Correlation of near-infrared spectral tomography (NIRST) with residual cancer burden (RCB) for prognostic assessment subsequent to neoadjuvant chemotherapy (NAC) in breast cancer.

Abstract

e12655 Background: NIRST provides biophysical data including oxy- and deoxy-hemoglobin, water, lipid, and scatter components, useful in distinguishing cancer from normal breast tissue. Unlike anatomic changes characterized by conventional radiographic techniques, these functional changes are uniquely suited to monitor early response to NAC. The Residual Cancer Burden (RCB) is a reproducible pathologic analysis to evaluate residual disease after treatment with NAC and proven useful prognostically. Methods: 19 women with locally advanced breast cancer were enrolled in this NIRST pilot study to evaluate tumor response to NAC. Patients underwent NIRST prior to NAC (HbT-pre), after cycle 1, after cycle 2, at the mid-point of NAC, and at the conclusion of NAC. RCB scores were determined in 16 of the 19 final tissue specimens. RCB scores for two patients were excluded due to positive excisional lymph node biopsy prior to NAC. One patient underwent surgery at an outside hospital and was excluded as pathology slides were unavailable. The RCB class was compared to NIRST biophysical data at time points throughout NAC. Ratios relative to the pretreatment average of the normal contralateral breast were used for statistical analysis to minimize inter-subject variability of breast density and the effect on biophysical data. Results: The best predictor for RCB class was the early change in total hemoglobin (HbT-Δ) (percent change when comparing the first cycle of NAC to HbT-pre), as analyzed by Pearson correlation coefficient of 0.71. The significance of this correlation was evaluated using a two-sided t-test. The resulting P-value of 0.0019 demonstrates a statistically significant correlation. Conclusions: In this pilot study of NIRST, we demonstrate a statistically significant correlation between HbT-Δ change during the first cycle of NAC and the RCB. These findings suggest the potential utility of this approach, and warrant further evaluation in a larger study. The ability to determine response within the first cycle of NAC would be of great clinical utility, and additionally of potential use in clinical trials involving novel neoadjuvant therapies.