Autofluorescence patterns in short-term cultures of normal cervical tissue

Optical imaging of the cervix

Recent advances in fiber optics, sources and detectors, imaging, and computer-controlled instrumentation have stimulated a period of unprecedented growth in the development of photonics technologies for a wide variety of diagnostic and therapeutic clinical applications. These include the application of quantitative optical spectroscopy and imaging for the detection of precancerous lesions in the uterine cervix, a topic of interest at the Second International Conference on Cervical Cancer, which was held April 11-14, 2002. Investigators have applied the Littenberg method of emerging technology assessment to new optical methods used to detect cervical neoplasia. Currently, such technologies as fluorescence spectroscopy (the combination of fluorescence and diffuse reflectance spectroscopy), tri-modal spectroscopy, and light-scattering spectroscopy that probe the spectral characteristics of tissue are being investigated. Optical technologies that create images of subcellular structure without biopsy subsequent to pathology that currently are under investigation include in vivo confocal imaging and optical coherence tomography. Numerous small studies have demonstrated the potential of these optical technologies. What remains to be elucidated are the fundamental biophysical origins of variations in remitted optical signals between normal and dysplastic tissue. Large multicenter randomized controlled trials are needed to confirm the detection and imaging capabilities of optical technology. Furthermore, the development of contrast agents that could boost detection with these technologies is needed, and basic biologic characterization of signals should be pursued. Applying the Littenberg assessment will help ensure that superior, not simply alternative, technologies are implemented.

Effects of biographical variables on cervical fluorescence emission spectra

Diagnostic algorithms can classify tissue samples as diseased or nondiseased based on fluorescence emission collected from the intact cervix. Such algorithms can distinguish high-grade squamous intraepithelial lesions from low-grade squamous intraepithelial lesions. An understanding of the effects of the values of biographical covariates, such as age, race, smoking, or menopausal status on the emission spectra for each patient could improve diagnostic efficiency. The analysis described was performed using data collected from two previously published clinical trials; one study measured spectra from 395 sites in 95 patients referred to a colposcopy clinic with abnormal Pap smears, and the second study measured spectra from 204 sites in 54 patients self-referred for screening and expected to have a normal Pap smear. For this analysis, data about age, race, menstrual cycle, and smoking were collected. The principal components from normalized data were compared. There are clear intensity differences observed with age and menopausal status; postmenopausal patients exhibit higher emission intensities. Differences associated with biographical variables need to be tested in larger studies, which stratify adequately for these variables. The addition of these biographical variables in the preprocessing of data could dramatically improve algorithm performance and applicability.

Relationship between collagen autofluorescence of the human cervix and menopausal status

The goal of this study was to evaluate the effect of different menopausal states (pre- and post-) on the endogenous fluorescence of normal cervical tissues. In particular, the average fluorescence as well as the interpatient and intrasample variability in the average fluorescence of the epithelium and stroma were evaluated as a function of pre- and postmenopausal states. High-resolution fluorescence images at excitation-emission wavelengths of 440, 520 nm and 365, 465 nm were obtained from epithelia and stroma of freeze-trapped cervical tissue blocks maintained at -196 degrees C. The fluorescence images were recorded using a low temperature optical scanner. Fluorescence images from a normal sample population (n = 27) were quantitatively analyzed, and the average epithelial and stromal fluorescence intensities were obtained. Data grouped according to menopausal status (pre- vs post-) showed statistically significant differences (P < 0.002) in stromal fluorescence. In particular, the cervical stroma of postmenopausal women showed (1) significantly greater average fluorescence and (2) greater interpatient and intrasample variability in the fluorescence, relative to that of premenopausal women. These results provide evidence for changes in collagen cross-linking with menopause.

Microanatomical and biochemical origins of normal and precancerous cervical autofluorescence using laser-scanning fluorescence confocal microscopy

