Tobacco Documents Online

Detection of Human Papillomavirus DNA in C.vtologic Specimens Derived from Esoph ageal Precancer Lesions and Cancer F. CHANG. O. SHEN. J. ZHOU. C. WANG. D. WANG, S. S'YR..1/~NEN & K. SYR~.-~NEN Dept. of Pathology. Kuopio Cancer Research Comte. Univc~'~ity of Kuoplo. Kuopio. Finland. and Dept. of Pathology. Henna Medical Uni'.'ersity. Zhen~hou. Henan. People's Republic of China Chang F. Shen O. Zhou J. Wang C, Wang D. Syrjanen S, Syrj~lnen K. Detection of human papillomavims DNA in H'tologlc specimens derived from esophageal pro- cancer lesions and cancer. Stand ] Gaslroenterol 199(I. 2_¢. 3~_~3_~.~ A series of 80 esophageal H'~ologic specimens derived from the same number of patients with previously din_noosed squamous cell dysplasia of the e~ophagus ,*'ere examined for the presence o1" human papillomavirus I'HPV) infection by filter in sltu hybridization (FISH). using a mixed DNA probe conlalning HPV ~'pes 11.16. and 18. All the patients came from an arc~ at high risk for esophageal cancer in China. A total of 53 cases (66.3c~J wer~ d~-m'~l~a'at,'d as HPV-DNA-positive. HPV DNA was detected in 22.2% (2 of 9} d the parenls without cylologic at.vpia, in 50% 13 of 6) with mild d.~plasia, in 80.6%. {25 o1"31) with moderate dysplasia, in 67.92 (19 of 28) with severe dysplada, and in 66.7% {4 of 6) with an invasive ~uamous ,','11 carcinoma. The present results eon~m ~ recent findings on HPV invovlemant in esophageal squamous c¢!1 lesions. They stEoport tic ~'pothcsis that HPV is a possible etiologic agent in nsoph~eal carcinogenesis, most probably acting synergistically with physical, chemical, and/or nutritional factors that ha','e previo~ly been related with this malignant" in the high-risk are~ of C~ina. K¢.r words: Esophageal carcinogenesis: filter in situ hybridization: human papil- Iomavirus Fuju Chang. M.D.. Dept of Pathology, Unive~sily of Kuopio. P.O. Box 6, SF-70211 Kuopto. Finland Esophageal cancer shows a striking geographic variation in the world. Its incidence remains remarkably high,in certain romans, including China. Iran, and South Africa (I.-~). The reasons |or~he high morbidity and mortality in these areas are still unknown. It is currently accepted that some of the recog~.~ized risk factors for this cancer. such as alcohol and. tobacco, play a minor etiologic role only {I. 4). !In ..Chin~:..suspicion has been focused on spe~cific nutrition'-q deficiencies. includin.z those o.f ~'ilamins A. B. and C and certain minerals, a~nd on nitrosamines formed in mouldy foodstuffs (l h In lran the same nu- trilional deficiencies were also rimed, and. in addition, opium tat was blamed (2.4h ltowever. it seems extremely unlikely that these factors alone could offer a satisfactoq" explanation for such a high mortallt2,." and moffridity of this malig- nant' in these high-risk areas. In many instances, a high incidence o1: par- ticularly malignant neoplasms in ~'ell-defined .¢raphic areas seems to be associated with viral infections as a possible etiolo~c factor. This has been shown to be the case with hepatitis B virus (HBV). Epstein-Bert ~'irus {EBV). and human T- lymphotroptc v~rus (HTLV- 1 ) (3.51. The strikint, ~eographic distribution of esophageal cancc~ ~i~t reprc.~nt another example of a malignant ~mor with possible viral etiology t6-9). ~Strong ~vldence has been accumulated in the BATCo document for Mayo Clinic 28 March 2002

