Nucleolar organizer regions (AgNOR) and subepithelialvascularization as field cancerization markers in oralmucosa biopsies of alcoholic and smoking patientsSilvia A. López-Blanc, DDS, PhD,a Ana M. Collet, DDS, PhD,b

Mariana S. Gandolfo, DDS, PhD,b Fabian Femopase, DDS, PhD,a Silvia L. Hernández, DDS,a

Victor H. Tomasi, Ht,b Maria L. Paparella, DDS, PhD,b and Maria Elina Itoiz, DDS, PhD,b

Córdoba and Buenos Aires, ArgentinaNATIONAL UNIVERSITY OF CÓRDOBA AND UNIVERSITY OF BUENOS AIRES

Objective. The aim of this study was to show that variations in nucleolar organizer regions (AgNOR) and the increasein subepithelial vascularization could reveal changes related to markers of field cancerization in alcoholic andsmoking patients who have not yet expressed clinical or histological malignant lesions.Study design. Quantitative variations in epithelial AgNOR and in the vascularization of the underlying connectivetissue were assessed by image analysis in histologically normal biopsy specimens from alcohol drinkers and smokingpatients (DS). AgNORs were evidenced by silver staining and vessel walls were labeled by immunohistochemicaldemonstration of the CD34 antigen. Samples of oral mucosa of nonalcoholic, nonsmoking patients (NDS) obtainedduring surgical procedures served as controls. Eight parameters related to number, volume, and shape of nuclei andAgNORs, and 4 parameters related to number and diameter of vascular sections were evaluated. Differences betweenDS and NDS groups were statistically evaluated by means of ANOVA test and posterior Bonferroni comparisons.Results. The morphometric analysis revealed more irregular-shaped AgNORs in the superficial and suprabasal layersof the oral mucosa of DS patients. The suprabasal layers also exhibited a significantly larger number of AgNORs. Thenormal oral mucosa of DS patients exhibited a greater vascular density, with predominance of small-caliber bloodvessels underlying the basement membrane.Conclusion. The variations in AgNOR and epithelial vascularization would be practical biomarkers to evaluatechanges underlying the augmented risk of cancerization in oral mucosa. (Oral Surg Oral Med Oral Pathol Oral Radiol

Endod 2009;108:747-753)

Alcohol consumption and tobacco smoking are themost important risk factors in oral mucosa cancer. Themutagenic capacity of tobacco has been extensivelystudied.1

Conversely, the carcinogenic mechanisms of alco-hol are poorly understood. Epidemiological studieshave demonstrated a synergistic effect between tobaccosmoking and alcohol consumption in increasing the riskof oral cancer.2-4 This synergism has also been demon-strated in experimental models.5-7 However, chronic alco-hol consumption has been shown to act as an independent

This work was supported by grants from the National ResearchCouncil of Argentina (CONICET) and the University of BuenosAires, Argentina (UBACyT Program).aDepartment of Oral Pathology, Clinical Stomatology, Faculty ofDentistry, National University of Córdoba, Córdoba, Argentina.bDepartment of Oral Pathology, Faculty of Dentistry, University ofBuenos Aires, Argentina.Received for publication Jul 31, 2008; returned for revision May 14,2009; accepted for publication Jun 26, 2009.1079-2104/$ - see front matter© 2009 Published by Mosby, Inc.

doi:10.1016/j.tripleo.2009.06.028

risk factor.8,9 At an experimental level, alcohol was shownto impair DNA repair mechanisms10-13 and directly affectcell membranes, increasing their permeability and absorp-tion capacity.14,15 Furthermore, the primary metabolicproduct of ethanol, acetaldehyde, is mutagenic.8,16,17

The particular action of these carcinogens that exerttheir effect from the surface of the mucosa induces thephenomenon known as “field cancerization.” The con-cept was introduced by Slaughter et al.18 to explain theincreased risk of malignant transformation in large ar-eas of the epithelial lining of the upper aerodigestivetract. This hypothesis was based on the high incidenceof second primary tumors or multifocal cancer and wasproved by the demonstration of molecular changes inclinically healthy mucosa of smoking patients.19,20 Fur-thermore, the sequential or simultaneous developmentof oral premalignant and malignant lesions in a singlepatient evidences progressive genotypic and pheno-typic alterations associated to field cancerization.21

The search for markers of field cancerization beforethe appearance of premalignant morphological alter-ations is of biological interest and clinically relevant in

terms of early diagnosis and prevention of oral cancer.

