Management of recurrent gastrointestinal stromal tumors

Different radiomics models in predicting the malignant potential of small intestinal stromal tumors

Objectives

To explore the feasibility of different radiomics models for predicting the malignant potential of small intestinal stromal tumors (SISTs), and to select the best radiomics model.

Methods

A retrospective analysis of 140 patients with SISTs was conducted. Radiomics features were extracted from CT-enhanced images. Support vector machine (SVM), Decision tree (DT), Conditional inference trees (CIT), Random Forest (RF), K-nearest neighbors (KNN), Back-propagation neural network (BPNet), and Bayes were used to construct different radiomics models. The clinical data and CT performance were selected using univariate analysis and to construct clinical model. Nomogram model was developed by combining clinical data and radiomics features. Model performances were assessed by using the area under the receiver operator characteristic (ROC) curve (AUC). The models' clinical values were assessed by decision curve analysis (DCA).

Results

A total of 1132 radiomics features were extracted. Among radiomics models, SVM was better than DT, CIT, RF, KNN, BPNet, Bayes because it had the highest AUC with a significant difference (P<0.05). The AUC of the clinical model was 0.781. The AUC of the radiomics model was 0.910. The AUC of nomogram model was 0.938. Clinical models had the lowest AUC. Nomogram AUC were slightly higher than radiomics model, but the difference was not significant (P=0.48). The DCA of the nomogram model and radiomics model showed optimal clinical efficacy.

Conclusions

The model constructed with SVM method was the best model for predicting the malignant potential of SISTs. Radiomics model and nomogram model showed high predictive value in predicting the malignant potential of SISTs.

Computed tomography-based radiomics model for discriminating the risk stratification of gastrointestinal stromal tumors

PURPOSE

The pathological risk degree of gastrointestinal stromal tumors (GISTs) has become an issue of great concern. Computed tomography (CT) is beneficial for showing adjacent tissues in detail and determining metastasis or recurrence of GISTs, but its function is still limited. Radiomics has recently shown a great potential in aiding clinical decision-making. The purpose of our study is to develop and validate CT-based radiomics models for GIST risk stratification.

METHODS

Three hundred and sixty-six patients clinically suspected of primary GISTs from January 2013 to February 2018 were retrospectively enrolled, among which data from 140 patients were eventually analyzed after exclusion. Data from patient CT images were partitioned based on the National Institutes of Health Consensus Classification, including tumor segmentation, radiomics feature extraction and selection. A radiomics model was then proposed and validated.

RESULTS

The radiomics signature demonstrated discriminative performance for advanced and nonadvanced GISTs with an area under the curve (AUC) of 0.935 [95% confidence interval (CI) 0.870-1.000] and an accuracy of 90.2% for validation cohort. The radiomics signature demonstrated favorable performance for the risk stratification of GISTs with an AUC of 0.809 (95% CI 0.777-0.841) and an accuracy of 67.5% for the validation cohort. Radiomics analysis could capture features of the four risk categories of GISTs. Meanwhile, this CT-based radiomics signature showed good diagnostic accuracy to distinguish between nonadvanced and advanced GISTs, as well as the four risk stratifications of GISTs.

CONCLUSION

Our findings highlight the potential of a quantitative radiomics analysis as a complementary tool to achieve an accurate diagnosis for GISTs.

Pattern Classification for Gastrointestinal Stromal Tumors by Integration of Radiomics and Deep Convolutional Features

Predicting malignant potential is one of the most critical components of a computer-aided diagnosis system for gastrointestinal stromal tumors (GISTs). These tumors have been studied only on the basis of subjective computed tomography findings. Among various methodologies, radiomics, and deep learning algorithms, specifically convolutional neural networks (CNNs), have recently been confirmed to achieve significant success by outperforming the state-of-the-art performance in medical image pattern classification and have rapidly become leading methodologies in this field. However, the existing methods generally use radiomics or deep convolutional features independently for pattern classification, which tend to take into account only global or local features, respectively. In this paper, we introduce and evaluate a hybrid structure that includes different features selected with radiomics model and CNNs and integrates these features to deal with GISTs classification. The Radiomics model and CNNs are constructed for global radiomics and local convolutional feature selection, respectively. Subsequently, we utilize distinct radiomics and deep convolutional features to perform pattern classification for GISTs. Specifically, we propose a new pooling strategy to assemble the deep convolutional features of 54 three-dimensional patches from the same case and integrate these features with the radiomics features for independent case, followed by random forest classifier. Our method can be extensively evaluated using multiple clinical datasets. The classification performance (area under the curve (AUC): 0.882; 95% confidence interval (CI): 0.816-0.947) consistently outperforms those of independent radiomics (AUC: 0.807; 95% CI: 0.724-0.892) and CNNs (AUC: 0.826; 95% CI: 0.795-0.856) approaches.

