Materials and methods

Fifty five conjunctival papillomas were collected from the files of the eye pathology institute (33 males, 22 females, age range 18–84 years, mean age 41 years). The specimens were registered during the period 1988–97. The papillomas were reviewed histologically on haematoxylin and eosin stained sections for confirmation of the diagnoses, and it was noted whether or not koilocytosis was present in the tumour.

POLYMERASE CHAIN REACTION

Three sections from each paraffin block were cut. The sections were placed in 1.5 ml tubes and 50 μl digestion buffer (10 mM TRIS pH 7.0, 1 mM EDTA, proteinase K 200 μg/ml) was added. The specimens were digested at 65°C for 3 hours, spun, and the aqueous phase was transferred to a new tube. Boiling for 10 minutes inactivated the proteinase K. The samples, together with appropriate positive and negative controls, were amplified with primers targeting a 288 bp fragment of the single copy β globin gene in order to ensure the integrity of the DNA. β Globin negative papillomas were excluded from further analysis. PCR was performed with the universally accepted consensus primers GP5+/GP6+ and CPI/CPII.1819 Samples that were HPV negative with these primers, were further subjected to PCR with MY9/MY11 consensus primers.20 These primers will amplify the majority of the known mucosatropic HPV types. (Primer GP5+/GP6+; HPV type: 6, 7, 10, 11, 13, 14D, 16, 18, 30, 31, 32, 33, 34, 35, 39, 40, 42, 43, 44, 45, 51, 52, 54, 55, 56, 58, 59, 61, 62, 66, 72, 73. Primer CPI/CPII; HPV type: 2A, 4, 5, 6, 7, 8, 10, 11, 16, 17, 18, 21, 24, 31, 33, 34, 35, 36, 44, 45, 47, 51, 52, 53, 54, 55, 56, 58, 59, 66, 68. Primer MY9/MY11; HPV type: 2A, 6, 7, 8, 9, 10, 11, 13, 15, 16, 17, 18, 21, 22, 23, 24, 25, 28, 29, 31, 32, 33, 34, 35, 36, 38, 40, 42, 44, 45, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 59, 61, 62, 63, 64, 66, 67, 68, 69, 70, 72, 73.)21 A volume of 15 μl of the PCR product was run in a 4% submerged 3:1 NueSieve agarose (Medinova, Hellerup, Denmark) in TAE buffer, stained with GelStar and viewed in ultraviolet transillumination.

The sample was considered HPV positive when a DNA sample of a predicted size (for example, ∼140 bp for the GP5+/GP6+ primer) was identified. HPV positive samples were analysed by dot-blot hybridisation with alkaline phosphatases labelled oligoprobes for HPV types 6/11, 16, 18, 31, 33, 35, and 45.21 Visualisation was performed by chemiluminescence (CSPP, Roche Molecular Biochemicals).

Results

Of the 55 papillomas investigated, the β globin fragment was amplified from 52 (94.5%). The three β globin negative papillomas were excluded from further analysis. HPV was detected in 48 of the 52 (92%) investigated papillomas. The GP5+/GP6+ primer identified 32 HPV positive samples. Using the CPI/CPII primer a further eight HPV positive papillomas were found and the MY9/MY11 primer identified eight further HPV positive specimens.

Dot-blot analysis was conducted on 47 of the 48 papillomas and demonstrated HPV 6/11 in 40 papillomas and HPV types 6/11 and 16 in one papilloma.

Koilocytosis was present in 22 of the 55 papillomas (40%).

The results of the PCR analysis for HPV were compared with the results of the histological examination for koilocytosis. Nineteen of the 48 (40%) papillomas that tested positive for HPV with PCR contained areas with koilocytosis (Table 1).

Discussion

We found like other investigators that the presence of koilocytosis has a low diagnostic value for HPV infection.2223 Consequently, increasing interest has been focused on methods of direct detection of HPV.

Various authors have analysed conjunctival papillomas for presence of HPV, and HPV types 6, 11, and 16 have been demonstrated (Table 2). The previously published material has either been small or less sensitive techniques have been applied (Table 2). Using the PCR technique with consensus primers we demonstrated HPV in 48 of 52 papillomas (92%). The consensus primers employed in this study each identified different proportions of the papillomas as being HPV positive. This may well depend upon the fact that the employed consensus primers have different sensitivity and specificity for HPV types, and this stresses the need to use more than one consensus primer when investigating archival paraffin embedded tissue for HPV.

This study revealed that the HPV types most commonly found in papillomas of the conjunctiva in Denmark are HPV 6/11. HPV infection with multiple types is known to occur24 and indeed we were able to demonstrate a single papilloma with a double infection with HPV types 6/11 and 16. Naturally, we cannot rule out the possibility of other HPV types also being present.

HPV types can be classified according to their potential for inducing anogenital carcinomas.25 HPV types rarely found in cancers are defined as low risk types and those often found in cancers are high risk types.25 In our material only one of 48 papillomas contained high risk HPV (HPV 16) and this might substantiate the fact that carcinoma rarely develops from a conjunctival papilloma.26 Furthermore, the high risk HPV DNA was detected in a papilloma also harbouring low risk HPV. Perhaps simultaneous infection of conjunctiva with low and high risk HPV types is necessary for the rare development of carcinoma in a papilloma.

HPV is rarely present in normal oral and nasal mucosa.2728 However, HPV has been found in cytomorphologically normal cervical smears with a prevalence up to 25%.29 In the only study of normal conjunctiva HPV was found in 32% of the samples, but this study only looked for two different HPV types.10 The finding of HPV in normal conjunctival tissue may indicate that the neoplastic development of a papilloma is initiated by multiple factors—among these HPV. Immune suppressed organ transplanted patients suffer particularly high incidences of HPV infections that are directly associated with the extent and duration of immune suppression, and an association between HIV and an increase in the incidence of conjunctival carcinomas has been found.3031

The transmission route of HPV to the conjunctiva has not yet been clarified.

Fetal passage through an infected birth canal might be one of the more common methods of direct transmission. Nine out of nine papillomas from patients aged 1–10 years contained HPV in one study.6Autoinoculation certainly seems as a probable transmission mode and McDonnell et al found HPV in ocular swabs from women with HPV induced lesions of the genital tract.32

HPV vaccines with both prophylactic and therapeutic potential are being developed and are presently undergoing phase I and phase II clinical trials.1314 This potential treatment would especially benefit patients experiencing multiple recurrences of conjunctival papilloma and patients with malignant tumours associated with HPV such as conjunctival carcinoma and squamous cell tumours of the lacrimal system.33-35 Moreover, the development of HPV vaccines may lead to a decrease in HPV associated tumours.