747

OOOOE748 López-Blanc et al. November 2009

Within this context, those markers that can be demon-strated in routine biopsies processed for histopathologicaldiagnosis are of particular significance. Among these,image analysis of variations in silver-stained nucleolarorganizer regions (AgNORs) in epithelia has proved tobe a marker of field cancerization. Nucleolar organizerregions (NOR) are loops of DNA that code for ribo-somal RNA. DNA transcription and ribosomal RNAsynthesis take place at these sites. Argyrophilic proteinsassociated with NORs are selectively identified by asilver colloid staining technique and visualized as darkintranuclear dots at the optical level.22 Variations innumber, volume, and/or shape of AgNORs in inter-phase nuclei reflect an increased activation of proteinsynthesis as occurs in malignant transformed cells.23

The simplicity of the staining procedure has led to itswidespread application to the evaluation of degree ofmalignancy in human pathology.24 We have shown theefficacy of AgNORs in the detection of early fieldcancerization in the hamster cheek pouch oral cancermodel25 and in epithelia devoid of histological alter-ations in the vicinity of human oral squamous cellcarcinoma.26 Variations in AgNORs have been demon-strated in cytological smears of alcoholic patients27,28

and in sections of oral mucosa of alcoholized mice.29

Thus, it is contributory to evaluate this marker in rou-tine biopsies of human oral mucosa at a greater risk ofmalignant transformation that has not yet developedtumors.

Angiogenesis induced by factors secreted by epithe-lia undergoing malignant transformation occurs in earlydysplastic lesions and in the oral mucosa of tumor-bearing patients.30-32 Thus, in absence of inflammation,an increase in vascularization would indicate an in-

Table I. Clinical data of patients included in the study

Patients Age

A

Grams/day

1 68 2402 30 8003 44 2104 65 4005 49 3856 51 1207 63 1608 51 1639 46 235

10 23 21311 42 16012 37 228

Mean � SD 47.42 � 13.07 276.17 � 185.73

All patients were male.

creased risk of malignant transformation.33

MATERIALS AND METHODSFifty-five biopsies of alcoholic and smoking patients

who attended the Instituto Provincial de Alcoholismo yDrogadicción (Institute for the Treatment of Alcohol-ism and Drug Addiction) of the province of Córdoba,Argentina, were evaluated. The biopsies were per-formed for histopathological diagnosis of different oralmucosa lesions during treatment for addiction underinformed consent. Patients bearing oral tumors wereexcluded from the study. Because inflammatory reac-tions can induce epithelial and vascular alterations,only a group of 12 buccal mucosa specimens containingat least 2.5 mm in length of epithelium without histo-logical changes or underlying inflammatory infiltrationwere collected and used for the present study as amodel of possible early cancerized field. Table I showsclinical data and alcohol and tobacco consumption val-ues for each patient. The control group comprised 15biopsies of healthy buccal mucosa taken from non-alcoholic, nonsmoking patients during surgery fordeep-seated lesions. Thus biopsies of both control andalcoholic-tobacco groups were of normal microscopicappearance. Mean age of the alcoholic-smoking groupwas 47.42 � 13.07 years and 40.75 � 11.20 in controlgroup.

All the biopsies were fixed in buffered formalin(pH 7.0) and embedded in paraffin. Serial, 5- to7-�m thick sections were obtained for hematoxylinand eosin (H&E) staining and AgNOR silver staining.Fifteen-micrometer sections were obtained for vascu-larization studies.

