PET/Computed Tomography Scanning and Precision Medicine: Esophageal Cancer

Preoperative prediction of clinical and pathological stages for patients with esophageal cancer using PET/CT radiomics

BACKGROUND

Preoperative stratification is critical for the management of patients with esophageal cancer (EC). To investigate the feasibility and accuracy of PET-CT-based radiomics in preoperative prediction of clinical and pathological stages for patients with EC.

METHODS

Histologically confirmed 100 EC patients with preoperative PET-CT images were enrolled retrospectively and randomly divided into training and validation cohorts at a ratio of 7:3. The maximum relevance minimum redundancy (mRMR) was applied to select optimal radiomics features from PET, CT, and fused PET-CT images, respectively. Logistic regression (LR) was applied to classify the T stage (T1,2 vs. T3,4), lymph node metastasis (LNM) (LNM(-) vs. LNM(+)), and pathological state (pstage) (I-II vs. III-IV) with features from CT (CT_LR_Score), PET (PET_LR_Score), fused PET/CT (Fused_LR_Score), and combined CT and PET features (CT + PET_LR_Score), respectively.

RESULTS

Seven, 10, and 7 CT features; 7, 8, and 7 PET features; and 3, 6, and 3 fused PET/CT features were selected using mRMR for the prediction of T stage, LNM, and pstage, respectively. The area under curves (AUCs) for T stage, LNM, and pstage prediction in the validation cohorts were 0.846, 0.756, 0.665, and 0.815; 0.769, 0.760, 0.665, and 0.824; and 0.727, 0.785, 0.689, and 0.837 for models of CT_LR_Score, PET_ LR_Score, Fused_ LR_Score, and CT + PET_ LR_Score, respectively.

CONCLUSIONS

Accurate prediction ability was observed with combined PET and CT radiomics in the prediction of T stage, LNM, and pstage for EC patients.

CRITICAL RELEVANCE STATEMENT

PET/CT radiomics is feasible and promising to stratify stages for esophageal cancer preoperatively.

KEY POINTS

• PET-CT radiomics achieved the best performance for Node and pathological stage prediction. • CT radiomics achieved the best AUC for T stage prediction. • PET-CT radiomics is feasible and promising to stratify stages for EC preoperatively.

Acceptance and disparities of PET/CT use in patients with esophageal or gastro-esophageal junction cancer: Evaluation of mature registry data

Background/rationale

PET/CT plays a crucial role in esophageal (EC) and gastroesophageal junction cancer (GEJ) diagnosis and management. Despite endorsement in clinical guidelines, variation in acceptance of PET/CT exists. The aim of this study was to assess the early use of PET/CT among EC and GEJ patients in a regionalized setting and identify factors contributing to disparity in access.

Materials and methods

Retrospective cohort study of adults with EC or GEJ between 2012 and 2014 from the Population Registry of Esophageal and Stomach Tumours of Ontario and Ontario Health (Cancer Care Ontario). Receipt of PET/CT and relevant demographics were collected, and statistical analysis performed. Continuous data were analysed with t-tests and Wilcoxon rank sum test. Categorical data were analysed with chi-square test. Kaplan-Meier methods were used to estimate median survival.

Results

Fifty-five percent of patients diagnosed with EC or GEJ between 2012 and 2014 received PET/CT (1321/2390). Eighty-four percent of patients underwent surgical resection (729/870), and 80% receiving radical treatment (496/622) underwent PET/CT. The use of PET/CT increased from 2012 to 2014. Male patients received more PET/CT than females (85% vs.78% p < 0.001).Median survival for the overall cohort was 11.1 months, 17.2 vs. 5.2 months among those who did and did not receive PET/CT and 35 vs. 27 months among the surgical cohort (p = 0.16).

