We investigated the prognostic value of 18F-FDG uptake in the supraclavicular lymph node (SCLN) on PET/CT in breast cancer patients with clinical ipsilateral SCLN metastasis (cN3c).
Fifty-five female patients with initial 18F-FDG PET/CT were treated with curative intent. For semiquantitative analysis, the SUVmax of the primary tumor, axillary lymph node, and SCLN were normalized by the SUVmean of the liver (defined as SUVR-tumor, SUVR–axillary lymph node, and SUVR-SCLN, respectively). Cox proportional hazards models were used to predict disease-free survival (DFS) and overall survival (OS). Differences in DFS and OS were assessed by Kaplan-Meier analysis.
Twenty-three patients (41.8%) experienced recurrence, and 13 (23.6%) died during follow-up (median, 70.0 months; range, 6–128 months). In multivariate analysis, SUVR-tumor greater than 3.26 (hazards ratio, 7.26; 95% confidence interval, 1.58–33.31; P = 0.01) and SUVR-SCLN greater than 1.05 (hazards ratio, 8.47; 95% confidence interval, 1.09–65.87; P = 0.04) were prognostic for OS. No clinicopathologic or PET/CT parameters were prognostic for DFS. The patients were divided into 3 groups: group 1 (n = 11, SUVR-tumor ≤3.26 and SUVR-SCLN ≤1.05); group 2 (n = 27, SUVR-tumor >3.26 or SUVR-SCLN >1.05); and group 3 (n = 17, SUVR-tumor >3.26 and SUVR-SCLN >1.05). The 5-year OS rates were 100% in group 1, 85.2% in group 2, and 51.0% in group 3. Group 3 showed worse prognosis than group 1 (P < 0.01) and group 2 (P < 0.01).
In addition to SUVR-tumor, SUVR-SCLN seemed to play an important role in selecting patients with the worst prognosis.