AgNOR demonstration was performed in keepingwith Howat et al.’s technique,34 slightly modified by usto improve the accuracy of image analysis.35 Briefly,

Tobacco

Years Cigarettes/day Years

48 20 258 0 0

28 5 3048 40 5133 20 3821 30 2044 15 4831 11 3128 17 4416 10 1121 14 2919 20 19

28.75 � 12.81 6.83 � 10.68 28.83 � 15.13

lcohol

sections were immersed in a solution of 7% nitric acid

OOOOEVolume 108, Number 5 López-Blanc et al. 749

for 5 minutes. The acid was then removed by washingin running water. The AgNOR staining solution wasprepared by mixing 1:2 volumes of 2% aqueous solu-tion of gelatine and 50% aqueous silver nitrate solution.Sections were stained under safe light conditions for 40to 45 minutes. After washing, sections were dehydratedand mounted.

The vessel walls were labeled by immunohistochem-ical demonstration of CD 34 antigen. Antigen recoverywas performed with citrate buffer (pH 6.0) in a micro-wave oven for 2 minutes. The sections were incubatedwith an antihuman CD 34 antibody raised in mouse asthe primary antibody (QBEnd/10; Biogenex, San Ramon,CA, USA) for 24 hours at 4°C in a wet chamber, followedby labeling with a biotin-streptavidin-peroxidase detectionkit (Biogenex). A light nuclear counterstain was per-formed with hematoxylin. Sections of biopsies fromstrongly vascularized human tongue muscle were usedas positive controls. Negative controls were performedby replacing the primary antibody by mouse normalserum.

Quantitative evaluationQuantitative evaluation of AgNORs was performed

with image analysis software previously developed “adhoc” by our laboratory36 and validated as a useful toolin the detection of variations in terms of number, shape,and size of AgNORs in tissue sections.25,26 The soft-ware was programmed in keeping with the standardizedinternational protocols in which clusters of AgNORswithin a nucleolus are treated as a single structure so asthe number of intranuclear AgNORs is the sum ofnucleoli and dispersed AgNORs.23,37

In each case, AgNORs were evaluated separately in100 nuclei in each of the basal, suprabasal, and remain-ing layers of the epithelia, excluding the stratum cor-neum. The digital images were obtained with an im-mersion objective in a Zeiss MPM800 microscope(Carl Zeiss, Jena, Germany), online with an imageanalyzer (IBAS, Zeiss-Kontron, Carl Zeiss).

The following morphometric parameters were eval-uated: nuclear volume (VNUC), single AgNOR volume(VNOR), total AgNOR volume per nucleus (TVNOR),proportion of nuclear volume occupied by AgNORs(TVNOR/VNUC), number of AgNORs per nucleus(nNOR), and shape index (perimeter/�area) of nuclei(SNuc) and of AgNOR (SNOR), which is an expressionof shape irregularities (the minimum value of 3.54corresponds to a perfect circle).

Another software was developed “ad hoc” for theevaluation of the vascularization in terms of numberand diameter of vascular sections in predeterminedareas of 155 � 155 �m of connective tissue and,

separately, in a 30-�m band of connective tissue, im-

mediately below the epithelium. These fields were mea-sured separately to detect potential neovascularizationinduced by epithelial factors. The software also allowsfor discrimination of blood vessels with a diameterbelow 8 �. This value was arbitrarily chosen as repre-sentative of the smallest capillaries on the basis of theerythrocyte diameter, which in histologic sections isapproximately of 7 �, and assuming that the smallestvessels are those of more recent formation. Accordingto the amount of tissue available in each biopsy spec-imen, 15 to 20 areas were measured in each case.

The following parameters were evaluated:

–Nt (total number of vessels): number of vesselsections per 10,000 �m2 of connective tissue.

–% sV (percentage of small vessels): percentage oftotal vessel sections with a diameter below 8 �m.

