Persistent infection with high-risk human papillomavirus (HR-HPV) is considered a necessary factor for the development of cervical cancer and its precursor lesions (1,2,3). At least 13 genotypes have been found associated with high risk of developing cervical cancer and have been defined as carcinogenic (4). Worldwide, HPV16 has been identified as the most frequent virus associated with squamous cell carcinoma and adenocarcinoma, followed by HPV18 (5). The carcinogenic mechanism involves integration of viral DNA into the cellular genome. Subsequently, there is viral DNA replication with particular importance of the expression of the E6 and E7 genes, which alters the cellular pathways of the host cell (6).

The interaction of the viral genome with fragile sites of the infected cell seems to be crucial for malignant progression. Functional inactivation of oncosuppressive genes such as p53 and pRb (retinoblastoma gene) determines the alteration of the pathways leading to the progression to cancer (7,8,9), through a proliferative state, genetic instability and formation of a malignant cell clone (9,10). The expression of E7 determines the inactivation of pRb, which increases the free form of E2F in the cell, which in turn increases the cyclin-dependent kinase inhibitor p16 (p16INK4a) and an aberrant proliferation, which can be observed by the increase in the cell proliferation marker Ki-67 (11). HR-HPV-induced molecular alterations can be studied using biomarkers used on cytology samples, as they represent a complementary tool of diagnostic and prognostic value in infected patients (7,10). These allow to differentiate the infections that are altering the cell cycle, which are the objective in the programs for early detection of cancer.

Immunocytochemistry P16INK4/Ki-67 dual staining allows recognizing neoplastic cells with altered cell cycle and uncontrolled proliferation in cytology samples of the cervix (12,13,14). Dual staining of biomarkers with immunocytochemistry is used in samples reported as atypical squamous cells of undetermined significance (ASC-US) and low-grade squamous intraepithelial lesion (LSIL) by conventional cytology or liquid-based cytology (LBC) and positive DNA HPV HR (15,16,17,18). Likewise, dual staining identifies high-grade intraepithelial lesions, with good sensitivity and specificity (96% and 83%, respectively). However, there may be inter-observer variability in these diagnoses, and the high-risk HPV test may become invalidated due to hypocellularity; in such circumstances, immunocytochemical staining helps define the diagnosis (19,20,21,22,23).

Staining is particularly useful in the differential diagnosis between non-neoplastic lesions (such as inflammatory processes or cellular changes associated with atrophy) and truly premalignant lesions (13,24,25,26,27). Cases of atrophy, tubal metaplasia, endometrial and endocervical epithelia, or squamous metaplasia (28,29,30) show a nonspecific positivity at p16 (25,31,32); while p16INK4a has been highly specific, with statistical significance in the screening of women with ASC-US, due to a relative sensitivity of 0.95 (95%; CI: 0.89-1.01) and a relative specificity of 1.82 (95%; CI: 1.57-2.12) (4).

In the present work, dual p16/Ki-67 staining was observed in cytology smears with a report of squamous atypia: ASC-US, LSIL and high-grade squamous intraepithelial lesion (HSIL), which allowed to standardize the methodology and interpret the immunocytochemistry. These results were complemented with colposcopy and biopsy in the relevant cases. In Colombia, double staining in LBC had not been used, so this study is a pioneer in the country.

Type of study. Cross-sectional analytical observational study with LBC supplemented with immunocytochemistry (p16/Ki-67) in cases diagnosed as ASC-US, an atypia that does not allow ruling out high-grade lesions (ASC-H), LSIL and HSIL. The usefulness of the immunocytochemical test (p16/Ki-67) to detect patients at risk of progression to cancer was explored. Positive cytology studies were subjected to complementary immunocytochemical studies, a second review by a gynecologist, colposcopy and sampling for histological study. The study was approved by the Institutional Ethics Committee.

Population. Samples were taken from a universe of 51 women who voluntarily attended a cytology campaign at the Thirteenth Occupational Health Conference of the Hospital San José and the Fundación Universitaria de Ciencias de la Salud (Colombia) in 2013. Women over 18 years of age were included. Pregnant women (33) and those who had had a hysterectomy were excluded. The participants were explained the purpose of the study, after which they signed an informed consent agreeing to be part of the research.

