Dataset Information

Prediction of post-radiotherapy locoregional progression in HPV-associated oropharyngeal squamous cell carcinoma using machine-learning analysis of baseline PET/CT radiomics.

ABSTRACT: Locoregional failure remains a therapeutic challenge in oropharyngeal squamous cell carcinoma (OPSCC). We aimed to devise novel objective imaging biomarkers for prediction of locoregional progression in HPV-associated OPSCC. Following manual lesion delineation, 1037 PET and 1037 CT radiomic features were extracted from each primary tumor and metastatic cervical lymph node on baseline PET/CT scans. Applying random forest machine-learning algorithms, we generated radiomic models for censoring-aware locoregional progression prognostication (evaluated by Harrell's C-index) and risk stratification (evaluated in Kaplan-Meier analysis). A total of 190 patients were included; an optimized model yielded a median (interquartile range) C-index of 0.76 (0.66-0.81; p?=?0.01) in prognostication of locoregional progression, using combined PET/CT radiomic features from primary tumors. Radiomics-based risk stratification reliably identified patients at risk for locoregional progression within 2-, 3-, 4-, and 5-year follow-up intervals, with log-rank p-values of p?=?0.003, p?=?0.001, p?=?0.02, p?=?0.006 in Kaplan-Meier analysis, respectively. Our results suggest PET/CT radiomic biomarkers can predict post-radiotherapy locoregional progression in HPV-associated OPSCC. Pending validation in large, independent cohorts, such objective biomarkers may improve patient selection for treatment de-intensification trials in this prognostically favorable OPSCC entity, and eventually facilitate personalized therapy.

SUBMITTER: Haider SP

PROVIDER: S-EPMC7568193 | biostudies-literature | 2020 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Prediction of post-radiotherapy locoregional progression in HPV-associated oropharyngeal squamous cell carcinoma using machine-learning analysis of baseline PET/CT radiomics.

Haider Stefan P SP Sharaf Kariem K Zeevi Tal T Baumeister Philipp P Reichel Christoph C Forghani Reza R Kann Benjamin H BH Petukhova Alexandra A Judson Benjamin L BL Prasad Manju L ML Liu Chi C Burtness Barbara B Mahajan Amit A Payabvash Seyedmehdi S

Translational oncology 20201016 1

Locoregional failure remains a therapeutic challenge in oropharyngeal squamous cell carcinoma (OPSCC). We aimed to devise novel objective imaging biomarkers for prediction of locoregional progression in HPV-associated OPSCC. Following manual lesion delineation, 1037 PET and 1037 CT radiomic features were extracted from each primary tumor and metastatic cervical lymph node on baseline PET/CT scans. Applying random forest machine-learning algorithms, we generated radiomic models for censoring-awar ...[more]

PMID: 33075658

Similar Datasets

Project description:PurposeTo develop and validate a clinico-biological features and 18F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) radiomic-based nomogram via machine learning for the pretherapy prediction of discriminating between adenocarcinoma (ADC) and squamous cell carcinoma (SCC) in non-small cell lung cancer (NSCLC).MethodsA total of 315 NSCLC patients confirmed by postoperative pathology between January 2017 and June 2019 were retrospectively analyzed and randomly divided into the training (n = 220) and validation (n = 95) sets. Preoperative clinical factors, serum tumor markers, and PET, and CT radiomic features were analyzed. Prediction models were developed using the least absolute shrinkage and selection operator (LASSO) regression analysis. The performance of the models was evaluated and compared by the area under receiver-operator characteristic (ROC) curve (AUC) and DeLong test. The clinical utility of the models was determined via decision curve analysis (DCA). Then, a nomogram was developed based on the model with the best predictive efficiency and clinical utility and was validated using the calibration plots.ResultsIn total, 122 SCC and 193 ADC patients were enrolled in this study. Four independent prediction models were separately developed to differentiate SCC from ADC using clinical factors-tumor markers, PET radiomics, CT radiomics, and their combination. The DeLong test and DCA showed that the Combined Model, consisting of 2 clinical factors, 2 tumor markers, 7 PET radiomics, and 3 CT radiomic parameters, held the highest predictive efficiency and clinical utility in predicting the NSCLC subtypes compared with the use of these parameters alone in both the training and validation sets (AUCs (95% CIs) = 0.932 (0.900-0.964), 0.901 (0.840-0.957), respectively) (p < 0.05). A quantitative nomogram was subsequently constructed using the independently risk factors from the Combined Model. The calibration curves indicated a good consistency between the actual observations and nomogram predictions.ConclusionThis study presents an integrated clinico-biologico-radiological nomogram that can be accurately and noninvasively used for the individualized differentiation SCC from ADC in NSCLC, thereby assisting in clinical decision making for precision treatment.