–Nbv (number of sub-basal vessels): number of sub-basal blood vessel sections in a 30-�m band ofconnective tissue, immediately below the basalmembrane. Values are expressed as number ofvascular sections per 100 �m of basal membrane.

–% sbv (percentage of small sub-basal vessels): per-centage of total sub-basal vessel sections with adiameter below 8 �m.

The first 2 parameters afford information on the totalvascularization in the selected sample, whereas the last2 parameters afford information on the vascularizationimmediately below the basal membrane.

Differences between DS and NDS groups were eval-uated by means of analysis of variance (ANOVA) testand posterior Bonferroni comparisons.

RESULTSThe light microscopy analysis of routine prepara-

tions (stained with H&E) did not show significant dif-ferences between alcoholic and smoking patients andcontrols. However, some cases exhibited a more vas-cularized subepithelial connective tissue with no con-gestion, edema, or leukocyte infiltration.

Conversely, the subjective analysis of silver-stainedsections revealed more irregular shapes and more abun-dant intranuclear dots (Fig. 1).

The morphometric analysis revealed statistically sig-nificant differences between alcoholic and smoking pa-tients and controls in the shape index of AgNORs in thesuperficial and suprabasal epithelial layers. The supra-basal layers exhibited a statistically significant increasein number of AgNORs. The parameters single AgNORvolume and total AgNOR volume per nucleus exhibitedhigher values in alcoholic and smoking patients. How-ever, this difference did not reach statistical signifi-

cance (Table II).

n DS a

OOOOE750 López-Blanc et al. November 2009

CD34 immunolabeling clearly identified the contoursof the vessel walls. The faint nuclear stain provides ade-

Fig. 1. AgNORs of normal oral mucosa of nondrinker/nonsmSuperficial (c) and suprabasal layers (d) exhibited more nummagnification �1000.

Table II. AgNOR values in oral mucosa of drinker/smBasal layer

DS NDS

VNUC 22.061 � 5.144 25.949 � 5.263 26.440VNOR 0.749 � 0.273 0.927 � 0.265 1.875TVNOR 1.396 � 0.440 1.527 � 0.546 2.031TVNOR/VNUC 0.009 � 0.005 0.008 � 0.003 0.007nNOR 1.962 � 0.373 1.754 � 0.650 2.301SNuc 4.195 � 0.097 4.285 � 0.149 4.249SNOR 3.974 � 0.157 3.998 � 0.092 4.313

Values are means � SD.AgNOR, nucleolar organizer regions; VNVC, nuclear volume; VNOTVNOR/VNUC, proportion of nuclear volume occupied by AgNOR;shape index of AgNOR.*P � .05. Analysis of variance and Bonferroni comparisons betwee

quate contrast for image analysis. All vascular contours

were labeled (Fig. 2). Quantitative evaluation revealed anoverall increase in vascularization in the mucous mem-

patients (a and b) and of drinker/smoker patients (c and d).and irregular AgNORs in drinker/smoker patients. Original

(DS) and non-drinker/smoker (NDS) patientsprabasal layer Superficial layer

NDS DS NDS

78 29.398 � 6.402 33.522 � 9.972 36.621 � 11.04753 0.962 � 0.358 1.270 � 0.384 1.192 � 0.41665 2.015 � 1.527 2.688 � 0.791 2.229 � 0.67003 0.007 � 0.003 0.062 � 0.188 0.008 � 0.00394* 1.747 � 0.455* 2.278 � 0.588 1.977 � 0.36802 4.264 � 0.156 4.314 � 0.167 4.297 � 0.19452* 4.111 � 0.201* 4.438 � 0.230* 4.232 � 0.182*

gle AgNOR volume; TVNOR, total AgNOR volume per nucleus;number of AgNOR per nucleus; SNuc, shape index of nuclei; SNOR,

nd NDS groups.

okererous

okerSu

DS

� 6.4� 3.3� 0.9� 0.0� 0.6� 0.1� 0.2

R, sinnNOR,

branes of alcoholic and smoking patients (Table III, Nt).

magni

OOOOEVolume 108, Number 5 López-Blanc et al. 751

These differences became highly statistically significantfor vessels with a diameter below 8 �m (Table III, % sv).Furthermore, total subepithelial vessels and small-calibersub-basal vessels exhibited a statistically significant in-crease (Table III, Nbv and % sbV).

