Detection of HSV-2 antigen in carcinoma cervix and premalignant conditions by immuno-cytochemistry.
One hundred and twelve cases of cervical pathology (58 paraffin sections and 54 cervical smears) were assessed by PAP technique and 30 cases by indirect immunofluorescent technique. Forty-two normal cervical smears from the age matched controls were stained by indirect immuno-fluorescent technique. HSV-2 antigen was detected by PAP method in 86 out of 112 cases (78.57%) i.e. 50/55 squamous cell carcinoma, 13/13 carcinoma in situ, 11/15 severe dysplasia, 3/4 moderate dysplasia and 9/16 mild dysplasia. The amount of antigen was maximum in squamous cell carcinoma and decreased in carcinoma in situ, severe, moderate and mild dysplasia in descending order. Three cases of adenocarcinoma cervix were negative. Only one case out of 42 controls was positive. HSV-2 antigen was detected by an indirect IF technique in 8/9 squamous cell carcinoma, 2/3 carcinoma in situ, 3/7 dysplasia, O/1 adenocarcinoma and 4/10 inflammatory cases. The above findings support the association between HSV-2 and squamous cell carcinoma cervix, as well as carcinoma in situ which is statistically significant.
Squamous cell carcinoma of the cervix is the commonest neoplasm in Indian women, constituting 21.04% of the total malignances. The survey, conducted at Bombay, by Indian Cancer Society reveals its incidence 22.3/1,00,000 in the general population,  whereas in our institute, the incidence is 4.73/1,000 of the gynaecologic cases. An association between Herpes simplex virus Type 2 (HSV-2) and cervical carcinoma has been well established by various parameters like sero-epidemiological studies,, animal experiments, tissue culture and immuno-fluorescent technique,,. However, we have observed only 46 cases of herpetic cervicitis (judged only by Papanicolaou smears), out of 1, 84, 849 cervical smears examined during the period of last 12 years. The incidence of herpetic cervicitis reported in North America (judged by PAP smears) is 87-217/1,00,000 in general population. The high incidence of squamous cell carcinoma of the cervix and the low incidence of herpetic cervicitis prompted us to undertake this study, to ascertain the association between HSV-2 and carcinoma of the cervix. PAP and indirect IF techniques are used as they are highly sensitive and specific methods for detection of HSV-2 antigen.
This immunocytochemical study comprises of three groups (1) retrospective study group, (2) prospective study group and (3) normal control group. For retrospective study, 112 cases of cervical pathology (54 cervical smears and 58 paraffin sections) were selected as follows: Invasive squamous cell carcinoma (55), carcinoma in situ (13), severe dysplasia (15), moderate dysplasia (4), mild dysplasia (16), adenocarcinoma of cervix (3), non-herpetic cervicitis (5) and lymph node metastasis (1). The sections were deparaffinised before subjecting them to PAP technique.
The Papanicolaou stained smears were subjected to PAP technique, after removing the coverslips and without prior destaining.
The reagents used for staining in this technique were, (1) Rabbit HSV-2 antibody (Dakopatts Cat. No. B 116) 1:10 dilution.
The antigen used for immunization has been prepared by sonication and extraction of HSV-2 infected rabbit cornea cells. In the antigen, all the virus proteins are present. The antisera react with type specific as well as with type common antigen. (2) swine anti-rabbit IgG (Sigma Laboratories) 1:10 dilution; (3) soluble complex of horse-radish peroxidase rabbit anti horse-radish peroxidase (Dakopatts Cat. No. Z-113) 1:20 dilution.
The positive and negative controls were processed with every batch of slides. To determine the specificity of the reaction, one smear in each batch was not covered by HSV-2 antibody (The step of specific antibody was omitted)
Interpretation of results:
Presence of brownish granules in the cytoplasm of the cells was considered as immuno-peroxidase (IP) positive result (See [Figure - 1] and [Figure:2). The intensity of the reddish brown granules was graded as follows: 4 +: very dark brown colour, 3 +: dark brown colour, 2 +: brown colour of less intensity, 1 +: faint brown colour. The brownish granules were not seen in the negative cases.
In the prospective study, 30 fresh cervical smears were collected from cases of. squamous cell carcinoma (9), carcinoma in situ (3), severe dysplasia (1), moderate dysplasia (3), mild dysplasia (3), adenocarcinoma of cervix (1), inflammatory cases which did not show cytological evidence of herpetic cervicitis (10).
The smears were immediately fixed in cold acetone for 10 minutes and then preserved at - 20?C.
The standard indirect immuno-fluorescent technique was performed, using the following reagents.
(1) Normal swine serum (L 10 dilution in phosphate buffer saline-PBS;
(2) rabbit HSV-2 antibody (Dakopatts Cat. No. B-116) 1:10 dilutions in PBS.
(3) Swine anti-rabbit immunoglobulin FITC conjugate (Dakopatts Cat. No. F 235) 1:10 dilution in PBS. pH 7.2, containing 0.01% Evons blue.
The chromatographically purified immunoglobulin fraction of antiserum is conjugated with fluorescein isothiocyanate isomer 1 (FITC). After conjugation, unreacted FITC is completely removed by gel filtration on Sephadex G 25. A further purification is carried out by ion exchange chromatography. This process removes unconjugated antibody molecules and antibody molecules to which more than 4 molecules of FITC are attached. The smears were mounted in buffered glycerol and examined immediately in the fluorescent microscope. With every batch of smears, one positive and one negative control were processed. The positive control was the smear of herpetic cervicitis and the negative control was cytologically normal cervical smear. To determine the specificity of the reaction, one smear in each batch was not covered by Rabbit HSV-2 antibody (The step of specific antibody was omitted).
