Factors influencing progression-free survival in gastrointestinal stromal tumors with special reference to pathologic features, cytogenetics, and radiologic response

• Published in: Gastrointestinal cancer research Vol. 4; no. 5-6; pp. 173 - 177
• Main Authors: Attili, Suresh V, Ananda, Bb, Mandapal, T, Anjaneyulu, V, Sinha, Sudha, Reddy, Obula C
• Format: Journal Article
• Language: English
• Published: United States International Society of Gastrointestinal Oncology 01.09.2011
• Subjects: Original Research
• ISSN: 1934-7820; 1934-7987

Treatment of gastrointestinal stromal tumors (GISTs) changed significantly with the advent of targeted therapy with imatinib. Clear markers predictive of response to imatinib therapy and disease-free survival in patients with GIST have not been identified. Even RECIST criteria are inadequate for predicting response to therapy, especially in patients with stable disease. Data collected at a tertiary care cancer center from 2003 to 2005 in south India were analyzed retrospectively to assess clinical, pathologic, and cytogenetic profiles of patients with GIST. In addition, radiologic responses to therapy were evaluated for correlation with the disease-free survival. GIST was defined as a mesenchymal spindle/epitheloid cell lesion arising in the GI tract with CD117 positivity. Only data from patients with locally advanced or metastatic disease were analyzed. Clinical and pathologic details of the patients were noted from case records. NCCN guidelines were followed for the treatment. Radiologic response to therapy was reassessed according to RECIST, and progression-free survival calculated for all analyzed patients using intent-to-treat analysis. The mean age of presentation was 42.8 ± 5.3 years (24-60), with a male-to-female ratio of 1.5:1. Small intestine was the most common disease site (60%), followed by stomach (20%), mesentery (7.2%), colorectal regions (7.2%), and other sites (5.6%). The most frequent pathologic finding in patients having recurrence was high mitotic rate. Initial tumor size (either in the metastatic setting or in local recurrence) had no bearing on progression-free or overall survival, nor did initial anatomic location or site of metastasis. Histologically, however, patients with a mixed-cell morphology had shorter survival compared to the other morphologies. Those patients having any cytogenetic abnormality had worse outcome compared to those with normal karyotype. Similarly, among patients who achieved remission, those who did so within 12 weeks had better overall survival than did those with a delayed time to remission. Overall survival of patients having stable disease and late partial responses (after 3 months) was similar and superior to survival for patients whose disease progressed while on therapy. GISTs characterized by a high mitotic rate and mixed-cell morphology and any cytogenetic abnormalities are associated with poorer outcome. Similarly, shorter time to response was more important than the actual response to therapy. Initial disease site, the site of metastasis, and tumor size had no bearing on outcomes to therapy.