Transcutaneous in vivo Raman spectroscopy: Detection of age-related changes in mouse breast

Abstract

The risk of female breast cancer increases as age progresses. This can be explained by Pike's model, which suggests that the process of aging in breast is not uniform. ‘Breast ageing’ is most rapid during menarche, slows with each pregnancy, slows further during perimenopause, and is least after the menopause. In this study, the feasibility of using transcutaneous in vivo Raman spectroscopy to detect age-related changes in mouse breast and its effect on tumor detection were explored. Spectra acquired transcutaneously from breast of 2 (menarche), 4–6 (mid reproductive phase), 10–12 (perimenopause), 13–15 month (menopause) old mice and frank breast tumors were analyzed using principal component-linear discriminant analysis (PC-LDA). A classification efficiency of ∼80% was achieved for different age groups. Further, it was observed that the number of misclassifications among age groups increase as age progresses. For example, 3% spectra from menarche misclassify with other age groups, while 19–28% from mid-reproductive, perimenopause and menopause misclassify with each other. Misclassifications between groups indicate homogeneity in tissues. Thus, results suggest that menarche breast is biochemically distinct while breast during mid-reproductive, perimenopause and menopause is relatively homogenous. This probably indicates that the rate of aging is rapid during menarche, but slows down during other phases. Thus, spectroscopic data correlate with Pike's model. Although sensitive to age-related changes, RS could classify tumors with 95% efficiency. Overall, results suggest possibility of distinguishing age-related changes using RS without affecting ability to classify tumors.