Materials. Seventy-seven cases of early stage invasive carcinoma of the uterine cervix were studied. The mean age of the patients was 41 (range 24-66) years. All neo plasms were limited to the uterine cervix, without involvement of the parametrial tissue or vaginal wall. In 64 cases (83.1%), the greatest diameter of the neoplasm was 4 cm or less (pT1b1 category), whereas in 13 cases (16.9%), it was greater than 4 cm (pT1b2 category). Regional lymph nodes were involved by the tumor in 22 cases (28.6%). Sixty-seven cases (87%) were squamous cell carcinomas, six cases (7.8%) had adenocarcinomas, and the remaining four cases (5.2%) had mixed carcinomas (adeno-squamous and mucoepidermoid carcinomas). The patients were surgically treated with radical hysterectomy and regional lymphadenectomy (Wertheim-Maigs) at the Clinic for Gynecology and Obstetrics or the Special Gynecology and Obstetrics Hospital, Skopje, Republic of Macedonia. and postoperatively received adjuvant radiation therapy at the Institute of Oncology and Radiotherapy in Skopje, Republic of Macedonia. The operative materials were subjected to uniform and detailed histopathological evaluation at the Department of Histopathology and Clinical Cytology at the Institute of Radiotherapy and Oncology in Skopje, Republic of Macedonia. Methods. All materials were routinely processed using the protocol for formalin-fixed paraffin embedded tissues and the standard hematoxylin-eosin staining procedure. Other histochemical stains were used in some cases to confirm certain histological types of tumor. The HPV status was evaluated by CISH and ISH TSA in all cases. Six 4 µ-thick sections were cut from the selected paraffin blocks containing a representative portion of the neoplasm. In some cases, mainly because of poor fixation, paraffin blocks from previous biopsy or diagnostic conization material were used. Commercially available (Enzo Diagnostics, Farmingdale, NY, USA) single and double well slides, especially designed for ISH, were used for the CISH. For the ISH TSA, poly-L-lysine pretreated slides on which the wells were circled using a pen, were used (DAKO, Glostrup, Denmark). Conventional In Situ Hybridization. Detection and typing of HPV in tumor tissues was performed in two steps: pretreatment of tissue and hybridization and detection of HPV DNA. The procedure followed the instructions given in the commercially available in situ Patho Gene® kit “Typing Assay for Human Papillomavirus” from Enzo Diagnostics, with minor modifications in the pretreatment and in the hybridization step. Mixed DNA probes for HPV types 6/11, 16/18 and 31/33/51 from Enzo Diagnostics were used. Preparation and Pretreatment of Samples. Three 4 µ-thick sections were cut from each paraffin block and applied on pretreated single and double well slides. The slides were then baked overnight in an oven at 60-80°C, to enable better adhesion of the tissue sections to the slides. Sections were then deparaffinated in xylol and hydrated at graded alcohol concentrations. In order to increase the permeability of cells to the probes, 0.35-0.5 mL of freshly prepared 1X Proteinase K (Enzo Diagnostics) was applied for 15 min. at 37°C, followed by washing in water and dehydration in graded alcohols. Hybridization and Detection. One drop (~40 µL) of each biotinylated HPV DNA probe was applied to the matching well in each case. Slides were then transferred to a preheated heating block (slide warmer) at 95 ± 3°C for 8-10 min. For the hybridization step, sections were transferred to a slide warmer preheated at 37°C for 6 hours. After hybridization, slides were washed in SignaSure® buffer (Enzo Diagnostics) at a pH of ~8.0, followed by incubation in a hybridization reagent (buffered formamide) at 37°C. Afterwards, streptavidin-alkaline phosphatase (AP-detection reagent from Enzo Diagnostics) was applied. Finally, application of nitroblue tetrazolium (NBT) as a chromogen and bromochloroindolyl phosphate (BCIP) as a substrate (Enzo Diagnostics), enabled visualization of the whole complex as a dark purple precipitate in the nuclei of the infected cells. For visualization of the cellular morphology, a counter stain with eosin was used. At the end of the procedure, the sections were dehydrated at graded alcohol concentrations and xylol, and were mounted with