Search

PurposeTo establish and validate a machine learning based radiomics model for detection of perineural invasion (PNI) in gastric cancer (GC).MethodsThis retrospective study included a total of 955 patients with GC selected from two centers; they were separated into training (n=603), internal testing (n=259), and external testing (n=93) sets. Radiomic features were derived from three phases of contrast-enhanced computed tomography (CECT) scan images. Seven machine learning (ML) algorithms including least absolute shrinkage and selection operator (LASSO), naïve Bayes (NB), k-nearest neighbor (KNN), decision tree (DT), logistic regression (LR), random forest (RF), eXtreme gradient boosting (XGBoost) and support vector machine (SVM) were trained for development of optimal radiomics signature. A combined model was constructed by aggregating the radiomic signatures and important clinicopathological characteristics. The predictive ability of the radiomic model was then assessed with receiver operating characteristic (ROC) and calibration curve analyses in all three sets.ResultsThe PNI rates for the training, internal testing, and external testing sets were 22.1, 22.8, and 36.6%, respectively. LASSO algorithm was selected for signature establishment. The radiomics signature, consisting of 8 robust features, revealed good discrimination accuracy for the PNI in all three sets (training set: AUC = 0.86; internal testing set: AUC = 0.82; external testing set: AUC = 0.78). The risk of PNI was significantly associated with higher radiomics scores. A combined model that integrated radiomics and T stage demonstrated enhanced accuracy and excellent calibration in all three sets (training set: AUC = 0.89; internal testing set: AUC = 0.84; external testing set: AUC = 0.82).ConclusionThe suggested radiomics model exhibited satisfactory prediction performance for the PNI in GC.

BackgroundVisceral sarcomas are a rare form of soft tissue sarcoma. This study aimed to evaluate the survival and prognostic factors and effective treatments for visceral sarcomas.MethodsAll patients with visceral sarcoma referred to our center between January 2010 and December 2021 were retrospectively analyzed. The Kaplan-Meier method and a log-rank test were used for survival analysis.ResultsA total of 53 patients with visceral sarcoma were analyzed in this study with the median age at diagnosis of 57 (range, 24-77) years. Among them, 37 (69.8%) and 16 (30.2%) patients had localized and metastatic diseases at the initial presentation, respectively, and 44 patients underwent surgical resection. The median follow-up, event-free survival (EFS) and overall survival (OS) were 63.0 (range, 2-130), 42.0 months (95% confidence interval [CI] 10.879-73.121) and 45.0 months (95% CI 9.938-80.062), respectively. The 5-year EFS and OS rates were 44% and 46%, respectively. Univariate analysis of prognostic indicators illustrated that metastasis at presentation, surgery, surgical margin and the types of surgery were significantly associated with OS and EFS. In this study, combined chemotherapy or radiotherapy had no effects on EFS and OS.ConclusionPrimary visceral sarcoma is an uncommon and aggressive malignant tumor with a higher rate of local recurrence. In the largest cohort of visceral sarcomas in China to date, we identified metastases at presentation, surgery, surgical margin, and the types of surgery as independent predictors of survival. The combination of chemotherapy and radiotherapy did not affect survival.

BackgroundTumorigenesis and progression are intimately associated with inflammation. However, the inflammatory landscape in soft tissue sarcoma (STS) and its clinical consequences are yet unknown, and more investigation is needed.MethodsRNA-seq expression data for STS and corresponding normal tissues were downloaded from The Cancer Genome Atlas database and the Genotype-Tissue Expression Portal. Differential and prognostic analyses were performed based on known inflammatory response genes from Gene Set Enrichment Analysis (GSEA). We utilized LASSO-Cox analysis to determine hub genes and built an inflammatory score (INFscore) and risk stratification model. Furthermore, a nomogram, including the risk stratification model, was established to predict the prognosis. We further elucidated the characteristics among different risk STS patients by GSEA, gene set variation analysis, and detailed immune infiltration analysis. Finally, the INFscore and risk stratification model in predicting prognosis and depicting immune microenvironment status were verified by pan-cancer analysis.ResultsFive hub genes (HAS2, IL1R1, NMI, SERPINE1, and TACR1) were identified and were used to develop the INFscore. The risk stratification model distinguished the immune microenvironment status and evaluated the efficacy of immunotherapy and chemotherapy in STS. The novel nomogram had good efficacy in predicting the prognosis of STS patients. Finally, a pan-cancer investigation verified the association of INFscore with prognosis and immunity.ConclusionsAccording to the present study, the risk stratification model can be used to evaluate STS prognosis, tumor microenvironment status, immunotherapy, and chemotherapy efficacy. The novel nomogram has an excellent predictive value. Thus, the INFscore and risk stratification model has potential value in assessing the prognosis and immune status of multiple malignancies.