Clinical variables: Age, parity, smoking, sexually transmitted diseases, age of sexual debut, number of sexual partners and planning methods. These variables correspond to the standardized history in the uterine cytology sampling format endorsed by the Ministry of Health and the Secretariat of Health of Bogotá. These variables are related to the population’s risk of developing cervical cancer.

Liquid based cytology samples. Initially, the cervix was assessed using a colposcope and subsequently the LBC was taken (directly to the vial), using a cytobrush (Rovers®). A smear was performed for each case, using the manual equipment (BD PrepMate™ System) to process the LBC, following the supplier’s instructions. The smears were processed and analyzed in the Faculty of Cytohistology of the Fundación Universitaria de Ciencias de la Salud. Papanicolau stain was used and its interpretation was based on the 2014 Bethesda criteria (34). The first reading was done by a cytohistology technologist, and the 16 cases reported with atypia (ASC-US, LSIL and HSIL) were confirmed by two pathologists. There was agreement in all cases. One of the pathologists also reviewed 10% of the negative cases, as is routinely done.

Sample for immunocytochemical staining. Residual samples interpreted as positive (ASC-US, LSIL and HSIL) were tested with vial immunocytochemistry (BD SurePath™) stored at 4 °C. Conventional immunocytochemistry was performed. The CINtec Plus Kit^R was used following the supplier’s instructions. The kit is designed to carry out a two-step immunocytochemical staining procedure on cervical cytology preparations. Positive and negative controls were used, as indicated by the supplier. These samples were analyzed by two blinded pathologists for the morphological results of each case. The criterion for determining a sample as positive was the finding of one or more epithelial cells that showed double reactivity: both brown cytoplasm and red nucleus (10,19,23).

In the cases of positive p16/Ki-67 tests, a second colposcopy was performed by a gynecologist with experience in colposcopy. Endocervical brushing and cervical biopsy were taken as reference standards for the study, one month after the first sampling. These histological samples were read by the two pathologists.

As for women with a negative cytology result, a three-year follow-up was indicated, using the 1-3-3 scheme recommended by the Ministry of Health and Social Protection of Colombia.

The universe of 51 women was included. The ages ranged from 18 to 60 years, with an average of 39.8 and a standard deviation of 10.3. Table 1 shows the demographic characteristics of patients with positive results.

Table 1 Table 1 Distribution of positive cases by cytology according to variables

FP: family planning; STD: sexually transmitted diseases; NSP: number of sexual partners; ASD: age of sexual debut.

Of the 51 cytologies studied, 35 were classified as negative, and 16, with atypia, which were classified as: ASCU-US (n=8), LSIL (n=7) and HSIL (n=1) (Figure 1). The immunocytochemical test was positive for 3 of the 7 cases reported as LSIL and for the case classified as HSIL. The 8 cases reported as ASC-US by cytology were negative for biomarkers. Figures 2 and 3 show positive and negative tests.

Figure 1 Figure 1 Classification of the cases

Figure 2 Figure 2 Positive sample

Figure 3 Figure 3 Negative sample

Those who had positive cytological results underwent a second colposcopic evaluation. When it was abnormal, biopsy or endocervical brushing was taken (Table 2). Brush cytology was abnormal in 3 of them (two LSIL and one HSIL), and biopsy, in 2 of the 4 patients (two cervical intraepithelial neoplasms I [CIN-I] associated with HPV). The subsequent treatment and follow-up required by the group of patients at risk was performed according to the procedure established by the health system.

Table 2 Table 2 Results with complementary test

AWE: acetowhite epithelium (biopsy was taken); IMD: insufficient material for diagnosis.

Worldwide, screening for early detection of cervical cancer has recently focused on the need to know if the infection is caused by a high-risk virus. More recently, studies have been to determine if the host cell genome control mechanisms are altered. Since viral infection is frequent, it is necessary to know in which cases it is a high-risk virus and when it has altered the cell regulation mechanisms, becoming a threat for cancer development.

