To Evaluate the Role of p16Ink4a Immunocytochemistry for Detection of CIN2+ in Women Detected Screen Positive by Visual Inspection Using Acetic Acid.

OBJECTIVE: To evaluate the role of p16Ink4a immunostaining for the detection of cervical intraepithelial neoplasia (CIN2+) in women who had a positive screening test using visual inspection with acetic acid (VIA).
METHODS: Opportunistic screening of women (30-50 years) coming to the gynecology clinic by VIA was performed; the screen-positive women were included in the study which had the institutional review board (IRB) approval. A cytology slide for p16Ink4a immunostaining, colposcopy, and biopsy was then performed sequentially. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of p16Ink4a immunocytochemistry were evaluated with histopathology as the gold standard.
RESULTS: p16Ink4a positivity showed a linear correlation with the increasing grade of CIN. p16Ink4a positivity was seen in 6% of CIN 1, 80% of CIN 2, 100% of CIN 3, and squamous cell carcinoma. The sensitivity and specificity of p16Ink4a immunocytochemistry for detecting CIN 2 or more was 87.5% (95%CI 61.65-98.45) and 97.06% (95%CI 84.67-99.93). Colposcopy had an equal sensitivity of 87.5% (95% CI 61.65-98.45) and specificity of 50% (95% CI 32.43-67.57), respectively.
CONCLUSION: With high sensitivity and specificity, p16Ink4a immunocytochemistry could be a viable option for triaging VIA-positive women. Copyright:

INTRODUCTION

Cervical cancer is the second commonest cause of deaths due to cancer in women aged 15–44 years worldwide with developing countries accounting for approximately 86% of cervical cancer deaths worldwide.[12] The age-standardized incidence rates (ASR) range from 5.5 per 100,000 women in Australia to 42.7 per 100,000 in Africa.[1] In India, the ASRs are as high as 28.6 in rural areas and 12.3 in urban areas.[3] Of the available screening tests, visual inspection using acetic acid (VIA) has been most widely investigated and accepted as an alternative to cytology in low resource settings.[45]

VIA-based screening is simple, effective, and economical with immediate results. Colposcopy followed by biopsy and treatment of any preinvasive lesion then follows. A dire shortage of trained colposcopists in developing countries forces women have to travel a long distance to get effective treatment for preinvasive cervical lesions. Identifying the VIA-positive women who definitely need colposcopy and treatment is a challenge.

Biomarkers used to improve the detection of cervical lesions at greatest risk for developing cervical cancer are p16Ink4a, Ki-67, BD ProEx C, and Cytoactiv HPV L1. Biomarkers can be used to facilitate the detection of abnormal cells within a Pap cytology sample based on simple immunocytochemistry assay. p16Ink4a, a cell cycle regulator has been shown to be expressed in 100% of high-grade lesions.[67] Benign squamous epithelium and endocervical epithelium are usually negative or very focally positive for p16Ink4a. Their role has not been tested in VIA screened population till now. This study aims to evaluate the usefulness of p16Ink4a biomarker for triaging VIA positive women prior to colposcopy and biopsy in a VIA-based screening program.

MATERIAL AND METHODS

We conducted a cross-sectional study in the outpatient department (OPD) in the gynecology clinic at a tertiary care government institution, New Delhi. Women coming to gynecology OPD for any complaint were screened by VIA. A statistically significant sample of VIA-positive women was derived as 50, as the prevalence of VIA-positive women in a hospital-based population in Delhi was 24.26%.[4] Exclusion criteria were women in whom squamocolumnar junction was not visualized on per speculum examination, pregnancy, prior treatment for cervical intraepithelial neoplasia (CIN)/cancer cervix, active vaginitis/cervicitis/vaginal bleeding and those not willing to adhere to the study protocol.

A thick cotton swab soaked with freshly prepared 5% acetic acid was applied over the cervix and inspected after 1 min. The detection of any acetowhite areas within the transformation zone was considered VIA positive. All VIA-positive patients were called for colposcopy.

