Prognostic factors associated with long-term survival in previously untreated metastatic renal cell carcinoma

PURPOSE

To identify prognostic factors (PF) for long-term survival in metastatic renal cell carcinoma (RCC) patients.

METHODS

We retrospectively reviewed a metastatic RCC database at the Cleveland Clinic Foundation consisting of 358 previously untreated patients who were enrolled in institutional review board-approved clinical trials of immunotherapy and/or chemotherapy at our institution from 1987 to 2002. In order to identify patient characteristics associated with long-term survival, we compared 226 'short-term' survivors [defined as overall survival (OS) <2 years] with 31 'long-term' survivors (OS >or=5 years).

RESULTS

Using logistic regression models, four adverse PF were identified as independent predictors of long-term survival: hemoglobin less than the lower limit of normal, greater than two metastatic sites, involved kidney (left), and Eastern Cooperative Oncology Group (ECOG) performance status (PS). Using the number of poor prognostic features present, three distinct risk groups could be identified. Patients with 0 or 1 adverse prognostic feature present had an observed likelihood of long-term survival of 32% (21/66) compared with 9% (8/91) for patients with two adverse features present and only 1% (1/93) for patients with more than two adverse features.

CONCLUSIONS

Independent predictors of long-term survival in previously untreated metastatic RCC include baseline hemoglobin level, number of involved sites, involved kidney, and ECOG PS. Incorporation of these factors into a simple prognostic scoring system enables three distinct groups of patients to be identified.

The predictive value of baseline fluoro-deoxy-glucose positron emission tomography (FDG-PET) standardized uptake value (SUV) for overall survival in patients (pts) with locally advanced esophageal and gastroesophageal junction (GEJ) cancer

4062

Background: The predictive value of FDG-PET SUV in pts with esophageal and GEJ cancer has been subject of recent interest. The heterogeneity of disease stages and treatments made interpretation of reported data difficult.We evaluated the value of baseline FDG-PET SUV in a homogeneous patient (pt) population treated in a uniform fashion. Methods: Retrospective analysis of 71 pts with stages II-IVa esophageal and GEJ cancer treated on 2 phase 2 trials at the Cleveland Clinic Foundation. Data was collected on baseline pt and tumor characteristics, baseline SUV uptake, and outcomes. Results: All pts were treated with preoperative concurrent fluorouracil/cisplatin chemoradiotherapy (CRT); 69 pts proceeded to esophagectomy and 58 pts received additional postoperative adjuvant CRT. Median pt age was 60 (range 33–75) years, 86% were male, 89% had adenocarcinoma, 35%, 41% and 24% had stage II, III and IVa disease respectively. Pts have been followed for a median of 14 (range 3–50) months. All pts underwent a baseline FDG-PET scan with a mean primary site SUV of 8.9 (range 0–28.2). Pathological response (complete or partial) was achieved after induction CRT in 54% and was more likely in those with a baseline primary site SUV ≥ 7.3 (OR: 3.95, 1.43–10.9, P=0.008). Recurrence developed in 33 pts (46.5%) with distant metastases identified in 31 of these 33. The Kaplan-Meier 2-year projected overall survival for all 71 pts is 58% with a median of 31 months. Mortality was less in pts with baseline SUV ≥ 5.0 (HR: 0.44, 0.20–0.94, P = 0.033). After adjusting for clinical stage at diagnosis, tumor location and histology, baseline SUV ≥ 5 was still predictive of improved survival in multivariate analysis (HR: 0.35, 0.15–0.85, P = 0.02). Conclusions: In this retrospective analysis, esophageal and GEJ cancer pts with a higher baseline primary site SUV were more likely to respond to induction CRT and had better overall survival. This observation suggests that the subset of pts with more metabolically active tumor may derive greater benefit from multimodality treatments that include CRT. A prospectively designed trial would be required to confirm this hypothesis.

No significant financial relationships to disclose.