pa~t few years implicating an etiolo~c role of HPV infection in the de~elopment of [nt~epi- .thclial neo~I~i~ and, squamou~ cell ~noma. pani~larly in the ]o~er genhal t~. in the skin ithat is. epide~odysplasla v¢~cifo~is), and HPV infe~ion of the esophagus wm fi~t su~- gos~ed by mo~ho]o~ic studies showing HPV- induced cytopathic changes (6. I~15). and la~er con~rm~ ~ immu~ohis~ochemi~l te~niques ~mons~rating HPV~andgens in squamous ~II papillomas [l~lS). Sub~quently. DNA ~'b~d- ~adon s~udies di~Io~ed HPV DNA sequences in ~h bcni~ and mallgnanz esophage~ lesio~ f[6. 17). ~m. subs~ntial es'idence h~ a~mu- la:ed m ~uppo~ the~concep~ that HPV infe~ion ~ mvoI~d in the pa~ho~enesis of ~ageal ~uamous cell c~cinoma as well (~9. To eiu~da~e fun~ ~h~ role of HPV infe~ion in the development Of e~phag~al caner, a ~es of es~hageal ~'tolo~c samples obtained ~om priggishly follo~ed up patients b~on~ng to ~he ~-~k ~pul~ion {or e~ph~eal ~ncer China were angled for the p~nce o~ HPV DNA ~" ~her in sire hyb~dization (~SH). MATERIALS AND METHODS Esophageal balloon, c.vtology examination All cytologic smears were obtained from patients who had: been diagnosed as hating dysplasia of esophageal squamous epithelium 3 3"ears earlier. All :the patients came from the suburbs of Hebi city. a known high-risk area for esophageal cancer in China. Most of them are the subiects of our.follow-up study of nutritional intervention in esophageal dysplasia and have been treated with L:~mpound riboflavin for 2 ye~Lrs [Q. ghen. D. Wan.~. C. Wang, F. Chang. Unpub- I/shed observationg). Esophageal ball.on cvtolo~, examination was carried out on eac,,It patient, and four smears from each of them we,re obtained for routine Papa- nicoiaou IPap) diagnosis. The esophageal epi- thelial changes w.'ere graded as described pre- viously (Ig. 19). Ih brief, negative represents the smears containing only morphologically normal cells or cells sh6wing benign changes due to inflammation. Mild d.~plasla indicates a mild d~gree of h.vperchromasia and convening of the chromatin: the nuclei of mildly dysplasdc cells are t~o or more buc less ~han three times greater than normal cells in the same layer. Severe dysplasia shows more striking hyperchromasia and coar- sening of ~he chromatin: the nuclei are four or more times greater tban normal. Th • nuorpholo~c features of moderate dysplasia fall betv,'een those of mild and severe dysplasias, gquamous cell car- cinoma was diagnosed when cellular morphology clearly indicated malignancy, such as marked hyperchromasia and coarsening of chromatin. ~hickening of the nuclear membrane, and varied sizes of the cells and their nuclei. Some of the patients, especially those cytologically showing cancerous changes, were further confirmed b.~ endoscopic examination with biopc-ies. Preparation of cell suspension After the direct smear, the balloon was washed in 10 ml phosphate-buffered saline ~PB$). pH 7,2. "['he cell suspension was centrifuged for 10 min at 30¢0 rpm. and cell pellets were resuspended in ! ml of PBS. Ceils in the suspension were counted. and 2-6 × I0"~ ceils were filtered onto a nitro- cellulose membrane. In addition, a mixture of HPV I1.16. and 18 DNA (each about 250pg) was spotted onto the membranes as a positive control. The filters were placed on a Whatman 3-ram paper, which was soaked with 1.5 M NaCI:0.5 M NaOH twice for 5 rain to lyse the ceils and de- nature the DNA. After neutralization in 0.5 .M Tris-HCI. pH 7.0/3 M NaCI for l 0 rain. the filters were baked at E~°C for .,t h. Probe preparation and hybridization HPV 11, 16. and 18 cloned in pBR322 were generously provided by Dr. Prof. H. zur Hausen. DKFZo Heidelberg, FRO. HPV insert sequences were separated from the vector by digestion appropriate endonuclease restriction and purified by agnrose gel electrophoresis. A mixture of HPV 11.16. and 18 DNA was labeled by nick trans- lation with ~-~P dCTP (3000 Ci/mmol: Amersham. U.K.) to a specific activity of greater than 2 x 10s cpm/~g DNA. 0 BATCo document for Mayo Clinic 28 March 2002

The membrnne.'~ were ~r+h.vbridi~,ed for 42=C in 20 ml bf hyh~diz~tion mixture contain- ing a final conc~nl+adon o~ ~(1~ fo~amidc. 5 x SSC. + ~ Dcn~ardt solution. ~d ~ ~g m] o~ denatured sal~on w~ pcfform~ under hi[h stringency conditions (Tm-20~ in 10 ml of hybridization mixture with 2~ 10~cpm/mi of ~diolabeled HPV DNA prob~ for ~ h at 42:C. Aher hybridizalion ~he filte~ were washed ~'icc for 15 sin a~ 25:C in 2 x SSC comai~ng 0.1% ~dium ~odecyi sulfate (SDS) and [hrce times for 30 min at 65~C in ~ x SSC containing 0.!%SDS. ~e memb~nes were auto~di~ ~phcd by using Koni~ X-film and imensi~ng ~ee~ for 5 d~ys RESULTS Four Pap smefirs from each patient were exam- ined for the pr,-sence of ~'tologic abnormalities in accordance with the previously described criteria (18. 19). Nine patients were dia~osed as having normal squarnous epithelia: 6 patients, mild dysplasia: 31 patients, moderate dysplasia: 28 patiems, severi., dysplasia: and 6 patients, squa- mous cell card~oma. As compared ,,~th rbe latest ~'tologic diagnosis, the lesions remained stable HPV IJ~.feenon m E~oph~.~us 3~"z. in 47 cas~s (58.8.~). r~gression ~cas noted in 19 cases (23.8.-~). and I~ons had progr~ed in 14 ~ far. ~m¢ clinical improvemen~ of ~fients after treatment with compound ri~flavin have been noted. ~e r~ult in this a~roach will ~ de~i~d in detail el~whcre (O. Sben. D. Wang. C. Wang. F. ~ang. Unpublished o~r- radon). Autoradiography aher zhe FISH with the c~k- tail pro~ is sho~ in F~. 1. Many of the ~mpl~ ~owed ~sitive hybridization as individual s~ts ~tber than as homogeneo~ signals. ~is is prob. ably card by ~he single or cl~t¢~d HPV DNA- ~ntaining epithelial ~lls. ~me of the ~mples showing a weak homogenous spot are regarded as background sign,s. Samples were read ~ ~si- tire when t~ s~ts could ~ clearly distinguished from the back~ound si~ah. A total of 53 ca~ (~.3%) w~ judged ~ HPV- DNA-~sitive. HPV DNA w~ detected in ~.2~ (2 of 9) of the patients c~ologi~lly showing nor- mal findings, in ~% (3 of 6) of ~e patients ~oMng mild dyspl~ia, in ~.6% (25 of 31) of the patients ~th m~era~e dy~lasia, in 67.9~ (19 of 28) of th~ ~th ~vcre d~pl~ia, and 67% (4 of 6) of the paden~ shoMng ~ involve ~mo~ cell ~rdnom. ~ correlation of HPV Fi.~. l. De~ec~ioa of human papilloma,,qrua (HPV) DNA in esophageal cy. tolo.fic gpecimens ~' filter in sire hybridization. Autoradiography of 40 samples after hybridiza:~on with "~:P-labeled probe cocktail of HPV ll, 16. and 18 DNA probes. The last spo~ m filter A is the HPV.DNA.positive control. 0 0 BATCo document for Mayo Clinic 28 March 2002

~ Chan~ e~ aL Table I Pre,~nce of huraan papillomav~rus ~ H P~I DNA in ¢-~ophagcal °'.'toi°~c ~mph-~-' cxmsistcnl wilh Precancer Cytologic diagnosis Normal Mild Moderato S~vcre Squamous epithelia dysplasla d.vsplasia dysplasia carcim)rna HPV DNA n c.f. n .e÷ n ~ n ~ n Prescn~ 2 2.22 3 ~[) ~ 80.6 19 67.9 4 .,..l ~ 33.3 Ab~nl : 7 T/. ,',¢,3 .~) .~ 19.4 9 ~" _ Total 9 I(X) b l~) 31 IIX) 2S IU0 6 DNA hybridization with the cytologic diagnosis is summarized in! Table I. There is a significant difference (chi-sq~aare = S.7774. P < 0.01 ) in the incidence of HP~" DNA between the Pap-nega- tive ~'_ of 9l and ~he Pap-positive cases--that is. dmse ~.'ith Pap abnormalities (51 of 71). DISCUSSION , HPV infection in the esophagus was first sug- gested in 1982 bi' Syrj~inen et al. (6. 7). who described condylo~matous changes (the ~'topathic changes of HPV) in esophageal cancer specimens and demonstrate~ group-specific HPV common antigens in a case ~f an esophageal squamous cell papilloma. These ~ndin~ have subsequently been confirmed by other workers demonstrating either condyiomatous changes. HPV antigens, or HPV DNA in benign and malignant esophageal squa- mous lesions (6. ~'. 