Evaluation of Clinical Plus Imaging Features and Multidetector Computed Tomography Texture Analysis in Preoperative Risk Grade Prediction of Small Bowel Gastrointestinal Stromal Tumors

OBJECTIVE

This study aimed to evaluate the prediction roles of clinical plus imaging features and multidetector computed tomography (MDCT) texture analysis in preoperative risk grade classification of small bowel (SB) gastrointestinal stromal tumors (GISTs).

METHODS

This study included 213 SB GIST patients. Clinical features and MDCT imaging findings were reviewed. Tumor risk stratifications were determined according to modified National Institutes of Health criteria. Random forest models were performed to evaluate the correlation of risk stratification.

RESULTS

The model of clinical plus imaging findings showed an area under receiver operating characteristic curve (AUC) of 92.0%. The AUC of texture analysis based on MDCT portal phase was 93.3%, without statistical difference from that of clinical plus imaging model (P = 0.378). The AUC of the model combined clinical plus imaging features and MDCT texture analysis was 94.3%, which was significantly higher than the AUC of clinical imaging model (P = 0.042).

CONCLUSION

Texture analysis may become an important comprehensive tool for preoperative risk stratification of SB GISTs.

Recurrent Gastrointestinal Stromal Tumors in the Imatinib Mesylate Era: Treatment Strategies for an Incurable Disease

Introduction

Recurrence of gastrointestinal stromal tumors (GISTs) after surgical resection and imatinib mesylate (IM) adjuvant therapy poses a significant treatment challenge. We present the case of a patient who underwent surgical resection after recurrence and review the current literature regarding treatment.

Case Presentation

A 58-year-old man with a large intra-abdominal jejunal GIST was treated with complete surgical resection followed by IM. The patient experienced disease recurrence 3.5 years later and underwent IM dose escalation and reresection.

Conclusion

Current strategies to treat recurrent GIST include dose escalation, modifying adjuvant tyrosine kinase inhibitor therapy, and surgery. High-level evidence will be required to better define the combinatory roles of tyrosine kinase inhibitor therapy, guided by molecular profiling, and surgery in the management of recurrent GIST.

Predictive features of CT for risk stratifications in patients with primary gastrointestinal stromal tumour

PURPOSE

To determine the predictive CT imaging features for risk stratifications in patients with primary gastrointestinal stromal tumours (GISTs).

MATERIALS AND METHODS

One hundred and twenty-nine patients with histologically confirmed primary GISTs (diameter >2 cm) were enrolled. CT imaging features were reviewed. Tumour risk stratifications were determined according to the 2008 NIH criteria where GISTs were classified into four categories according to the tumour size, location, mitosis count, and tumour rupture. The association between risk stratifications and CT features was analyzed using univariate analysis, followed by multinomial logistic regression and receiver operating characteristic (ROC) curve analysis.

RESULTS

CT imaging features including tumour margin, size, shape, tumour growth pattern, direct organ invasion, necrosis, enlarged vessels feeding or draining the mass (EVFDM), lymphadenopathy, and contrast enhancement pattern were associated with the risk stratifications, as determined by univariate analysis (P < 0.05). Only lesion size, growth pattern and EVFDM remained independent risk factors in multinomial logistic regression analysis (OR = 3.480-100.384). ROC curve analysis showed that the area under curve of the obtained multinomial logistic regression model was 0.806 (95 % CI: 0.727-0.885).

CONCLUSION

CT features including lesion size, tumour growth pattern, and EVFDM were predictors of the risk stratifications for GIST.

KEY POINTS

• CT features were of predictive value for risk stratification of GISTs. • Tumour size, growth patterns, and EVFDM were risk predictors of GISTs. • Large size, mixed growth pattern, or EVFDM indicated high risk GIST.

Treatment patterns, prescribing decision drivers, and predictors of complete response following disease recurrence in gastrointestinal stromal tumor patients-a chart extract-based approach

PURPOSE

The aim of this study was to investigate tumor characteristics, treatment patterns, and outcomes in recurrent KIT + GIST patients treated in a community practice setting.

METHODS

An online tool was used to retrieve data on 410 patients treated with adjuvant imatinib mesylate (IM) for primary resectable KIT + GIST, who discontinued, had a recurrence, and then restarted IM or initiated sunitinib. Tumor characteristics at recurrence, treatment patterns, and factors associated with post-recurrence complete response (CR) achievement were analyzed.

