Quantification of the spatial distribution of primary tumors in the lung to develop new prognostic biomarkers for locally advanced NSCLC.

The anatomical location and extent of primary lung tumors have shown prognostic value for overall survival (OS). However, its manual assessment is prone to interobserver variability. This study aims to use data driven identification of image characteristics for OS in locally advanced non-small cell lung cancer (NSCLC) patients. Five stage IIIA/IIIB NSCLC patient cohorts were retrospectively collected. Patients were treated either with radiochemotherapy (RCT): RCT1* (n = 107), RCT2 (n = 95), RCT3 (n = 37) or with surgery combined with radiotherapy or chemotherapy: S1* (n = 135), S2 (n = 55). Based on a deformable image registration (MIM Vista, 6.9.2.), an in-house developed software transferred each primary tumor to the CT scan of a reference patient while maintaining the original tumor shape. A frequency-weighted cumulative status map was created for both exploratory cohorts (indicated with an asterisk), where the spatial extent of the tumor was uni-labeled with 2 years OS. For the exploratory cohorts, a permutation test with random assignment of patient status was performed to identify regions with statistically significant worse OS, referred to as decreased survival areas (DSA). The minimal Euclidean distance between primary tumor to DSA was extracted from the independent cohorts (negative distance in case of overlap). To account for the tumor volume, the distance was scaled with the radius of the volume-equivalent sphere. For the S1 cohort, DSA were located at the right main bronchus whereas for the RCT1 cohort they further extended in cranio-caudal direction. In the independent cohorts, the model based on distance to DSA achieved performance: AUCRCT2 [95% CI] = 0.67 [0.55-0.78] and AUCRCT3 = 0.59 [0.39-0.79] for RCT patients, but showed bad performance for surgery cohort (AUCS2 = 0.52 [0.30-0.74]). Shorter distance to DSA was associated with worse outcome (p = 0.0074). In conclusion, this explanatory analysis quantifies the value of primary tumor location for OS prediction based on cumulative status maps. Shorter distance of primary tumor to a high-risk region was associated with worse prognosis in the RCT cohort.

Preselection of robust radiomic features does not improve outcome modelling in non-small cell lung cancer based on clinical routine FDG-PET imaging.

BACKGROUND: Radiomics is a promising tool for identifying imaging-based biomarkers. Radiomics-based models are often trained on single-institution datasets; however, multi-centre imaging datasets are preferred for external generalizability owing to the influence of inter-institutional scanning differences and acquisition settings. The study aim was to determine the value of preselection of robust radiomic features in routine clinical positron emission tomography (PET) images to predict clinical outcomes in locally advanced non-small cell lung cancer (NSCLC). METHODS: A total of 1404 primary tumour radiomic features were extracted from pre-treatment [18F]fluorodeoxyglucose (FDG)-PET scans of stage IIIA/N2 or IIIB NSCLC patients using a training cohort (n = 79; prospective Swiss multi-centre randomized phase III trial SAKK 16/00; 16 centres) and an internal validation cohort (n = 31; single centre). Robustness studies investigating delineation variation, attenuation correction and motion were performed (intraclass correlation coefficient threshold > 0.9). Two 12-/24-month event-free survival (EFS) and overall survival (OS) logistic regression models were trained using standardized imaging: (1) with robust features alone and (2) with all available features. Models were then validated using fivefold cross-validation, and validation on a separate single-centre dataset. Model performance was assessed using area under the receiver operating characteristic curve (AUC). RESULTS: Robustness studies identified 179 stable features (13%), with 25% stable features for 3D versus 4D acquisition, 31% for attenuation correction and 78% for delineation. Univariable analysis found no significant robust features predicting 12-/24-month EFS and 12-month OS (p value > 0.076). Prognostic models without robust preselection performed well for 12-month EFS in training (AUC = 0.73) and validation (AUC = 0.74). Patient stratification into two risk groups based on 12-month EFS was significant for training (p value = 0.02) and validation cohorts (p value = 0.03). CONCLUSIONS: A PET-based radiomics model using a standardized, multi-centre dataset to predict EFS in locally advanced NSCLC was successfully established and validated with good performance. Prediction models with robust feature preselection were unsuccessful, indicating the need for a standardized imaging protocol.

