Usefulness of p16/ki67 Immunostaining in the Triage of Women Referred to Colposcopy

Transcription

1 Usefulness of p16/ki67 Immunostaining in the Triage of Women Referred to Colposcopy Jaume Ordi, MD, PhD 1 ; Amaia Sagasta, MD 1 ; Meritxell Munmany, MD 2 ; Leonardo Rodrıguez-Carunchio, MD 1 ; Aureli Torne, MD, PhD 2 ; and Marta del Pino, MD, PhD 2 BACKGROUND: This study compared the performance of p16/ki67 dual-staining and human papillomavirus (HPV) testing in women referred to colposcopy and sought to determine the usefulness of a morphological evaluation of the doublestained cells. METHODS: This prospective study included 1123 women (mean age, years) referred to colposcopy from October 2009 to November 2012 due to positive HPV testing or abnormal cytology results (atypical squamous cells of unknown significance, or worse abnormalities). Liquid-based cytology specimens (PreservCyt, Hologic) were used for HPV detection (Hybrid Capture 2 [HC2]; Qiagen) and p16/ki67 dual-staining (CINtec Plus; Roche-mtm Laboratories). All women underwent histological study. After completion of the study, 18 patients were classified as having cervical cancer (CC), 378 had a high-grade squamous intraepithelial lesion (HSIL), 304 had a low-grade squamous intraepithelial lesion, and 423 were negative. RESULTS: The sensitivity and specificity of p16/ki67 dual-staining for HSIL/CC were 90.9% (95% confidence interval [CI] ) and 72.1 (95% CI ), respectively. For HC2, the figures were, respectively, 96.0% (95% CI ) and 41.4 (95% CI ). The values were high both in women < 30 and 30 years old (86.9% and 63.3% versus 92.3% and 77.8%, respectively). The addition of a morphological evaluation of the dual-stain positive cells with establishment of HSIL features as the threshold for a positive reaction increased the specificity (93.5%) but decreased the sensitivity (84.1%). CONCLUSIONS: Use of the molecular markers p16 and Ki67 has higher specificity than HC2 tests for HSIL or CC, which may support p16/ki67 dual-staining use in the triage of patients referred for abnormal screening results. Morphological evaluation of p16/ki67-positive cells may have some benefits in women younger than 30 years or with low-grade squamous intraepithelial lesion. Cancer (Cancer Cytopathol) 2014;122: VC 2013 American Cancer Society. KEY WORDS: human papillomavirus; p16/ki67; cervical intraepithelial neoplasia; cytology. INTRODUCTION Programs for cervical cancer (CC) screening based on the Papanicolaou test (Pap test) have shown a reduction of morbidity and mortality caused by this neoplasm. 1 However, there is overwhelming evidence that cytology presents suboptimal sensitivity and interobserver subjectivity. 2 In recent years, molecular tests of high-risk human papillomavirus (hr-hpv) detection have achieved a sensitivity of up to 95% to 97% for the detection of CC and its precursors, although with a slightly lower specificity than cytology. 2,3 Consequently, these Corresponding author: Jaume Ordi, MD, PhD, Department of Pathology, CRESIB-Hospital Clınic, Barcelona, Spain, C/Villarroel 170, Barcelona, Spain; Fax: (011) ; jordi@clinic.ub.es 1 Department of Pathology, CRESIB (Centre de Recerca en Salut Internacional de Barcelona)-Hospital Clınic, University of Barcelona, Barcelona, Spain; 2 Institut Clinic of Gynecology, Obstetrics and Neonatology, Hospital Clınic-Institut d Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), University of Barcelona, Barcelona, Spain Presented in part at the United States and Canadian Academy of Pathology Annual Meeting, Vancouver, BC, Canada, March 19-21, The authors are grateful to Lorena Marimon, Nadia Abu-Ligha, Roser Esteve, Silvia Alos, Francisco Manuel Perez, and Naiara Vega for their help with the evaluation of the p16/ki67 dual-stained slides. We thank Donna Pringle for the English revision of the manuscript. Received: August 12, 2013; Revised: September 9, 2013; Accepted: October 2, 2013 Published online November 7, 2013 in Wiley Online Library (wileyonlinelibrary.com) DOI: /cncy.21366, wileyonlinelibrary.com Cancer Cytopathology March