BackgroundThe rarity and complexity of soft tissue sarcoma (STS) make it a challenge to determine the incidence, survival, and metastasis rates. In addition, the clinicopathological risk factors for lymph node metastasis have rarely been reported.MethodsData on patients diagnosed with STS in the SEER database from 2000 to 2018 were extracted by SEER*Stat 8.3.9.1, and the incidence trend was calculated by Joinpoint 4.9 software. The KM method was used to calculate the survival curve, and the log-rank method was used to compare differences in the survival curves. The clinicopathological risk factors for lymph node metastasis were screened by logistic regression.ResultsAmong the 35987 patients, 4299 patients (11.9%) had distant metastasis. The overall lymph node metastasis rate was 6.02%, which included patients suffering from both lymph node and distant metastasis. Considering that some lymph node metastases might be accompanying events of distant metastasis, the rate of only lymph node metastasis in STS patients decreased to 3.42% after excluding patients with distant metastasis. Patients with only lymph node metastases (N1/2M0) had a significantly worse prognosis than those without metastases (N0M0) but a better prognosis than those with only distant metastases (N0M1) (p

PurposeOur previously study showed that recombinant human endostatin (Endostar) combined with chemotherapy had significant activity to increase the mPFS in patients with advanced sarcomas with tolerable side effects. However, the small cohort size and short follow-up time made it difficult to screen sensitive sarcoma subtypes and determine whether there is an overall survival benefit. With the largest sarcoma cohort to our knowledge, we try to confirm the efficacy and safety of chemotherapy combined with Endostar in stage IV sarcomas, with the specific purpose of finding out the sensitive sarcoma types for this combined treatment.MethodsAfter the exclusion of ineligible patients, 156 patients with stage IV bone and soft tissue sarcomas were included in this study according to the inclusion criteria.ResultsBy the end of follow-up, the ORR was 10.7% (9/84) vs 1.4% (1/72) (p=0.041), the DCR was 26.2% (22/84) vs 5.6% (4/72) (p=0.001) in the combined group and chemotherapy group, respectively. The mPFS of combined group was significantly longer than the chemotherapy group (10.42 vs 6.87 months, p=0.003). The mOS were 26.84 months and 23.56 months, without significant difference (p= 0.481). In osteogenic sarcoma, there was no statistically significant difference in the mPFS between the two groups (p=0.59), while in the soft tissue sarcoma, the mPFS in the combined group was significantly higher than that of the chemotherapy group (11.27 vs 8.05 months, p=0.004). Specifically, undifferentiated polymorphic sarcoma (UPS) was the possible sarcoma subtypes that benefited from the combined therapy. For the 38 UPS patients (28 patients in the combined group and 10 patients in the chemotherapy group), the mPFS in the combined group was up to 14.88 months, while it was only 7.1 months in the chemotherapy group, with a significant difference (p=0.006). The most common adverse events in the combined group were myelosuppression, gastrointestinal reactions and abnormal liver function, without significant difference in two groups.ConclusionChemotherapy plus Endostar could prolong mPFS and improve ORR and DCR in patients with stage IV soft tissue sarcoma, suggesting that the combined therapy could improve the patient prognosis in soft tissue sarcomas, especially the UPS patients.

BackgroundSurgery is an important treatment option for desmoid tumor (DT) patients, but how to decrease and predict the high recurrence rate remains a major challenge.MethodsDesmoid tumor patients diagnosed and treated at Tianjin Cancer Institute & Hospital were included, and a web-based nomogram was constructed by screening the recurrence-related risk factors using Cox regression analysis. External validation was conducted with data from the Fudan University Shanghai Cancer Center.ResultsA total of 385 patients were identified. Finally, after excluding patients without surgery, patients who were lost to follow-up, and patients without complete resection, a total of 267 patients were included in the nomogram construction. Among these patients, 53 experienced recurrence, with a recurrence rate of 19.85%. The 3-year and 5-year recurrence-free survival (RFS) rates were 82.5% and 78%, respectively. Age, tumor diameter, admission status, location, and tumor number were correlated with recurrence in univariate Cox analysis. In multivariate Cox analysis, only age, tumor diameter and tumor number were independent risk factors for recurrence and were then used to construct a web-based nomogram to predict recurrence. The concordance index (C-index) of the nomogram was 0.718, and the areas under the curves (AUCs) of the 3-year and 5-year receiver operating characteristic (ROC) curves were 0.751 and 0.761, respectively. In the external validation set, the C-index was 0.706, and the AUCs of the 3-year and 5-year ROC curves are 0.788 and 0.794, respectively.ConclusionsAge, tumor diameter, and tumor number were independent predictors of recurrence for DTs, and a web-based nomogram containing these three predictors could accurately predict RFS (https://stepforward.shinyapps.io/Desmoidtumor/).