Impact of CT convolution kernel on robustness of radiomic features for different lung diseases and tissue types.

OBJECTIVES: In this study, we aimed to assess the impact of different CT reconstruction kernels on the stability of radiomic features and the transferability between different diseases and tissue types. Three lung diseases were evaluated, i.e. non-small cell lung cancer (NSCLC), malignant pleural mesothelioma (MPM) and interstitial lung disease related to systemic sclerosis (SSc-ILD) as well as four different tissue types, i.e. primary tumor, largest involved lymph node ipsilateral and contralateral lung. METHODS: Pre-treatment non-contrast enhanced CT scans from 23 NSCLC, 10 MPM and 12 SSc-ILD patients were collected retrospectively. For each patient, CT scans were reconstructed using smooth and sharp kernel in filtered back projection. The regions of interest (ROIs) were contoured on the smooth kernel-based CT and transferred to the sharp kernel-based CT. The voxels were resized to the largest voxel dimension of each cohort. In total, 1386 features were analyzed. Feature stability was assessed using the intraclass correlation coefficient. Features above the stability threshold >0.9 were considered stable. RESULTS: We observed a strong impact of the reconstruction method on stability of the features (at maximum 26% of the 1386 features were stable). Intensity features were the most stable followed by texture and wavelet features. The wavelet features showed a positive correlation between percentage of stable features and size of the ROI (R2 = 0.79, p = 0.005). Lymph node radiomics showed poorest stability (<10%) and lung radiomics the largest stability (26%). Robustness analysis done on the contralateral lung could to a large extent be transferred to the ipsilateral lung, and the overlap of stable lung features between different lung diseases was more than 50%. However, results of robustness studies cannot be transferred between tissue types, which was investigated in NSCLC and MPM patients; the overlap of stable features for lymph node and lung, as well as for primary tumor and lymph node was very small in both disease types. CONCLUSION: The robustness of radiomic features is strongly affected by different reconstruction kernels. The effect is largely influenced by the tissue type and less by the disease type. ADVANCES IN KNOWLEDGE: The study presents to our knowledge the most complete analysis on the impact of convolution kernel on the robustness of CT-based radiomics for four relevant tissue types in three different lung diseases. .

Radiomics Feature Activation Maps as a New Tool for Signature Interpretability.

INTRODUCTION: In the field of personalized medicine, radiomics has shown its potential to support treatment decisions. However, the limited feature interpretability hampers its introduction into the clinics. Here, we propose a new methodology to create radiomics feature activation maps, which allows to identify the spatial-anatomical locations responsible for signature activation based on local radiomics. The feasibility of this technique will be studied for histological subtype differentiation (adenocarcinoma versus squamous cell carcinoma) in non-small cell lung cancer (NSCLC) using computed tomography (CT) radiomics. MATERIALS AND METHODS: Pre-treatment CT scans were collected from a multi-centric Swiss trial (training, n=73, IIIA/N2 NSCLC, SAKK 16/00) and an independent cohort (validation, n=32, IIIA/N2/IIIB NSCLC). Based on the gross tumor volume (GTV), four peritumoral region of interests (ROI) were defined: lung_exterior (expansion into the lung), iso_exterior (expansion into lung and soft tissue), gradient (GTV border region), GTV+Rim (GTV and iso_exterior). For each ROI, 154 radiomic features were extracted using an in-house developed software implementation (Z-Rad, Python v2.7.14). Features robust against delineation variability served as an input for a multivariate logistic regression analysis. Model performance was quantified using the area under the receiver operating characteristic curve (AUC) and verified using five-fold cross validation and internal validation. Local radiomic features were extracted from the GTV+Rim ROI using non-overlapping 3x3x3 voxel patches previously marked as GTV or rim. A binary activation map was created for each patient using the median global feature value from the training. The ratios of activated/non-activated patches of GTV and rim regions were compared between histological subtypes (Wilcoxon test). RESULTS: Iso_exterior, gradient, GTV+Rim showed good performances for histological subtype prediction (AUCtraining=0.68-0.72 and AUCvalidation=0.73-0.74) whereas GTV and lung_exterior models failed validation. GTV+Rim model feature activation maps showed that local texture feature distribution differed significantly between histological subtypes in the rim (p=0.0481) but not in the GTV (p=0.461). CONCLUSION: In this exploratory study, radiomics-based prediction of NSCLC histological subtypes was predominantly based on the peritumoral region indicating that radiomics activation maps can be useful for tracing back the spatial location of regions responsible for signature activation.