2 techniques are being increasingly used in the management of patients with cytological abnormalities and, more recently, in primary screening. Another limitation of the Pap test is that it detects a high number of patients with cell abnormalities such as atypical squamous cells of unknown significance (ASC- US) and low-grade squamous intraepithelial lesions (LSIL), with a relatively low risk of having an underlying high-grade lesion, consequently requiring further evaluation. Although the introduction of hr-hpv testing has been shown to be useful in the triage of women with ASC-US, this test provides very limited information in patients with LSIL, as demonstrated by the ALTS (Atypical Squamous Cells of Undetermined Significance/Low- Grade Squamous Intraepithelial Lesions Triage Study) trial, 4 because ALTS found that at least 80% of LSIL cytology samples were positive for hr-hpv. Most women with ASC-US, LSIL, or a positive hr-hpv test with normal cytology clear both the hr-hpv infection and the cytological abnormalities within a few months or years (transient infections), and only a minority have a persistently positive hr-hpv test, which is the backdrop from which a high-grade squamous intraepithelial lesion (HSIL) may eventually develop. 5,6 Therefore, there is an important need to explore new, more-specific techniques which may help in the triage of these women with mild cytologic abnormalities who are at a low risk of presenting underlying high-grade lesions. A number of molecular markers have recently been proposed that may provide a high sensitivity and specificity for detecting premalignant cervical lesions The protein p16 INK4a (p16) is a cell-cycle regulatory protein that induces cell-cycle arrest under normal physiological conditions. 9 This biomarker has shown to be effective in histology specimens and is widely used to improve the reproducibility of cervical biopsy evaluation. 8,11-14 In cytological analysis, p16 can improve the accuracy for detecting premalignant lesions The simultaneous detection of p16 and Ki67, a proliferation marker, within the same cervical epithelial cell has recently been proposed as a surrogate marker of cell-cycle deregulation mediated by transforming hr-hpv infections. This combination of biomarkers (dual staining) has provided promising results in cytology specimens. 19 However, its accuracy for the detection of CC and premalignant lesions has been analyzed in few studies so far. 10,19-25 According to the manufacturer, the evaluation of p16/ki67 immunostaining should be done independently of the morphology analysis. Nevertheless, it is not known whether the morphological evaluation of dual-positive cells provides valuable additional information. The aim of this study was: 1) to evaluate the clinical performance of p16/ki67 staining in a large series of women referred to colposcopy for abnormal CC screening results, and comparing this immunostaining with the results of hr-hpv testing; and 2) to determine whether morphological evaluation of the double-positive cells provides additional information. MATERIALS AND METHODS Study Subjects This prospective study included all women referred to the colposcopy clinic of the Department of Obstetrics and Gynecology of the Hospital Clınic of Barcelona, Spain, from October 2009 to November Referral to colposcopy was based on 1) a positive result in hr-hpv testing in women older than 30 years, 2) Pap test result of ASC- US, or LSIL persisting for at least 1 year, or 3) a Pap test result of atypical squamous cells cannot exclude highgrade lesion (ASC-H), atypical glandular cells (AGC), HSIL, or CC. All the referral tests (Pap and HPV testing) had been performed within the 6 months prior to the inclusion in the study. Patients previously treated for cervical disease or who were pregnant were excluded from the study. The study was approved by the Hospital Clınic Institutional Ethical Review Board, and all the participants provided informed consent. Patient Management On the visit to the colposcopy clinic, all the women underwent a Pap test, hr-hpv testing, p16/ki67 dualstaining cytology, colposcopy examination, and either colposcopically directed biopsy or endocervical curettage. Prior to the colposcopy procedure, a cervical sample was collected in all the women by use of a cytobrush, and the sample was then transferred to PreservCyt solution (Hologic Corp, Marlborough, Mass). The first part of the sample was used for ThinPrep liquid-based cytologic analysis. The residual material was first used for hr-hpv testing followed by p16/ki67 immunostaining. Colposcopy was performed using a Olympus Evis Exera II CV-180 colposcope (Olympus, Barcelona, Spain) after preparing the cervix with 5% acetic acid. A 228 Cancer Cytopathology March 2014

