Search

Introduction: COVID-19 is associated with characteristic lung CT findings, such as rounded ground-glass opacities in certain distributions Diagnosing COVID-19 is a particular concern in oncology care, since cancer patients are a vulnerable population who receive treatment in close proximity to other patients and staff Radiotherapy patients routinely undergo CT simulation before starting therapy We hypothesized that simulation CT scans obtained on patients treated during the pandemic would reveal characteristic COVID-19 findings and represent a tool to identify patients with asymptomatic COVID-19 Methods: We reviewed patients undergoing CT simulation during a six-week period (March 1 to April 13, 2020) at a major tertiary cancer center located in an early epicenter of the COVID-19 pandemic in the United States Most scans were done under free-breathing conditions, with slice thickness ≤3mm and without IV contrast All scans were reviewed according to the RSNA classification of COVID-19 lung CT findings (“typical,” “indeterminate,” “atypical,” or “negative” for COVID-19 pneumonia) by radiation oncologists who had been trained by a diagnostic radiologist All “typical” or “indeterminate” scans were considered suspicious and re-reviewed by a board-certified diagnostic radiologist Radiographic classifications were then compared with available COVID-19 PCR test results A one-tailed T test was used to compare the rate of positive COVID-19 tests in the radiographically suspicious vs non-suspicious groups Results: 414 CT simulation scans that included the lungs were performed on 400 patients during the study period 119 patients (corresponding to 130 scans, or 31 4%) had COVID-19 PCR test results available The most common cancer types were breast (37%), lung/thoracic (23%), and spine (21%) On initial review by radiation oncologists, 17 scans (4 1%), were deemed “typical” for COVID-19 pneumonia, 54 (13%) were “indeterminate,” 85 (21%) were “atypical,” and 258 (62 3%) were “negative ” Of the 71 suspicious (typical or indeterminate) scans, 23 had corresponding COVID-19 test results, of which 3 (15 7%) were positive for infection 107 non-suspicious (atypical or negative) scans had corresponding COVID-19 test results, and 9 were positive (8 4%) This difference in COVID-19 positivity between radiographically suspicious and non-suspicious groups was not statistically significant (p=0 23) Upon re-review by a diagnostic radiologist, 25 (35%) of the suspicious scans were still deemed suspicious while the majority (n=46, or 65%) were deemed “atypical ” Conclusion: Simulation CT scans obtained for radiation treatment planning can be reviewed for signs of COVID-19 pneumonia About 17% of patients simulated in our metropolitan pandemic epicenter demonstrated findings suspicious for COVID-19 when reviewed by radiation oncologists according to consensus criteria However, few of these patients proved to have COVID-19 infections based on PCR testing, and there was no significant correlation between radiographically suspicious simulation CT scans and COVID-19 positivity in this study Analysis was limited by the lack of available COVID-19 test results in many patients The concordance between radiographic classification by radiation oncologists vs diagnostic radiologists was also low These results suggest that routine review of radiotherapy simulation CT scans is of limited value in identifying asymptomatic COVID-19 infection Keywords: COVID-19, radiotherapy, CT

Background: Radiation with platinum-based chemotherapy is the standard of care for unresectable stage III non-small cell lung cancer (NSCLC). Despite aggressive treatment, progression-free survival and overall survival remain poor. It is unclear whether any tumor genetic mutations are associated with response to chemoradiation therapy. Methods: We retrospectively reviewed clinical outcomes of patients with stage III NSCLC treated with definitive radiation who had undergone tumor molecular profiling through a next-generation DNA sequencing platform. Cox proportional hazards model was used to investigate associations between clinical outcomes and genetic mutations detected by next-generation sequencing. Results: 110 patients were identified with stage III NSCLC and underwent definitive radiation between 2013 and 2017 and tumor molecular profiling. Concurrent or sequential chemotherapy was given in 104 patients (95%). Unbiased genomic analyses revealed a significant association between AKT2 mutations and decreased local-regional tumor control and overall survival (hazard ratios [HR] 12.5 and 13.7, P = .003 and P = .003, respectively). Analyses restricted to loss-of-function mutations identified KMT2C and KMT2D deleterious mutations as negative prognostic factors for overall survival (HR 13.4 and 7.0, P < .001 and P < .001, respectively). Deleterious mutations in a panel of 38 DNA damage response and repair pathway genes were associated with improved local-regional control (HR 0.32, P = .049). Conclusions: This study coupled multiplexed targeted sequencing with clinical outcome and identified mutations in AKT2, KMT2C, and KMT2D as negative predictors of local-regional control and survival, and deleterious mutations in damage response and repair pathway genes were associated with improved local-regional disease control after chemoradiation therapy. These findings will require validation in a larger cohort of patients with prospectively collected and detailed clinical information.