The smears were interpreted as positive (presence of HSV-2 antigen), when an apple green fluorescence was observed in the cytoplasm of the cells either as a diffuse cytoplasmic mass which may not be covering the nucleus or a distinct perinuclear ring. The negative area appeared as dull brick red in colour.
Duplicate cervical smears were collected from women aged between 30 and 55 years, who came for routine screening. Of these, 42 smears were selected as absolutely normal controls i.e. the smears were without any evidence of inflammation, dysplasia or malignancy. The duplicate smears of these cases were stained by an indirect immuno-fluorescent technique, as described above.
(1) Retrospective study group: Of the 112 cases, stained by PAP technique, 86 cases were positive for HSV-2 antigen (78.5%) As shown in [Table - 1] and [Table - 2], maximum number of cases of squamous cell carcinoma were PAP positive (50/55); many of them (26/50) revealed 3 + intensity. All the 13 cases of carcinoma in situ were positive and 8 of them had 2 + intensity. Of the 15 cases of severe dysplasia, 11 were positive, but only half of them had 2 + intensity. Three cases of moderate dysplasia were positive and of these 2 revealed 1 + intensity. Among 16 mild dysplasia, 9 were positive and 8 of them had 1 + intensity. All the 5 cases of non-herpetic cervicitis, 3 adenocarcinoma and one lymph node metastasis were negative.
It was observed that in one smear, two giant cells were present, adjacent to each other. One of them revealed ground glass appearance of the nucleus (Herpes giant cell), whereas the other showed features of histiocytic giant cell. The cytological observations were confirmed by PAP technique.
(2) Prospective study group:The results of 30 cases, processed by indirect immunofluorescence technique are shown in [Table - 1].
Diffuse cytoplasmic fluorescence covering the nucleus was seen in 5 cases, whereas, the nucleus was not covered by fluorescing mass in 6 cases. The perinuclear ring like fluorescence was seen in 4 smears and fluorescing granules in the cytoplasm were present in 2 smears. In two of the above-mentioned smears combination of the patterns was seen.
(3) Normal control group: Of the 42 smears, only 1 revealed an apple green fluorescence, covering the nucleus.
On statistical analysis, the association of HSV-2 and squamous cell carcinoma and that of HSV-2 and carcinoma in situ was found to be significant. (x = 14.57; p < 0.001)
The results of prospective study groups were compared with the similar studies done in past ,,, and are illustrated in [Table - 3] .
Squamous cell carcinoma of the cervix is one of the extensively studied human neoplasms. Its epidemiology , and natural history are well understood. They reveal that this carcinoma has many attributes of communicable diseases. The sexually transmitted infectious agents like HSV-2 and Human Papilloma virus are blamed for oncogenicity. The high incidence of cervical neoplasia is reported in patients with cytologically detectable herpetic cervicitis. However, the low incidence of cytologically detected herpetic cervicitis in our institute may be due to asymptomatic primary genital herpes infection.
Dressman et al have performed an indirect IP staining on cases of carcinoma and premalignant conditions of cervix. Their findings of percentage positivity are as follows: Squamous cell carcinoma and carcinoma in situ (40%), severe dysplasia (36%), moderate dysplasia (10%) and mild dysplasia (3%). The percentage positivity is much higher in the present study. It may be due to the use of commercially prepared specific reagents in this study. All the three cases of adenocarcinoma and five cases of non-herpetic cervicitis are IP negative. Menczer et al have suggested the association between HSV-2 and adenocarcinoma of the cervix, which is not confirmed in this study.
It is observed that the variation in the intensity of the reddish brown colour is proportional to the amount of antigen present. It is noticed in the present study that the squamous cell carcinoma cases have the maximum amount of antigen. The amount of antigen decreases in the cases of carcinoma in situ, severe dysplasia, moderate dysplasia and mild dysplasia in the descending order.
As illustrated in [Table - 3], Royston and Aurelian detected HSV-2 antigen in all the cases of carcinoma cervix, carcinoma in situ and dysplasia, and did not rind HSV-2 antigen in cases other than atypia. However, the proportion of dyskaryotic cells showing presence of antigen was less than 50%. Pacsa et al found that there was no difference between the percentage positivity of carcinoma cases and bland disorders (inflammation and erosion). They found 9.4% positivity in normal smears also. They further studied more number of cases i.e. 530 normal cases, 175 bland disorders, 52 dysplasia and 38 cervical carcinoma. In this study, they concluded that even though normal smears revealed HSV-2 antigen, the percentage positivity increased in bland disorders, dysplasia and carcinoma. They thought that since HSV-2 is venereally transmitted viral infection, the antigen is harboured, perhaps temporarily in normal looking cervical cells too. If the person has an effective immunological surveillance system, viral antigen carrying cells may be eliminated. The high percentage positivity in bland disorders can be explained in the same manner. The results of Borri et a1 and the present study are compable to Pacsa et al. The comparison of the results of squamous cell carcinoma and pre-malignant conditions versus age matched normal controls were statistically significant. Thus we conclude that carcinoma cervix and carcinoma in situ are associated with HSV-2 infection.
We thank Dean, Seth GS Medical College and King Edward Memorial Hospital for allowing us to publish this article. We are also grateful to Research Society, Seth GS Medical College & King Edward Memorial Hospital for the financial support.