Conclusions

We found that PET/CT use increased from 2012 to 2014 and that the majority of EC/GEJ patients being considered for curative therapy received PET/CT. There were also gender disparities identified. PET/CT appears to confer a potential survival benefit in our study, although our assessment is limited. Our findings may serve as learned lessons for other new imaging modalities, new indications for PET/CT or even for the introduction of new radiopharmaceuticals for PET/CT.

Nomogram for Predicting Occult Locally Advanced Esophageal Squamous Cell Carcinoma Before Surgery

Introduction

The limitations of preoperative examination result in locally advanced esophageal squamous cell carcinoma (ESCC) often going undetected preoperatively. This study aimed to develop a clinical tool for identifying patients at high risk for occult locally advanced ESCC; the tool can be supplemented with preoperative examination to improve the reliability of preoperative staging.

Materials and Methods

Data of 598 patients who underwent radical resection of ESCC from 2010 to 2017 were analyzed. Logistic multivariate analysis was used to develop a nomogram. The training cohort included patients who underwent surgery during an earlier period (n = 426), and the validation cohort included those who underwent surgery thereafter (n = 172), to confirm the model’s performance. Nomogram discrimination and calibration were evaluated using Harrell's concordance index (C-index) and calibration plots, respectively.

Results

Logistic multivariate analysis suggested that higher preoperative carcinoembryonic antigen levels (>2.43, odds ratio [OR]: 2.093; 95% confidence interval [CI], 1.233–2.554; P = 0.006), presence of preoperative symptoms (OR: 2.737; 95% CI, 1.194–6.277; P = 0.017), presence of lymph node enlargement (OR: 2.100; 95% CI, 1.243–3.550; P = 0.006), and advanced gross aspect (OR: 13.103; 95% CI, 7.689–23.330; P < 0.001) were independent predictors of occult locally advanced ESCC. Based on these predictive factors, a nomogram was developed. The C-indices of the training and validation cohorts were 0.827 and 0.897, respectively, indicating that the model had a good predictive performance. To evaluate the accuracy of the model, we divided patients into high-risk and low-risk groups according to their nomogram scores, and a comparison was made with histopathological data.

Conclusion

The nomogram achieved a good preoperative prediction of occult locally advanced ESCC; it can be used to make rational therapeutic choices.

18F-FDG PET/CT as predictive and prognostic factor in esophageal cancer treated with combined modality treatment

Purpose

[18F] fluorodeoxyglucose positron emission tomography/computed tomography ([18F] FDG-PET/CT) is used for diagnosis, staging, response assessment and prognosis prediction in different tumors, but its role in esophageal cancer is still debated. The aim of this study was to evaluate the role of semiquantitative baseline PET parameters as possible prognostic and predictive factors in a series of esophageal carcinomas treated with combined modalities.

Methods

43 patients with esophageal carcinoma were treated with chemoradiotherapy (CRT) followed by surgery in 20 cases and underwent pre-treatment 18F-FDG-PET/CT. Semiquantitative PET parameters were evaluated including Standardized Uptake Value (SUVmax e SUVmean), Metabolic Tumor Volume (MTV) and Total Lesion Glycolysis (TLG) with isocontour of 41 and 50%. Further variables analyzed were gender, primary tumor site, histological type, use of surgery, achievement of a radical resection and the type of chemotherapy regimen. The correlation of all variables with treatment response, loco-regional control (LR), Overall survival (OS) and Disease-Free Survival (DFS) was evaluated.

Results

SUVmax, SUVmean50 and SUVmean41 were significantly higher in node-positive cases and in squamous cell carcinomas. With respect to prognostic factors, MTV was found to be correlated with OS: patients with MTV41 < 11.32 cm³ and MTV50 < 8.07 cm³ (both p values = 0.04) showed better 3-year OS rates (33 vs. 20%). Further factors predicting a better prognosis were the use of surgery and radical resection (R0) (both p values < 0.01).

Conclusions

Pre-treatment MTV values were significant prognostic factors for OS, together with the use of surgery and R0 resection in esophageal cancers treated with multimodal therapies.