Clinical studies have shown that in vivo fluorescence spectroscopy can improve the diagnosis of cervical precancer. Recent work suggests that epithelial fluorescence increases, whereas stromal fluorescence decreases, with precancer. However, the microanatomic and biochemical sources of fluorescence in living cervical tissue have not yet been established. This study aims to characterize the origins of living normal and precancerous cervical fluorescence at microscopic levels using laser-scanning fluorescence confocal microscopy. Ten pairs of colposcopically normal and abnormal biopsies were obtained; transverse, 200 microm thick, short-term tissue cultures were prepared and imaged when viable with UV (351-364 nm) and 488 nm excitation before and after addition of the vital dye, Mitotracker Orange. In normal epithelium basal epithelial cells showed cytoplasmic fluorescence; parabasal, intermediate and superficial cells showed fluorescence only at the periphery of the cell. In low-grade precancers cytoplasmic fluorescence was visible in the bottom one-third of the epithelium; in high-grade precancers cytoplasmic fluorescence was visible throughout the lower two-thirds of the epithelium. Cytoplasmic fluorescence was colocalized with the MitoTracker probe and is attributed to mitochondrial reduced form of nicotinamide adenine dinucleotide at UV excitation and mitochondrial flavin adenine dinucleotide at 488 nm excitation. Stromal fluorescence originated from matrix fibers; with the development of precancer the density and fluorescence intensity of matrix fibers decrease. Autofluorescence properties of precancerous cervix reflect an increased number of metabolically active mitochondria in epithelial cells and a reduced stromal fluorescence, which can be an indicator for altered communication between precancerous epithelium and stroma. These changes can explain differences in in vivo fluorescence spectra of normal and precancerous cervical tissue.

Effect of fiber optic probe geometry on depth-resolved fluorescence measurements from epithelial tissues: a Monte Carlo simulation

Developing fiber optic probe geometries to selectively measure fluorescence spectra from different sublayers within human epithelial tissues will potentially improve the endogenous fluorescence contrast between neoplastic and nonneoplastic tissues. In this study, two basic fiber optic probe geometries, which are called the variable aperture (VA) and multidistance (MD) approaches, are compared for depth-resolved fluorescence measurements from human cervical epithelial tissues. The VA probe has completely overlapping illumination and collection areas with variable diameters, while the MD probe employs separate illumination and collection fibers with a fixed separation between them. Monte Carlo simulation results show that the total fluorescence detected is significantly higher for the VA probe geometry, while the probing depth is significantly greater for the MD probe geometry. An important observation is that the VA probe is more sensitive to the epithelial layer, while the MD probe is more sensitive to the stromal layer. The effect of other factors, including numerical aperture (NA) and tissue optical properties on the fluorescence measurements with VA and MD probe geometries, are also evaluated. The total fluorescence detected with both probe geometries significantly increases when the fiber NA is changed from 0.22 to 0.37. The sensitivity to different sublayers is found to be strongly dependent on the tissue optical properties. The simulation results are used to design a simple fiber optic probe that combines both the VA and MD geometries to enable fluorescence measurements from the different sublayers within human epithelial tissues.

Optimal methods for fluorescence and diffuse reflectance measurements of tissue biopsy samples

BACKGROUND AND OBJECTIVE

In developing fluorescence spectroscopy systems for the in vivo detection of pre-cancer and cancer, it is often necessary to perform preliminary testing on tissue biopsies. Current standard protocols call for the tissue to be immediately frozen after biopsy and later thawed for spectroscopic analysis, but this process can have profound effects on the spectroscopic properties of tissue. This study investigates the optimal tissue handling methods for in vitro fluorescence spectroscopy studies.

STUDY DESIGN/MATERIALS AND METHODS

The epithelial tissue of the Golden Syrian hamster cheek pouch was used in this study. Three specific experiments were carried out. First, the fluorescence properties of tissues in vivo and of frozen and thawed tissue biopsies were characterized at multiple excitation wavelengths spanning the ultraviolet-visible (UV-VIS) spectrum. Next, comparison of tissue fluorescence emission spectra in vivo, ex vivo (immediately after biopsy), and after the freeze and thaw process were systematically carried out at the excitation wavelengths corresponding to the previously identified fluorescence peaks. Lastly, intensities at the excitation and emission wavelength pairs corresponding to the fluorescence peaks were measured as a function of time after biopsy. Diffuse reflectance measurements over the UV-VIS spectrum were also made to evaluate the effects of oxygenation, blood volume, and scattering on the tissue fluorescence at these different excitation-emission wavelengths.

RESULTS

This study indicates that the freezing and thawing process produces a significant deviation in intensity and lineshape relative to the in vivo fluorescence emission spectral data over the entire UV-VIS range between 300 and 700 nm. By contrast, examination of ex vivo emission spectra reveals that it closely preserves both the intensity and lineshape of the in vivo emission spectra except between 500 and 700 nm. The observed deviations can be explained by the diffuse reflectance measurements, which suggest increased hemoglobin deoxygenation and wavelength dependent changes in scattering in ex vivo tissues, and increased total hemoglobin absorption in the frozen and thawed samples. Furthermore, it was found that over a time window of 1.5 hours, spectroscopic changes brought about by degradation of the tissue due to biopsy or other factors are significantly smaller (10-30% variations in intensity) than those associated with the freezing and thawing process (50-70% decrease in intensity).

