Integrating Anatomical, Molecular and Clinical Risk Factors in Gastrointestinal Stromal Tumor of the Stomach

Preoperative CT-based radiomics and deep learning model for predicting risk stratification of gastric gastrointestinal stromal tumors

BACKGROUND

Gastrointestinal stromal tumors (GISTs) are clinically heterogeneous with various malignant potential in different individuals. It is crucial to explore a reliable method for preoperative risk stratification of gastric GISTs noninvasively.

PURPOSE

To establish and evaluate a machine learning model using the combination of computed tomography (CT) morphology, radiomics, and deep learning features to predict the risk stratification of primary gastric GISTs preoperatively.

METHODS

The 193 gastric GISTs lesions were randomly divided into training set, validation set, and test set in a ratio of 6:2:2. The qualitative and quantitative CT morphological features were assessed by two radiologists. The tumors were segmented manually, and then radiomic features were extracted using PyRadiomics and the deep learning features were extracted using pre-trained Resnet50 from arterial phase and venous phase CT images, respectively. Pearson correlation analysis and recursive feature elimination were used for feature selection. Support vector machines were employed to build a classifier for predicting the risk stratification of GISTs. This study compared the performance of models using different pre-trained convolutional neural networks (CNNs) to extract deep features for classification, as well as the performance of modeling features from single-phase and dual-phase images. The arterial phase, venous phase and dual-phase machine learning models were built, respectively, and the morphological features were added to the dual-phase machine learning model to construct a combined model. Receiver operating characteristic (ROC) curves were used to evaluate the efficacy of each model. The clinical application value of the combined model was determined through the decision curve analysis (DCA) and the net reclassification index (NRI) was analyzed.

RESULTS

The area under the curve (AUC) of the dual-phase machine learning model was 0.876, which was higher than that of the arterial phase model or venous phase model (0.813, 0.838, respectively). The combined model had best predictive performance than the above models with an AUC of 0.941 (95% CI: 0.887-0.974) (p = 0.012, Delong test). DCA demonstrated that the combined model had good clinical application value with an NRI of 0.575 (95% CI: 0.357-0.891).

CONCLUSION

In this study, we established a combined model that incorporated dual-phase morphology, radiomics, and deep learning characteristics, which can be used to predict the preoperative risk stratification of gastric GISTs.

Impact of tumour rupture risk on the oncological rationale for the surgical treatment choice of gastrointestinal stromal tumours

Tumour rupture of gastrointestinal stromal tumours (GISTs) has been considered to be a remarkable risk factor because of its unfavourable impact on the oncological outcome. Although tumour rupture has not yet been included in the current tumor-node-metastasis classification of GISTs as a prognostic factor, it may change the natural history of a low-risk GIST to a high-risk GIST. Originally, tumour rupture was defined as the spillage or fracture of a tumour into a body cavity, but recently, new definitions have been proposed. These definitions distinguished from the prognostic point of view between the major defects of tumour integrity, which are considered tumour rupture, and the minor defects of tumour integrity, which are not considered tumour rupture. Moreover, it has been demonstrated that the risk of disease recurrence in R1 patients is largely modulated by the presence of tumour rupture. Therefore, after excluding tumour rupture, R1 may not be an unfavourable prognostic factor for GISTs. Additionally, after the standard adjuvant treatment of imatinib for GIST with rupture, a high recurrence rate persists. This review highlights the prognostic value of tumour rupture in GISTs and emphasizes the need to carefully take into account and minimize the risk of tumour rupture when choosing surgical strategies for GISTs.

Prediction of the mitotic index and preoperative risk stratification of gastrointestinal stromal tumors with CT radiomic features

OBJECTIVE

The objective is to develop a mitotic prediction model and preoperative risk stratification nomogram for gastrointestinal stromal tumor (GIST) based on computed tomography (CT) radiomic features.

METHODS

A total of 267 GIST patients from 2009.07 to 2015.09 were retrospectively collected and randomly divided into (6:4) training cohort and validation cohort. The 2D-tumor region of interest was delineated from the portal-phase images on contrast-enhanced (CE)-CT, and radiomic features were extracted. Lasso regression method was used to select valuable features to establish a radiomic model for predicting mitotic index in GIST. Finally, the nomogram of preoperative risk stratification was constructed by combining the radiomic features and clinical risk factors.

RESULTS

Four radiomic features closely related to the level of mitosis were obtained, and a mitotic radiomic model was constructed. The area under the curve (AUC) of the radiomics signature model used to predict mitotic levels in training and validation cohorts (training cohort AUC = 0.752; 95% confidence interval [95%CI] 0.674-0.829; validation cohort AUC = 0.764; 95% CI 0.667-0.862). Finally, the preoperative risk stratification nomogram combining radiomic features was equivalent to the clinically recognized gold standard AUC (0.965 vs. 0.983) (p = 0.117). The Cox regression analysis found that the nomogram score was one of the independent risk factors for the long-term prognosis of the patients.