Development of targeted therapy of NRF2high esophageal squamous cell carcinoma

Esophageal squamous cell carcinoma (ESCC) is a deadly disease and one of the most aggressive cancers of the gastrointestinal tract. As a master transcription factor regulating the stress response, NRF2 is often mutated and becomes hyperactive, and thus causes chemo-radioresistance and poor survival in human ESCC. There is a great need to develop NRF2 inhibitors for targeted therapy of NRF2high ESCC. In this review, we mainly focus on three aspects, NRF2 inhibitors and their mechanisms of action, screening novel drug targets, and evaluation of NRF2 activity in the esophagus. A research strategy has been proposed to develop NRF2 inhibitors using human ESCC cells and mouse models.

Surgical Management of Early Esophageal Cancer

Esophageal cancer is the eighth most common cancer worldwide, and its incidence has been increasing over the past several decades. Esophagectomy currently is the standard of care for more advanced early esophageal cancer and should be performed at centers of excellence with high volumes, appropriate supportive staff, and multidisciplinary expertise.

Predicting aggressive histopathological features in esophageal cancer with positron emission tomography using a deep convolutional neural network

Background

The presence of lymphovascular invasion (LVI) and perineural invasion (PNI) are of great prognostic importance in esophageal squamous cell carcinoma. Currently, positron emission tomography (PET) scans are the only means of functional assessment prior to treatment. We aimed to predict the presence of LVI and PNI in esophageal squamous cell carcinoma using PET imaging data by training a three-dimensional convolution neural network (3D-CNN).

Methods

Seven hundred and ninety-eight PET scans of patients with esophageal squamous cell carcinoma and 309 PET scans of patients with stage I lung cancer were collected. In the first part of this study, we built a 3D-CNN based on a residual network, ResNet, for a task to classify the scans into esophageal cancer or lung cancer. In the second stage, we collected the PET scans of 278 patients undergoing esophagectomy for a task to classify and predict the presence of LVI/PNI.

Results

In the first part, the model performance attained an area under the receiver operating characteristic curve (AUC) of 0.860. In the second part, we randomly split 80%, 10%, and 10% of our dataset into training set, validation set and testing set, respectively, for a task to classify the scans into the presence of LVI/PNI and evaluated the model performance on the testing set. Our 3D-CNN model attained an AUC of 0.668 in the testing set, which shows a better discriminative ability than random guessing.

Conclusions

A 3D-CNN can be trained, using PET imaging datasets, to predict LNV/PNI in esophageal cancer with acceptable accuracy.

Hypofractionated Radiotherapy in Combination With Chemotherapy Improves Outcome of Patients With Esophageal Carcinoma Tracheoesophageal Groove Lymph Node Metastasis

This study investigated the efficiency and safety of hypofractionated radiotherapy (HFR) combined with paclitaxel chemotherapy for the treatment of postsurgery tracheoesophageal groove lymph node (TGLN) metastasis in patients with esophageal cancer (EC). Fifty-three EC patients with TGLN metastasis after surgery admitted to the Yancheng Third People's Hospital from January 2013 to June 2015 were included in this study. They were randomly divided into the HFR group (n = 25) and conventional fractioned radiotherapy (CFR) group (n = 28) based on the random grouping method. Patients in the HFR group received treatment with radiation of 60 Gy (5 fractions per week, total 20 fractions) combined with paclitaxel chemotherapy at a dose of 50 mg once per week for 4 weeks. Patients in the CFR group received radiation of 60 Gy (5 fractions per week, total 30 fractions) combined with paclitaxel chemotherapy at a dose of 50 mg once per week for 6 weeks. The adverse events and treatment outcomes in these two groups were analyzed. It was found that there was no significant difference in the incidence of radiation esophagitis in the HFR group compared with that of the CFR group (grades 3-4, 44.0 vs. 25.0%; P = 0.149). There was no statistical difference in the incidence of radiation pneumonitis between these two groups (grades 3-4, 16.0 vs. 7.1%; P = 0.314). No statistical difference was noticed in complete response (CR), partial response (PR), and no response (NR) between these two groups. The median overall survival (OS) in the HRF group was significantly longer compared with that of the CRF group (24.2 months (95% CI, 16.2-32.1 months) vs. 11.8 months (95% CI, 9.2-14.4 months); P = 0.024). Our results indicated that the combination of HFR and chemotherapy improved the prognosis of EC patients with TGLN metastasis with no increased adverse events.