Acute esophagitis (AE) occurs among a significant number of patients with locally advanced lung cancer treated with radiotherapy. Early prediction of AE, indicated by esophageal wall expansion, is critical, as it can facilitate the redesign of treatment plans to reduce radiation-induced esophageal toxicity in an adaptive radiotherapy (ART) workflow. We have developed a novel machine learning framework to predict the patient-specific spatial presentation of the esophagus in the weeks following treatment, using magnetic resonance imaging (MRI)/ cone-beam CT (CBCT) scans acquired earlier in the 6 week radiotherapy course. Our algorithm captures the response patterns of the esophagus to radiation on a patch level, using a convolutional neural network. A recurrence neural network then parses the evolutionary patterns of the selected features in the time series, and produces a predicted esophagus-or-not label for each individual patch over future weeks. Finally, the esophagus is reconstructed, using all the predicted labels. The algorithm is trained and validated by means of ∼ 250 000 patches taken from MRI scans acquired weekly from a variety of patients, and tested using both weekly MRI and CBCT scans under a leave-one-patient-out scheme. In addition, our approach is externally validated using a publicly available dataset (Hugo 2017). Using the first three weekly scans, the algorithm can predict the condition of the esophagus over the succeeding 3 weeks with a Dice coefficient of 0.83 ± 0.04, estimate esophagus volume highly (0.98), correlated with the actual volume, using our institutional MRI/CBCT data. When evaluated using the external weekly CBCT data, the averaged Dice coefficient is 0.89 ± 0.03. Our novel algorithm may prove useful in enabling radiation oncologists to monitor and detect AE in its early stages, and could potentially play an important role in the ART decision-making process.

The presence of two intact lungs makes it challenging to reach a tumoricidal dose with hemithoracic pleural intensity-modulated radiation therapy (IMRT) in patients with malignant pleural mesothelioma (MPM) who underwent pleurectomy/decortications or have unresectable disease. We developed an anatomy-based model to predict attainable prescription dose before starting optimization. Fifty-six clinically delivered IMRT plans were analyzed regarding correlation of prescription dose and individual and total lung volumes, planning target volume (PTV), ipsilateral normal lung volume and ratios: contralateral/ipsilateral lung (CIVR); contralateral lung/PTV (CPVR); ipsilateral lung /PTV (IPVR); ipsilateral normal lung /total lung (INTLVR); ipsilateral normal lung/PTV (INLPVR). Spearman's rank correlation and Fisher's exact test were used. Correlation between mean ipsilateral lung dose (MILD) and these volume ratios and between prescription dose and single lung mean doses were studied. The prediction models were validated in 23 subsequent MPM patients. CIVR showed the strongest correlation with dose (R=0.603,p

Purpose Local recurrence is a significant problem after surgical resection of thoracic tumors. As intraoperative radiotherapy (IORT) can deliver radiation directly to the threatened margin, we have used this therapy in an attempt to reduce local recurrence, using high-dose-rate (HDR) as well as low-dose-rate (LDR) techniques. Methods and Materials We performed a retrospective review of patients undergoing LDR ( 125 I) mesh placement or HDR ( 192 Ir) afterloading therapy during lung tumor resection between 2001 and 2013 at our institution. Competing risks methods were used to estimate the cumulative incidence of local failure. We also assessed possible predictive factors of local failure. Results Fifty-nine procedures (41 LDR and 18 HDR) were performed on 58 patients. Median follow-up was 55.1 months. Cumulative incidence of local failure at 1, 2, and 3 years was 28.5%, 34.2%, and 34.2%, respectively. Median overall survival was 39.9 months. There was no significant difference in local failure according to margin status, HDR vs. LDR, use of adjuvant external beam radiotherapy, or metastatic vs. primary tumor. Two patients (3.4%) experienced Grade 3+ toxicities likely related to brachytherapy. Additionally, 7 patients experienced Grade 3+ postsurgical complications unlikely related to brachytherapy. Conclusions IORT is associated with good local control after resection of thoracic tumors otherwise at very high risk for local recurrence. There is a low incidence of severe toxicity attributable to brachytherapy. HDR-IORT appears to have equivalent outcomes to LDR-IORT. HDR or LDR-IORT can, therefore, be considered in situations where the oncologic completeness of thoracic tumor resection is in doubt.