RESULTS

About 72.7 % of patients did not have surgery post-recurrence as majority of them had unresectable (45 %), metastatic (40 %), or multifocal tumors (62.4 %). Following recurrence, 76.6 % of patients were re-started on IM and 23.4 % on sunitinib; patients were 7.37 times more likely to re-start IM if initial treatment duration was ≤18 months (p < 0.001). Patients were also more likely to re-start IM if recurrence occurred >12 months post-discontinuation, or they had a recurrence inside the GI system, lower or unknown Fletcher risk score at primary diagnosis, or lower mitotic rate, (odds ratio (OR) = 3.54, p < 0.001; OR = 2.64, p = 0.006; OR = 2.55, p = 0.007; and OR = 2.45, p = 0.002, respectively). About 22.4 % achieved CR; patients were more likely to achieve CR if they had unifocal tumor at recurrence, inside the GI system, of ≤2 cm, or had lower mitotic rate (OR = 2.61, p < 0.001; OR = 2.27, p = 0.036; OR = 2.16, p = 0.023, OR = 1.87, p = 0.017, respectively).

CONCLUSIONS

IM treatment duration at primary diagnosis, time to develop recurrence after IM discontinuation, tumor location, and mitotic rate at recurrence were the main prescribing decision drivers. Tumor characteristics were the most important factor in achieving CR following c-KIT inhibitor retreatment.

[Gastrointestinal stromal tumor of the ileum with metastatic relapse developed in the mesentery]

Nous rapportons le cas d'un homme de 29 ans admis pour une masse hypogastrique très douloureuse. La laparotomie exploratrice réalisée en urgence permettait de mettre en évidence l'origine et la localisation iléale de la masse tumorale. Le traitement chirurgical consistait en une tumorectomie complète avec respect des marges carcinologiques. Les suites opératoires étaient simples et l'examen histologique confirmait la nature stromale de la tumeur iléale réséquée. Les marges de résection passaient en tissus sains. Le patient était perdu de vue et n'ayant reçu aucun traitement adjuvant. Un an plus tard, il est revenu pour ballonnement et masse pelviens d’évolution rapidement progressive et dont l'exploration chirurgicale révélait l'existence d'une tumeur mésentérique. Une tumorectomie était réalisée mais incomplète à cause d'une perte sanguine avec instabilité hémodynamique induite par l'exérèse tumorale. Après analyse histologique et immunohistochimique de la pièce opératoire, le diagnostic d'une tumeur stromale était confirmé. Il s'agissait d'une rechute métastatique à localisation mésentérique d'une tumeur stromale digestive d'origine iléale à fort potentiel malin. La réduction tumorale suivie d'un complément thérapeutique par l'Imatinib (glivec^®) permettaient d'obtenir un résultat satisfaisant. Avec un recul de 12 mois, le patient était asymptomatique et aucune récidive locale ni de métastase à distance n'a été observée.

Primary mesenteric gastrointestinal stromal tumour

Primary mesenteric gastrointestinal stromal tumours (GISTs) are rare tumours and can be included as a differential for an expanding intraabdominal mass. We present the case, in our institution, of a 72-year-old male who presented with non-specific symptoms and was diagnosed with a primary mesenteric GIST following resection. We report his follow-up and discuss the current theories as to the origins of these rare tumours and current treatment modalities.

Ten-year review of gastrointestinal stromal tumours at a tertiary referral hospital in New Zealand

BACKGROUND

Gastrointestinal stromal tumours (GISTs) are the most common mesenchymal tumours of the gastrointestinal tract and make up 1-2% of all gastrointestinal malignancies. Traditionally, the treatment of choice for primary disease is surgical resection; however, no single surgeon or institution gets extensive exposure to these patients so appropriate decision-making is difficult, particularly since the introduction of the tyrosine kinase inhibitor imatinib, which has become an important additional management tool.

METHOD

All patients were diagnosed and treated for GISTs in Christchurch Hospital (Christchurch, New Zealand) between 1 January 2000 and 31 December 2010. We maintain a prospective database of all patients with GISTs. Data on clinical and histopathological variables, management and survival outcomes were recorded. These were then reviewed.

RESULTS

There were 93 patients in this study. Fifty were women. Median age was 69 (interquartile range (IQR) 59-76) years. Fifty-one tumours were located in the stomach, 27 in the small bowel, six in the colon, three in the oesophagus, one in the rectum and five were extra-gastrointestinal. In total, 22 patients received imatinib therapy; four patients with metastatic disease had imatinib as sole therapy. The median follow-up was 58 (IQR 30-90) months. The 5-year overall survival and disease-free survival (DFS) for the entire study population was 69% and 64%, respectively. The 5-year DFS was higher for all patients who have localized disease when compared with those who have metastatic disease (76% versus 28%, P-value 0.001).

