Prediction of the probability of metastasis to the sentinel lymph node in breast cancer by ATR-FTIR spectroscopy and chemometrics

Abstract

The status of axillary lymph nodes is considered one of the most critical prognostic indicators for patients diagnosed with invasive breast cancer (BC). The Breast Service of Memorial Sloan Kettering Cancer Center (MSKCC) nomogram was developed to obtain the likelihood of metastasis to the sentinel lymph node (SLN). Blood plasma from 28 patients with BC stages I, II, and III were collected and analyzed by Attenuated total reflectance-Fourier transform infrared spectroscopy (ATR-FTIR). Therefore, the aim of the study is to use ATR-FTIR spectroscopy and chemometrics in blood plasma to predict the probability of SLN metastases in breast cancer in ≥ 50 % (Model 1) or > 30 % (Model 2) determined by the MSKCC nomogram. The chemometric models were performed with orthogonal partial least squares discriminant analysis (OPLS-DA). Both Models 1 and 2 obtained sensitivity and specificity of 100 %. Model 1 was built with 7 latent variables (LV) having a cumulative variance of 92.56 % and root mean square error of cross-validation (RMSECV) of 0.0096. Model 2 for 6 LV had a cumulative variance of 94.17 % and RMSECV of 0.0031. External validation was only tested for Model 1, obtaining 100 % accuracy and RMSE of validation (RMSEV) of 0.274. Proteins and nucleic acids were the biomolecules with the highest contribution to this prediction. Therefore, ATR-FTIR spectroscopy generates valuable insights into the tumor aggressiveness and the probability of metastasis to the SLN.