13-17). The presem res.ults indicate that HPV infection o| the esophagus does occur in Chinese patients as well. |ndecd. it seems to be particularly prevalent among the population at high risk for esophageal cancer in China. ;especially in the padents with q..'tolo~c evidenci" of precancerous lesions and cancer. Using a mixed HPV I1.16. and 18 probe. we showed that 7;.3% (47 of ~5) of the patients with precancerou.~ lesions (that is. mild. moder- ate. and severe d.~!splasia) and 67~ (4 of 6) of the esophageal canc.~r patients had HPV DNA sequences in their cytologic samples. These re- suits are consistent with our recent findings dem- onstrating HPV-shggestive morphologic changes in 49% ~ of 511 iof the esophageal cancer speci- mens and HPV 6. I1.16. and 18 DNA sequences in 43.1% (~.9 of 51) of the esophageal precancer and cancer lesions by DNA in situ hybridization il~). As discussed above, the etiology of esophageal cancer is unknown as yet. Although some high- risk factors have been proposed, no common etiologic agents have been found which could explain the highly distinm geogaphic distribution of this malignancy (3. 5). During recent years HPV infections have been linked with the patho- genesis of various human squamous call curci- nomas, including those of the genital, respir- ator', and gastrointestinal tract--for example. oral mucosa (g.-12L It has been well established that bovine papillomavirus fBPV 4) commonly infects the esophagus in cattle developing pap- illomas, in which mali~ant transformation may occur after exposure to carcinogens and immuno- suppressants in ingested bracken fern (20. 21). Similarly. esophageal HPV infection has been reported among the high-risk populations both in South Africa 114) and in China (17). These data suggest that HPV inf~ions might be involved in esophageal carcinogenesis also in human beings. most probably acting synergistically with other carcinogens, such as those of physical factors Icoarse and hot food intake), chemical car- cinogens (nitrosamines. aflatoxin, and cigarette smoking), and nutritional factors (excessive alco- hol intake and deficiencies of vitamins A. B. and C and certain minerals) which have been related to esophageal cancer in the high-risk areas (1-4). HPV 16 and 18 infections have been implicated in the pathogenesis of intraepithelial neoplasia and squamous cell carcinoma, especially in the i BATCo document for Mayo Clinic 28 March 2002

lower geniza| tract (IL~12). Our prcvlous study ck:monstrated these high-rlsk HPV L~-pes in 31.-1.~ 06 of 51 } of esophageal precancer and cancer lesions {17). HPV 6 and II DNA originally iso- lated from _ocnital~cond.vloma and la~'ngeal pap- illoma are known to be frequently associated with benign genita! lesions, and thus regarded as the "[ow-Hsk" HPV types (IO--12L However. recent studies suggest th,at infections with HPV 6 or may be related to~crodigestive carcinogenesis as well (8. 9. 22-24!. Accordingly. HPV 6 and were recently detected in benign and malignant esophageal lesions {I~. observed an esophageal and bronchial pap- illomatosls with a clinically malignant course, in which HPV II DNA was demonstrated. Using the polymerase c[~a~n reaction (PCR) technique. we recently foun.d HPV II DNA in 1.7.~.% (9 of 52) of esophagca~ cancer samples collected from the high-dsk arc.a in China (unpublished data). These data suggested that the higher prevalence of ~ese viral ty~s among the FI-'PV infec[Jons of ~.be esophagus in China may be correlated with the higher morbi~lin.., of esophageal cancer in this ar~a. However. f~rt bur studies on the relationship between HPV i~fection and the development of esophageal canc.