DISCUSSIONMorphological variations in AgNORs indicate changes

in cellular activity in terms of protein synthesis. Thisexplains that, under different stimuli, early cellularactivity can be detected before any morphological vari-ations appear in the routine histologic preparations.This fact was clearly observed during the experimentalproduction of epithelial hyperplasia reactive to skinlocal irradiation.38 The validity of AgNOR as a marker

Fig. 2. Immunohistochemical expression of antigen CD24 insubjects (a) and drinker/smoker patients (b). The drinker/smoof small-caliber vessels below the basal membrane. Original

Table III. Values of parameters related to vascularityin oral mucosa of drinker/smoker (DS) and non-drink-er/smoker (NDS) patients

DS NDS

Nt 6.52 � 2.52* 2.92 � 3.27*% sV 95.01 � 5.06† 40.50 � 30.21†Nbv 2.26 � 0.63† 1.00 � 0.78†% sbv 95.01 � 5.06† 58.87 � 35.15†

Values are means � SD.Analysis of variance and Bonferroni comparisons between DS andNDS groups.Nt, number of total vessels in 10.000 �2; % sV, % of blood vessels� 8 �m; Nvb, number of sub-basal blood vessels/100 �m basalmembrane; % sbv, % sub-basal blood vessels � 8 �m of totalsub-basal blood vessels.*P � .05.†P � .0005.

of malignant transformation and grade of malignancy

has been widely demonstrated in different neoplasticand preneoplastic entities.39-42 Similar to virtually allthe histochemical markers of malignancy available,changes in AgNOR evidence alterations in prolifera-tion, differentiation, and/or other metabolic alterationsassociated with malignant transformation. The identi-fication of features that are exclusive to malignanttransformation in routine biopsies is virtually re-stricted to the detection of aneuploid cells.43 How-ever, this technique is considerably more complex.Quantitative evaluation of AgNOR has allowed us todetect incipient alterations associated with malignanttransformation in experimental models25 and in areasof field cancerization in human oral mucosa.26 Inboth studies, AgNORs were evaluated in histologi-cally normal epithelia in the vicinity of carcinomas.These areas are widely accepted as models of fieldcancerization.44,45 However, it has been argued thatthe malignant tumor can produce paracrine effectson the mucosa, inducing metabolic changes that re-flect the influence of the tumor rather than the actualstatus of field cancerization.46 The present studyaddresses this issue further, demonstrating that Ag-NORs mark subclinical and subhistological alter-ations caused by alcohol and tobacco in the oralmucosa of patients without tumor. Similarly, Paiva etal.27 and Gedoz et al.28 described AgNOR numberand area increments in exfoliated cytology samplesfrom alcoholic and smoker patients that were inter-preted as increments of proliferation. Interestingly,the most significant variations in our biopsies werefound in parabasal cells, in contrast to our findings intumor-bearing oral mucosa26,27,36 and in other con-

lood vessel walls. Normal mucosa of nondrinker/nonsmokertients exhibited greater vascularization with a predominancefication �400.

the bker pa

ditions of epithelial injury38,47 in which the greatest

OOOOE752 López-Blanc et al. November 2009

differences in AgNOR were detected in basal cells.In basal cells, the alterations would be mainly attrib-utable to the action of proliferative stimuli, whereasin the suprabasal layers the alterations in number andshape of AgNOR could reflect metabolic alterationsinduced from the surface by tobacco and alcohol.