CONCLUSIONS

It was found that the effects of freezing and thawing on the fluorescence properties of tissue are greater than any changes brought about by degradation of tissue over a time frame of 90 minutes after biopsy. Performing ex vivo fluorescence measurements within a reasonable time window has the advantage of more accurately reproducing the clinically relevant in vivo conditions in the case of the hamster cheek pouch tissue. Therefore, in tissue biopsy studies, the tissue sample should ideally be maintained in an unfrozen state prior to measurement.

Optical spectroscopy for detection of neoplasia

Fluorescence and reflectance spectroscopy provide the ability to assess tissue structure and metabolism in vivo in real time, providing improved diagnosis of pre-cancerous lesions. Reflectance spectroscopy can probe changes in epithelial nuclei that are important in pre-cancer detection, such as mean nuclear diameter, nuclear size distribution and nuclear refractive index. Fluorescence spectroscopy can probe changes in epithelial cell metabolism, by assessing mitochondrial fluorophores, and epithelial-stromal interactions, by assessing the decrease in collagen crosslink fluorescence that occurs with pre-cancer. Thus, fluorescence and reflectance spectroscopy provide complementary information useful for pre-cancer diagnosis. Tissue engineering provides three-dimensional cell cultures that can be used to further explore the relationship between tissue structure and biological events important in cancer development and progression. In the future, improving our understanding of the biological changes that can be assessed using spectroscopy will not only improve optical techniques but also provide new tools to better understand cancer biology.

Autofluorescence spectroscopy for the diagnosis of cervical intraepithelial neoplasia

OBJECTIVE

To assess the feasibility of autofluorescence spectroscopy in the diagnosis of cervical intraepithelial neoplasia (CIN) using broadband light excitation.

DESIGN

Feasibility study.

SETTING

Colposcopy clinic of an university hospital.

POPULATION

Sixty-eight patients at risk for CIN.

METHODS

After excitation with a broadband light between 375 and 440 nm, spectral distribution of native tissue fluorescence (autofluorescence) was acquired from 685 cervical sites for the localisation and differentiation of CIN, and compared with colposcopically directed biopsy and human papillomavirus (HPV) DNA testing.

MAIN OUTCOME MEASURE

Detection of CIN.

RESULTS

The evaluation of spectral measurements revealed significantly lower autofluorescence values for CIN 3 lesions compared with normal tissue (P < 0.001), and compared with CIN 1 or CIN 2 (P < 0.002). High grade CIN lesions (CIN 2/3) presented with a significant reduced autofluorescence compared with CIN 1 (P < 0.002). Patients with a positive HPV DNA testing showed a significantly lower autofluorescence than patients tested negative for HPV DNA (P < 0.05). Severe inflammation such as chronic cervicitis may lead to false positive results.

CONCLUSIONS

Autofluorescence spectroscopy represents an interesting approach for the detection of cervical neoplasia. Using an excitation wavelength band between 375 and 440 nm, significant differences between normal and precancerous lesions of the cervix can be seen.

Realistic three-dimensional epithelial tissue phantoms for biomedical optics

We introduce new realistic three-dimensional tissue phantoms which can help to understand the optical properties of human epithelium as well as the optical signatures associated with the dysplasia to carcinoma sequence. The phantoms are based on a step by step multilayer reconstitution of the epithelial tissue using main components characteristic for the human epithelium. Each consecutive step is aimed to increase the similarity between real tissue and a phantom. We began by modeling the stromal layer which predominantly consists of a network of collagen bundles. Phantoms consisting of a collagen matrix alone and in the presence of embedded cervical cells were created. Their morphology and fluorescence properties were studied and were compared with those of cervical epithelium. We show that the phantoms resemble the microstructure and the optical properties of the human epithelial tissue. We also demonstrate that the proposed phantoms provide an opportunity to study changes in optical properties of different tissue components as a result of their interactions with each other or exogenous factors.

Fluorescence spectroscopy for detection of malignancy: H-ras overexpressing fibroblasts as a model

Autofluorescence from intracellular chromophores upon illumination of cells by monochromatic light has been studied towards the development of novel noninvasive and sensitive technology for the early detection of cancer. To investigate the relationship between biochemical and morphological changes underlying malignant disease and resulting fluorescence spectra, an in vitro model system of a paired normal and malignant murine fibroblasts cell lines, differing in cancer-associated H-ras expression was employed. A comparison of fluorescence excitation and emission spectra of proliferative cells revealed that fluorescence intensity of malignant cells was significantly less than that of normal cells upon excitation at 290 nm. Fluorescence of both cell lines decreased with decreasing cell concentration, but at each concentration, normal cells had higher fluorescence intensity than malignant cells. Similar differences between the cell lines were observed when brought to quiescence or at stationary phase. Results suggested that the chromophore contributing most significantly to these spectra is tryptophan and its moieties in proteins. This model system demonstrates the specific contribution of H-ras to subcellular chromophores, resulting in a significant difference in their autofluorescence intensity, and implies the potential use of the technique for cancer detection. This model system is potent for analysis of the contribution of other oncogenes and their combinations towards spectral detection of cancer.