CONCLUSION

Surgery aiming at an R0 resection remains the mainstay of treatment. We propose the most effective way to grow the knowledge base in New Zealand is the establishment of a national register, thereby allowing better clinical decision-making by interpretation of a larger data set.

An unexpected but interesting response to a novel therapy for malignant extragastrointestinal stromal tumor of the mesoileum: a case report and review of the literature

BACKGROUND

Gastrointestinal stromal tumors (GISTs) are the most common mesenchymal tumors of the gastrointestinal tract. Extragastrointestinal stromal tumors (eGISTs) of the mesoileum are extremely rare and are usually treated with surgery combined with imatinib therapy.

CASE PRESENTATION

We present the case of a 43-year-old man who developed a large eGIST in the mesoileum. Abdominal/pelvic computed tomography revealed a large heterogeneous mass with cystic and solid components that measured 20.0 × 12.0 × 8.0 cm. Three cycles of neoadjuvant chemotherapy with epirubicin, cyclophosphamide and hydroxycamptothecin; en bloc resection; and three more cycles of adjuvant chemotherapy with the same regimen and drugs resulted in five years of disease-free survival without any symptoms.

CONCLUSIONS

Although imatinib treatment is usually chosen for eGISTs, resistance to imatinib remains a concern; these patients may receive neoadjuvant or adjuvant chemotherapy. In case of the former, further treatment, that is, surgery or adjuvant chemotherapy, depends on tumor response to the neoadjuvant chemotherapy. In addition, this treatment for eGIST is not only beneficial but also economical for patients compared with imatinib. A novel treatment approach that combined neoadjuvant chemotherapy, surgery and adjuvant chemotherapy resulted in long-term survival in our patient, thus showing promise as a potential therapy for eGISTs.

Secondary recurrent multiple EGIST of the mesentary: A case report and review of the literature

INTRODUCTION

Gastrointestinal stromal tumors (GISTs) are rare intra-abdominal tumors arising from mesenchymal stromal cells. EGISTs are mesenchymal tumors that originate outside the GI tract and tend to have similar characteristics to GISTs. To the best of our knowledge, few cases of long standing recurrent EGIST have been reported.

PRESENTATION OF CASE

We present the case of a rare recurrent EGIST in the mesentery of a 39 year old female patient. The tumor was symptomatic at the time of complaint and measured 8.4cm×7.7cm×7.6cm. Histological analysis revealed a spindled pattern with fusiform cells arranged in long fascicles and little atypia. Immunochemistry showed positivity for CD117 and was negative for CD34, S-100, Desmin, and MSA. B-catenin was weakly positive. A Ki-67 staining shows approximately 5% positivity revealing a low proliferative rate. The patient was doing well postoperatively and was discharged on 400mg imanitib regimen.

DISCUSSION

While GISTs are the most common tumors of the GI tract, recurrent EGISTs of the mesentery are extremely rare. Factors that indicate poor prognosis include tumor size greater than 5cm, mitotic rate greater than 1-5/10 HPF, presence of tumor necrosis or metastasis and most recently the c-kit mutation. Our patient had a very long time between recurrence of disease.

CONCLUSION

The current literature on EGISTs is limited. Our patient presents a very interesting case due to the time elapsed between disease recurrence and lack of metastasis or excessive growth.

Gastrointestinal stromal tumors

INTRODUCTION

The gastrointestinal stromal tumor (GIST) is the most common mesenchymal tumor of the intestinal tract, known to be refractory to conventional chemotherapy or radiation. Its pathogenesis is defined by mutations within the KIT and PDGFRA gene, which constitutively activate KIT and PDGFRA oncoproteins, and serve as crucial diagnostic and therapeutic targets.

DISCUSSION

Besides surgery, therapy with imatinib mesylate, which inhibits KIT kinase activity, represents the other cornerstone for the treatment of GIST. Still, the only curative option for GIST is given after complete surgical removal even in a metastatic setting, but recurrence is common, and the risk can be defined by surgical factors like incomplete resection, intraperitoneal rupture, or bleeding and tumor associated factors like tumor size, mitotic index, or localization.

CONCLUSION

Consequently, adjuvant therapy with imatinib mesylate or other tyrosine kinase inhibitors is recommended for high-risk patients after complete resection. For unresectable and advanced GIST, a partial response or stable disease can be achieved in about 80% of patients with imatinib mesylate.