Prior to colposcopy, a cytology slide was taken for p16 immunostaining with an Ayre spatula rotated circumferentially to obtain an ectocervical sample, followed by an endocervical brush sample. The specimen was then smeared (spatula) and rolled (brush) uniformly on to a slide and rapidly fixed using spray fixative (ethyl alcohol) and then stored in a solution containing 50% methanol and 50% acetone kept at 4° centigrade (cold acetone solution). After taking a sample for cytology, normal saline, 5% acetic acid and Lugol's iodine were applied sequentially to the ectocervix, any acetowhite or iodine negative lesion was visualized under magnification using a colposcope and then scored using Swede scoring system and suspicious areas biopsied.

p16Ink4a Immunocytochemistry was done using the p16Ink4a antibody (Rabbit Polyclonal Antibody). The smears were placed in 3% hydrogen peroxide in methanol (hydrogen peroxide block) for 30 min. Three washes with Tris buffer were given. Nonspecific proteins were blocked with 5% milk block. Three washes with Tris buffer were given. The tissue was incubated with the primary antibody (p16Ink4a) overnight at 4°C. Three washes with Tris buffer were given. Secondary horseradish peroxidase (HRP) primer was added for 60 min. Three washes with Tris buffer were given. Diamino Benzidine (DAB) was applied on the slides and the reaction monitored under a microscope. The slides were immersed in water as soon as crisp golden brown nuclear staining was seen. The slides were counterstained with hematoxylin. They were then dehydrated in graded alcohol solutions. They were then passed through xylene which acted as a clearing agent. Distyrene plasticizer xylene (DPX) and a coverslip was put. Specific brown color cytoplasmic staining for the p16Ink4a antibody was taken as positive.

The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of p16Ink4a immunostaining were compared with colposcopy with histopathology as the gold standard. Categorical variables were presented in number and percentage (%) and continuous variables were presented as mean ± SD and median. Qualitative variables were correlated using the Chi-Square test/Fisher's exact test. A diagnostic test was used to find out sensitivity, specificity, NPV and PPV. A P value of < 0.05 was considered statistically significant. The data were entered in MS Excel spreadsheet and analysis was done using Statistical Package for Social Sciences (SPSS) version 21.0.

RESULTS

We took 50 VIA-positive women in our study; their mean age was 37.28 years (range 30–50 years, SD = 6.4 years). Age at marriage was ≤20 years in 86% of women; only 14% were >20 years at marriage.

Colposcopy was performed in all of them; which showed lesions suggestive of low-grade CIN in 19 (38%) and high-grade CIN lesion in 31 (62%). All these women had a biopsy, the histopathology showed benign (squamous metaplasia and chronic cervicitis) in 18 (36%), CIN 1 in 16 (32%), CIN 2 in 10 (20%), CIN 3 in 4 (8%), and squamous cell carcinoma (SCC) in 2 women (4%) [Table 1].

Table 1

Correlation of p16 with biopsy

Biopsy findings	p16 findings		Total
	Positive	Negative
Squamous metaplasia	0 (0.00%)	5 (14.29%)	5 (10.00%)
Chronic cervicitis	0 (0.00%)	13 (37.14%)	13 (26.00%)
CIN 1	1 (6.67%)	15 (42.86%)	16 (32.00%)
CIN 2	8 (53.33%)	2 (5.71%)	10 (20.00%)
CIN 3	4 (26.67%)	0 (0.00%)	4 (8.00%)
SCC	2 (13.33%)	0 (0.00%)	2 (4.00%)
Total	15 (100.00%)	35 (100.00%)	50 (100.00%)

Comparing colposcopy results with histopathology, sensitivity, specificity, PPV, and NPV of colposcopy for predicting CIN 1 or more lesions was 75%, 61.11%, 77.42%, and 57.89%, respectively and for CIN 2 or more lesions was 87.50%, 50%, 45.16%, and 89.47%, respectively [Table 2].