CONCLUSION

Preoperative CT radiomic features can effectively predict the level of mitosis in GIST, and combined with preoperative tumor size, accurate preoperative risk stratification can be performed to guide clinical decision-making and individualized treatment.

C-kit-negative transmural gastrointestinal stromal tumor in the stomach: A rare case of upper GI bleeding in Ghana

Upper GI bleeding from GIST is rare in Africa. The C-kit-negative variant is uncommon worldwide. We report a case of a 42-year-old woman managed for upper GI bleeding secondary to stomach GIST, negative for CD117 but positive for DOG 1 and CD 34. The 1-year postsurgery review is uneventful.

Genomic Complexity as a Biomarker to De-Escalate Adjuvant Imatinib Treatment in High-Risk Gastrointestinal Stromal Tumor

PURPOSE

Adjuvant imatinib treatment is recommended for patients with localized gastrointestinal stromal tumor (GIST) at high risk of recurrence. Almost half of high-risk patients are cured by surgery alone, indicating a need for improved selection of patients for adjuvant therapy. The aim of this study was to investigate if genomic tumor complexity could be used as a prognostic biomarker.

METHODS

The discovery cohort consisted of patients who underwent resection of primary GIST at Oslo University Hospital between 1998 and 2020. Karyotypes were categorized as simple if they had ≤ 5 chromosomal changes and complex if there were > 5 chromosomal aberrations. Validation was performed in an independent patient cohort where chromosomal imbalances were mapped using comparative genomic hybridization.

RESULTS

Chromosomal aberrations were detected in 206 tumors, of which 76 had a complex karyotype. The most frequently observed changes were losses at 14q, 22q, 1p, and 15q. The 5-year recurrence-free survival (RFS) in patients classified as very low, low, or intermediate risk was 99%. High-risk patients with a simple tumor karyotype had an estimated 5-year RFS of 94%, and patients with a complex karyotype had an estimated 5-year RFS of 51%. A complex karyotype was associated with poor RFS in patients with and without adjuvant imatinib treatment and in multivariable analysis adjusted for tumor site, size, mitotic count, and rupture. The prognostic impact of genomic complexity was confirmed in the validation cohort. In both cohorts, the 5-year disease-specific survival was > 90% for high-risk patients with genomically simple tumors.

CONCLUSION

Genomic tumor complexity is an independent prognostic biomarker in localized, high-risk GIST. Recurrences were infrequent for tumors with simple karyotypes. De-escalation of adjuvant imatinib treatment should be explored in patients with cytogenetically simple GISTs.

Prognostic significance of serum CA125 in the overall management for patients with gastrointestinal stromal tumors

BACKGROUND

Carbohydrate antigen 125 (CA125) is elevated as a tumor marker in many carcinomas, but its association with gastrointestinal stromal tumor (GIST) has received less attention. This study intends to evaluate whether CA125 level can predict tumor progression and overall survival (OS) of GIST patients.

METHODS

We retrospectively analyzed the clinical data and follow-up records of GIST patients who underwent surgical resection in Nanjing Drum Tower Hospital from August 2010 to December 2020. All patients were classified according to serum CA125 level. The relationship between CA125 and clinical outcomes was then examined.

RESULTS

A total of 406 GIST patients were enrolled in this study, among which 46 patients had preoperative elevated serum CA125 level and 13 patients with high CA125 level both preoperative and postoperative were observed. Preoperative CA125 concentration was significantly related to rupture status, resection style, tumor site, tumor size, mitotic index, NIH risk grade and c-kit exons. According to Kaplan-Meier curve analysis, high expression of postoperative CA125 was significantly correlated with worse progression-free survival (PFS) and OS among patients with preoperative elevated CA125 level. Ultimately, Cox proportional regression model analysis revealed that increase of preoperative and concurrent postoperative CA125 concentration was an independent predictive factor for PFS.

CONCLUSIONS

The concurrent abnormality of serum CA125 before and after operation was an independent risk factor for GIST progression, suggesting its significance as a serum biomarker in the overall management of GIST patients.

Prognostic Nomogram for Gastrointestinal Stromal Tumors after Surgery Based on the SEER Database

