Mechanism of Resistance in Gastrointestinal Stromal Tumors

Use of a Convolutional Neural Network to Predict the Malignant Potential of Gastrointestinal Stromal Tumors in Transabdominal Ultrasound Images: Visualization of the Focus of the Prediction Model

OBJECTIVE

We established a deep convolutional neural network (CNN) model based on ultrasound images (US-CNN) for predicting the malignant potential of gastrointestinal stromal tumors (GISTs).

METHODS

A total of 980 ultrasound images from 245 pathology-confirmed GIST patients after surgical operation were retrospectively collected and divided into a low (very-low-risk, low-risk) and a high (medium-risk, high-risk) malignant potential group. Eight pre-trained CNN models were used to extract the features. The CNN model with the highest accuracy in the test set was selected. The model's performance was evaluated by calculating accuracy, sensitivity, specificity, positive-predictive value (PPV), negative-predictive value (NPV) and the F1 score. Three radiologists with different experience levels also predicted the malignant potential of GISTs in the same test set. US-CNN and human assessments were compared. Subsequently, gradient-weighted class activation diagrams (Grad-CAMs) were used to visualize the model's final classification decisions.

RESULTS

Among the eight transfer learning-based CNNs, ResNet18 performed best. The accuracy, sensitivity, specificity, PPV, NPV and F1 score were 0.88, 0.86, 0.89, 0.82, 0.92 and 0.90, respectively, which were significantly better than those achieved by radiologists (resident doctor: 0.66, 0.55, 0.79, 0.74, 0.62 and 0.69; attending doctor: 0.68, 0.59, 0.78, 0.70, 0.69 and 0.73; professor: 0.69, 0.63, 0.72, 0.51, 0.80 and 0.76). Model interpretation with Grad-CAMs revealed that the activated areas mainly focused on cystic necrosis and margins.

CONCLUSION

The US-CNN model predicts GIST malignant potential well, which can assist in clinical treatment decision-making.

Surgery for Gastrointestinal Stromal Tumors: State of the Art of Laparoscopic Resection and Surgery for M1 Tumors

BACKGROUND

The principles of surgery for primary gastrointestinal stromal tumor (GIST) are resection with clear margins without lymphadenectomy, thus enabling organ-preserving resection in many cases. Tumor rupture needs to be avoided, because it entails peritoneal sarcomatosis. Given these preconditions, there is a rationale for laparoscopic resection of primary GIST. In metastatic GIST, surgery cannot achieve cure, but constitutes an important cornerstone of treatment. Resection can be performed either for residual lesions in responding patients, for focally progressive lesions, or as palliative measure in generalized progression.

METHODS

Selective literature review.

RESULTS

There is ample evidence on laparoscopic resection of gastric GIST, but none from randomized trials. The studies show favorable perioperative outcomes and suggest adequate oncological results; however, a direct comparison to open resection is difficult. For surgery in M1 stages, survival outcomes are better for resection in generalized response than in focal progression. Perioperative morbidity is acceptable. Surgery in generalized progression does not prolong survival and bears relevant morbidity risks.

CONCLUSION

Laparoscopy seems an adequate surgical approach for primary gastric GIST, as long as the surgical principles relevant for this entity are respected. For other tumor sites, there is no sufficient evidence available. In M1 stages, selected patients may benefit from surgery. Survival differences between resection at different stages probably mirror the biology of the disease rather than the effect of the timing of the operation.

Histone demethylase KDM4D promotes gastrointestinal stromal tumor progression through HIF1β/VEGFA signalling

BACKGROUND

Gastrointestinal stromal tumour (GIST) is the most common soft tissue sarcoma. The identification of the molecular mechanisms regulating GIST progression is vital for its treatment and prevention. Increasing reports have demonstrated that epigenetic alterations play critical roles in GIST development. However, the role of the histone demethylase KDM4D in GIST progression is poorly understood.

METHODS

In clinically matched GIST tissues, KDM4D protein levels were measured by Western blot and immunohistochemical (IHC) staining. KDM4D mRNA levels were examined by quantitative real-time PCR (qRT-PCR). Bioinformatics analysis was used to examine KDM4D expression. The biological effects of KDM4D were investigated in vitro using CCK-8, BrdU/PI, wound healing, colony formation, tube formation and Transwell assays and in vivo using a xenograft mice model. Luciferase assays were used to assess regulation of HIF1β gene promoter activity by KDM4D. ChIP assays were performed to assess KDM4D, H3K36me3 and H3K9me3 occupancy on the HIF1β gene promoter.

RESULTS

We observed a significant upregulation of KDM4D in GIST tissue compared with matched normal tissue and further explored the oncogenic function of KDM4D both in vitro and in vivo. Furthermore, we demonstrated that KDM4D directly interacted with the HIF1β gene promoter and regulated its activity, promoting tumour angiogenesis and GIST progression both in vitro and in vivo. Finally, we demonstrated that KDM4D transcriptionally activates HIF1β expression via H3K9me3 and H3K36me3 demethylation at the promoter region.

CONCLUSIONS

Our findings reveal the important roles of the KDM4D/HIF1β/VEGFA signalling pathway in GIST progression, and this pathway may act as a potential therapeutic target for GIST patients.

Inhibition of stromal-interacting molecule 1-mediated store-operated Ca2+ entry as a novel strategy for the treatment of acquired imatinib-resistant gastrointestinal stromal tumors

Imatinib has revolutionized the treatment of gastrointestinal stromal tumors (GIST); however, primary and secondary resistance to imatinib is still a major cause of treatment failure. Multiple mechanisms are involved in this progression. In the present study, we reported a novel mechanism for the acquired resistance to imatinib, which was induced by enhanced Ca²⁺ influx via stromal‐interacting molecule 1 (STIM1)‐mediated store‐operated Ca²⁺ entry (SOCE). We found that the STIM1 expression level was related to the acquired resistance to imatinib in our studied cohort. The function of STIM1 in imatinib‐resistant GIST cells was also confirmed both in vivo and in vitro. The results showed that STIM1 overexpression contributed to SOCE and drug response in imatinib‐sensitive GIST cells. Blockage of SOCE by STIM1 knockdown suppressed the proliferation of imatinib‐resistant GIST cell lines and xenografts. In addition, STIM1‐mediated SOCE exerted an antiapoptotic effect via the MEK/ERK pathway. The results from this study provide a basis for further research into potential novel therapeutic strategies in acquired imatinib‐resistant GIST.