In Colombia, the determination of high-risk HPV was introduced as a primary screening test; however, it is still not done uniformly. This should be accompanied by LBC, which has the advantage of allowing molecular tests. However, in our country this test is more than twice as expensive as conventional cytology. Problems also persist in the quality of the sampling, since they may lack a transformation zone.

Among the post-analytical problems is the fact that some patients do not pick up the result of the test, and the lack of coordination between some institutions providing health services to actively seek out patients with positive results; in addition, laboratories do not have mechanisms to inform treating physicians and patients about positive results for cervical lesions.

By tending to promote a more effective screening, in this work a first approach was made to LBC with the application of dual staining with biomarker (immunocytochemistry) for high-grade lesions. The advantage of the test is that it shows high-grade lesions that have not been identified morphologically as such, allowing early diagnosis of progression to high-risk lesions.

Multiple studies highlight the benefits of this technique. In the study by Wentzensen et al. (20), based on a pilot project (2007-2008) with 425 women who had negative Papanicolau tests but positive HPV, dual p16/Ki-67 staining was included in order to identify cases of CIN-2 that were corroborated by biopsy. The study showed a sensitivity of 91.9% to detect CIN-2 and 96.4% for CIN-3. The specificity was 82.1% for CIN-2 and 76.9% for CIN-3. The study concluded that dual-staining cytology can identify women with a high possibility of CIN-2 and can complement cervical carcinoma screening programs. In the study by Petry et al. (19), the sensitivity and specificity in the ASC-US group was 95% and 84%, and for LSIL cases, 100% and 81%, respectively. Based on the findings, the study authors concluded that the use of a biomarker allows the identification of HSIL (18,19,23,26,35,36,37). The accuracy of the p16INK4a test has been tested in cytology with a diagnosis of ASC-US and LSIL (38,39).

The advantage of this technique is that it showed that the coexpression of p16INK4a and Ki-67 only occurred in the presence of dysplasia (40). In addition, it reduces intra- and interobserver variability, which occurs in the evaluation of cervical-uterine cytologies (16) and in histopathological interpretation (15,36,41,42,43). The test is recommended to differentiate between truly dysplastic lesions.

Although the present study had a small sample controlled by conventional cytology screening, positive cases were found with dual staining (p16/Ki-67), which led to further studies with a second colposcopy and histological studies. Patients with low-grade lesions who had not been directed to subsequent studies (colposcopy: biopsy or endocervical brushing) underwent additional examinations, thanks to the result of the positive biomarker that identified cell cycle deregulation and, therefore, the potential risk of progression of the infection to cancer. Immunocytochemistry is aimed at high-grade lesions not detected in the morphology of the screening cytology, which would have been reported as ASC-US or LSIL.

One of the limitations of the study is that it had a closed population, probably with little exposure to risk factors. Nor was there another valuable diagnostic tool for all women, such as the high-risk HPV DNA test. More studies are necessary in the future with a more representative population in terms of risk factors and number of participants.

Specific HPV tests are very important for screening, and are an exceptional complement in cases of abnormal vaginal citologies; they will also be important for monitoring future generations to evaluate vaccine effectiveness. In cases such as the ones described, premalignant lesions were detected, not by the presence of a high-risk virus, but by evidence of an alteration in the cell cycle regulation, which draws attention to the close surveillance of these patients.

This tool allows eliminating unnecessary surveillance of infections caused by high-risk viruses that patients would eliminate without treatment, and also of high-risk cytological lesions that would return spontaneously. With this type of tests, epidemiological surveillance can be optimized, emphasizing where the true risk of progression to cancer is.

Globally, cervical cancer remains an enemy to women’s health. Molecular tests have been complementing screening programs and helping to understand the infection and its mechanisms of progression to cancer, both from the virus and the host point of view. We have in our country this first experience in the implementation of p16/Ki-67 in LBC, which detected women whose early management changed due to the molecular result, modifying what would have been the “natural history of the illness.” We expect that this experience will allow new research to expand the experience of cytologists and pathologists with these techniques, bringing these tests closer to the population at risk in our country.