Computer Aided Measurement of Tumor Area in Positive Lymph Nodes Is an Alternative Way To Predict Prognosis of Metastatic Breast Cancer
Y Li, J Jackson, S Karan, P Rajan, A Salhadar, K Albequerque, CH Ersahin. Loyola University Medical Center, Maywood, IL
Background: The number of positive lymph nodes (LNs) is the only node-related factor recognized by AJCC for prognostic evaluation of breast cancer. Counting the number of LNs may be problematic in cases of matted LNs. In addition, the number of positive LNs may not completely reflect the degree of tumor involvement. For example, microinvasion or complete tumor replacement in a LN are both counted as one positive LN after the diagnosis of the first macrometastatic LN.
Design: From 1998-2000, 127 patients were diagnosed with breast cancer metastasized to regional LNs in our institution. Areas of metastatic cancer in the LNs were identified and measured using Olympus MicroSuite5 software. Survival and prognosis of patients were compared in relation to the LN tumor area vs. number of positive LNs according to AJCC's classification (N1, N2 and N3).
Results: The patients are divided into 3 groups according to the LN tumor areas: 1) <100mm2, 2) 100-500mm2, 3) >500mm2. Survival analysis of the 3 groups showed statistical difference with a 5-year survival of 75%, 53% and 44%, respectively (Fig 1).
The survival between N1 (1-3 LNs), N2 (4-9), and N3 (>9) is also different with a 5-year survival of 75%, 52%, and 40% respectively. Interestingly, 15% of the deceased N3 patients showed better prognosis (with <350 mm2 tumor areas) compared to other N3 patients with mean tumor area of 711mm2.
Conclusions: Our retrospective study used imaging analysis software to measure tumor areas of positive LNs as an estimation of tumor volume. Our results suggest that LN tumor area is accurate in depicting the extent of metastatic cancer involvement and the prognosis independent of the number of positive LNs. Patients with tumor area <100mm2 showed better prognosis, and those >500mm2 showed worst prognosis, with mean survival of 4.9 and 3.4 years, respectively. Although the prognosis is comparable between LN tumor areas and AJCC LN stages, 15% of N3 patients showed better survival due to small tumor areas. Hence, computer aided tumor area analysis can be used as an alternative way to predict prognosis in breast cancer patients.
Tuesday, March 23, 2010 1:30 PM
Platform Session: Section B, Tuesday Afternoon