The capacity to induce vascularization has been rec-ognized as one of the most important characteristics oftumor cells. The development of angiogenic activitycan occur at any time during the neoplastic process. Insome entities, the expression of angiogenic activityoccurs in the early stages of malignant transformation,anticipating the development of numerous preneoplas-tic populations. Within this context, angiogenic activitycould be an early marker of malignant transforma-tion.48-50

El-Gazzar et al.33 reported vascularity-based evi-dence of field cancerization using biopsies of mouthmucosa from patients with oral carcinomas. They alsofound an increased amount of vessels in patients whowere smokers/drinkers, but failed to demonstrate dif-ferences between nonsmokers/drinkers and smokers/drinkers in the control group of no-cancer patients. Thepresent study shows that the use of tobacco and alcoholcan increase vascularization “per se,” conceivably bystimulating epithelial cells to produce angiogenic fac-tors. Because the software we used allowed us to dis-criminate and evaluate separately the vascularizationimmediately below the epithelium in areas not sub-jected to inflammatory factors, we were able to obtaindata that supports this hypothesis.

Detection of AgNOR and vessel walls requires sim-ple techniques that are easy to perform in conventionalsurgical pathology laboratories. Even without imageanalysis, subjective evaluation by trained pathologistscould contribute to suspect field cancerization and thesubsequent increment of transformation risk. In thisway, these biomarkers would be a valuable diagnosticand prognostic tool and might constitute an element ofpersuasion for patients with harmful habits to adoptpreventive measures.

REFERENCES1. DeMarini DM. Genotoxicity of tobacco smoke and tobacco

smoke condensate: a review. Mutat Res 2004;567:447-74.2. Schottenfeld D. Alcohol as a co-factor in the etiology of cancer.

Cancer 1979;43:1962-6.3. Weinberg M, Estefan D. Assessing oral malignancies. Am Fam

Physician 2002;65:1379-84.4. Harris C, Warnakulasuriya K, Cooper D, Peters T, Gelbier.

Prevalence of oral mucosal lesions in alcohol misusers in southLondon. J Oral Pathol Med 2004;33:253-9.

5. Elzay R. Local effect of alcohol in combination with DMBA onhamster cheek pouch. J Den Res 1966;45:1788-95.

6. Lin Y, Ho I, Lee T. Ethanol and acetaldehyde potentiate the

clastogenicity of ultraviolet light, methyl methanesulfonate, mi-

tomycin C and bleomycin in Chinese hamster ovary cells. MutatRes 1989;216:93-9.

7. Simanowsky U, Stickel F, Maier H, Gärtner U, Seitz K. Effectof alcohol on gastrointestinal cell regeneration as a possible mech-anism in alcohol-associated carcinogenesis. Alcohol 1995;12:111-5.

8. Riedel F, Goessler U, Hörmann K. Alcohol-related diseases ofthe mouth and throat. Best Pract Res Clin Gastroenterol2003;17:543-55.

9. Wight A, Ogden G. Possible mechanisms by which alcohol mayinfluence the development of oral cancer. A review. Oral Oncol1998;34:441-7.

10. Garro A, Espina N, Farinati F, Salvagnini S. The effects ofchronic ethanol consumption on carcinogen metabolism and onO6- methylguanine transferase-mediated repair of alkylatedDNA. Alcohol Clin Exp Res 1986;10:73-7S.

11. Mufti S, Salvagnini M, Lieber C, Garro A. Chronic ethanol con-sumption inhibits repair of dimethylnitrosamine-induced DNA al-kylation. Biochem Biophys Res Comm 1988;152:423-31.

12. Hsu T, Furlong C, Spitz M. Ethyl alcohol as a cocarcinogen withspecial reference to the aerodigestive tract: a cytogenetic study.Anticancer Res 1991;11:1097-1101.

13. Mufti S. Alcohol acts to promote incidence of tumours. CancerDetect Prev 1992;16:157-62.

14. Axford S, Ogden G, Stewart A, Saleh A, Ross P, Hopwood D.Fluid phase within buccal mucosal cells of alcohol misusers. OralOncol 1999;35:86-92.