3 RESULTS We recruited 1169 women for the study. In 46 of these women (3.7%), the study could not be completed because of unsatisfactory Pap tests (n 5 12), insufficient material for HC2 (n 5 19), or insufficient/unsatisfactory cellularity for the p16/ki67 dual-staining test (n 5 15). Thus, a final total of 1123 women with a mean age 6 standard deviation of years (range, years) were included in the study. The referral abnormality was a positive result in hr-hpv testing with normal cytology in 36 women and an abnormal Pap test in 1087 (HSIL in 543 women, LSIL in 384, ASC-H in 29, AGC in 6, and ASCp16/Ki67 in Women Referred to Colposcopy/Ordi et al colposcopically directed biopsy was taken on identification of an abnormal area. 6,26 When the transformation zone was not completely visible, endocervical curettage was also performed using a Kevorkian curette. In women with a completely visible transformation zone with no colposcopic abnormalities, a random biopsy was performed from the transformation zone. Liquid-Based Cytology, hr-hpv Testing and p16/ki67 Dual-Staining Cytology Thin-layer cytology slides were prepared using the Thin- Prep T2000 slide processor (Hologic) and stained using the Papanicolaou method. Cytology slides were evaluated by a cytotechnologist and confirmed by a pathologist using the revised Bethesda nomenclature. 27 Detection of hr-hpv (serotypes 16, 18, 32, 34, 36, 39, 45, 51, 52, 56, 58, 59, and 68) was performed using the Hybrid Capture 2 (HC2) system (Qiagen, Gaithersburg, Md) of the material collected in liquid-based media (PreservCyt). A relative light unit of 1 (1.0 pg/ml) was used as the cutoff to classify a specimen as positive for hr- HPV. 28 Finally, the residual material was used from the PreservCyt vial. A slide was prepared for each case using the T2000 processor and was subsequently subjected to p16/ Ki67 dual-staining using the CINtec Plus kit (Rochemtm Laboratories, Heidelberg, Germany) according to manufacturer instructions. 10,19,23 Staining was performed using a Dako Autostainer Link 48 (Dako, Glostrup, Denmark). The slides were run in batches of 24, and all cases were evaluated by a trained cytotechnologist. An initial evaluation was directed at confirming the presence of the minimum criteria for squamous cellularity as defined by the Bethesda 2001 terminology. 27 Subsequently, the slide was reviewed to detect the presence of doubleimmunoreactive cervical epithelial cells, ie, cells with simultaneous brown cytoplasmic p16 immunostaining and red nuclear Ki67 immunostaining, which were interpreted as positive by dual-stained cytology analysis, independently of the morphological interpretation. All cases with dual-stain positive cell(s) were reviewed by a pathologist to confirm the result. In this final review, a morphological evaluation was performed and the dual-stained cells were classified as ASC (including ASC-US and ASC- H), LSIL, or HSIL depending on the nuclear abnormalities and the nuclear-to-cytoplasm ratio. Figures 1A and 1B show an example of an LSIL with dual-stain positive cells in LSIL cells. Figures 1C and 1D show an example of an HSIL with dual-stain positive cells in HSIL cells. For all the morphologic techniques (Pap test, p16/ Ki67 dual-staining, and histology), the observers were unaware of the results of the other techniques and of the hr-hpv testing. Final Diagnosis Diagnosis of CC, HSIL, and in situ adenocarcinoma was established in all cases after histological confirmation. Diagnosis of LSIL was determined based on either histological confirmation or an LSIL result in the Pap test with a negative biopsy. Women with negative biopsy and Pap test results of normal, ASC-US, ASC-H, or AGC were classified as negative for intraepithelial lesion or malignancy (negative). Data Analysis Statistical analysis was performed using SPSS, version 18.0, software (SPSS, Inc, Chicago, Ill). The chi-square test was used for analyses of nominal variables. Sensitivity, specificity, positive (PPV) and negative predictive values (NPV), area under the receiver operating characteristic curve, and likelihood ratio were determined by comparing HC2 and p16/ki67 dual-staining with the final diagnoses. For these values, 95% confidence intervals were assessed using either binomial or normal distribution according to the data. For p16/ki67 dual-staining, these values were calculated considering 2 different thresholds for a positive reaction: 1) any positive dual-stained cell independently of the morphological characteristics, and 2) only positive dual-stained cells with cytomorphological features of HSIL. Cancer Cytopathology March