Comparison of robust to standardized CT radiomics models to predict overall survival for non-small cell lung cancer patients.

BACKGROUND: Radiomics is a promising tool for the identification of new prognostic biomarkers. Radiomic features can be affected by different scanning protocols, often present in retrospective and prospective clinical data. We compared a computed tomography (CT) radiomics model based on a large but highly heterogeneous multicentric image dataset with robust feature pre-selection to a model based on a smaller but standardized image dataset without pre-selection. MATERIALS AND METHODS: Primary tumor radiomics was extracted from pre-treatment CTs of IIIA/N2/IIIB NSCLC patients from a prospective Swiss multicentric randomized trial (npatient  = 124, ninstitution  = 14, SAKK 16/00) and a validation dataset (npatient  = 31, ninstitution  = 1). Four robustness studies investigating inter-observer delineation variation, motion, convolution kernel, and contrast were conducted to identify robust features using an intraclass correlation coefficient threshold >0.9. Two 12-months overall survival (OS) logistic regression models were trained: (a) on the entire multicentric heterogeneous dataset but with robust feature pre-selection (MCR) and (b) on a smaller standardized subset using all features (STD). Both models were validated on the validation dataset acquired with similar reconstruction parameters as the STD dataset. The model performances were compared using the DeLong test. RESULTS: In total, 113 stable features were identified (nshape  = 8, nintensity  = 0, ntexture  = 7, nwavelet  = 98). The convolution kernel had the strongest influence on the feature robustness (<20% stable features). The final models of MCR and STD consisted of one and two features respectively. Both features of the STD model were identified as non-robust. MCR did not show performance significantly different from STD on the validation cohort (AUC [95%CI] = 0.72 [0.48-0.95] and 0.79 [0.63-0.95], p = 0.59). CONCLUSION: Prognostic OS CT radiomics model for NSCLC based on a heterogeneous multicentric imaging dataset with robust feature pre-selection performed equally well as a model on a standardized dataset.

Recurrence Patterns and Long-term Results After Induction Chemotherapy, Chemoradiotherapy, and Curative Surgery in Patients With Locally Advanced Esophageal Cancer.

OBJECTIVE: The long-term follow up data of 2 prospective phase II trials is reported (NCT00072033, NCT00445861), which investigated neoadjuvant chemoradiation followed by surgery in patients with esophageal carcinoma. Postoperative complications as well as prognostic factors and patterns of relapse during long-term observation are shown. SUMMARY OF BACKGROUND DATA: Long-term follow-up is often missing in the complex setting of multimodal treatments of esophageal carcinoma; this leads to rather undifferentiated follow-up guidelines for this tumor entity. METHODS: In the first trial, patients received induction chemotherapy followed by chemoradiation and surgery. In the second trial, cetuximab was added to the same neoadjuvant treatment concomitant with induction chemotherapy and chemoradiation. RESULTS: Eighty-two patients underwent surgery; the median follow-up time was 6.8 and 6.4 years, respectively. Fifty-five percent were diagnosed with adenocarcinoma, 80% clinically node-positive, 68% received transthoracic esophagectomy, and 32% transhiatal or transmediastinal resection. Five patients died postoperatively in-hospital due to complications (6%). The median overall survival was 4.3 years, and the median event-free survival was 2.7 years. Patients with adenocarcinoma rarely relapsed after a 3-year event-free survival. Whereas patients with residual tumor cells after neoadjuvant therapy primarily experienced relapse within the first 2 postoperative years, this in contrast to several patients with complete remission who also experienced late relapses 4 years after surgery. CONCLUSION: After curative surgery in a multimodal setting, the histological type and the response to neoadjuvant therapy predicted the time frame of relapse; this knowledge may influence further follow-up guidelines for esophageal carcinoma.

Multimodal Treatment in Operable Stage III NSCLC: A Pooled Analysis on Long-Term Results of Three SAKK trials (SAKK 16/96, 16/00, and 16/01).

INTRODUCTION: Long-term data on outcomes of operable stage III NSCLC are scarce. METHODS: Individual patient data from 368 patients enrolled in one phase III and two phase II trials were pooled and outcomes after applying the eighth (denoted with an asterisk [*]) versus the sixth TNM staging edition were compared. Patients were treated with either preoperative radiotherapy following 3 cycles of induction chemotherapy (trimodal) or neoadjuvant chemotherapy alone (bimodal). RESULTS: With the sixth version, the 5- and 10-year survival rates were 38% and 28% for stage IIIA, respectively, and 36% and 24% for stage IIIB, respectively. Factors associated with improved 5-year overall survival were younger age, R0 resection, and pathologic complete remission (pCR) (p = 0.043, p < 0.001 and p = 0.009). With the eighth TNM staging version, 162 patients were moved from stage IIIA to IIIB*. The 5- and 10-year survival rates were 41% and 29% for stage IIIA*, respectively, and 35% and 27% for stage IIIB* patients, respectively. There was no difference in the bi- versus trimodal group with regard to median overall survival (28 months [95% confidence interval (CI): 21-39 months] and 37 months [95% CI: 24-51 months], p = 0.9) and event-free survival (12 months [95% CI: 9-15 months] versus 13 months [95% CI: 10-22 months], p = 0.71). CONCLUSIONS: We showed favorable 10-year survival rates of 29% and 27% in stage IIIA* and IIIB*, respectively. Younger age, R0 resection, and pathologic complete response were associated with improved long-term survival. Outcomes using the sixth versus eighth edition of the TNM classification were similar in operable stage III NSCLC.

Induction chemoradiation in stage IIIA/N2 non-small-cell lung cancer: a phase 3 randomised trial.

BACKGROUND: One of the standard options in the treatment of stage IIIA/N2 non-small-cell lung cancer is neoadjuvant chemotherapy and surgery. We did a randomised trial to investigate whether the addition of neoadjuvant radiotherapy improves outcomes. METHODS: We enrolled patients in 23 centres in Switzerland, Germany and Serbia. Eligible patients had pathologically proven, stage IIIA/N2 non-small-cell lung cancer and were randomly assigned to treatment groups in a 1:1 ratio. Those in the chemoradiotherapy group received three cycles of neoadjuvant chemotherapy (100 mg/m(2) cisplatin and 85 mg/m(2) docetaxel) followed by radiotherapy with 44 Gy in 22 fractions over 3 weeks, and those in the control group received neoadjuvant chemotherapy alone. All patients were scheduled to undergo surgery. Randomisation was stratified by centre, mediastinal bulk (less than 5 cm vs 5 cm or more), and weight loss (5% or more vs less than 5% in the previous 6 months). The primary endpoint was event-free survival. Analyses were done by intention to treat. This trial is registered with ClinicalTrials.gov, number NCT00030771. FINDINGS: From 2001 to 2012, 232 patients were enrolled, of whom 117 were allocated to the chemoradiotherapy group and 115 to the chemotherapy group. Median event-free survival was similar in the two groups at 12·8 months (95% CI 9·7-22·9) in the chemoradiotherapy group and 11·6 months (8·4-15·2) in the chemotherapy group (p=0·67). Median overall survival was 37·1 months (95% CI 22·6-50·0) with radiotherapy, compared with 26·2 months (19·9-52·1) in the control group. Chemotherapy-related toxic effects were reported in most patients, but 91% of patients completed three cycles of chemotherapy. Radiotherapy-induced grade 3 dysphagia was seen in seven (7%) patients. Three patients died in the control group within 30 days after surgery. INTERPRETATION: Radiotherapy did not add any benefit to induction chemotherapy followed by surgery. We suggest that one definitive local treatment modality combined with neoadjuvant chemotherapy is adequate to treat resectable stage IIIA/N2 non-small-cell lung cancer. FUNDING: Swiss State Secretariat for Education, Research and Innovation (SERI), Swiss Cancer League, and Sanofi.