a permanent mounting medium (Enthelan; Merck, Darmstadt, Germany). Controls. Two types of control slides were used: i) for the pretreatment, hybridization and detection steps, slides with formalin-fixed, paraffin-embedded tissue sections infected with HPV 6/11 were used; ii) slides containing commercially available cervical carcinoma cells (CaSki cells, Figure 1a) which have 400-600 copies of HPV type 16 integrated in at least 11 different places in the chromosomes [16]. In Situ Hybridization Tyramide Signal Amplification. The necessary reagents were provided in the Gen Point kit from DAKO. Biotinylated HPV DNA probes for HPV types 6/11, 16/18 and 31/33 were used. For four cases, HPV DNA probes from Enzo Diagnostics were used. In two others, which were positive for HPV types 31/33/51 when using 44 CISH but negative for HPV types 31/33 using ISH TSA, the latter procedure was repeated using the mixed probe for HPV types 31/33/51 from Enzo Diagnostics. Preparation and Pretreatment of the Samples. Preparation of the slides was the same as for CISH, but with modifications in the pretreatment step. To increase the permeability of the cells for the probes, the slides were incubated in a demasking solution, commercially available from DAKO as a “target retrieval solution” (sodium citrate buffer, pH 6.0) in a water bath at 95°C for 50 min. They were then immersed in 0.01% pepsin solution for 10 min. to enable proteolytic digestion of the tissue. After washing in distilled water, the slides were incubated in 0.3% hydrogen peroxide in methanol for 30 min. to block the endogenous avidin-binding ability of the tissue. Hybridization and Detection. Detection and typing of HPV was the same as for CISH, but with modifications. We introduced the signal amplification step in the post hybridization and detection step, with biotinylated tyra mide as a reporter molecule. The detection step begins with application of conjugated primary streptavidin and HRP (DAKO). After the initial binding of streptavidin-peroxidase to the biotinylated probe, the peroxidase oxidizes the biotinyl tyramide, which immediately forms covalent bonds with aromatic amino acids of proteins in the specimen. The additional biotin is then used to capture more streptavidin-peroxidase. The amplified hybridization signal uses diaminobenzidine (DAB) as a chromogen, which when oxidized by HRP, turns into a black-brown precipitate in cell nuclei. A counter stain with hematoxylin was used for the visualization of the cellular morphology. At the end of the procedure, the slides were dehydrated and mounted with a permanent mounting medium (Enthe lan; Merck). Controls. Control slides were included in the kit, and contained SiHa cells infected with one or two copies of HPV type 16 (Figure 1b), integrated at chromosome 13q21-q31. A negative control probe (biotinylated plasmid DNA probe) was also used on the SiHa cells, to determine the specificity of the hybridization reaction. To visualize different types of hybridization signal, control slides containing CaSki cells (Figure 1c) from Enzo Diagnostics were also used. Evaluation of the Human Papillomavirus Status. This included detection of HPV DNA, identification of the type of HPV, determination of the number of infected cells and the type of hybridization signal. The results were evaluated using light microscopy with 400X magnification. A result was considered to be positive if at least one cell contained a purple or brown precipitate in the nucleus. All positive cases were photodocumented. According to the number of positive cells in the tissue sample, cases were divided into groups that contained 1-5, 6-10, or more than 10 positive cells per sample. The cases were categorized into dot, mixed and diffuse types for evaluation of the type of hybridization signal as was initially recommended [27] and is widely accepted [22,24,28]. Statistical Analysis. We used McNemar’s test for the analysis of correlated proportions [29].

Figure 1. Positive control slides: CaSki cells containing 400-600 copies of HPV DNA type 16 (a, CISH, counter stained with eosin, 400X magnification); SiHa cells containing 1-2 copies of HPV DNA type 16 (b, ISH TSA,counter stained with hematoxylin, magnification, arrow);CaSki cells (c, ISH TSA, counter stained with hematoxylin,400X magnification).