Page 39 - ARNM-3-2

Advances in Radiotherapy
& Nuclear Medicine Radiomics for gastric cancer

Table 3. (Continued)
Authors Year Image Purpose Sample Features Result
Li et al. 71 2020 CT Prediction of DFS and 208 (Single center) Radiomics+clinical The C-index of the nomogram was
chemotherapy response in significantly higher than that of
stage II/III GC TNM stage (0.768 vs. 0.639) in the
validation cohort
Wang et al. 72 2020 CT To predict the HER2 460 (Single center) Radiomics+clinical The maximum diameter of tumor,
expression before radical differentiation degree, CT attenuate
gastrectomy in patients value in non-contrast, arterial, and
with gastric ADC portal phases, and corrected CT
attenuate value in the arterial phase
were the predictive factors of HER2
expression status
Wang et al. 73 2020 CT Prediction of HER2 status 132 (Single center) Radiomics The AP radiomic model had an AUC
in patients with GC of 0.830 for HER2-negative cases in
the test cohort
Binnewies et al. 74 2018 CT Prediction of GC 1,591 (Multi-center) Radiomics+clinical Incorporating the radiomic signature
survival and benefit of into the radiomic-based nomograms
chemotherapy resulted in better performance for the
estimation of DFS and OS than the
clinicopathological nomograms and
TNM staging system
Jiang et al. 75 2020 CT To develop a non-invasive 1,778 (Multi-center) Radiomics+clinical The radiomic signature is a
radiomic-based predictor reliable tool for the evaluation
of ImmunoScore for GC of immunoscore and retains the
prognostic significance in GC
Lin et al. 76 2022 CT Evaluation of 642 (Multi-center) Radiomics Three independent cohorts
the non-invasive showed good differentiation
immunosuppression of immunosuppressive states
scoring system and (AUC=0.840, 0.809, and 0.843,
treatment outcomes respectively)
based on radiological
characteristics
Gao et al. 77 2020 CT To develop and validate 165 (Multi-center) Radiomics The radiomic signature showed
a radiomic signature good ability for estimating tumor
for evaluation of tumor infiltrating regulatory T cells in the
infiltrating regulatory training, validation, and testing
T cells and outcome cohort, with AUC of 0.884, 0.869, and
prediction of GC patients 0.847, respectively
Li et al. 78 2022 PET-CT To assess the tumor 230 (Single center) Radiomics The radiomic tumor immune
immune microenvironment microenvironment score signature
and predict survival and was significantly associated with
chemotherapy benefits in DFS and OS both in the training and
patients with GC based on validation cohort (p<0.001)
imaging omics
Abbreviations: ADC: Adenocarcinoma; AP: Arterial phase; AUC: Area under the curve; CT: Computed tomography; DFS: Disease-free survival;
DR: Detection radiomics; GC: Gastric cancer; HER2: Human epidermal growth factor 2; NAC: Neoadjuvant chemotherapy; OS: Overall survival;
PET-CT: Positron emission tomography-computed tomography; PR: Prediction radiomics; TNM: Tumor-node-metastasis.
survival for locally advanced GC patients who underwent response, as shown in Table 3. 17,70-73 In a multicenter
curative resection without neoadjuvant therapy. retrospective study conducted by Jiang et al. using
17
Radiotherapy and chemotherapy are common cancer portal venous phase-enhanced CT of 1,591 patients, 19
treatment methods, which can inhibit the proliferation of stable radiomic features were extracted and selected to
cancer cells, effectively control the development of tumors, construct radiomic signature, which effectively improved
and reduce the probability of other complications. the accuracy of survival classification in the form of
69
Although there are few studies on radiotherapy response, nomogram (C-index: DFS = 0.850; OS = 0.860). In studies
we found four studies that predicted the chemotherapy conducted by Wang et al. and Wang et al., 72,73 they used

Volume 3 Issue 2 (2025) 31 doi: 10.36922/arnm.8350