4 FIGURE 1. (A) Low-grade squamous intraepithelial lesion (LSIL), Papanicolaou-stained smear; (B) p16/ki67 dual-staining positive cell with morphological features of LSIL; (C) High-grade squamous intraepithelial lesion (HSIL), Papanicolaou-stained smear; (D) p16/ki67 dual-staining positive cells with morphological features of HSIL. US in 125). The final diagnosis after the completion of the study was invasive carcinoma in 18 women (1.6%, 16 squamous cell carcinomas, 2 adenocarcinomas), HSIL in 377 women (33.6%), in situ adenocarcinoma in 1 woman (0.1%), LSIL in 304 women (27.0%), and negative in 423 women (37.7%). Of the 304 women classified as LSIL, 165 women (54.3%) had a histological diagnosis, whereas in 139, the diagnosis was established on the basis of LSIL cytology in patients with negative biopsy. Of 423 women classified as negative because of a biopsy showing no SIL lesions, 68 (16.1%) showed ASC-US (51 women) or ASC-H (17 women) with the Pap test results. Table 1 shows the percentage of women with positive results for HC2 and p16/ki67 dual-stained cytology in each final diagnostic category. Table 2 shows the results of the morphological evaluation of the dual-stain positive cells in each final diagnostic category. Table 3 shows the clinical accuracy for HSIL/CC of hr-hpv testing and p16/ki67 dual-staining cytology using the 2 different thresholds (any positive cell versus positive cells with HSIL morphology). Compared with hr-hpv testing, p16/ki67 dual-staining had a slightly reduced sensitivity and NPV, but markedly increased specificity and PPV. The main benefit of the use of HSIL morphology as a threshold for positive p16/ki67 dualstain reaction was a marked increase in specificity and PPV. Table 4 shows the clinical performance of hr-hpv testing and p16/ki67 dual-stained cytology for HSIL/CC using the 2 different thresholds (any positive cell versus positive cells with HSIL morphology) after grouping the women according to age (younger and older than 30 years). The sensitivity and specificity for HSIL/CC were higher in women older than 30 years for both tests, but, in 230 Cancer Cytopathology March 2014

5 p16/ki67 in Women Referred to Colposcopy/Ordi et al Table 1. Percentage of Women With a Positive Result by Hybrid Capture 2 (HC2) and p16/ki67 Dual-Stained Cytology in Each Final Diagnostic Category Final Diagnosis No. of Women HC2-Positive No. (%) p16/ki67-positive No. (%) Negative (38.8%) 34 (8.0%) <.0001 LSIL (86.2%) 169 (55.6%) <.0001 HSIL a (96.6%) 344 (91.0%).002 Carcinoma (83.3%) 16 (88.9%).991 Negative indicates negative for intraepithelial neoplasia or malignancy. Abbreviations: HSIL, high-grade squamous intraepithelial lesion; LSIL, lowgrade squamous intraepithelial lesion. a Includes 1 woman with in situ adenocarcinoma. the group of women younger than 30 years, the specificity of p16/ki67 was double that of hr-hpv. The use of HSIL morphology as a threshold for p16/ki67 dual-stain positivity resulted in a marked increase in specificity and PPV. Table 5 shows the clinical performance of hr-hpv testing and p16/ki67 dual-stained cytology for HSIL/CC using the 2 different thresholds (any positive cell versus positive cells with HSIL morphology) after grouping the women according the referral Pap test result: 1) minor abnormalities: positive hr-hpv testing with negative cytology, or a Pap test result of ASC, AGC, or LSIL (n 5 580); and 2) women with referral Pap test of HSIL. In both groups, p16/ki67 dual-staining markedly increased the specificity and slightly reduced the sensitivity for the identification of HSIL or carcinoma when compared with HC2. DISCUSSION This is the largest prospective study evaluating the performance of p16/ki67 dual-staining for the identification of HSIL/CC in a colposcopy referral population. In comparison with hr-hpv testing using HC2, p16/ki67 dual staining showed a slightly reduced sensitivity and NPV, with an almost 2-fold higher specificity and PPV. Interestingly, the addition of cytomorphological evaluation of the p16/ki67 dual-stained cells significantly increased the specificity of the test to more than 90% while slightly reducing the sensitivity. These results indicate that p16/ Ki67 testing could reduce the number of patients referred to colposcopy by approximately half. Remarkably, p16/ Ki67 dual-staining showed a much lower variability in specificity and sensitivity related to the age of the patients, in contrast with HC2 testing that showed a significantly reduced specificity in patients younger than 30 years. In P Table 2. Results of the Morphological Evaluation of the Dual-Stain Positive Cells in Each Final Diagnostic Category Final Diagnosis No. of p16/ki67-positive Cases Morphology of Dual-Stain Positive Cells LSIL/ASC-US No. (%) HSIL No. (%) Negative (70.6%) 10 (29.4%) LSIL (78.1%) 37 (21.9%) HSIL a (7.6%) 318 (92.4%) Carcinoma 16 1 (6.2%) 15 (93.8%) Negative indicates negative for intraepithelial neoplasia or malignancy. Abbreviations: ASC-US, atypical squamous cells of unknown significance; HSIL, high-grade squamous intraepithelial lesion; LSIL, low-grade squamous intraepithelial lesion. a Includes 1 woman with in situ adenocarcinoma. the group of women usually referred to colposcopy according to the current clinical protocols such as patients with a Pap test result showing ASC, AGC, or LSIL, or those with HPV-positive testing with normal cytology, the use of p16/ki67 dual-staining maintained a marked increase in the specificity for the identification of HSIL or carcinoma with slight reduction of the sensitivity when compared with HC2. A possible limitation of our study is that the women included do not reflect a conventional primary screening population. In the present series, there is a low number of patients with positive hr-hpv testing and a normal Pap test, a high rate of women with ASC-US referred to the colposcopy clinic without previous triage with hr-hpv testing, and finally, there is a high percentage of women referred because of a Pap test result of HSIL. The infrequent use in our health care region of hr-hpv testing in primary screening and in the triage of women with ASC- US result in the Pap test, and the referral criteria for persistent LSIL and HSIL cases, contribute to explain these biases. Nevertheless, the number of patients included in each group allows reaching an adequate approximation of the accuracy of the p16/ki67 test for HSIL or cervical cancer compared with HC2 performance in the different scenarios. The Pap test, the most common method used in the screening of CC, has shown limited sensitivity to detect HSIL/CC. On the contrary, it does detect a significant number of patients with ASC-US and LSIL with a low risk of having an underlying high-grade HSIL/CC. 29 The recognition of hr-hpv as a necessary factor for developing CC has led to a paradigm shift in the prevention Cancer Cytopathology March

6 Table 3. Sensitivity, Specificity, and Positive and Negative Predictive Values (PPV and NPV) for High-Grade Intraepithelial Lesion or Carcinoma (HSIL/CC) of Human Papillomavirus (HPV) Detected with Hybrid Capture 2 and p16/ki67 Dual-Stained Cytology in the Whole Series of women (n ) a p16/ki67 Dual-Staining Hybrid Capture 2 Any Positive Cell Positive HSIL Cells % (95% CI) % (95% CI) % (95% CI) Sensitivity 96.0 ( ) 90.9 ( ) 84.1 ( ) Specificity 41.4 ( ) 72.1 ( ) 93.5 ( ) PPV 47.1 ( ) 63.9 ( ) 87.6 ( ) NPV 94.9 ( ) 93.6 ( ) 91.5 ( ) AUC 0.69 ( ) 0.82 ( ) 0.89 ( ) LR Abbreviations: AUC, area under the ROC curve; CI, confidence interval; LR1, positive likelihood ratio. a The clinical performance of p16/ki67 dual-stained cytology for HSIL/CC is presented using 2 different thresholds (any positive cell versus positive cells with HSIL morphology). Table 4. Sensitivity, Specificity, Positive and Negative Predictive Values (PPV and NPV), Area Under the Receiver Operating Characteristic Curve (AUC) and Positive Likelihood Ratio (LR1) for High-Grade Intraepithelial Lesion or Carcinoma (HSIL/CC) of Hybrid Capture 2 (HC2) and p16/ki67 Dual-Stained Cytology in Women Younger and Older Than 30 Years <30 Years (n 5 367) 30 Years (n 5 756) HC2 p16/ki67 (Any Positive Cell) p16/ki67 (Positive HSIL Cells) HC2 p16/ki67 (Any Positive Cell) p16/ki67 (Positive HSIL Cells) Sensitivity a 95.1% 86.9% 78.0% 96.4% 92.3% 86.5% ( ) ( ) ( ) ( ) ( ) ( ) Specificity a 30.2% 63.3% 91.0% 47.0% 77.8% 94.8% ( ) ( ) ( ) ( ) ( ) ( ) PPV a 40.4% 54.1% 81.4% 50.9% 70.3% 90.5% ( ) ( ) ( ) ( ) ( ) ( ) NPV a 92.5% 90.6% 89.6% 95.8% 94.7% 92.5% ( ) ( ) ( ) ( ) ( ) ( ) AUC ( ) ( ) ( ) ( ) ( ) ( ) LR The clinical performance of p16/ki67 dual-stained cytology for HSIL/CC is presented using 2 different thresholds (any positive cell versus positive cells with HSIL morphology). a Values in parenthesis are 95% confidence intervals. programs, with hr-hpv testing increasingly being considered in primary screening. Although hr-hpv testing is highly sensitive, it cannot discriminate between common transient infections and premalignant lesions, which are much less prevalent, 30 and increases the number of women referred to colposcopy. In recent years, several molecular markers thought to be more specific for HPVassociated transformation have been developed with the goal of improving the detection of premalignant lesions. p16/ki67 dual-staining on cytological slides has been shown to be one of the most promising tool for the triage of patients with abnormal screening results. 19 To date, only a few studies have investigated this test, with appreciable differences in the inclusion criteria (Table 6). 10,19-25 In most studies, the sensitivity and specificity are very high, similar to those observed in our study. 10,19,21-23 Only 2 studies showed a suboptimal sensitivity. In a small series performed in slides prepared with SurePath, Edgerton et al 20 reported a sensitivity of 64%. A recent report by Zappacosta et al described similar sensitivity values (62.3%), partly attributed to a high rate of insufficient residual specimen for p16/ki67 dual-staining after Pap slide processing. 24 In our series, the percentage of insufficient samples for the dual-staining was very low. 232 Cancer Cytopathology March 2014

7 p16/ki67 in Women Referred to Colposcopy/Ordi et al Table 5. Sensitivity, Specificity, Positive and Negative Predictive Values (PPV and NPV), Area under the ROC curve (AUC) and Positive Likelihood Ratio (LR1) for High-Grade Intraepithelial Lesion or Carcinoma (HSIL/CC) Using the 2 Different Thresholds (Any Positive Cell vs Positive Cells With HSIL Morphology) Women With Minor Abnormalities (Positive hr-hpv Testing With Negative Cytology, or a Pap Test Result of ASC, AGC, or LSI); and in Women With Referral Pap Test of HSIL Referral Pap Test of ASC/AGC/LSIL or HPV1 (n 5 580) Referral Pap test of HSIL (n 5 543) HC2 p16/ki67 (Any Positive Cell) p16/ki67 (Positive HSIL cells) HC2 p16/ki67 (Any positive cell) p16/ki67 (Positive HSIL Cells) Sensitivity a 95.6% 88.9% 83.3% 96.1% 94.5% 85.3% ( ) ( ) ( ) ( ) ( ) ( ) Specificity a 36.7% 72.9% 93.5% 51.1% 73.4% 89.9% ( ) ( ) ( ) ( ) ( ) ( ) PPV a 21.7% 37.4% 24.2% 71.7% 81.3% 91.6% ( ) ( ) ( ) ( ) ( ) ( ) NPV a 97.8% 97.3% 99.6% 91.0% 91.6% 82.6% ( ) ( ) ( ) ( ) ( ) ( ) AUC ( ) ( ) ( ) ( ) ( ) ( ) LR a Values in parenthesis are 95% confidence intervals. Table 6. Sensitivity and Specificity of p16/ki67 Dual-Staining for the Detection of High-Grade Intraepithelial Lesion or Carcinoma (HSIL/CC) in the Studies Published Author Year Preservation Referral cytology n Sensitivity Specificity Schmidt et al. (19) 2011 ThinPrep ASC-US and LSIL Petry et al. (22) 2011 ThinPrep Negative, HPV Wentzensen et al. (23) 2012 ThinPrep ASC-US, HPV1, and LSIL Waldstrom et al. (10) 2012 ThinPrep LSIL Loghavi et al. (21) 2012 ThinPrep ASC-US and LSIL Edgerton et al. (20) 2013 Surepath ASC-US Zappacosta et al. (24) 2013 ThinPrep LSIL, HPV Present study 2013 ThinPrep ASC-US1 1, Abbreviations: ASC-US, atypical squamous cells of unknown significance; HPV1, positive result for human papillomavirus testing; LSIL, low-grade squamous intraepithelial lesion; LSIL1, LSIL or HSIL; ASC-US1: ASC-US, atypical squamous cells, cannot exclude HSIL (ASC-H), atypical glandular cells (AGC), LSIL, or HSIL. However, it should be noted that in our series as well as in all previously published studies there is a small but significant reduction in sensitivity for HSIL or carcinoma between p16/ki67 dual-staining and HC2. In our study, all women referred because of a positive result in the hr-hpv testing or with ASC-US or LSIL in the referral Pap test were evaluated by colposcopy and biopsy, allowing the performance of p16/ki67 in the triage of these patients to be analyzed. For these women, who have low-risk of underlying high-grade lesions, p16/ Ki67 achieved a significantly better specificity compared with hr-hpv testing, representing a reduction of referral to colposcopy by almost half compared with the current practice. These findings support p16/ki67 as a viable option as a triage test for women with mild cytologic abnormalities. In the present study, p16/ki67 dual-staining showed a higher sensitivity and specificity for the detection of lesions among women 30 years of age and older, which is in keeping with previously published data. 23 Although slightly lower figures were observed in women younger than 30 years, the specificity doubled the results obtained with the HC2 test. This indicates that p16/ki67 dualstaining may be particularly useful in this subset of women, who frequently have transient hr-hpv infections and mild abnormalities. 6,23 According to the manufacturer, the p16/ki67 dualstain assay does not require adjunct morphologic interpretation of positive cells. 19,23 Thus, the test is considered positive when at least one cell exhibits p16/ki67 costaining. In this study, we evaluated, for the first time, the possible usefulness of adding morphological evaluation to the Cancer Cytopathology March

