Anal intraepithelial neoplasia: part of a multifocal disease process

1271

25. Dawood MY, Spellacy WN, Dmowski WP, et al. A comparison of the efficacy and safety of buserelin vs danazol in the treatment of endometriosis. In Chandra DR, Buttram VC, eds. Current concepts in endometriosis. New York: Alan R Liss, 1990: 253-67. 26. American Fertility Society. Revised American Fertility Society classification of endometriosis. Fertil Steril 1985; 43: 351-52. 27. Doberl A, Bergquist A, Jeppson S, et al. Regression of endometriosis following shorter treatment with or lower dose or danazol. Acta Obstet Gynaecol Scand 1984; 123 (suppl): 51-58. 28. Jelley RJ. Multicentre open comparative study of buserelin and danazol in the treatment of endometriosis. Br J Clin Pract 1986; 41 (suppl 48) 64-68. 29. Matta WH, Shaw RW, Hesp R, Evans R. Reversible trabecular bone density loss following induced hypo-oestrogenism with the GnRH analogue buserelin in premenopausal women. Clin Endocrinol 1988; 29: 45-51.

30. Tummon IS, Radwanska E, Ali A. Bone mineral density in women with endometriosis and during ovarian suppression with gonadotropinreleasing hormone agonists or danazol. Fertil Steril 1988; 49: 792-96. 31. Sutton C. Advances in surgical management of endometriosis. In: Shaw RTW, ed. Advances in reproductive endocrinology I, endometriosis. Carnforth: Parthenon Publishing Group, 1990: 209-24. 32. Henderson AF, Studd JWW, Watson N. A retrospective study of oestrogen replacement therapy following hysterectomy for the treatment of endometriosis. In Shaw RW, ed. Advances in reproductive endocrinology I, endometriosis. Carnforth: Parthenon Publishing Group, 1990: 131-40. 33. Donnez J, Nisolle M, Grandjean P, Gillerot S, Clerckx F. The place of GnRH agonists in the treatment of endometriosis and fibroids by advanced endoscopic techniques. Br J Obstet Gynaecol 1992; 99 (suppl 7): 31-33.

EPIDEMIOLOGY Anal

intraepithelial neoplasia: part of a multifocal disease process

Invasive carcinomas of the anogenital epithelium share a common aetiological factor—human papillomavirus (HPV) type 16. Although genital may be multifocal, there studies of the prevalence of anal intraepithelial neoplasia in women with intraepithelial neoplasia of the genital tract. We tested the hypothesis that women with high-grade cervical intraepithelial neoplasia are at higher risk of disease in the anus than are control women of similar age with no history of anogenital neoplasia. 29 (19%) of 152 women with cervical intraepithelial neoplasia grade III had histological evidence of anal intraepithelial neoplasia. Of the 29 patients, 11 had grade III anal lesions; 2 of those women had concomitant invasive anal squamouscell carcinomas. Only 7% (8/115) women with high-grade lesions of the cervix alone had evidence of anal intraepithelial neoplasia; by contrast, 57% (21/37) of those with more than one focus of

intraepithelial neoplasia have been

The role of anal examination in the assessment of with any focus of genital intraepithelial neoplasia requires further investigation.

women

Lancet 1992; 340: 1271-73.

no

intraepithelial neoplasia (cervix plus vulva, vagina, or both) had anal lesions. HPV 16 DNA was identified in 18 (51%) of 35 anal biopsy samples in the study group. No evidence of anal intraepithelial neoplasia was found in the control group (50 women), although 2 patients had grade I cervical lesions. HPV 16 DNA was identified in 12 (24%) of biopsy samples from the cervix and 7 (14%) from the anus in the control group; all 7 women with anal H PV 16 had concomitant cervical infection.

Introduction There is increasing evidence that human papillomavirus (HPV) type 16 has a role in the aetiology of cervical squamous-cell carcinoma.HPV 16 has also been implicated in vaginal and vulval squamous-cell carcinoma and anal cancer.2 It appears that HPV infection of one area of anogenital epithelium can be rapidly followed by infection of adjacent areas by direct spread of the infecting agent;3a large area of epithelium thereby becomes susceptible to neoplastic change. Vulval, vaginal, and anal cancers are much less common than cervical cancer.S Although many gynaecologists acknowledge that vaginal or vulval squamous-cell carcinoma may develop in some women with cervical cancer, anal examination is not routinely done. There are a few case-reports of involvement of the anal areabut no study has tried to find out the prevalence of anal involvement in women with genital intraepithelial neoplasia. One study, on the prevalence of

ADDRESSES Departments of Surgery and Gynaecology, Clinical Sciences Centre (J H Scholefield, ChM, W. G. E. Hickson, MPhil, Prof K. Rogers, MD, Prof F. Sharp, MD) and Department of Histopathology (J H F. Smith, MRCPath), Northern General Hospital, Sheffield, UK. Correspondence to Mr J. H Scholefield, Department of Surgery, Clinical Sciences Centre, Northern General Hospital, Sheffield S5 7AU,UK.

