CONCLUSIONS: Substantial and anisotropic position variability of esophageal tumors was observed during radiation therapy, and nonuniform margins should be considered. Cranial-caudal PTV margins need to be larger than those commonly used. Target positioning during image-guided radiotherapy could be improved with a CTV registration-based correction strategy.

Subgroup Survival Analysis in Stage I-II NSCLC Patients With a Central Tumor Partly Treated With Risk-Adapted SBRT.

PURPOSE: Stereotactic body radiation therapy has been associated with increased toxicity when delivered to patients with early-stage non-small cell lung cancer with a tumor within 2 cm of the proximal bronchial tree (PBT). We investigated noncancer deaths for these patients as related to gross tumor volume (GTV) proximity to the PBT, compared with peripheral tumors. METHODS AND MATERIALS: We included 765 patients with early-stage non-small cell lung cancer who were treated with stereotactic body radiation therapy to a median of 3 × 18 Gy. Central tumors were treated with a risk-adapted (less-intense) schedule (mostly 8 fractions) in 55% of the patients in the first-centimeter group and 27% of the patients in the second-centimeter group. An average anatomy with contouring of PBT and organs at risk (OARs) was deformed onto each patient to obtain the distance of the GTV to the PBT and doses to OARs. Log-rank, 1-way analysis of variance, and Cox regressions were performed to assess differences in the first-centimeter, second centimeter, and peripheral groups and associations with noncancer deaths. RESULTS: The median overall survival was 42.7 months, the median noncancer death occurred in 57.3 months, and the median follow-up was 34.8 months. Noncancer death in the first-centimeter group (31 patients) was significantly different from noncancer death in the other groups, with a hazard ratio of 3.175 (P < .001). Noncancer death in the second-centimeter group (71 patients) was not different from noncancer death in the peripheral group (P = .53). Doses to OARs were higher in the first- and second-centimeter groups than in the peripheral group for all OARs. High dose to the PBT was associated with noncancer death (D1%; hazard ratio, 1.006 Gy-1; P = .003). CONCLUSIONS: Patients with a GTV in the first centimeter surrounding the PBT died more often from causes other than cancer compared with other patients. Noncancer death in patients with a GTV in the second centimeter, who partly received a risk-adapted schedule, was comparable to that in patients with a peripheral tumor.

Heart dose associated with overall survival in locally advanced NSCLC patients treated with hypofractionated chemoradiotherapy.

Association of heart dose and overall survival was investigated in a cohort including 469 locally-advanced NSCLC patients receiving daily low-dose hypofractionated chemo-radiotherapy. Significant associations were found over a range of dose parameters. Multivariate analysis showed significant associations of heart_V2Gy:HR=1.007%-1 (95% CI:1.002-1.013; p=0.006), age:HR=1.026year-1 (1.011-1.042; p=0.001) and GTV volume:HR=1.001cc-1 (1.000-1.002; p=0.006) with overall survival.

Dose to heart substructures is associated with non-cancer death after SBRT in stage I-II NSCLC patients.

BACKGROUND AND PURPOSE: To investigate potential associations between dose to heart (sub)structures and non-cancer death, in early stage non-small cell lung cancer (NSCLC) patients treated with stereotactic body radiation therapy (SBRT). METHODS: 803 patients with early stage NSCLC received SBRT with predominant schedules of 3×18Gy (59%) or 4×12Gy (19%). All patients were registered to an average anatomy, their planned dose deformed accordingly, and dosimetric parameters for heart substructures were obtained. Multivariate Cox regression and a sensitivity analysis were used to identify doses to heart substructures or heart region with a significant association with non-cancer death respectively. RESULTS: Median follow-up was 34.8months. Two year Kaplan-Meier overall survival rate was 67%. Of the deceased patients, 26.8% died of cancer. Multivariate analysis showed that the maximum dose on the left atrium (median 6.5Gy EQD2, range=0.009-197, HR=1.005, p-value=0.035), and the dose to 90% of the superior vena cava (median 0.59Gy EQD2, range=0.003-70, HR=1.025, p-value=0.008) were significantly associated with non-cancer death. Sensitivity analysis identified the upper region of the heart (atria+vessels) to be significantly associated with non-cancer death. CONCLUSIONS: Doses to mainly the upper region of the heart were significantly associated with non-cancer death. Consequently, dose sparing in particular of the upper region of the heart could potentially improve outcome, and should be further studied.

Dose-effect analysis of radiation induced rib fractures after thoracic SBRT.