Stereotactic body radiation therapy (SBRT) is associated with excess toxicity following treatment of central lung tumors. Risk-adapted fractionation appears to have mitigated this risk, but it remains unclear whether SBRT is safe for all tumors within the central lung zone, especially those abutting the proximal bronchial tree (PBT). We investigated the dependence of toxicity on tumor proximity to PBT and whether tumors abutting the PBT had greater toxicity than other central lung tumors after SBRT.A total of 108 patients receiving SBRT for central lung tumors were reviewed. Patients were classified based on closest distance from tumor to PBT. Primary endpoint was SBRT-related death. Secondary endpoints were overall survival, local control, and grade 3+ pulmonary adverse events. We compared tumors abutting the PBT to nonabutting and those ≤1 cm and1 cm from PBT.Median follow-up was 22.7 months. Median distance from tumor to PBT was 1.78 cm. Eighty-eight tumors were primary lung and 20 were recurrent or metastatic; 23% of tumors were adenocarcinoma and 71% squamous cell. Median age was 77.5 years. Median dose was 4500 cGy in 5 fractions prescribed to the 100% isodose line. Eighteen patients had tumors abutting the PBT, 4 of whom experienced SBRT-related death. No other patients experienced death attributed to SBRT. Risk of SBRT-related death was significantly higher for tumors abutting the PBT compared with nonabutting tumors (P.001). Two patients with SBRT-related death received anti-vascular endothelial growth factor therapy and experienced pulmonary hemorrhage. Patients with tumors ≤1 cm from PBT had significantly more grade 3+ events than those with tumors1cm from PBT (P = .014).Even with risk-adapted fractionation, tumors abutting PBT are associated with a significant and differential risk of SBRT-related toxicity and death. SBRT should be used with particular caution in central-abutting tumors, especially in the context of anti-vascular endothelial growth factor therapy.

e21058 Background: Definitive concurrent chemoradiation (cCRT) and durvalumab is a standard therapy for patients with unresectable stage III non-small cell lung cancers (NSCLC). Data is limited on outcomes with this regimen outside of clinical trials. Local-regional control rates to date remain undefined. Methods: We reviewed patients with stage III unresectable NSCLC treated between November 2017 and February 2019 with cCRT. Patients that received at least one cycle of durvalumab were further assessed for 12-month progression free survival (PFS), overall survival (OS), and the incidence and pattern of local-regional and metastatic failures. Disease-relapse was characterized to determine patients potentially eligible for metastasis-directed ablative therapies. Toxicities leading to durvalumab discontinuation were evaluated using CTCAE v.5.0. Results: Of the 83 patients with stage III NSCLC treated with cCRT (median 60Gy), 62 received durvalumab and were evaluable (median follow-up: 12 months). Patients (n = 21, 25%) did not receive durvalumab largely related to metastatic progression (n = 9) or persistent cCRT toxicity (n = 10). In the 62 durvalumab treated patients the median age was 66 (range: 49 - 86), 73% had stage IIIB (n = 33) or IIIC (n = 12) disease, and 58% (n = 36) had adenocarcinoma. The median time from cCRT end to durvalumab start was 1.5 months. Patients received a median of 8 months of durvalumab; 35% (n = 22) of patients completed 12 months of therapy. Common reasons for discontinuing durvalumab included disease progression (32%, 20/62) and toxicity (24%, 15/62). The estimated 12-month PFS and OS were 65% (95% CI: 51 - 79%) and 85% (95% CI: 75 - 95%), respectively. High TMB (≥ 8.8 mt/Mb) or PD-L1 (≥ 1% or PD-L1 ≥ 50%) did not predict improved PFS. Patients who discontinued durvalumab due to toxicity did not have inferior PFS. The cumulative 12-month incidence of local-regional and distant metastatic failures were 18% (95% CI: 5.9 - 30%) and 30% (95% CI: 16.3 - 44.5%), respectively. Of the 17 patients with distant metastases, 9 had oligometastases and would have been potential candidates for comprehensive ablative therapies. Conclusions: Outcomes and toxicities outcomes with cCRT and durvalumab in clinical practice align with the PACIFIC trial. Analysis of disease-relapse suggests a substantial minority of patients with disease progression may be potential candidates for metastasis-directed therapies. Local regional outcomes appear improved to historical data of cCRT alone.