8 common dual-staining-only evaluation. Interestingly, when the threshold for a positive reaction was based on the presence of p16/ki67 dual-stain positive cells with HSIL morphology, we observed a marked increase in the specificity to more than 90%. This strategy could have some benefit in women with a low risk of presenting an underlying HSIL/CC such as in women under 30 years of age or patients with minor abnormalities in the Pap test. Neither this study nor previous reports have provided any information on the possible prognostic significance of p16/ki67 dual-stain positive results in cytological specimens from patients with LSIL or a negative biopsy. Evidence indicating that patients with histologically confirmed LSIL positive for p16 are at increased risk of progression 31,32 should encourage follow-up studies to assess the significance of the finding of p16/ki67 dual-stained cells in cytology specimens in this group of women. Moreover, there are no data to estimate the prospective risk of HSIL/CC in women testing positive by hr-hpv but with a negative result in the p16/ki67 dualstain. Further studies are required to evaluate the clinical significance of this finding. In summary, we show that p16/ki67 cytology has a high specificity for detecting cervical precancer in a large population of women referred to colposcopy with accurate disease ascertainment. Compared with HC2, this triage biomarker could reduce colposcopy referral, with a low risk of missing high-grade cervical lesions. The addition of a morphological evaluation to the p16/ki67 dualstain positive cells with a threshold based on HSIL morphology significantly increases the specificity. This strategy can be particularly useful in women under the age of 30 years or with LSIL result in the Pap test who are frequently positive for hr-hpv, but have a low risk of harboring HSIL/CC. FUNDING SOURCES This work was supported in part by the grants PI09/1084, PI09/ 1524, PI12/01231, and PI12/01165 from the Fondo de Investigaciones Sanitarias. CONFLICT OF INTEREST DISCLOSURE Dr. Ordi and Dr. Torne have been temporary advisors to Rochemtm Laboratories. All other authors made no disclosure. REFERENCES 1. Arbyn M, Raifu AO, Weiderpass E, Bray F, Anttila A. Trends of cervical cancer mortality in the member states of the European Union. Eur J Cancer. 2009;45: Cuzick J, Clavel C, Petry KU, et al. Overview of the European and North American studies on HPV testing in primary cervical cancer screening. Int J Cancer. 2006;119: Ordi J, Alonso I, Torne A, et al. Human papillomavirus load in Hybrid Capture II assay: does increasing the cutoff improve the test? Gynecol Oncol. 2005;99: The Atypical Squamous Cells of Undetermined Significance/Low- Grade Squamous Intraepithelial Lesions Triage Study (ALTS) Group. Human papillomavirus testing for triage of women with cytologic evidence of low-grade squamous intraepithelial lesions: baseline data from a randomized trial. J Natl Cancer Inst. 2000;92: Cuzick J, Sasieni P, Davies P, et al. A systematic review of the role of human papilloma virus (HPV) testing within a cervical screening programme: summary and conclusions. Br J Cancer. 2000;83: del Pino M, Torne A, Alonso I, et al. Colposcopy prediction of progression in human papillomavirus infections with minor cervical lesions. Obstet Gynecol. 2010;116: Cuzick J, Bergeron C, von Knebel Doeberitz M, et al. New technologies and procedures for cervical cancer screening. Vaccine. 2012;30(suppl 5):F107-F Galgano MT, Castle PE, Atkins KA, Brix WK, Nassau SR, Stoler MH. Using biomarkers as objective standards in the diagnosis of cervical biopsies. Am J Surg Pathol. 2010;34: von Knebel Doeberitz M, Reuschenbach M, Schmidt D, Bergeron C. Biomarkers for cervical cancer screening: the role of p16(ink4a) to highlight transforming HPV infections. Expert Rev Proteomics. 2012;9: Waldstrom M, Christensen RK, Ornskov D. Evaluation of p16(ink4a)/ki-67 dual stain in comparison with an mrna human papillomavirus test on liquid-based cytology samples with low-grade squamous intraepithelial lesion. Cancer Cytopathol. 2013; 121: Bergeron C, Ordi J, Schmidt D, Trunk MJ, Keller T, Ridder R. Conjunctive p16ink4a testing significantly increases accuracy in diagnosing high-grade cervical intraepithelial neoplasia. Am J Clin Pathol. 2010;133: Darragh TM, Colgan TJ, Thomas CJ, et al. The Lower Anogenital Squamous Terminology Standardization project for HPVassociated lesions: background and consensus recommendations from the College of American Pathologists and the American Society for Colposcopy and Cervical Pathology. Int J Gynecol Pathol. 2013;32: Ordi J, Garcia S, del Pino M, et al. p16 INK4a immunostaining identifies occult CIN lesions in HPV-positive women. Int J Gynecol Pathol. 2009;28: Dijkstra MG, Heideman DA, de Roy SC, et al. p16(ink4a) immunostaining as an alternative to histology review for reliable grading of cervical intraepithelial lesions. J Clin Pathol. 2010;63: Carozzi F, Cecchini S, Confortini M, et al. Role of P16(INK4a) expression in identifying CIN2 or more severe lesions among HPV-positive patients referred for colposcopy after abnormal cytology. Cancer. 2006;108: Carozzi F, Confortini M, Dalla Palma P, et al. Use of p16-ink4a overexpression to increase the specificity of human papillomavirus testing: a nested substudy of the NTCC randomised controlled trial. Lancet Oncol. 2008;9: Carozzi F, Gillio-Tos A, Confortini M, et al. Risk of high-grade cervical intraepithelial neoplasia during follow-up in HPV-positive women according to baseline p16-ink4a results: a prospective analysis of a nested substudy of the NTCC randomised controlled trial. Lancet Oncol. 2013;14: Denton KJ, Bergeron C, Klement P, Trunk MJ, Keller T, Ridder R. The sensitivity and specificity of p16(ink4a) cytology vs HPV 234 Cancer Cytopathology March 2014

9 p16/ki67 in Women Referred to Colposcopy/Ordi et al testing for detecting high-grade cervical disease in the triage of ASC-US and LSIL pap cytology results. Am J Clin Pathol. 2010; 134: Schmidt D, Bergeron C, Denton KJ, Ridder R. p16/ki-67 dual-stain cytology in the triage of ASCUS and LSIL papanicolaou cytology: results from the European equivocal or mildly abnormal Papanicolaou cytology study. Cancer Cytopathol. 2011;119: Edgerton N, Cohen C, Siddiqui MT. Evaluation of CINtec PLUS(R) testing as an adjunctive test in ASC-US diagnosed Sure- Path(R) preparations. Diagn Cytopathol. 2013;41: Loghavi S, Walts AE, Bose S. CINtec(R) plus dual immunostain: a triage tool for cervical pap smears with atypical squamous cells of undetermined significance and low grade squamous intraepithelial lesion. Diagn Cytopathol. 2013;47: Petry KU, Schmidt D, Scherbring S, et al. Triaging Pap cytology negative, HPV positive cervical cancer screening results with p16/ Ki-67 dual-stained cytology. Gynecol Oncol. 2011;121: Wentzensen N, Schwartz L, Zuna RE, et al. Performance of p16/ Ki-67 immunostaining to detect cervical cancer precursors in a colposcopy referral population. Clin Cancer Res. 2012;18: Zappacosta R, Caraceni D, Ciccocioppo L, et al. Implementing specificity of HPV-DNA primary screening in a successful organised cervical cancer prevention programme. Gynecol Oncol. 2013; 128: Dona MG, Vocaturo A, Giuliani M, et al. p16/ki-67 dual staining in cervico-vaginal cytology: correlation with histology, Human Papillomavirus detection and genotyping in women undergoing colposcopy. Gynecol Oncol. 2012;126: Walker P, Dexeus S, De Palo G, et al. International terminology of colposcopy: an updated report from the International Federation for Cervical Pathology and Colposcopy. Obstet Gynecol. 2003;101: Solomon D, Davey D, Kurman R, et al. The 2001 Bethesda System: terminology for reporting results of cervical cytology. JAMA. 2002;287: Terry G, Ho L, Londesborough P, Cuzick J, Mielzynska-Lohnas I, Lorincz A. Detection of high-risk HPV types by the hybrid capture 2 test. J Med Virol. 2001;65: Stoler MH, Schiffman M. Interobserver reproducibility of cervical cytologic and histologic interpretations: realistic estimates from the ASCUS-LSIL Triage Study. JAMA. 2001;285: Schiffman M, Wentzensen N. From human papillomavirus to cervical cancer. Obstet Gynecol. 2010;116: del Pino M, Garcia S, Fuste V, et al. Value of p16(ink4a) as a marker of progression/regression in cervical intraepithelial neoplasia grade 1. Am J Obstet Gynecol. 2009;201:488.e1-e Hariri J, Oster A. The negative predictive value of p16ink4a to assess the outcome of cervical intraepithelial neoplasia 1 in the uterine cervix. Int J Gynecol Pathol. 2007;26: Cancer Cytopathology March