1272

anal intraepithelial neoplasia in men and women presenting with anogenital wartssuggested that women who had intraepithelial neoplasia of the cervix were at risk of anal lesions.

Patients and methods Two groups of women were studied. The study group comprised

series of consecutive patients presenting to the colposcopy clinic at the Northern General Hospital, Sheffield (a tertiary referral centre) with histologically confirmed cervical intraepithelial neoplasia grade III on punch biopsy samples. They underwent genital and anal examination with a colposcope and proctoscope (Graham Anderson pattern) after application of 5% aqueous acetic acid to the whole anogenital epithelium.’ Before the examination an anal smear was collected for exfoliative cytology. Biopsy samples were taken for histology from abnormal lesions seen by colposcopy on the cervix, vulva, and anus.7 Adjacent tissue sections were examined for HPV DNA by the polymerase chain reaction. A consecutive series of 50 women, of similar age and parity to the study group, formed the control group, from whom we sought to estimate the prevalence of symptomless cervical and anal HPV carriage. All had a previously negative cervical cytology screening history and were electively undergoing laparoscopic sterilisation. They also underwent thorough genital and anal colposcopy under general anaesthesia before the sterilisation procedure. Smear preparations were taken from the anus and cervix. A biopsy sample was taken from any colposcopically abnormal area. Biopsy samples were taken from the transitional zones of the anus and cervix for histology and for HPV DNA studies. Local ethical committee approval to recruit patients and to take biopsy samples was obtained before the study. All patients gave informed consent. Anal smears were taken with a Cytobrush (Scanda, UK), moistened with 0-9% saline and introduced into the anal canal without further lubrication. Cervical smears were taken in the standard way with an Ayres spatula. All smears were fixed immediately in 95 % methanol, 5% acetic acid, and stained by the Papanicolaou method.8 All biopsy samples were fixed in buffered formalin and embedded in paraffin wax for histology. 5 um tissue sections were stained with haematoxylin and eosin. One consultant histopathologist (J. H. F. S.) examined and reported on all smears and tissue sections. For HPV DNA studies four adjacent 20 pm tissue sections were cut from tissue blocks of each cervical, anal, and (where possible) vulval and vaginal lesion. They were dewaxed with xylene, rehydrated through graded ethanol solutions, and dried. The tissue was digested in "tris"-edetic acid buffer with proteinase K at 37°C for 72 h. DNA was extracted from the digest with phenol and chloroform and precipitated in absolute ethanol. The DNA was resuspended in "tris"-edetic acid and its concentration measured by fluorimetry at 365 nm. 200 ng purified DNA was used as a template DNA for the polymerase chain reaction with primers derived from the E6 and E7 open reading frames of the HPV type 16 genome.9,10 DNA samples were subjected to forty amplification cycles of denaturation at 95°C, annealing at 45°C, and polymerisation at 72°C. p-globin primers were used as controls for each sample to ensure that template DNA amplification was taking place." The amplified gene products were visualised after electrophoresis in 1-5% agarose gels by ethidium bromide staining and ultraviolet a

MULTIFOCAL INTRAEPITHELIAL NEOPLASIA IN WOMEN PRESENTING WITH CERVICAL INTRAEPITHELIAL NEOPLASIA GRADE III

Separate suites of rooms were used for solution preparation, template DNA isolation, and amplification reactions. illumination.

Positive control tissue derived from CaSki and SiHa cells was used contained sterile water in