Project description:Background and purposeChemoradiotherapy is the standard treatment for moderate and advanced oesophageal cancer. The aim of this study was to establish a predictive model based on enhanced computed tomography examination, and to evaluate its clinical value for detecting locoregional recurrence-free survival (LRFS) in cases of oesophageal squamous cell carcinoma after radiotherapy.Materials and methodsIn total, 218 patients with pathologically diagnosed oesophageal squamous cell carcinoma who received radical chemoradiotherapy from July 2016 to December 2017 were collected in this study. Patients were randomly divided into either a training group (n=153) or a validation group (n=65) in a 7:3 ratio. Clinical patient information was then recorded. The enhanced computed tomography scan images of the patients were imported into 3D-slicer software (version 4.8.1), and the radiomic features were extracted by the Python programme package. In the training group, the dimensionality reduction of the radiomic features was implemented by Lasso regression, and then a radiological label, the model of predicting LRFS, was established and evaluated. To achieve a better prediction performance, the radiological label was combined with clinical risk factor information to construct a radiomics nomogram. A receiver operating characteristic curve was used to evaluate the efficacy of different models. Calibration curves were used to assess the consistency between the predicted and observed recurrence risk, and the Hosmer-Lemeshow method was used to test model fitness. The C-index evaluated the discriminating ability of the prediction model. Decision curve analysis was used to determine the clinical value of the constructed prediction model.ResultsOf the 218 patients followed up in this study, 44 patients (28.8%) in the training group and 21 patients (32.3%) in the validation group experienced recurrence. There was no difference in LRFS between the two groups (χ2 = 0.525, P=0.405). Lasso regression was used in the training group to select six significant radiomic features. The radiological label established using these six features had a satisfactory prediction performance. The C-index was 0.716 (95% CI: 0.645-0.787) in the training group and 0.718 (95% CI: 0.612-0.825) in the validation group. The radiomics nomogram, which included the radiological label and clinical risk factors, achieved a better prediction than the radiological label alone. The C-index was 0.742 (95% CI: 0.674-0.810) in the training group and 0.715 (95% CI: 0.609-0.820) in the validation group. The results of the calibration curve and decision curve analyses indicated that the radiomics nomogram was superior in predicting LRFS of oesophageal carcinoma after radiotherapy.ConclusionsA radiological label was successfully established to predict the LRFS of oesophageal squamous cell carcinoma after radiotherapy. The radiomics nomogram was complementary to the clinical prognostic features and could improve the prediction of the LRFS after radiotherapy for oesophageal cancer.

Project description:The global rise of HPV(+) oropharyngeal squamous cell carcinoma (OPSCC) has generated considerable interest underlying its etiology and management. Despite an overall decline in head and neck malignancies, the incidence in OPSCC has by contrast sharply risen, with HPV(+) subtypes now comprising 80% of all OPSCC.1,2 Relative to HPV(-) OPSCC, HPV(+) patients are more responsive to chemoradiation and harbor a 52% risk-of-death reduction.3 Despite this distinct outcome, treatment regimens remain the same for all OPSCC subtypes (smoking-driven and virus-driven), rather than an adaptive approach to what most consider distinct diseases. The short-term effects (e.g., mucositis, odynophagia) and long-term toxicities (e.g., xerostomia, dysphagia, ototoxicity) from treatment substantially affect quality of life, and rival the impact of the cancer itself. Recently published as well as ongoing trials are actively examining deintensification approaches2,4-9, with the goal of diminishing treatment sequelae for HPV(+) subtypes. While deintensification may decrease chemoradiation-related toxicities, it nonetheless may also undertreat a meaningful percentage of HPV(+) patients who may then recur. In examining national trials, 19% of HPV(+) patients had disease progression after therapy, with a two-year survival rate of 60%.10 Current methods to ascertain HPV status involve p16 staining. However, on comparison with gold standard E6/E7 expression by qPCR, p16 harbored a 15% false positivity rate11, suggesting limited utility as a sole biomarker for deintensification. Improved molecular stratification would greatly enhance the clinician’s ability to precisely tailor treatment while minimizing the risk of jeopardizing outcomes. One approach encompasses in-depth proteomic profiling of HPV(+) OPSCC to reveal distinct protein expression profiles and delineate clinically relevant upstream pathways. In turn, these proteomic differences may distinguish higher-risk disease (cases predisposed to recurrence that may benefit from treatment intensification) from lower-risk phenotypes (cases whose treatment response is sufficiently robust to warrant deintensification). Here, two HPV(+) OPSCC cohorts stratified by treatment response are compared via a hybrid data dependent acquisition/data independent acquisition (DDA/DIA) approach via mass spectrometry. We focused on detection of low-abundance proteins to highlight proteomic signatures that can be potentially exploited for treatment stratification.

Dataset Information

Prediction of post-radiotherapy locoregional progression in HPV-associated oropharyngeal squamous cell carcinoma using machine-learning analysis of baseline PET/CT radiomics.

Publications

Prediction of post-radiotherapy locoregional progression in HPV-associated oropharyngeal squamous cell carcinoma using machine-learning analysis of baseline PET/CT radiomics.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets