Predicting pathological complete response in rectal cancer after chemoradiotherapy with a random forest using 18F-fluorodeoxyglucose positron emission tomography and computed tomography radiomics

Abstract

Background: Neoadjuvant chemoradiotherapy (NCRT) followed by surgery is the standard treatment for patients with locally advanced rectal cancer. This study developed a random forest (RF) model to predict pathological complete response (pCR) based on radiomics derived from baseline 18 F-fluorodeoxyglucose ([ 18 F] FDG)-positron emission tomography (PET)/computed tomography (CT).Methods: This study included 169 patients with newly diagnosed rectal cancer. All patients received 18 F[FDG]-PET/CT, NCRT, and surgery. In total, 68 radiomic features were extracted from the metabolic tumor volume. The numbers of splits in a decision tree and trees in an RF were determined based on their effects on predictive performance. Receiver operating characteristic curve analysis was performed to evaluate predictive performance and ascertain the optimal threshold for maximizing prediction accuracy.Results: After NCRT, 22 patients (13%) achieved pCR, and 42 features that could differentiate tumors with pCR were used to construct the RF model. Six decision trees and seven splits were suitable. Accordingly, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 81.8%, 97.3%, 81.8%, 97.3%, and 95.3%, respectively. Conclusions: By using an RF, we determined that radiomics derived from baseline 18 F[FDG]-PET/CT could accurately predict pCR in patients with rectal cancer. Highly accurate and predictive values can be achieved but should be externally validated.

A Systematic Review of PET Textural Analysis and Radiomics in Cancer

“…We collected 47 publications on “Other” cancer types. This category included studies on gastrointestinal (11) [ 257 , 258 , 259 , 260 , 261 , 262 , 263 , 264 , 265 , 266 , 267 ], pancreatic (8) [ 268 , 269 , 270 , 271 , 272 , 273 , 274 , 275 ], sarcoma (8) [ 276 , 277 , 278 , 279 , 280 , 281 , 282 , 283 ], neuroendocrine (5) [ 284 , 285 , 286 , 287 , 288 ], prostate (4) [ 289 , 290 , 291 , 292 ], thyroid (3) [ 293 , 294 , 295 ], thymic (2) [ 296 , 297 ], skin (2) [ 298 , 299 ], liver (2) [ 300 , 301 ], and renal carcinomas (1) [ 302 ]. The average number of patients was 84 (median = 70, range, 26–214) and the average number of textural features extracted was 29 (median = 17, ...…”

Development and validation of magnetic resonance imaging- based radiomics models for preoperative prediction of microsatellite instability in rectal cancer

“…This method has recently gained attention as a promising tool for predicting prognosis and guiding clinical decision-making (17,18). Recent studies on rectal cancer have provided encouraging evidence that radiomics can be applied to predict characteristics including therapeutic responses (19,20), lymph node metastasis (21,22), and KRAS mutations (23,24). Three recent studies, including all left-and right-sided colorectal cancers, reported that radiomics analysis based on computed tomography (CT) data could predict MSI in colorectal cancer (25)(26)(27); however, MSI prevalence in colorectal cancer is not evenly distributed, and it gradually varies from the proximal colon to the rectum (28).…”

Novel biomarkers and prediction model for the pathological complete response to neoadjuvant treatment of triple-negative breast cancer

“…To facilitate the introduction of a practical prediction model, we conducted analyses on the basis of random forest analysis and Lasso analysis to retrieve the intersected proportion for dimensionally reduction and features selection, which has been widely used for characteristics of cancer diagnosis and therapy [24][25][26]. After this group of genes identified, the respective prediction values for pCR were successively evaluated by logistic analysis, and the six-gene panel comprising ATP4B, FBXO22, FCN2, RRP8, SMERK2, TET3 were finally recognized.…”