TPS9079 Background: Pleurectomy/Decortication (P/D) with neoadjuvant or adjuvant chemotherapy has become a common lung-sparing surgical approach for MPM. Adjuvant hemithoracic IMPRINT was developed at Memorial Sloan Kettering Cancer Center and safe in a multi-institutional phase II study, with promising survival outcomes. The National Cancer Institute (NCI) sponsored this phase III randomized cooperative group trial to test the efficacy of this lung-sparing trimodality approach for resectable MPM. Methods: Patients with newly diagnosed MPM amenable to P/D are enrolled and undergo P/D followed by adjuvant platinum/pemetrexed (preferred) or neoadjuvant chemotherapy followed by P/D. Patients are stratified by histologic subtype, resection status (R0/1 vs. R2), and center patient volume (≤10 vs. > 10 P/Ds per year). Within 8 weeks after completion of the second modality patients are randomized 1:1 to undergo hemithoracic IMPRINT vs. no further therapy. All IMPRINT contours and treatment plans will be centrally reviewed. A contouring atlas and treatment planning constraints for target structures and organs at risk including acceptable and unacceptable variations and deviations were developed. Photon and proton therapy are permitted. The primary endpoint of the study is overall survival. Secondary endpoints include local failure-free, distant-metastases-free and progression-free survival, treatment-related toxicities (CTCAE v5.0) and change in quality-of-life (EORTC QLQ-C30 mean score changes at 9 months post randomization). The target accrual is 150 patients. This study was activated on January 29, 2020. Over 20 institutions have already committed to opening the study which is open to all National Clinical Trials Network (NCTN) sites. Treatment planning guidelines and helpful hints for photon and proton therapy will be presented. Conclusions: NRG LU-006 (clinicaltrials.gov: NCT04158141 ) is open to accrual. This is the first NRG Oncology randomized phase III trial on MPM and evaluates the use of IMPRINT following lung-sparing P/D and chemotherapy. This project was supported by grants U10CA180868 (NRG Oncology Operations), U10CA180822 (NRG Oncology SDMC), U24CA180803 (IROC) from the National Cancer Institute (NCI). Clinical trial information: NCT04158141.

Objective Treatment of patients with thymic malignancies metastatic to the pleura or pericardium is challenging, and benefits of aggressive treatment are unclear. We sought to characterize the long-term outcomes in this population. Methods We retrospectively identified patients who underwent resection for de novo thymic malignancies metastatic to the pleura between May 1997 and December 2017. Patients with pleural recurrence after prior thymectomy were excluded. Patient demographics, perioperative treatments, pathologic findings, and postoperative outcomes were collected. Descriptive statistics and Kaplan–Meier analyses were performed. Results Seventy-two patients were included (median age, 51 years [range 25-80]; 36/72 women [50%]). Pathologic diagnosis was thymoma in 57 patients (79%) and thymic carcinoma in 15 patients (21%). Most patients (67/72; 93%) received chemotherapy, radiation, or both. Forty-eight patients underwent thymectomy with pleurectomy, 7 patients underwent extrapleural pneumonectomy, 10 patients underwent thymectomy alone, and 7 patients were unresectable. Macroscopic complete resection was achieved in 52 patients (73%). Five-, 10-, and 15-year overall survivals were 73%, 51%, and 18%, respectively, and median overall survival was 11 years (median follow-up, 5.9 years). Forty-six patients (64%) had disease progression (median time to progression, 2.2 years). Repeat episodes of progression and treatment were common (median, 3 episodes/patient). The longest disease-free interval was 12.4 years. Thirteen patients (18%) remain disease-free; 7 patients (10%) were disease-free for more than 5 years. The longest ongoing survival without progression or reintervention is 9.9 years. Conclusions Prolonged survival and, in some cases, cure can be achieved in patients with thymic malignancies metastatic to the pleura or pericardium. Aggressive multimodality therapy may be appropriate for select patients.

We present an adversarial domain adaptation based deep learning approach for automatic tumor segmentation from T2-weighted MRI. Our approach is composed of two steps: (i) a tumor-aware unsupervised cross-domain adaptation (CT to MRI), followed by (ii) semi-supervised tumor segmentation using Unet trained with synthesized and limited number of original MRIs. We introduced a novel target specific loss, called tumor-aware loss, for unsupervised cross-domain adaptation that helps to preserve tumors on synthesized MRIs produced from CT images. In comparison, state-of-the art adversarial networks trained without our tumor-aware loss produced MRIs with ill-preserved or missing tumors. All networks were trained using labeled CT images from 377 patients with non-small cell lung cancer obtained from the Cancer Imaging Archive and unlabeled T2w MRIs from a completely unrelated cohort of 6 patients with pre-treatment and 36 on-treatment scans. Next, we combined 6 labeled pre-treatment MRI scans with the synthesized MRIs to boost tumor segmentation accuracy through semi-supervised learning. Semi-supervised training of cycle-GAN produced a segmentation accuracy of 0.66 computed using Dice Score Coefficient (DSC). Our method trained with only synthesized MRIs produced an accuracy of 0.74 while the same method trained in semi-supervised setting produced the best accuracy of 0.80 on test. Our results show that tumor-aware adversarial domain adaptation helps to achieve reasonably accurate cancer segmentation from limited MRI data by leveraging large CT datasets.

On March 28- 29, 2017, the National Cancer Institute (NCI) Thoracic Malignacy Steering Committee, International Association for the Study of Lung Cancer, and Mesothelioma Applied Research Foundation convened the NCI-International Association for the Study of Lung Cancer- Mesothelioma Applied Research Foundation Mesothelioma Clinical Trials Planning Meeting in Bethesda, Maryland. The goal of the meeting was to bring together lead academicians, clinicians, scientists, and the U.S. Food and Drug Administration to focus on the development of clinical trials for patients in whom malignant pleural mesothelioma has been diagnosed. In light of the discovery of new cancer targets affecting the clinical development of novel agents and immunotherapies in malignant mesothelioma, the objective of this meeting was to assemble a consensus on at least two or three practice-changing multimodality clinical trials to be conducted through NCI's National Clinical Trials Network.

BACKGROUND: Stereotactic Body Radiotherapy (SBRT) has been adopted as the standard of care for inoperable early-stage non-small cell lung cancer (NSCLC), with local control rates consistently > 90%. However, data directly comparing the outcomes of SBRT with those of conventionally fractionated radiotherapy (CONV) is lacking. MATERIAL AND METHODS: Between 1990 and 2013, 497 patients (525 lesions) with early-stage NSCLC (T1-T2N0M0) were treated with CONV (n=127) or SBRT (n=398). In this retrospective analysis, five endpoints were compared, with and without adjusting for clinical and dosimetric factors. Competing risks analysis was performed to estimate and compare the cumulative incidence of local failure (LF), nodal failure (NF), distant failure (DF) and disease progression. Overall survival (OS) was estimated by the Kaplan-Meier method and compared by the Cox regression model. Propensity score (PS) matched analysis was performed based on seven patient and clinical variables: age, gender, Karnofsky performance status (KPS), histology, T-stage, biologically equivalent dose (BED), and history of smoking. RESULTS: The median dose delivered for CONV was 75.6 Gy in 1.8 to 2.0 Gy fractions (range 60 to 90 Gy; median BED = 89.20 Gy) and for SBRT 48 Gy in 4 fractions (45 to 60 Gy in 3 to 5 fractions; median BED = 105.60 Gy). Median follow up was 24.4 months, and 3-year LF rates were 34.1% with CONV and 13.6% with SBRT (p