throughout. Negative control samples place of template DNA. Results

The median age of the 152 study group women was 38 (range 18-70) years and that of the 50 women in the control group 31 (21-45) years (p=0’5). The two groups were of similar parity (2 [0-8] vs 2 [1-5]). Of the 152 study group women with grade III intraepithelial neoplasia of the cervix, 11had synchronous grade III anal lesions (table). In all but 2 of those 11, lesions of the vulva, vagina, or both were also found at the same time or at other times. In 5 patients the anal lesions were circumferential, extending from the pectinate line proximally to the perianal skin and contiguous with vulval lesions. 4 patients with circumscribed anal lesions had predominantly perianal disease, which extended to the pectinate line in only 2. Only 2 women had lesions predominantly in the anal canal-1 had cervical disease alone, and 1 had cervical and vulval lesions. In 1 patient with grade III lesions of cervix, vulva, and anus, the anal lesions progressed to squamous-cell carcinoma 2 months after presentation. Another patient with grade IIIintraepithelial neoplasia of cervix, vagina, and anus and grade II vulval lesions had had a squamous-cell carcinoma excised from the anal margin 12 months before referral to our colposcopy clinic because of an abnormal cervical smear. 18 patients had anal intraepithelial neoplasia of grades I and II (table). We are following up these patients, but no progression has yet been recorded. Among the remaining 123 patients in the study group, 46 (37%) had colposcopic evidence of perianal aceto-white change. Biopsy samples were taken from the perianal area of 20 of these patients during general anaesthesia for cone biopsy; features of HPV infection without dysplastic change were confirmed in 18. 1 patient’s biopsy sample showed non-caseating granulomata; she had complained of abdominal pain and had previously been diagnosed as having an irritable bowel. Small-bowel radiographs showed ileal Crohn’s disease, which was treated medically. Cytological assessment was possible for 110 (72%) anal smears; 36 smears had inadequate numbers of cells and 6 were heavily contaminated with faeces, so cytological assessment was impossible. Cytological evidence of dyskaryosis was found in 10 of the 29 patients with histologically confirmed anal intraepithelial neoplasia. Cytology raised suspicion of histologically confirmed anal lesions in only 5 of the 11 women with grade III lesions. Cytology did not suggest that any high-grade anal lesions had been missed clinically. Cytological features of anal HPV infection were seen in 25 (23%) smears, although koilocytes were present in only 2. Among the 50 control women, 2 had aceto-white lesions on the cervix, which biopsy samples showed to be subclinical papillomavirus infection. No anal aceto-white lesions were detected in the control group. Cytology of cervical and anal smears showed evidence of dyskaryosis in 3 cases. The 2 women with subclinical papillomavirus infections showed moderate dyskaryosis.1 other smear from a colposcopically normal cervix showed mild dyskaryosis. Cytological assessment showed inflammatory changes in two other patients. Histology of biopsy samples from the 2

1273

moderately dyskaryotic lesions showed them to be intraepithelial neoplasia grade I. Superficial papillomavirus infection was found on subsequent examination of the patient with the mildly dyskaryotic smear. HPV 16 DNA was detected in 18 (51 %) of the 35 anal biopsy samples from women with anal intraepithelial neoplasia. In the control group, HPV 16 DNA was detected in 12 (24%) of 50 cervical biopsy samples and in 7 (14%) of 50 from the anus. In 3 cases both cervical and anal biopsy samples were positive for HPV 16 DNA. Discussion

Although it is not uncommon for the vagina and vulva to be examined in women who present with high-grade cervical lesions, the anus is rarely examined. Our finding that 19% of such women also have anal intraepithelial neoplasia suggests that the need for anal examination should be studied further. Perianal intraepithelial neoplasia is similar to vulval disease in that it is often shown by changes in the pigmentation of the anal and perianal epithelium. Vascular patterns seen by colposcopy on the non-keratinised cervical epithelium are not commonly seen on perianal skin, but they do occur in the anal canal. Anal colposcopy should include long acetic acid "soaks" of the perianal skin and proctoscopy of the anal canal.12 Since diagnosis of anal intraepithelial neoplastic lesions may be new to many colposcopists, a low threshold for anal biopsy is required. We have found that anal biopsy under local anaesthetic can be done in an

outpatient setting. Anal intraepithelial neoplasia may be contiguous with vulval disease, but foci of high-grade anal disease may occur in the absence of vulval disease. Colposcopists should be aware of the possible discontinuous nature of these changes. Evidence of perianal papillomavirus infection should raise the possibility of intraepithelial neoplasia in the anal canal. The natural history of anal intraepithelial neoplasia is uncertain. However, cervical intraepithelial neoplasia progresses to invasive cervical cancer in about 25% of cases during 15 years;13asimilar proportion of grade I cervical lesions progresses to grade III during 5 years.6 By contrast, the natural history of vulval, vaginal, and anal intraepithelial neoplasia is poorly understood. Since anal cancer is much less common than cervical cancer, the progression rate may be lower for grade III intraepithelial neoplasia of the anus than of the cervix. The recognition and treatment of clinically identifiable and intraepithelial neoplastic lesions may provide an opportunity to reduce the mortality and morbidity of anal cancer. Until more is known of the natural history of the disorder, it is difficult to know how best to treat these lesions. Our own experience of 2 patients with invasive anal cancers arising in areas of intraepithelial neoplasia grade IIII leads us to believe that these lesions are best treated. At present, excision is probably the best approach since it permits full histopathological examination. Cytological assessment of anal HPV infection is difficult, owing to the low cell yield and contamination by faecal material. We used the method previously shown to be the most effective for anal epithelial cell sampling.l4 5 of 11 high-grade anal intraepithelial neoplastic lesions were diagnosed on cytology-those extensively involving the anal canal. Smears from perianal lesions are not always satisfactory, which may explain why several lesions were not

detected by cytology.