~r are still urgently needed. FISH seems to offer a:simple and effective method for lace-scale population studi~s on this problem. ACIG~OWLEDiSEMENT We wish to thank Prof. Dr. Harald ~ur H'ausen, Demsehes Kreb~forschun~zentrum. Held¢lberg. FRG. for placin."g the HPV DNA probes a~ our disposal. REFERENCES !, Yang CS. Research on e~ophageal cancer in China: a review. Cancer Rcs 1980. 40. 263.~.2644 2. Cook-Mozaff~ri PJ. Azordegan F. Day NE. Ressi- caud A. Saba~ C, Aramesh B. Ocsophageal cancer studies in the: Caspian littoral of han: results o| a cas~ control ~udy. Br J Cancer 197o. 3. Morris HHB~, Price SK. Langerhan'~- cells, papil- lomavlruses ~d oesophageal carcinoma--a hypoth- esis, S Aft Mud. J 198-',. 71(supp,l) l~-l,g. 4. Joint [ran/IARC Stud)" Group. Oesophageal cancer studies in lh~ Caspian littoral of Iran: results of population st~dies--a prodrome. J Nad Cancer [nst 1977. $9. I127-II38 HPt" [n.'e~..on in Esopf,,,gus 387 -~. Books LA. Steyn L.M. Renan ML O~ophage~I carcinoma--~earch .:~r a ~ible ~ff~l acz~lo~ ~ing molecular h~d~at~on ~cchniqu~. S ~r Mud J ]9~. 7l(suppl~. ~ b. S}~j~ncn KJ. Hism]og~cal chang~ iden~i~/to o[ ~ndylomatom ]csmns found in e~pha$~l ~us Cell ~r~nOme~ Arch G~hwu~ffo~h 19~ Z Syrj~n~n K. ~vrhbn~n S. Aokce S. K~kcla E. 5q~o~ col} pap~]loma of ihe esophagus: a I~ pt~aEy cau~d ~ human papilloma ~qrus (HPV}. Diag H~topalhoi ~9~. 5. 291-2~ 8. S~j~ncn ~. Papi}/omavi~ infeczions and ca~er. In: Syrj~ncn K. Gissmann L K~s LG. e~. Pap/l- loma*~ and human d~ase. Springer-VeHag. He/del~. 1987. ~503 9, Ka~ima H. Mounzs P. Tumors of the head n~k. I~x. lun£ and e~phagus and their ~sib~ ~lation m HPV. ln: Syri~ncn K, G~ssm~n. K~ LG. ads. Papillom~ses ~d human ~. Springcr-Verlag. He:~e}beue. 1987, 13~157 10. ~r Ha~cn H. Haman papillomaviru~ and ~blc role in~uamous cell ~nomas. ~ 1. ~mann L, Papillomavi~ses and Iheir wi~ ~ncer in ammals and in man. Cancer S~ I~ S~jinen K. Human papil~mavi~s (HPV) al~ns with imraepitheli~ ne~l~a and ~ ~noma, Pa~hol Annu 1~. 21.5~98 13. Winkler B, Ca~ V. Reumann W. el al. H~ papill~avi~ in~ec~ion of lhc ~g~: a ~pathologic s~udy wixh demonsltatio~ of ~pil- l~a~ antigen ~" the ~hnique, Ca~e~ 1985. 55. 14. ~]le JJ. M~goli~ KA. Marko~z S. I~n H~an papillomatous inf~Iion relaled to ~ ~aSeal ~xcinom~ in black Sooth Africans. S Air McdJ 1~6. 69. 15. de Borg~ ~. Acev~o F. Mir~les E. Mijar~ P. ~u~ous papilloma of lhe ~hag~ dia[~ ~lology: re~ o[ a case with con~ffem ~i~oiO carcinoma. A~a Cytol 19~. ~. ~7- 4~ 16. Kulski J. Demeter T. Starter GF. ShilOh ~. Human papilloma~s { HPV) DNA in ~noma. ~ncc: 19~. 2. ~ 17. ~ang F, S~nen $. Shun ~, Ji ~. ~rj~mn K. Human papillomavi~ (HPV) DNA in es~geal ~ancer lesions and ~uamo~ cell from ~jna. Im J Cancer 1~ fin 18. Shun Q. Diagnostic ~'mlogy. In: Huang GJ. Wu YK. cds. Cardnoma of the e~phagm ~d ~ia. Spdnger-Verlag. Be~in. 1~. 19. Shcn O. Wang D. C~ng F. WangC. C~opalholo~" of early canc~: of the ~phagu~ and l~ lesions. J Hcna~ Mud Univ 1999. 2, 12-19 20. Jaffeu W~. Envimnmema[ ~rci~ge~ and ~pil- Ioma~s~ in :he pa~hogcncsis of ~ncer. P~ ~Lond B 19S'. ~l. 1-11 2L C~ MS. Papillomas and ~nccr in ca10e. C~er Su~- 19B7, 6.3~ ~. Syrjfinen S. Syr~an~n K. M~my)~ R. Col~n Y. BATCo document for Mayo Clinic 28 March 2002

Received |1 Au_c, ust 1989 Accepzed 20 Oc~obcr 1989 24. de: Vil|[¢~ E-M. Papilloma vir',~ in cancc~ and papillomas of ~hc acrodih'csd~e ~a¢~. Biomcd Ph~r- maco| 1989. 43. 3|-36 ~. HoMing M. Hordlng U. Daugaard $. Norrild Fabvr V. Human papilloma vlr~s type 11 in a fata~ case of esophag,'al and bronchial papillomatosis. Stand J fnfec~ Dis 1989. 21. BATCo document for Mayo Clinic 28 March 2002