We aimed to determine prognostic factors and develop an effective and practical nomogram for predicting cancer-specific survival in gastrointestinal stromal tumor (GIST) patients. Postoperative data were obtained from the SEER database (2000-2018). Patients were divided into training and validation cohorts at random (7 : 3). Prognostic factors were screened, and a prognostic nomogram was established using log-rank testing and Cox regression. We used DCA, ROC curves, C-index, and calibration curves to evaluate our model's predictive performance. The clinical value of the nomogram and the modified National Institute of Health (M-NIH) classification were compared using the NRI and IDI. The Kaplan-Meier method was applied to examine survival by risk group, and log-rank tests were applied to compare variations in survival curves. Independent prognostic risk factors associated with cancer-specific survival on multivariate Cox proportional hazards regression analysis were age, race, and tumor location, size, grade, and stage. Clinically relevant variables need to be considered in addition to statistically significant variables when developing prognostic models to aid clinical decision-making. We included two additional variables (mitotic rate and chemotherapy) when constructing the prognostic model. The C-index was 0.766 (95% confidence interval (CI): 0.737-0.794) in the training cohort and 0.795 (95% CI: 0.754-0.836) in the internal validation group suggesting robustness. The areas under the ROC curve for three-year and five-year survival were >0.700, indicating satisfactory discrimination. The calibration curves showed good agreement between the predictions of the nomogram and the actual results. The NRI (0.346 for 3-year and 0.265 for 5-year cancer-specific survival for patients with GIST (GSS) prediction; validation cohort: 0.356 for 3-year and 0.246 for 5-year GSS prediction) and IDI values (0.047 for 3-year and 0.060 for 5-year GSS prediction; validation cohort: 0.071 for 3-year and 0.084 for 5-year GSS prediction) suggested that the established nomogram performed significantly better than the M-NIH classification. The DCA indicated that the nomogram was clinically useful and had a high discriminative ability in identifying patients who were at high risk of poor outcomes. According to nomogram findings, patients were divided into three groups (high, moderate, and low risk), with significantly different prognoses in both cohorts. Our nomogram satisfactorily predicted survival in postsurgical GIST patients, which may assist clinicians to evaluate the postoperative status and guide subsequent treatments.

Cytogenetic and molecular analyses of 291 gastrointestinal stromal tumors: site-specific cytogenetic evolution as evidence of pathogenetic heterogeneity

Gastrointestinal stromal tumor (GIST) is a mesenchymal neoplasm with variable behavior. An increased understanding of the tumor pathogenesis may improve clinical decision-making. Our aim was to obtain more data about the overall chromosome aberrations and intratumor cytogenetic heterogeneity in GIST. We analyzed 306 GIST samples from 291 patients using G-banding, direct sequencing, and statistics. Clonal chromosome aberrations were found in 81% of samples, with 34% of 226 primary tumors demonstrating extensive cytogenetic heterogeneity. 135 tumors had simple (≤5 changes) and 91 had complex (>5 changes) karyotypes. The karyotypically complex tumors more often were non-gastric (P < 0.001), larger (P < 0.001), more mitotically active (P = 0.009) and had a higher risk of rupture (P < 0.001) and recurrence (P < 0.001). Significant differences between gastric and non-gastric tumors were found also in the frequency of main chromosome losses: of 14q (79% vs. 63%), 22q (38% vs. 67%), 1p (23% vs. 88%), and 15q (18% vs. 77%). Gastric PDGFRA-mutated tumors, compared with gastric KIT-mutated, had a lower incidence of 22q losses (18% vs. 43%) but a higher rate of 1p losses (42% vs. 22%). The present, largest by far karyotypic study of GISTs provides further evidence for the existence of variable pathogenetic pathways operating in these tumors’ development.

Relationship Between Nutritional Status and Clinical Outcome in Patients With Gastrointestinal Stromal Tumor After Surgical Resection

Background

Currently, gastrointestinal stromal tumors (GIST) are the most common mesenchymal tumors in the gastrointestinal tract, and surgical resection is the main treatment. Malnutrition after gastrointestinal surgery is not uncommon, which may have adverse effects on postoperative recovery and prognosis. However, the nutritional status of GIST patients after surgical resection and its impact on clinical outcomes have received less attention. Therefore, the aim of this study was to dynamically evaluate the nutritional status of GIST patients undergoing surgical resection, and to analyze the correlation between nutritional status and clinical outcomes.

Methods

We retrospectively analyzed the clinical data of GIST patients who underwent surgical resection in the Fourth Hospital of Hebei Medical University from January 2016 to January 2020. Nutritional risk screening 2002 (NRS2002) and Patient-Generated Subjective Global Assessment (PG-SGA) were used to assess the nutritional status of all patients at admission and discharge, and the correlation between nutritional risk and clinical outcomes was analyzed.

Results

A total of 413 GIST patients were included in this study, among which 114 patients had malnutrition risk at admission (NRS2002 score ≥ 3), and 65 patients had malnutrition (PG-SGA score ≥ 4). The malnutrition risk rate (27.60 vs. 46.73%, p &lt; 0.001) and malnutrition incidence (15.73 vs. 37.29%, p &lt; 0.001) at admission were lower than those at discharge. Compared with the laboratory results at admission, the albumin, prealbumin, and total protein of the patients at discharge were significantly lower (all p &lt; 0.05). And there was a negative correlation between PG-SGA and clinical outcome (all p &lt; 0.05).

Conclusion

The nutritional status of GIST patients after surgical resection at discharge was worse than that at admission, and malnutrition is an important risk factor leading to poor clinical outcomes.