[66] P63 Is Not a Helpful Marker in the Diagnosis of Giant Cell Tumor of Bone

G de la Roza. SUNY-Upstate Medical University, Syracuse, NY

Background: Giant cell tumor of bone (GCT) is a locally aggressive neoplasm with high recurrence rate that occurs mainly in long bones of young adults. While the diagnosis is often straightforward, it can be challenging with small core needle biopsies, particularly when dealing with unusual sites or skeletally immature patients. The differential diagnosis includes both benign and malignant neoplasms of bone and soft tissue that are rich in giant cells. Recent studies using immunohistochemistry and molecular methods on tissue microarray samples have demonstrated overexpression of p63 in the stromal cells of most giant GCTs and advocate the use of p63 as diagnostic marker of GCT.
Design: 87 cases of giant cell lesions of bone and soft tissues, including 23 giant cell tumors of bone, 8 primary aneurysmal bone cysts (ABC), 15 tenosynovial giant cell tumors (TSGCT), 12 chondroblastomas (CHB), 4 giant cell reparative granulomas (GCRG), 6 non-ossifying fibromas (NOF), 2 fibrous dysplasias (FD), 4 pigmented villonodular synovitis (PVNS), and 4 osteosarcomas with giant cells (OS), were selected from our pathology files. Only cases with abundance of giant cells were selected for the study. Except for one needle biopsy, all cases were resection samples, open biopsies, or bone curettages. All routine histologic sections were reviewed and one or two paraffin blocks were selected for p63 immunohistochemistry. Immunoreactivity was scored by intensity of nuclear staining as weak (1+), moderate (2+), and strong (3+) and also by percentage of staining cells as focal (<10%), intermediate (10-50%), and diffuse (>50%).
Results: p63 reactivity was identified in 20 of 23 GCTs (86.9%), 5 of 8 primary ABCs (62.5%), 10 of 12 CHBs (83.3%), 4 of 4 GCRGs (100%), 2 of 4 OS (50%), 1 of 15 TSGCT (6.6%), 1 of 6 NOF (16.6%), and 1 of PVNS (25%). The immunoreactivity was only seen in stromal cells. While the intensity and percentage of cells staining was variable, strong and moderate staining was seen in most GCTs and some non-GCT cases. In CHBs, significant reactivity was seen in chondroid areas. The sensitivity, specificity, positive predictive value and negative predictive value of p63 immunohistochemistry for the diagnosis of GCT was 86.95%, 62.50%, 45.45%, and 93.02%, respectively.
Conclusions: These findings indicate that p63 immunohistochemistry is not a helpful diagnostic marker in the diagnosis of GCT since a significant number of other giant cell lesions express this marker.
Category: Bone & Soft Tissue

Tuesday, March 23, 2010 9:30 AM

Poster Session III # 4, Tuesday Morning


Close Window