prevalence of colposcopic, cytological, and histological abnormalities in the anogenital tracts of the control group accords with other studies in populations of similar age.9,1S There was good correlation among colposcopic, cytological, and histological diagnoses in this study. The proportions of intraepithelial neoplastic lesions of the cervix and anus with HPV 16 DNA detected in this study are similar to those reported elsewhere, in both the control and study groups. The discontinuous nature of multifocal intraepithelial neoplasia may reflect a discontinuous epithelial response to papillomavirus infection. The

An increase in the incidence of anal cancer has been reported in the United States/6 though as yet there is no evidence of such an increase in the United Kingdom. To reduce mortality and morbidity of invasive anal cancer, identification of patients at high risk of the disease is important. This study identified one risk factor in womennamely, the presence of other foci of genital intraepithelial neoplasia. Further investigation of the natural history of anal intraepithelial neoplasia in a multicentre study now seems

appropriate. We thank Trent Regional Health Authority for funding the project, the staff of the Colposcopy and Laser clinic at the Northern General Hospital for their help and support, and Dr T. Crooks and Dr D. Wrede, Ludwig Institute, St Mary’s Hospital, London, for their help with the polymerase chain reaction.

REFERENCES Hausen H. Papillomaviruses of human genital cancer. Med Oncol Tumour Pharmacother 1987; 4: 187-92. 2. Palmer JG, Scholefield JH, Coates PJ, Shepherd NA, Crawford LV, Northover JMA. Anal cancer and human papillomaviruses. Dis Colon Rectum 1989; 32: 1016-22. 3. Goorney BD, Waugh MA, Clarke J. Anal warts in heterosexual men. Genitourin Med 1987; 63: 216. 4. Sonnex C, Scholefield JH, Kocjan G, et al. Anal human papillomavirus infection in heterosexuals with genital warts: prevalence and relation with sexual behaviour. BMJ 1991; 303: 1243-44. 5. Waterhouse J, Muir C, Shanmagaratnam K, eds. Cancer incidence in five continents. Lyon: International Agency for Research on Cancer, 1982; 675-793. 6. Campion MJ, McCance DJ, Cuzick J, Singer A. Progressive potential of mild cervical atypia: prospective cytological, colposcopic, and virological study. Lancet 1986; ii: 237-40. 7. Scholefield JH, Sonnex C, Talbot IC, Whatrup C, Mindel A, Northover JMA. Anal and cervical intraepithelial neoplasia: possible parallel. Lancet 1989; ii: 765-69. 8. Bancroft JD, Stevens A. In: Theory and practice of histological techniques, 2nd ed. Edinburgh: Churchill Livingstone, 1982. 9. Young LS, Bevan IS, Johnson MA, et al. The polymerase chain reaction: a new epidemiological tool for investigating cervical human papillomavirus infection. BMJ 1989; 298: 14-18. 10. Cornelissen MTE, Van Den Tweel JG, Struyk APHB, et al. Localisation of human papillomavirus type 16 DNA using the polymerase chain reaction in the cervix uteri of women with cervical intra-epithelial neoplasia. J Gen Virol 1989; 70: 2555-62. 11. Saiki RK, Gelfand DH, Stoffel S, et al. Primer directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 1989; 239: 487-89. 12. Sharp F. Integrated colposcopy of the vulva and anus. In: Anderson MC, Morse AR, Jordan JA, Sharp F, eds. A text and atlas of integrated colposcopy. London: Chapman Hall, 1991: 186. 13. McIndoe WA, McLean MR, Jones RW, Mullins PR. The invasive potential of carcinoma in situ of the cervix. Obstet Gynaecol 1984; 64: 1.

zur

451-58. 14. Scholefield

JH. Studies in the pathogenesis of HPV associated anal squamous cell carcinoma. University of Liverpool: ChM thesis, 1991.

15.

Kiviat NB, Koutsky LA, Paavonen JA, et al. Prevalence of genital human papillomavirus infection among women attending a college student health clinic or sexually transmitted disease clinic. J Infect Dis 1989;

159: 293-302. 16. Wexner SD, Milsom JW, Dailey TH. The demographics of anal cancer are changing. Dis Colon Rectum 1988; 30: 942-46.