Intense therapy in patients with locally advanced esophageal cancer beyond hope for surgical cure: a prospective, multicenter phase II trial of the Swiss Group for Clinical Cancer Research (SAKK 76/02).

BACKGROUND: There is no standard treatment for patients with locally advanced esophageal carcinoma without systemic metastasis in whom surgery is no longer considered a reasonable option. PATIENTS AND METHODS: Patients with cervical esophageal tumors, locally very advanced stage (T4 and/or M1a) or locally advanced (T3 and/or N+) with comorbidities were included. THERAPY: 2 cycles of induction chemotherapy (cisplatin and docetaxel, both 75 mg/m(2) 3-weekly) followed by chemoradiation therapy (CRT) comprising a total radiation dose of 59.4 Gy together with docetaxel 15 mg/m(2) and cisplatin 25 mg/m(2) (5 weekly doses). Primary endpoint: Histologically proven freedom from local failure 6 months after CRT completion. RESULTS: 21 patients were included: 12 had locally very advanced tumors, 3 had cervical esophagus tumors, and 6 were medically unfit for surgery. 18 patients completed therapy per protocol. Grade 3/4 toxicities during CRT were thrombopenia (10%) and dysphagia (15%). 1 patient died due to herpes simplex infection. The primary endpoint was achieved by 4 patients, 6 were alive after median follow-up of 34 months, and median survival was 16 months. Most patients experienced lasting improvement of dysphagia following induction chemotherapy. CONCLUSIONS: This regimen is feasible, showed clinically meaningful, long-lasting improvements in quality of life and resulted in long-term survival in 29% of the patients.

Neoadjuvant chemotherapy and radiotherapy followed by surgery in selected patients with stage IIIB non-small-cell lung cancer: a multicentre phase II trial.

BACKGROUND: Stage IIIB non-small-cell lung cancer (NSCLC) is usually thought to be unresectable, and is managed with chemotherapy with or without radiotherapy. However, selected patients might benefit from surgical resection after neoadjuvant chemotherapy and radiotherapy. The aim of this multicentre, phase II trial was to assess the efficacy and toxicity of a neoadjuvant chemotherapy and radiotherapy followed by surgery in patients with technically operable stage IIIB NSCLC. METHODS: Between September, 2001, and May, 2006, patients with pathologically proven and technically resectable stage IIIB NSCLC were sequentially treated with three cycles of neoadjuvant chemotherapy (cisplatin with docetaxel), immediately followed by accelerated concomitant boost radiotherapy (44 Gy in 22 fractions) and definitive surgery. The primary endpoint was event-free survival at 12 months. Efficacy analyses were done by intention to treat. This trial is registered with ClinicalTrials.gov, number NCT00030810. FINDINGS: 46 patients were enrolled, with a median age of 60 years (range 28-70). 13 (28%) patients had N3 disease, 36 (78%) had T4 disease. All patients received chemotherapy; 35 (76%) patients received radiotherapy. The main toxicities during chemotherapy were neutropenia (25 patients [54%] at grade 3 or 4) and febrile neutropenia (nine [20%]); the main toxicity after radiotherapy was oesophagitis (ten patients [29%]; nine grade 2, one grade 3). 35 patients (76%) underwent surgery, with pneumonectomy in 17 patients. A complete (R0) resection was achieved in 27 patients. Peri-operative complications occurred in 14 patients, including two deaths (30-day mortality 5.7%). Seven patients required a second surgical intervention. Pathological mediastinal downstaging was seen in 11 of the 28 patients who had lymph-node involvement at enrolment, a complete pathological response was seen in six patients. Event-free survival at 12 months was 54% (95% CI 39-67). After a median follow-up of 58 months, the median overall survival was 29 months (95% CI 16.1-NA), with survival at 1, 3, and 5 years of 67% (95% CI 52-79), 47% (32-61), and 40% (24-55). INTERPRETATION: A treatment strategy of neoadjuvant chemotherapy and radiotherapy followed by surgery is feasible in selected patients. Toxicity is considerable, but manageable. Survival compares favourably with historical results of combined treatment for less advanced stage IIIA disease. FUNDING: Swiss Group for Clinical Cancer Research (SAKK) and an unrestricted educational grant by Sanofi-Aventis (Switzerland).