BACKGROUND AND PURPOSE: To determine a dose-effect relation for radiation induced rib fractures after stereotactic body radiation therapy (SBRT) in early stage non-small cell lung cancer (NSCLC). Automatic rib delineation has enabled the analysis of a large patient group. MATERIAL AND METHODS: Four-hundred and sixty-six patients with stage I/II NSCLC received SBRT with a median of 54Gy in 3 fractions. The optimal EQD2-corrected dose parameter to predict (a)symptomatic fractures was found using Cox regression. Three normal tissue complication probability (NTCP) models based on this optimal parameter were constructed: (1) at a median follow up (FU) of 26months, (2) for all data, with time to toxicity taken into account and (3) at a FU of 26months, excluding low dose ribs. RESULTS: The median time to fracture was 22 (range 5-51) months. Maximum rib dose best predicted fractures. The TD50 (dose with 50% complication) of the second NTCP model was 375Gy. The TD50 was significantly higher for the other models indicating an under-estimation of the dose effect at the median follow-up time and/or when excluding low dose ribs. CONCLUSIONS: The risk of symptomatic rib fractures after SBRT was significantly correlated to dose, and was <5% at 26months when Dmax<225Gy.

Validation of automatic segmentation of ribs for NTCP modeling.

BACKGROUND AND PURPOSE: Determination of a dose-effect relation for rib fractures in a large patient group has been limited by the time consuming manual delineation of ribs. Automatic segmentation could facilitate such an analysis. We determine the accuracy of automatic rib segmentation in the context of normal tissue complication probability modeling (NTCP). MATERIALS AND METHODS: Forty-one patients with stage I/II non-small cell lung cancer treated with SBRT to 54 Gy in 3 fractions were selected. Using the 4DCT derived mid-ventilation planning CT, all ribs were manually contoured and automatically segmented. Accuracy of segmentation was assessed using volumetric, shape and dosimetric measures. Manual and automatic dosimetric parameters Dx and EUD were tested for equivalence using the Two One-Sided T-test (TOST), and assessed for agreement using Bland-Altman analysis. NTCP models based on manual and automatic segmentation were compared. RESULTS: Automatic segmentation was comparable with the manual delineation in radial direction, but larger near the costal cartilage and vertebrae. Manual and automatic Dx and EUD were significantly equivalent. The Bland-Altman analysis showed good agreement. The two NTCP models were very similar. CONCLUSIONS: Automatic rib segmentation was significantly equivalent to manual delineation and can be used for NTCP modeling in a large patient group.

How the blood pool properties at onset affect the temporal behavior of simulated bruises.

The influence of initial blood pool properties on the temporal behavior of bruises is currently unknown. We addressed this important issue by utilizing three typical classes of bruises in our three-layered finite compartment model. We simulated the effects of their initial shapes, regularity of boundaries and initial blood concentration distributions (gaussian vs. homogeneous) on the hemoglobin and bilirubin areas in the dermal top layer. Age determination of bruises with gaussian hemoglobin concentration was also addressed. We found that the initial blood pool properties strongly affect bruise behavior. We determined the age of a 200-h simulated bruise with gaussian hemoglobin concentration with 3 h uncertainty. In conclusion, bruise behavior depends non-intuitively on the initial blood pool properties; hence, a model that includes shape, area and concentration distribution at onset is indispensable. Future age determination, including inhomogeneous hemoglobin distributions, will likely be based on the presented method for gaussian distributions.

Can color inhomogeneity of bruises be used to establish their age?

Bruises become spatially inhomogeneous during the healing process; a smaller red-blue core area, caused by hemoglobin, is surrounded by a larger yellow area, caused by bilirubin, which is enzymatically formed from hemoglobin. These two areas develop at different rates and hence carry information about the age of the bruise. We present a proof of principle demonstration that the age of bruises can be determined via an inverse procedure using a mathematical model and daily measurements of these two areas using a hyperspectral imaging system. The inaccuracy found is 2.3% for fresh bruises and 3 to 24% for bruises up to 3 days old. In conclusion, color inhomogeneity of bruises can be used to determine their age. We expect that future age determination of bruises by the inverse procedure described here, possibly also including the distribution of concentrations in the areas will open up a new phase in clinical bruise classification.

3D finite compartment modeling of formation and healing of bruises may identify methods for age determination of bruises.

Simulating the spatial and temporal behavior of bruises may identify methods that allow accurate age determination of bruises to assess child abuse. We developed a numerical 3D model to simulate the spatial kinetics of hemoglobin and bilirubin during the formation and healing of bruises. Using this model, we studied how skin thickness, bruise diameter and diffusivities affect the formation and healing of circular symmetric bruises and compared a simulated bruise with a natural inhomogeneous bruise. Healing is faster for smaller bruises in thinner and less dense skin. The simulated and natural bruises showed similar spatial and temporal dynamics. The different spatio-temporal dynamics of hemoglobin and bilirubin allows age determination of model bruises. Combining our model predictions with individual natural bruises may allow optimizing our model parameters. It may particularly identify methods for more accurate age determination than currently possible to aid the assessment of child abuse.