Assessment of morphological CT imaging features for the prediction of risk stratification, mutations, and prognosis of gastrointestinal stromal tumors
Roberto Cannella, Emanuele Tabone, Giorgia Porrello, Giovanni Cappello, Cecilia Gozzo, Lorena Incorvaia, Giovanni Grignani, Alessandra Merlini, Lorenzo D’Ambrosio, Giuseppe Badalamenti, Daniele Regge, Tommaso Vincenzo Bartolotta
Eur Radiol31, 8554–8564 (2021). https://doi.org/10.1007/s00330-021-07961-3
Dr. Charikleia Triantopoulou, Head of the Radiology Department and Anastasia Doukopoulou, Third-Year Radiology Resident ‘Konstantopouleion’ General Hospital of Nea Ionia, Athens, Greece.
Gastrointestinal stromal tumors (GISTs) are the most common mesenchymal neoplasms of the gastrointestinal tract. Most commonly originate from the stomach (50-60%) and small bowel (30-35%) and occasionally are found in colorectum (5%) and esophagus (<1%). Nevertheless, they can arise anywhere along the gastrointestinal tract and rarely even in extra-gastrointestinal locations [1].
In order to achieve the appropriate treatment and patient’s management it is important to ensure an accurate preoperative risk stratification of gastrointestinal stromal tumors. The National Institutes of Health (NIH) consensus classification system is currently used to assess patient prognosis after surgical resection and is based on tumor location, size, mitotic count and rupture [2].
In this retrospective dual-institution study, the authors intended to assess the correlation between CT imaging features of GISTs with risk stratification, prediction of mutation status, and prognosis.
Eighty-eight patients with pathologically proven GISTs were included in the study. All of them had preoperative contrast-enhanced CT performed between 2008 and 2019, received no treatment before imaging and had available pathological analysis. All CT studies were reviewed by two readers with experience in abdominal imaging in a blinded fashion.
Firstly, they evaluated the location of the lesion, as well as the size and the growth (endophytic, exophytic or mixed). Also, the following morphological imaging features of GISTs were assessed: contour, margins, enhancement pattern, enhancement degree, presence of intralesional calcifications, intralesional hemorrhage, surface ulceration, cystic degeneration, tumor necrosis, intratumoral vessels, enlarged feeding vessels or draining veins and direct invasion of adjacent abdominal organs.
The analysis of CT studies revealed that the primary tumor was located in the stomach in 52 cases, and small bowel and colorectum in 32 and 4 cases, respectively. GISTs most commonly presented as large lesions measuring ≥ 5 cm with exophytic growth, well-defined margins, and variable amount of intralesional necrosis. Distant metastases were identified in 14 cases. According to NIH 2008 risk stratification criteria, 3 patients had very low-risk GISTs, 21 low-risk, 19 intermediate risk and 45 high-risk. Mutational analysis presented 50 KIT 11, 11 KIT 9, 14 PDFGRα mutations, and 13 wild-type GISTs.
After statistical analysis, it was shown that there are differences in imaging features according to the risk stratification of GISTs. The percentage of high-risk tumors demonstrating lobulated contour, heterogeneous enhancement, hypoenhancement on portal venous phase, presence of necrosis, severe intralesional necrosis, intratumoral vessels and enlarged feeding vessels was remarkably higher compared to low-to-intermediate risk GISTs. Independent predictors of high-risk GISTs were lesions size ≥ 5 cm and enlarged feeding vessels, whereas in the cases of GISTs without metastases, only lesions size ≥ 5 cm and lobulated contour independently predicting high-risk tumors.
The identification of specific mutations genes has led to more targeted treatments with exceptional increase in survival rates [3]. This study investigates the correlation between tumor enhancement and different mutation genes. As a result, hyperenhancement on portal venous phase was significantly more frequent in PDGFRα-mutated/wild type GISTs compared to GISTs with KIT mutations.
During a median follow-up time of 44.7 months, progression and postoperative recurrence was observed in 25 patients, while 5 patients died due to the GIST. High-risk GISTs had remarkably shorter progression-free survival compared to low-to-intermediate-risk GISTs. Despite that several imaging features were associated with poorer PFS (including size, contour, margins, enhancement pattern, calcifications, necrosis, amount of necrosis, cystic degeneration, intratumoral vessels, enlarged feeding vessels, and direct organ invasion), only ill-defined margins were associated with poorer PFS at multivariate analysis. In addition, ill-defined margins and intralesional hemorrhage were independently associated with poorer overall survival.
In this study, the authors try to indicate that multiple CT imaging features are significantly different between low-to-intermediate-risk and high-risk GISTs and investigate the predictive values of these features for the progression-free survival and overall survival. They also refer to other studies, i.e. Zhou et al [4], with which interestingly share some common results, that tumor size and enlarged vessels feeding or draining the mass are independent predictors of high-risk tumors.
In conclusion, this efficient study may have used a relatively small sample, but it proved that morphologic contrast-enhanced CT features are considerably different depending on the risk status or mutations. These findings may help in the prognostic prediction of GISTs and if supported by further studies with radiologic-pathologic correlations may lead to more appropriate and targeted therapies.
References:
- Imaging of Gastrointestinal Stromal Tumors: From diagnosis to evaluation of therapeutic response. Federica Vernuccio et al. Anticancer Research Jun 2016, 36 (6): 2639-2648
- Risk stratification of patients diagnosed with gastrointestinal stromal tumor. Heikki Joensuu et al. Human Pathology 2008, 39: 1411-1419
- Gastrointestinal stromal tumours: ESMO-EURACAN Clinical Practice Guidelines for diagnosis, treatment and follow-up. Casali P.G. et al. Annals of Oncology 2018, 29: iv68–iv78
- Predictive features of CT for risk stratifications in patients with primary gastrointestinal stromal tumour. Zhou C. et al. Eur Radiol 2016, 26:3086–3093
Dr. Anastasia Doukopoulou is a third-year radiology resident at ‘Konstantopouleion’ General Hospital of Nea Ionia, Athens, Greece. She completed her undergraduate medical degree at Comenius University in Bratislava in 2018 and after that she joined the Medical Imaging Department at General Hospital of Naoussa in Greece. Abdominal imaging is one of the main fields she is interested in.
Comments may be sent to: anas.doukopoulou@gmail.com