15. Howie N, Trigkas T, Cruchley AT, Wertz P, Squier CA, Wil-liams DM. Short-term exposure to alcohol increases the perme-ability of human oral mucosa. Oral Dis 2001;7:349-54.

16. Korte A, Obe G. Influence of chronic ethanol uptake and acuteacetaldehyde treatment on the chromosomes of bone-marrowcells and peripheral lymphocytes of Chinese hamsters. Mutat Res1981;88:389-95.

17. Obe G, Jonas R, Schmidt S. Metabolism of ethanol in vitroproduces a compound which induced sister-chromatid exchangesin human peripheral lymphocytes in vitro: acetaldehyde notethanol is mutagenic. Mutat Res 1986;174:47-51.

18. Slaughter D, Southwick H, Smejtal W. “Field cancerization” inoral stratified squamous epithelium: clinical implications of mul-ticentric origin. Cancer 1953;6:963-8.

19. Brennan JA, Boyle JO, Koch WM, Goodman SN, Hruban RH,Eby YS, et al. Association between cigarette smoking and mu-tation of the p53 gene in squamous cell carcinoma of the headand neck. N Engl J Med 1995;332:712-7.

20. Braakhuis BJM, Tabor MP, Kummer JA, Leemans CR, Braken-hoff RH. A genetic explanation of Slaughter’s concept of fieldcancerization. Cancer Res 1999;152:113-8.

21. Tabor M, Brakenhoff H, Ruijter-Schippers H, Van Der Wal J,Snow G, Leemans C, et al. Persistence of genetically alteredfields in head and neck cancer patients: biological and clinicalimplications. Clin Cancer Res 2001;7:1523-32.

22. Crocker J, Skilbeck N. Nucleolar organizer region associatedproteins in cutaneous melanotic lesions: a quantitative study.J Clin Pathol 1987;40:885-9.

23. Crocker J, Boldy A, Egan M. How should we count AgNORs?Proposals for a standardized approach. J Pathol 1989;158:185-8.

24. Derenzini M, Ploton D. Interphase nucleolar organizer regions incancer cells. Int Rev Exp Pathol 1991;32:149-92.

25. Schwint A, Folco A, Morales A, Cabrini R, Itoiz M. AgNORmark epithelial foci in malignant transformation in hamstercheek pouch carcinogenesis. J Oral Pathol Med 1996;25:20-4.

26. Schwint A, Savino T, Lanfranchi H, Marschoff E, Cabrini R,

Itoiz M. Nucleolar organizer regions in lining epithelium adja-

OOOOEVolume 108, Number 5 López-Blanc et al. 753

cent to squamous cell carcinoma of human oral mucosa. Cancer1994;73:2674-9.

27. Paiva RL, Sant’Ana MF, Bohrer IS, Rados PV. AgNOR quantifi-cation in cells of normal oral mucosa exposed to smoking andalcohol. A cytopathologic study. Anal Quant Cytol Histol 2004;26:175-80.

28. Gedoz L, Lauxen IS, Sant’Ana MF, Rados PV. Proliferativeactivity in clinically healthy oral mucosa exposed to tobaccosmoking and alcohol: a longitudinal study using the AgNORstaining technique. Anal Quant Cytol Histol 2007;29:231-8.

29. Carrard VC, Filho MS, Rados PV, Chaves AC, Lauxen IS.Quantification of silver-staining nucleolar organizer region inepithelial cells of tongue of mice after exposure to, or intake of,alcohol. Alcohol 2004;34:233-8.

30. Macluskey M, Chandrachud L, Pazouki S, Green M, Chisholm,Odgen G, et al. Apoptosis, proliferation and angiogenesis in oraltissues. Possible relevance to tumour progression. J Pathol2000;191:368-75.

31. Lopez de Cicco R, Watson J, Bassi D, Litwin S, Klein-Szanto A.Simultaneous expression of furina and vascular endothelialgrowth factor in human oral tongue squamous cell carcinomaprogression. Clin Cancer Res 2004;10:4480-8.