Purpose To cross-validate and expand a predictive atlas that can estimate geometric patterns of lung tumor shrinkage during radiation therapy using data from 2 independent institutions and to model its integration into adaptive radiation therapy (ART) for enhanced dose escalation. Methods and Materials Data from 22 patients at a collaborating institution were obtained to cross-validate an atlas, originally created with 12 patients, for predicting patterns of tumor shrinkage during radiation therapy. Subsequently, the atlas was expanded by integrating all 34 patients. Each study patient was selected via a leave-one-out scheme and was matched with a subgroup of the remaining 33 patients based on similarity measures of tumor volume and surroundings. The spatial distribution of residual tumor was estimated by thresholding the superimposed shrinkage patterns in the subgroup. A Bayesian method was also developed to recalibrate the prediction using the tumor observed on the midcourse images. Finally, in a retrospective predictive treatment planning (PTP) study, at the initial planning stage, the predicted residual tumors were escalated to the highest achievable dose while maintaining the original prescription dose to the remainder of the tumor. The PTP approach was compared isotoxically to ART that replans with midcourse imaging and to PTP-ART with the recalibrated prediction. Results Predictive accuracy (true positive plus true negative ratios based on predicted and actual residual tumor) were comparable across institutions, 0.71 versus 0.73, and improved to 0.74 with an expanded atlas including 2 institutions. Recalibration further improved accuracy to 0.76. PTP increased the mean dose to the actual residual tumor by an averaged 6.3Gy compared to ART. Conclusion A predictive atlas found to perform well across institutions and benefit from more diversified shrinkage patterns and tumor locations. Elevating tumoricidal dose to the predicted residual tumor throughout the entire treatment course could improve the efficacy and efficiency of treatment compared to ART with midcourse replanning.

OBJECTIVES: Occult primary non-small cell lung cancer (OP-NSCLC) involving mediastinal lymph nodes without an identifiable primary tumor is a rare presentation, with little known about how outcomes compare to typical Stage III NSCLC. We reviewed our experience treating OP-SCLC with definitive radiotherapy and compared outcomes to a contemporary cohort of stage III NSCLC patients. MATERIALS AND METHODS: We reviewed 605 patients with stage III NSCLC staged with PET-CT and treated with definitive radiotherapy between 1998 and 2013. Overall survival, intrathoracic control, and freedom from distant metastasis were computed using Kaplan-Meier method and logrank comparison. Cox hazard ratios were used to perform univariate and multivariate analyses. RESULTS: Twenty-one patients were identified with OP-NSCLC (3.5%). Patients with OP-NSCLC, as compared to known primary NSCLC, had significantly better 5-year rates of intrathoracic control (83.5% vs. 24.2%, P

Purpose The purpose of this study was to evaluate the quality of automatically propagated contours of organs at risk (OARs) based on respiratory‐correlated navigator‐triggered four‐dimensional magnetic resonance imaging (RC‐4DMRI) for calculation of internal organ‐at‐risk volume (IRV) to account for intra‐fractional OAR motion. Methods and Materials T2‐weighted RC‐4DMRI images were of 10 volunteers acquired and reconstructed using an internal navigator‐echo surrogate and concurrent external bellows under an IRB‐approved protocol. Four major OARs (lungs, heart, liver, and stomach) were delineated in the 10‐phase 4DMRI. Two manual‐contour sets were delineated by two clinical personnel and two automatic‐contour sets were propagated using free‐form deformable image registration. The OAR volume variation within the 10‐phase cycle was assessed and the IRV was calculated as the union of all OAR contours. The OAR contour similarity between the navigator‐triggered and bellows‐rebinned 4DMRI was compared. A total of 2400 contours were compared to the most probable ground truth with a 95% confidence level (S95) in similarity, sensitivity, and specificity using the simultaneous truth and performance level estimation (STAPLE) algorithm. Results Visual inspection of automatically propagated contours finds that approximately 5–10% require manual correction. The similarity, sensitivity, and specificity between manual and automatic contours are indistinguishable (P > 0.05). The Jaccard similarity indexes are 0.92 ± 0.02 (lungs), 0.89 ± 0.03 (heart), 0.92 ± 0.02 (liver), and 0.83 ± 0.04 (stomach). Volume variations within the breathing cycle are small for the heart (2.6 ± 1.5%), liver (1.2 ± 0.6%), and stomach (2.6 ± 0.8%), whereas the IRV is much larger than the OAR volume by: 20.3 ± 8.6% (heart), 24.0 ± 8.6% (liver), and 47.6 ± 20.2% (stomach). The Jaccard index is higher in navigator‐triggered than bellows‐rebinned 4DMRI by 4% (P