Background: Proteomic expression profiling has been suggested as a potential tool for the early diagnosis of cancer and other diseases. The objective of our study was to assess the feasibility of this approach for the detection of breast cancer.
Materials and methods: In a randomized block design pre-operative serum samples obtained from 78 breast cancer patients and 29 controls were used to generate high-resolution MALDI-TOF protein profiles. The spectra generated using C8 magnetic beads assisted mass spectrometry were smoothed, binned and normalized after baseline correction. Linear discriminant analysis with double cross-validation, based on principal component analysis, was used to classify the protein profiles.
Results: A total recognition rate of 99%, a sensitivity of 100%, and a specificity of 97.0% for the detection of breast cancer were shown. The area under the curve of the classifier was 98.3%, which demonstrates the separation power of the classifier. The first 2 principal components account for most of the between- group separation.
Conclusions: Double cross-validation showed that classification could be attributed to actual information in the protein profiles rather than to chance. Although preliminary, the high sensitivity and specificity indicate the potential usefulness of serum protein profiles for the detection of breast cancer.