32. Johnstone S, Logan R. The role of vascular endothelial growthfactor (VEGF) in oral dysplasia and oral squamous cell carci-noma. Oral Oncol 2006;42:337-42.

33. El-Gazzar R, Macluskey M, Odgen G. Evidence for a fieldchange effect based on angiogenesis in the oral mucosa. A briefreport. Oral Oncol 2005;41:25-30.

34. Howat A, Giri D, Cotton D, Slater D. Nucleolar organizerregions in Spitz nevi and malignant melanomas. Cancer 1989;63:474-8.

35. Orrea S, Tomasi VH, Schwint A, Itoiz M. Modifications to thesilver staining technique for nucleolar organizer regions to im-prove the accuracy of image analysis. Biotech Histochem2001;76:67-73.

36. Cabrini R, Schwint A, Mendez A, Femopase F, Lanfranchi H,Itoiz M. Morphometric study of nucleolar organizer regions inhuman oral normal mucosa, papilloma and squamous cell carci-noma. J Oral Pathol Med 1992;21:275-9.

37. Aubele M, Biesterfeld S, Derenzini M, Hufnagl P, Martín H,Öfner D, et al. Guidelines of AgNOR quantitation. Committee onAgNOR Quantitation within the European Society of Pathology.Zentralbl Pathol 1994;140:107-8.

38. Schwint A, Gomez E, Itoiz M, Cabrini R. Nucleolar organizerregions as a marker of incipient cellular alterations in squamous

epithelium. J Dent Res 1993;72:1233-6.

39. Reeves B, Casey, Harris H. Variations in the activity ofnucleolar organizer in different tissues, demonstrated by silverstaining in human normal and leukemic cells. Cancer GenCytogen 1982;6:223-30.

40. Murty V, Mitra A, Sharma J, Luthra U. Nucleolar organizerregions in patients with precancerous and cancerous lesions ofthe uterine cervix. Cancer Gen Cytogen 1985;18:275-9.

41. Egan M, Crocker J. Nucleolar organizer regions in cutaneoustumors. J Pathol 1988;154:247-53.

42. Crocker J. Nucleolar organizer regions. Curr Top Pathol1990;82:91-149.

43. Böcking A. DNA-cytometric diagnosis of prospective malig-nancy in borderline lesions of the uterine cervix. Cancer Res1991;122:106-15.

44. Greaves P, Filipe M, Branfoot A. Transitional mucosa and sur-vival in human colorectal cancer. Cancer 1980;46:764-70.

45. Lawson M, White L, Coyle P, Butler R, Roberts-Thomson I,Conyers A. An assessment of proliferative and enzyme activ-ity in transitional mucosa adjacent to colonic cancer. Cancer1989;64:1061-6.

46. Coffey R, Shipley G, Moses H. Production of transforminggrowth factors by human colon cancer lines. Cancer Res1986;46:1164-9.

47. Ielmini M, Heber E, Schwint A, Cabrini R, Itoiz M. AgNOR aresensitive markers of radiation lesions in squamous epithelia. JDent Res 2000;79:850-6.

48. Gimbrone M, Gullino P. Angiogenic capacity of preneoplasticlesions of the murine mammary gland as a marker of neoplastictransformation. Cancer Res 1976;36:2611-20.

49. Pazouki S, Chisholm D, Adi M, Carmichael G, Fraquharson M,Odgen G, et al. The association between tumour progression andvascularity in the oral mucosa. J Pathol 1997;183:39-43.

50. Sawatsubashi M, Yamada T, Fukushima N, Mizokami H, Toku-naga O, Shin T. Association of vascular endothelial growthfactor and mast cell with angiogenesis in laryngeal squamous cellcarcinoma. Virchows Arch 2000;436:243-8.

Reprint requests:

Maria Elina Etoiz, PhDUniversity of Buenos AiresOral PathologyM.T.de Alvear 21421428 Buenos AiresBuenos Aires, Argentina

mitoiz@odon.uba.ar