A probability-based multivariate statistical algorithm for autofluorescence spectroscopic identification of oral carcinogenesis.

C Y Wang; C T Chen; C P Chiang; S T Young; S N Chow; H K Chiang

Home/Archive/PubMed

PubMed · 10212579

A probability-based multivariate statistical algorithm for autofluorescence spectroscopic identification of oral carcinogenesis.

C Y Wang·C T Chen·C P Chiang·S T Young·S N Chow·H K Chiang

View original source ↗Export metadata ↓

Abstract

A probability-based multivariate statistical algorithm combining partial least-squares (PLS) and logistic regression was developed to identify the development stages of oral cancer through analysis of autofluorescence spectra of oral tissues. Tissues were taken from a 7,12-dimethylbenz[a]anthracene-induced hamster buccal pouch carcinogenesis model. Analyses were conducted at various excitation wavelengths, ranging from 280 nm to 400 nm in 20 nm increments, to assess classification performance at different excitations. For each excitation the PLS analysis and logistic regression were combined, on the basis of cross validation, to calculate the posterior probabilities of samples belonging to four stages of cancer development: normal tissues, hyperplasia, dysplasia and early cancers and frankly invasive cancers. Results showed that the 320 nm excitation wavelength optimally classified the cancer development stages: the accuracy rates for identifying samples at that excitation were 91.7%, 83.3%, 66.7% and 83.3% for the four respective stages. The average accuracy rate was 81.3%. These results suggest that the algorithm described in this study might be useful for the detection of human oral cancers.

Explore related subjects

9,10-Dimethyl-1,2-benzanthracene Algorithms Animals Carcinogens Cricetinae Humans Male Mesocricetus Mouth Neoplasms Multivariate Analysis Photobiology Spectrometry, Fluorescence Journal Article

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

C Y Wang, C T Chen, C P Chiang, S T Young, S N Chow, H K Chiang. 1999. A probability-based multivariate statistical algorithm for autofluorescence spectroscopic identification of oral carcinogenesis.. https://pubmed.ncbi.nlm.nih.gov/10212579/

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed1999

A functional and quantitative mutational analysis of p53 mutations in yeast indicates strand biases and different roles of mutations in DMBA- and BBN-induced tumors in rats.

In order to analyze the mutational events and to understand the biological significance of the p53 gene in chemical carcinogenesis, we applied a new yeast-based p53 functional assay to ovarian tumors induced by 7, 12-dimethylbenz[a]anthracene (DMBA), as well as to transitional cell carcinomas of the urinary bladder induced by N-butyl-N-(4-hydroxybutyl) nitrosamine (BBN) in rats. The assay demonstrated that 15 of 19 DMBA induced tumors harbored clonal p53 mutations, which is consistent with the expectations of the "clonal expansion" hypothesis. The majority of the mutations were purine (AG) to pyrimidine (CT) transversions (12/19) on the non-transcribed (sense) strand (NTS), which is likely to be due to depurination created by DMBA adduct formation on the NTS. In contrast, we found no pyrimidine to purine [corrected] transversion on the NTS. After cessation of BBN treatment, BBN-induced multifocal lesions in the bladder contained heterogeneous p53 mutations at an early stage. In the later stage, however, clonal p53 mutations were identified in 4 out of 7 bladders analyzed, conforming with the concept of "field cancerization". The observed base substitutions were G-->A (1/6) or C -->T transitions (2/6), and mutations at T (3/6) on the NTS in clonal mutations, together with non-clonal mutations, showing a preference of C-->T to G-->A (17 vs. 0). Thus, preferential repair was found in the transcribed strand of the p53 gene, whether modified by DMBA or by BBN carcinogens. Very similar mutation patterns were observed between clonal and non-clonal mutations in the DMBA- and BBN-induced tumors, indicating that the rat yeast p53 functional assay can be a potential tool for the characterization of in vivo mutation patterns of p53, when modified by chemical carcinogens.

9,10-Dimethyl-1,2-benzanthracene↗

PubMed1999

Skin tumorigenesis by initiators and promoters of different chemical structures in lines of mice selectively bred for resistance (Car-r) or susceptibility (Car-s) to two-stage skin carcinogenesis.

Carcinogenesis-resistant (Car-R) and carcinogenesis-susceptible (Car-S) mice were obtained applying a bi-directional selective breeding approach to a two-stage skin carcinogenesis protocol, using 9,10-dimethyl-1,2-benzanthracene (DMBA) as initiator and 12-O-tetradecanoylphorbol-13-acetate (TPA) as promoter. Sixteen generations of selection produced a remarkable interline difference in responsiveness to two-stage skin carcinogenesis between Car-R and Car-S: identical DMBA (25 microgram) and TPA (5 microgram) doses induced papillomas in 100% of Car-S compared with 3.3% of Car-R mice and maximal responses of 14.3 or 0.03 papillomas/mouse, respectively, despite the shorter promotion applied to Car-S (49 vs. 208 days). To define the factors determining this great difference, Car-R and Car-S mice were challenged by initiators/promoters chemically unrelated to those used for selection. Both lines were subjected to either initiation by N-methyl-N-nitrosourea (MNU) followed by TPA promotion, or promotion by benzoyl peroxide, or 1,8-dihydroxy-3-methyl-9-anthrone (chrysarobin) following DMBA initiation. Initiation with MNU induced a 10-fold tumour incidence in Car-S compared with Car-R mice, and a 32-fold difference in tumour induction rate. The 2 lines also differed markedly in susceptibility to benzoyl peroxide promotion: Car-S mice initiated with 25 microgram DMBA and promoted with 7.5 mg benzoyl peroxide showed a 12-fold tumour incidence and a 103-fold tumour induction rate compared with the corresponding Car-R group. Both lines, however, were refractory to chrysarobin promotion. The progression of papillomas to carcinomas was examined in all Car-S groups. The incidence of mice that developed carcinomas was 57% in MNU-initiated mice. Benzoyl peroxide was also able to promote carcinoma development in Car-S mice, though with a lower incidence (30.4%) than TPA.

9,10-Dimethyl-1,2-benzanthracene↗

PubMed1999

Inhibitory effects of chlorophyllin on 7,12-dimethylbenz[a]anthracene-induced bacterial mutagenesis and mouse skin carcinogenesis.

Chlorophyllin (CHL), a water-soluble derivative of chlorophyll, has been used for the treatment of several abnormal human conditions without apparent toxicity. Recent studies have revealed that CHL has the excellent chemopreventive potential. In the present investigation, we have found the inhibitory activities of CHL against 7,12-dimethylbenz[a]anthracene (DMBA)-induced mutagenesis in Salmonella typhimurium TA100 and also on DMBA-initiated and 12-O-tetradecanoyl-phorbol-13-acetate (TPA)-promoted mouse skin tumor formation. The incidence and the multiplicity of skin tumors were not significantly decreased in mice by a single topical application of CHL prior to the DMBA treatment, but there was a marked suppression of papillomagenesis in mice treated with CHL during the promotional stage. Furthermore, the formation of DMBA-induced papillomagenesis was reduced in all mice that had received CHL for 6 weeks following treatment with TPA for 6, 18 and 24 weeks. These results indicate that CHL can inhibit both tumor promotion and the progression of papillomagenesis in the two-stage mouse skin carcinogenesis induced by DMBA and TPA.

9,10-Dimethyl-1,2-benzanthracene↗