A correlation research of Ki67 index, CT features, and risk stratification in gastrointestinal stromal tumor

Recurrence and metastasis are the most important factors affecting the quality of life and survival rate of patients with gastrointestinal stromal tumors (GISTs). Accurate preoperative determination of the malignant degree of GISTs and the development of a reasonable treatment plan can effectively reduce the recurrence rate. CT is currently considered the preferred imaging modality for initial assessment. Until now, there have only been a few studies investigating the relationship between CT features and recurrence of GISTs. However, the value of CT features in prognostic assessment is still unclear. In this study, we attempted to investigate the prognostic significance of CT features and the Ki67 index in GISTs.

We retrospectively analyzed the clinicopathological and imaging data for 151 patients with a histopathological diagnosis of GIST who had received contrast-enhanced CT examination and surgical resection at XinHua Hospital from October 2008 to December 2015 or Sir Run Run Shaw Hospital in 2017. Then, we explored the correlation among CT features, the Ki67 index, and risk stratification of GISTs. The correlation among CT features, the Ki67 index, and risk stratification was mainly analyzed using the Spearman rank correlation.

The incidence of high-risk disease or metastasis was clearly higher in the group with Ki67 > 5% than that in the group with Ki67 ≤ 5% (P < 0.001). The Ki67 index was positively correlated with risk stratification (r = 0.558) or mitotic index (r = 0.619). CT imaging features including size, contour, and margin of the tumor were associated with the Ki67 index (r = 0.332, 0.333, and 0.302, respectively). The multivariate logistic regression analysis revealed that the tumor size [P = 0.043 Exp (B) = 1.150] and the presence of ulceration [P = 0.011, Exp (B) = 3.669] were effective variables in distinguishing between the groups with Ki67 ≤ 5% and >5%. The presence of necrosis or cystic degeneration, tumor contour, tumor margin, and pattern of enhancement were associated with risk stratification (r = 0.530, 0.501, 0.419, and 0.447, respectively).

Our findings suggest that the Ki67 index is an effective complementation in predicting the prognosis of GISTs, and CT features including size, contour, and margin of the tumor, presence of necrosis or cystic degeneration, and pattern of enhancement provide evidence to support the importance of preoperative assessment.