SMALL-INTESTINAL STROMAL TUMORS - A CLINICOPATHOLOGICAL STUDY OF 31 TUMORS

Citation
Ms. Chang et al., SMALL-INTESTINAL STROMAL TUMORS - A CLINICOPATHOLOGICAL STUDY OF 31 TUMORS, Pathology international, 48(5), 1998, pp. 341-347
Citations number
25
Categorie Soggetti
Pathology
Journal title
ISSN journal
13205463
Volume
48
Issue
5
Year of publication
1998
Pages
341 - 347
Database
ISI
SICI code
1320-5463(1998)48:5<341:SST-AC>2.0.ZU;2-N
Abstract
The biologic behavior of gastrointestinal stromal tumors is difficult to predict, and they can be best studied in a site-specific fashion. T he aims of this study are to analyze the clinicopathologic parameters and assess the prognostic value of p53 (DO-7) and Ki-67 (MIB-1) immuno reactivities in small intestinal stromal tumors (SIST). The histopatho logic features of 31 SIST were assessed and categorized into two group s as follows. Group A (clinically aggressive) in which death due to tu mor, metastasis, recurrence or relapsed melena were seen (n = 15) and group B (clinically benign; n = 16). For both groups, the period of fo llow-up was 30-144 months, p53 overexpression was observed in four tum ors (31%) in group A, and in none in group B. For groups A and B, the mean Ki-67 index was 16.8 +/- 12.5 and 8.4 +/- 12.6, respectively. Sta tistical analysis revealed that the significant predictors of malignan cy were high cellularity (odds ratio (OR) = 999; 95% confidence interv al(CI) = 0-999); p53 overexpression (OR = 999; CI = 0-999); size of tu mor greater than or equal to 5 cm (OR = 18.0; CI = 1.9-171.9); greater than or equal to 5 mitoses/50 high-power fields (HPF) (OR = 17.1;CI = 1.8-165.9); pleomorphism (OR = 17.1; Cl = 1.8-165.9); and necrosis (O R = 11.9; CI = 2.2-65.1; P < 0.05). High Ki-67 index (greater than or equal to 8.4) had a marginal impact on risk (OR = 4.1; Cl = 0.8-20.2; P = 0.08). In conclusion, high cellularity, p53 overexpression, size o f tumor greater than or equal to 5 cm, greater than or equal to 5 mito ses/50 HPF, pleomorphism and necrosis are important parameters for the prediction of malignancy in SIST.