Table 2

Clinical performance of p16 testing and colposcopy for triaging VIA-positive women

	p16 immunostaining CIN 1 or more	p16 immunostaining CIN 2 or more	Colposcopy CIN 1 or more	Colposcopy CIN 2 or more
Sensitivity (95% CI)	46.88% (29.09-65.26%)	87.50% (61.65%-98.45%)	75% (56.6-88.54%)	87.50% (61.65-98.45%)
Specificity (95% CI)	100% (81.47-100%)	97.06% (84.67-99.93%)	61.11% (35.75-82.7%)	50 % (32.43-67.57%)
PPV (95% CI)	100% (78.2-100%)	93.33% (68.05-99.83%)	77.42% (58.9-90.4%)	45.16% (27.32-63.97%)
NPV (95% CI)	51% (33.99-68.02%)	94.29% (80.84-99.3%)	57.89% (33.5-79.75%)	89.47% (66.8-98.7%)

p16Ink4a positivity was seen in immunocytochemistry in 30% (15/50), 70%(35/50) were p16Ink4a negative. p16Ink4a showed a linear correlation with increasing grade of squamous dysplasia. p16Ink4a positivity was seen in 6% cases of CIN1, 80% of cases of CIN2, 100% cases of CIN 3 and 100% cases of SCC. None of the benign histologies stained positive for p16Ink4a [Table 1; Figures 1 and 2].

Figure 1

(a) Cytology smear showing Intermediate squamous cells with nuclear and cytoplasmic positivity for p16 corresponded to CIN 2 on histopathology. Immunocytochemistry ×400, (b) Cytology smear showing small parabasal cells with nuclear and cytoplasmic positivity for p16 expression corresponded to CIN 3 on histopathology. Immunocytochemistry ×400, (c) Cytology smear showing malignant squamous cells showing nuclear and cytoplasmic positivity for p16 corresponded to squamous cell carcinoma on histopathology. Immunocytochemistry ×100

Figure 2

(a) Cervical intraepithelial neoplasia 2 (CIN 2) on histopathology. H and E, ×400, (b) Cervical intraepithelial neoplasia 3 (CIN 3) on histopathology. H and E, ×200, (c) Keratinizing squamous cell carcinoma on histopathology. H and E, ×200

Comparing p16Ink4a immunostaining results with histopathology; sensitivity, specificity, PPV, and NPV for predicting CIN 1 or more lesions was 46.88%, 100%, 100%, and 51%, respectively and for CIN 2 or more lesions was 87.50%, 97.06%, 93.33%, and 94.29%, respectively [Table 2].

DISCUSSION

Various studies have evaluated the usefulness of the immunocytochemical detection of biomarkers as potential markers of dysplasia in cervical cytology preparations, however, the literature is scant on its usefulness in a VIA screened population.[8910]

The mean age of women in our study was 37.3 years; the mean age of those with CIN 1 was 34.8 years, CIN 2 was 38.9 years, CIN 3 was 40 years, and SCC was 46.5 years. In a large study done by Luthra et al. screened 68,436 women and found that the mean ages of mild, moderate, and severe cervical dysplasia were 33.8, 35.2, and 40.2 years, respectively.[11]

The sensitivity of colposcopy for predicting CIN lesions is variable. Mitchell et al. in their study showed that sensitivity and specificity for predicting CIN 1 or more was 85% and 69%, respectively; for CIN 2 or more was 96% and 48%, respectively.[12] However, Pretorius et al. found that the sensitivity and specificity of colposcopy for CIN 2 or more were 62.4% and 93.7%.[13] In our study, the sensitivity and specificity of colposcopy were 75% and 61%, respectively for prediction of CIN 1 or more; for CIN 2 or more, the sensitivity was better at 87.5% with a specificity of 50%.