Fluorescence spectroscopy for in vivo characterization of ovarian tissue

BACKGROUND AND OBJECTIVE

The objective of this study was to explore whether fluorescence spectroscopy signatures differed between normal variations within the ovary, benign neoplasms, and ovarian cancer.

STUDY DESIGN/MATERIALS AND METHODS

Ovarian tissue fluorescence emission spectra were collected sequentially at 18 excitation wavelengths ranging from 330 to 500 nm from 11 patients undergoing oophorectomy and assembled into fluorescence excitation emission matrices (EEMs); biopsies corresponded to the area interrogated. Spectral areas that could differentiate normal ovary, benign neoplasms, and cancers were evaluated, using histopathology as the reference standard.

RESULTS

The most promising measurements are (1) the integrated fluorescence intensity from 400 to 430 nm excitation at 460 nm emission, and (2) the ratios of fluorescence intensities at 330 nm excitation, 385 and 500 nm emission, and at 375 and 415 nm excitation, 460 nm emission. Simple systems to visualize these optical signatures at laparoscopy could be designed.

CONCLUSION

Fluorescence spectroscopy may have the ability to distinguish ovarian cancers from normal ovarian structures and benign neoplasms, as well as differentiate between normal variations and metaplastic structures and should be further explored as a device for the early detection of ovarian cancers.

Understanding the contributions of NADH and collagen to cervical tissue fluorescence spectra: modeling, measurements, and implications

OBJECTIVE

At 380 nm excitation, cervical tissue fluorescence spectra demonstrate characteristic changes with both patient age and the presence of dysplasia. A Monte Carlo model was developed in order to quantitatively examine how intrinsic NADH and collagen fluorescence, in combination with tissue scattering and absorption properties, yield measured tissue spectra.

METHODS

Excitation-emission matrices were measured for live cervical cells and collagen gel phantoms. Fluorescence microscopy of fresh tissue sections was performed to obtain the location and density of fluorophores as a function of patient age and the presence of dysplasia. A Monte Carlo model was developed which incorporated measurements of fluorophore line shapes and spatial distributions.

RESULTS

Modeled spectra were consistent with clinical measurements and indicate that an increase in NADH fluorescence and decrease in collagen fluorescence create clinically observed differences between normal and dysplastic tissue spectra. Model predictions were most sensitive to patient age and epithelial thickness.

CONCLUSIONS

Monte Carlo techniques provide an important means to investigate the combined contributions of multiple fluorophores to measured emission spectra. The approach will prove increasingly valuable as a more sophisticated understanding of in vivo optical properties is developed.

Autofluorescence characteristics of immortalized and carcinogen-transformed human bronchial epithelial cells

Tissue autofluorescence has been explored as a potential method of noninvasive pre-neoplasia (pre-malignancy) detection in the lung. Here, we report the first studies of intrinsic cellular autofluorescence from SV40 immortalized and distinct tobacco-carcinogen-transformed (malignant) human bronchial epithelial cells. These cell lines are useful models for studies seeking to distinguish between normal and pre-neoplastic human bronchial epithelial cells. The cells were characterized via spectrofluorimetry and confocal fluorescence microscopy. Spectrofluorimetry revealed that tryptophan was the dominant fluorophore. No change in tryptophan emission intensity was observed between immortalized and carcinogen-transformed cells. Confocal autofluorescence microscopy was performed using a highly sensitive, spectrometer-coupled instrument capable of limiting emission detection to specific wavelength ranges. These studies revealed two additional endogenous fluorophores, whose excitation and emission characteristics were consistent with nicotinamide adenine dinucleotide (NADH) and flavins. In immortalized human bronchial epithelial cells, the fluorescence of these species was localized to cytoplasmic granules. In contrast, the carcinogen-transformed cells showed an appreciable decrease in the fluorescence intensity of both NADH and flavins and the punctate, spatial localization of the autofluorescence was lost. The observed autofluorescence decrease was potentially the result of changes in the redox state of the fluorophores. The random cytoplasmic fluorescence pattern found in carcinogen-transformed cells may be attributed to changes in the mitochondrial morphology. The implications of these results to pre-neoplasia detection in the lung are discussed.