Deep Convolutional Neural Network-Based Positron Emission Tomography Analysis Predicts Esophageal Cancer Outcome

In esophageal cancer, few prediction tools can be confidently used in current clinical practice. We developed a deep convolutional neural network (CNN) with 798 positron emission tomography (PET) scans of esophageal squamous cell carcinoma and 309 PET scans of stage I lung cancer. In the first stage, we pretrained a 3D-CNN with all PET scans for a task to classify the scans into esophageal cancer or lung cancer. Overall, 548 of 798 PET scans of esophageal cancer patients were included in the second stage with an aim to classify patients who expired within or survived more than one year after diagnosis. The area under the receiver operating characteristic curve (AUC) was used to evaluate model performance. In the pretrain model, the deep CNN attained an AUC of 0.738 in identifying patients who expired within one year after diagnosis. In the survival analysis, patients who were predicted to be expired but were alive at one year after diagnosis had a 5-year survival rate of 32.6%, which was significantly worse than the 5-year survival rate of the patients who were predicted to survive and were alive at one year after diagnosis (50.5%, p < 0.001). These results suggest that the prediction model could identify tumors with more aggressive behavior. In the multivariable analysis, the prediction result remained an independent prognostic factor (hazard ratio: 2.830; 95% confidence interval: 2.252–3.555, p < 0.001). We conclude that a 3D-CNN can be trained with PET image datasets to predict esophageal cancer outcome with acceptable accuracy.

Radiomics in esophageal and gastric cancer

Esophageal, esophago-gastric, and gastric cancers are major causes of cancer morbidity and cancer death. For patients with potentially resectable disease, multi-modality treatment is recommended as it provides the best chance of survival. However, quality of life may be adversely affected by therapy, and with a wide variation in outcome despite multi-modality therapy, there is a clear need to improve patient stratification. Radiomic approaches provide an opportunity to improve tumor phenotyping. In this review we assess the evidence to date and discuss how these approaches could improve outcome in esophageal, esophago-gastric, and gastric cancer.

Impact of hybrid FDG-PET/CT on gross tumor volume definition of cervical esophageal cancer: reducing interobserver variation

Intensity-modulated radiation therapy is being increasingly used to treat cervical esophageal cancer (CEC); however, delineating the gross tumor volume (GTV) accurately is essential for its successful treatment. The use of computed tomography (CT) images to determine the GTV produces a large degree of interobserver variation. In this study, we evaluated whether the use of [18F]-fluoro-2-deoxy-D-glucose positron emission tomography (FDG-PET)/CT fused images reduced interobserver variation, compared with CT images alone, to determine the GTV in patients with CEC. FDG-PET/CT scans were obtained for 10 patients with CEC, imaged positioned on a flat tabletop with a pillow. Five radiation oncologists independently defined the GTV for the primary tumors using routine clinical data; they contoured the GTV based on CT images (GTVCT), followed by contouring based on FDG-PET/CT fused images (GTVPET/CT). To determine the geometric observer variation, we calculated the conformality index (CI) from the ratio of the intersection of the GTVs to their union. The interobserver CI was compared using Wilcoxon's signed rank test. The mean (±SD) interobserver CIs of GTVCT and GTVPET/CT were 0.39 ± 0.15 and 0.58 ± 0.10, respectively (P = 0.005). Our results suggested that FDG-PET/CT images reduced interobserver variation when determining the GTV in patients with CEC. FDG-PET/CT may increase the consistency of the radiographically determined GTV in patients with CEC.