Many studies have evaluated the role of p16Ink4a immunostaining for triaging cytology smears and reducing referrals for colposcopy. Carrozi et al. in 2008 did a multicenter randomized controlled trial involving 24,661 women comparing Pap cytology with p16 immunostaining for triaging HPV+ve women and found that the sensitivity and specificity of CIN 2 or more was 88% and 61%, respectively and for CIN 3 or more was 61% and 59%, respectively concluding that the sensitivity was better with p16Ink4a immunostaining than with Pap cytology for triage.[14] Tsoumpou et al. in 2009 did a meta-analysis of 61 studies and found that there is good evidence that p16Ink4a immunostaining correlates with the severity of cytological/histological abnormalities.[15] The Primary ASC-US and LSIL Marker Study (PALMS) multicenter study carried out in five European countries for screening cervical cancer precursors with p16/ki-67 dual stain cytology involving 27,349 women of 18 years or older, found that dual stain cytology was more sensitive than Pap cytology (86.7% vs. 68.5%) for detecting CIN2+, with comparable specificity (95.2% vs. 84.7%).[16]

Comparing p16Ink4a immunostaining versus histopathology in VIA-positive women; we found p16Ink4a positivity was highest with SCC and CIN 3 (100%) followed by 80% positivity in CIN 2 and 6% in CIN 1. The sensitivity and specificity for predicting CIN 1 or more lesions was 46.88% and 100%, respectively. For predicting CIN 2 or more lesions, the sensitivity was higher at 87.50% with a specificity of 97.06%.

Of our VIA-positive women; 36% had normal histology, 32% had low-grade CIN, 32% had CIN 2 or more. The sensitivity of colposcopy for predicting CIN 2 or more was 87.5%; however, specificity was only 50%. p16Ink4a had equal sensitivity to colposcopy at 87.5% with a higher specificity of 97%. With a comparable sensitivity to colposcopy, p16Ink4a staining was accurately able to identify those VIA-positive women with high-grade CIN who needed definitive management.

India has the highest number of cervical cancer deaths in the world; logistically a cytology-based screening program is not practical. Therefore, the government of India's operational guidelines based on WHO recommendations recommend screening by auxiliary nurse midwives (ANMs) using VIA as a screening method.[1718] VIA-positive women are then to be referred to doctors at the health center for treatment using cryotherapy or loop electrosurgical excision procedure (LEEP) with or without colposcopy and biopsy depending on the expertise of the doctor.[16]

There is a dire shortage of doctors trained in colposcopy, cryotherapy, and LEEP. With the screening program being rolled out across India, there will be a large pool of VIA-positive women generated after screening who will need to be referred for treatment with or without colposcopy. Identifying the women who definitely need biopsy and treatment is a challenge. Here emerges the role of p16Ink4a as a triaging method. A slide for p16Ink4a can be taken at the peripheral health center by a health care worker and sent to a centralized lab. Only the p16Ink4a positive women could be called for cryotherapy or LEEP at specialized centers.

The limitations of our study are the small sample size; a limiting factor for a bigger study would be the availability of the dye, which is presently available in very few centers in India.

The interpretation of our findings could affect the way screened women are managed. Presently screen and treat by cryotherapy are recommended by the WHO. Availability of p16Ink4a immunostaining can correctly identify those VIA-positive women needing biopsy and treatment and avoid overtreatment. The VIA-positive women could have the p16Ink4a slide taken at the peripheral center, reducing the number of patients traveling to the specialized center for definitive management. Only the p16 Ink4a-positive VIA-positive women could then be referred to the tertiary center.

Interpretation of p16 slides can potentially be done by trained health care workers. Wentzensen et al. found a high interobserver agreement between a trained cytologist and one with minimal training in cytology.[19] Therefore, p16Ink4a immunostaining holds great promise in VIA-based screening programs.

The other question which this study raises is when p16Ink4a becomes more widely available, will colposcopy become redundant? Colposcopy will still be needed to identify the abnormal areas and tailor biopsy and treatment. The few colposcopists located in specialized centers could then be used for these selected patients. Triaging by p16Ink4a immunostaining could reduce the number of women needing travel for colposcopy and treatment. Presently, p16Ink4a immunostaining is available only at a few centers and needs a long procedure of staining. In the future, durable, simplified, cost-effective stains could be developed which could be followed widely.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.