Tri-Modal spectroscopy for early detection of cervical neoplasia in vivo

Motivation
In the current procedure for identifying cervical dysplasia (cervical intraepithelial neoplasia, CIN) and cervical cancer, exfoliative cytology, commonly referred as Pap smear, is performed annually. Patients with positive screening findings then undergo colposcopy, examination of the reflected light from uterine cervix under the low magnification microscopy, and biopsy of suspicious regions. The diagnostic sensitivity of colposcopy ranges from 37.5 %[1] to 90 %[2], meaning that early cervical pre-malignant lesions are often missed. Colposcopy also lacks specificity (<50 %)[3], leading to a large number of unnecessary biopsies.

It is important to develop efficient and reliable methods for detecting early precancerous lesions, since it is estimated that more than 90% of epithelial cancers can be prevented by early detection and treatment.[4] Spectroscopy has emerged as a minimally invasive tool with great diagnostic potential in organs such as the cervix and the esophagus, as well as the oral cavity, the colon and the lungs.

Technique
Tri-Modal Spectroscopy (TMS), a contact probe technique for spectral diagnosis developed by our lab, is a combination of three spectroscopic techniques: Diffuse Reflectance Spectroscopy (DRS), Intrinsic Fluorescence Spectroscopy (IFS), and Light Scattering Spectroscopy (LSS). They are combined since each of the spectroscopic techniques provides complementary information, thus leading to enhanced sensitivity and specificity. The instrument that facilitates the TMS studies, designed by the lab, is called Fast Excitation-Emission Matrix Instrument (FastEEM). This instrument has the ability to excite the tissue with white light and 10 different fluorescence excitation lights and collect the reflectance and fluorescence spectra in less than a second from the same tissue spot. TMS and FastEEM provide for the non-invasive in-vivo spectral diagnosis in real-time.

Clinical research
Clinical research in vivo had been performed in Brigham and Women’s Hospital in Boston. The spectroscopic measurements are taken during colposcopy examination. Suspicious sites are biopsied and the pathology results are compared to the spectral diagnosis of the same site. The pathology is used as a gold standard for developing the spectral analysis algorithm. The initial TMS clinical work in the cervix has produced promising results. Georgakoudi et al [5] were able to distinguish squamous intraepithelial lesions (SIL) from benign tissue with 92% sensitivity and 71% specificity, and SIL from non SIL with 92% sensitivity and 90% specificity.

Research goals
The main goal of cervical dysplasia research is to provide quantitative parameters to distinguish normal tissue from diseased tissue and to understand the underlying chemical changes. Diseased tissue should be correctly recognized as low-grade dysplasia, high-grade dysplasia, or cancer. Diseased tissue should also be differentiated from inflammation. This is of a great importance because inflammation is a frequent process in cervix that is not necessarily associated with dysplastic change.

Tissue structure and composition also change with age. The most striking difference relevant to our studies noted so far is that the amount of collagen cross-linking increases with age, which leads to an increase of collagen fluorescence.6 It will be important to determine if an age factor should be included into the diagnostic algorithm for cervical TMS study.

An important goal of cervical dysplasia research is not only to develop a reliable non-invasive clinical detection tool, but also to gain quantitative understanding of the chemical basis for the changes in tissue through exploring the role of NADH, collagen, porphyrins, tryptophan and FAD in tumorgenesis in future clinical work.

Recent Publications

1. Skehan, M. Brit J Obstet Gynaec 1990, 97(9), 811-816.
2. Javaheri, G.; Fejgin, M.D. Am J Obstet Gynecol 1980, 137(5), 588-594.
3. Follen, M. Obstet. Gynecol. 1998, 91, 626-631.
4. American Cancer Society 1998 Cancer Facts and Figures.
5. Georgakoudi, I.; Sheets, E. E.; Muller, M. G.; Backman, V.; Crum, C. P.; Badizadegan, K.; Dasari, R. R.; Feld, M. S. Am J Obstet Gynecol 2002, 186, 374-382.
6. Bailey, A. J.; Paul, R .G.; Knott, L. Mechanisms of Ageing and Development 1998, 106, 1-56.