MATERIALS AND METHODS

Golden Syrian hamster cheek pouch model is a well established animal model to study DMBA induced oral carcinogenesis. Male golden Syrian hamsters (Mesocricetus auratus, retired breeders 150-200 g) were procured from National Institute of Nutrition, Hyderabad and maintained at the Central Animal House, Rajah Muthiah Institute of Medical Science. The animals were housed in plastic cages under controlled environmental conditions with a 12 h light/dark cycle and had free access to water and standard food. Totally, eighteen animals were taken up for the study. They were grouped as control (n = 6), premalignant group (n = 6) and malignant group (n = 6). A 0.5% solution of 7,12-dimethylbenz(a)anthracene (DMBA, Sigma) in liquid paraffin oil was painted in the right buccal cheek pouch and the left buccal cheek pouch was left untreated which acted as internal control. The DMBA application was done two times a week for first two weeks and then three times a week for 12 weeks. As the painting was started there was increased inflammation with purulent discharge in the buccal pouch. Thus time was given for the inflammation to subside and for the animal to get conditioned to DMBA painting. For the premalignant group animals, as soon as the clinically white patch (leukoplakia) was observed some where around four to six weeks the application of the DMBA was stopped. For the malignant group the DMBA was applied till a clinically apparent tumor was observed at about the end of 14 weeks. Rajah Muthiah Medical College, Annamalai University Institutional Animal Ethics Committee Clearance was obtained.

Fluorescence measurement: The animals from the control, premalignant and malignant group were subjected to fluorescence spectroscopy analysis under ketaminium hydrochloride anesthesia. A fiber optic probe connected to the Varian Fluorescent Spectrophotometer, which was, in turn, connected to the computer, was used for the analysis. The right buccal pouch was pulled out and the fiber optic probe was gently placed over the mucosa of the animal in control group and over the lesion proper in premalignant (leukoplakia) and malignant group. The tissue was excited at 405 nm and was scanned over 420 to 750 nm range of wavelength for the fluorescence emission by the tissue. The scan thus obtained was emission spectra.

Statistical analysis: The emission curve of the control group, premalignant (leukoplakia) and malignant group were averaged to get individual average emission spectra of each group. To find the statistical difference between the average spectrum ratio parameters were introduced. With the obtained ratio parameters through ANOVA the statistical significance was seen between the groups.

RESULTS AND DISCUSSION

Figure 1 shows the typical autofluorescence spectra of the normal, premalignant (leukoplakia) and oral squamous cell carcinoma. The fluorescence intensities gradually decreased with the progression of the scan in normal and premalignant group. There is a sharp increase in the fluorescence intensity forming a prominent peak at around 635 nm and a small peak at around 700 nm seen in the malignant group. No such peaks were observed in the normal and premalignant group. In the normal, premalignant and malignant group a small additional peak around 490 nm was also observed. Mean ratio parameters between 490, 635 and 700 nm viz., I_490/635, I_490/700 and I_635/700 were introduced. An ANOVA test showed significant difference at p<0.05 in the ratio value among all categories. To further identify the relationship between groups, LSD post hoc test was performed (Table 1).

Fig. 1:	Average emission at 405 nm excitation

Table 1:	Results of LSD post hoc test
*Indicates statistically significant with p<0.05

The patient suffering from initial stages of oral cancer usually does not suffer from grave symptoms except for white or red patch with or without ulcerations. The patient presents himself to the dental office usually when the oral cancer is in the advanced stages, thus making it very difficult to treat and have a very poor prognosis. To reduce the mortality rate associated with this disease, it would be prudent to confront this disease at its premalignant or early stages of invasion. Autofluorescence spectroscopy is one such real time optical technique, which is under research for the early diagnosis of cancer. This technique exploits the usage of biomolecules called fluorophores, tissue metabolite like porphyrin which has the property to fluoresce when excited between 400-450 nm of light (Coghlan et al., 2000, 2001). When light of certain wavelength is absorbed by an atom or molecule, an electron is excited to the higher energy level. When these displaced electrons return to the original ground state it may emit a quantum of light, which is known as fluorescence. This technique has a many advantages compared with the non invasive procedures currently available. (1) It can detect subtle changes in the tissue based upon tissue metabolites (porphyrin), (2) both quantitative and qualitative analysis is possible thus avoiding bias and (3) it is simple and very easy to perform. The only disadvantage is that this technique cannot be used to differentiate different pathological types of oral cancers.

In the present in vivo study, the Syrian golden hamsters buccal cheek pouch were used for the induction of tumor by local application of 0.5% solution of 7,12-dimethylbenz(a)anthracene in liquid paraffin. This carcinogenesis animal model is a well established and is used in various chemoprevention experiments due to its advantages. Induction of tumor though various stages of occurrence such as inflammatory reaction, hyperkeratosis, papillary stage seen in human is possible in this animal model. Clinical distinction can be made between premalignant lesion (leukoplakic patch) and malignant oral cancer tumor. The duration of the induction of tumor is relatively less 14 weeks.

Bottiroli et al. (1995), Dhingra et al. (1996) and Vengadesan et al. (1998) worked in this area and have concluded that there is localization of porphyrins in the tumor tissue. But not many in vivo studies are available where autofluorescent property of the oral tissue has been studied comparing the normal, premalignant lesion (leukoplakia) and oral cancer tissue. This has led us to take up the study to compare the fluorescent characteristics of the tissue from normal to malignant transformation. Porphyrin is a metabolic product in the heme biosynthesis. It leads to the synthesis of protoporphyrin IX (PpIX), which acquires an atom form heme. δ-Aminolevulinate synthetase, the enzyme catalyzing the committed step in this pathway, is feedback inhibited by heme. Porphyrin produces a red fluorescence, with peaks at 590, 630 and 690 nm when excited in the blue spectral region between 400 and 450 nm. In this study, the results revealed that, when the cancer tissue was excited at 405 nm there was a prominent peak at 635 nm and another small peak at 700 nm. These two peaks can be attributed to porphyrin fluorescence. As stated above, there is an accumulation of the porphyrin due to the increase in the cellular metabolism, breakdown of RBC`s and increased neovascularization (Onizawa et al., 2002; Dhingra et al., 1996). When compared with the normal control and premalignant group no such peak was found. Vengadesan et al. (1998) and Onizawa et al. (2002) in their in vitro study of malignant animal tissue found prominent porphyrin peak at 630 nm when excited at 405 nm.

The peak around 490 nm which was observed in all the three groups may be attributed to nicotinamide adenine dinucleotide of reduced form (NADH) (Wu et al., 2006). When mean ratio parameters between the NADH and porphyrin peaks were introduced, statistical significance of p-value less than 0.05 was observed between groups and within groups. In the LSD post hoc test, it was evident that all the ratio parameter, I_490/635, I_490/700 and I_635/700 showed significant difference between the normal-malignant groups and premalignant-malignant groups. But the normal-premalignant group showed no difference. Probably some other excitation wavelengths have to be studied to show the difference between the normal-premalignant groups. However, from the results obtained from the present study it is evident that autofluorescence technique shows significant difference between the normal and diseased (malignant) tissue.

CONCLUSION

Autofluorescence spectroscopic analysis of clinically subtle lesions, which occurs long before the apparent lesions, has to be explored. Since this technique is noninvasive, it can help the dental surgeon to detect early, choose the site for biopsy and to mark the boundaries of oral squamous cell carcinoma lesion. Further in vivo studies have to be done in order to exploit the potentiality of this technique for mass screening at the primary clinics.

ACKNOWLEDGMENT

The Authors thank the University Grants Commission, New Delhi, India to support this study under Major Research Project No. F.30-163/2004(SR).