Abstract Objectives This study aimed to evaluate the influence of prior knowledge of human papillomavirus (HPV) status in cervical cytopathology readings. Methods Participants comprised 2,376 women who underwent parallel cytology and HPV-DNA testing. Smears were read twice by the same team, first with previous knowledge of HPV-DNA status. Results Overall, 239 (10.2%) smears had their cytology classification altered by the HPV-informed review. Cytology readings with prior knowledge of the HPV status revealed 10.5% of abnormal smears (atypical squamous cells of undetermined significance or higher), while without prior knowledge, this rate dropped to 7.6%. When HPV status was informed, a significant increase in all categories of altered smears was observed. Cytology with prior knowledge of HPV status detected more cervical intraepithelial neoplasia grade 2 or higher (CIN 2+) compared with blinded: 86.7% vs 60.0%. Conclusions Our data indicate that cytology interpreted with prior knowledge of the HPV status provides higher sensitivity for CIN 2+ lesions while marginally reducing the overall specificity compared with HPV status blinded cytology. Cervical cytology, Screening, HPV-DNA, Papanicolaou test, Cervical cancer Cervical cancer (CC) remains the most important human papillomavirus (HPV)–related human cancer, with an estimated 530,000 new cases and 275,000 deaths annually.1 Cytology-based CC screening was introduced in the early 1950s and has led to substantial reductions in CC incidence and mortality, but it is well recognized that the interpretation of cervical cytology is relatively subjective and associated with high rates of interobserver and intraobserver variability. A single test has low sensitivity, of approximately 50% to 70%, requiring repeat testing throughout women’s lives, bringing complexity and costs to the strategy.2 Cytology laboratories are constantly making efforts to improve screening for the provision of better patient care, moving beyond the Papanicolaou (Pap) test.3 In the past decade, new and more effective technologies for cervical cancer control have emerged—namely, molecular testing for HPV, which was demonstrated to have great efficacy in large randomized screening trials,4,5 and liquid-based cytology (LBC), addressed to reduce cells’ overlapping and artifacts of preparation, potentially improving results while providing some practical advantages (eg, quicker interpretation, ancillary molecular testing).6,7 In addition, LBC analysis guided by a computerized system can select fields of view that facilitate the recognition of abnormal cells, potentially reducing the false-negative rates.8 In Brazil, the SUS (Public Unified Health System) recommends cytology-based CC screening, conducted with a 3-year interval after two consecutive annual negative examinations for women aged between 25 and 64 years. For several reasons, estimates of cancer incidence and mortality have remained high throughout the decades, with approximately 16,000 new cases and 5,430 deaths annually.9 Unlike cytology, testing for high-risk HPV (HR-HPV) DNA is objective, reproducible, and of superior sensitivity, displaying a higher negative predictive value than the Pap test. Nowadays, national programs worldwide are evaluating other algorithms such as HPV-based screening followed by cytology triage of women found to be HR-HPV positive or cytologic screening combined with HR-HPV testing.3 These strategies have been recognized as more efficient than cytology screening alone, improving the detection of cervical intraepithelial neoplasia grade 2 or higher (CIN 2+; 94% vs 65%)7; the HPV molecular test also offers a lower prospective (subsequent) risk of cervical intraepithelial neoplasia grade 3 (CIN 3) and invasive cervical cancer for women found HPV-DNA negative.5 Previous knowledge of HR-HPV DNA status may foster a reduction of cytology false negatives by leading to a better scrutiny of borderline abnormalities, consequently improving sensitivity.10,11 In a real-world scenario, with the expansion of the global use of HPV tests, it is expected that the cytopathologist will commonly have access to a patient’s HPV status at the time he or she evaluates the cytology specimen.12 Cytology reading samples with a positive HR-HPV test would probably depict a higher rate of abnormalities due to the attention paid by the readers, which may be an appropriate strategy in Brazil and countries with high incidences of cervical cancer and suboptimal cytology performance and control.13 Automated cytology evaluation based on the FocalPoint (BD, Burlington, NC) system may achieve a performance close to that of unaided microscopically interpreted cytology. This could lead to the deployment of integrated approaches using HPV testing and automated cytology, minimizing the subjectivity of cervical cancer screening.3 In this study, we aimed to evaluate the influence of HPV status reader awareness before computerized-guided cytologic screening vs cytology reviewed without prior knowledge of HPV status and compare the effectiveness and variation in the performance of the Pap test. Cytopathologists were blinded to their previous classification of the smear and also to the HPV result. They were informed that the second review was due to a quality control procedure foreseen in the study protocol. Materials and Methods Study Population Participants aged 14 to 86 years (mean [SD], 43 [14.6] years; median, 43 years) consisted of 2,376 women undergoing opportunistic routine CC screening at the Projeto Região Oeste, São Paulo, Brazil, in the following centers: Basic Health Units (supported by Faculdade de Medicina da Universidade de São Paulo) and the Interlagos Hospital and Maternity Center, from December 2014 to April 2015. Collection and Preparation of Cytologic Sample Cytology Cervical samples were collected from the endo-ectocervical junction. The whole brush head was transferred to the recipient containing BD SurePath liquid (BD Diagnostics—TriPath, Burlington, NC) and sent to the Fundação Oncocentro de São Paulo (FOSP) laboratory for preparation of slides for the cytology test and a residual aliquot for the molecular test Onclarity HPV Assay (BD), using the Totallys (BD) equipment. Slides were first scanned by the FocalPoint system (BD), and then 10 selected fields were revised by well-trained and well-experienced cytotechnologists and cytopathologists. Briefly, the FocalPoint system classifies 25% of all slides as no further review (NFR), in which the probability of intraepithelial lesions is extremely low. The other 75% are categorized in quintiles 1 to 5, in which quintile 1 has the highest probability of abnormality, based on slide scores. In this study, the NFR slides were manually screened to verify the reliability of the system in our experience. Smears were examined and classified according to the Bethesda system.14 Women with a cytologic diagnosis of atypical squamous cells of undetermined significance (ASC-US) or worse were classified as having abnormal cytology. Molecular Tests HPV was detected and genotyped by the Onclarity HPV Assay (BD), a real-time polymerase chain reaction (PCR) on the fully integrated Viper LT platform (BD) that targets E6/E7 genes. This assay provides individual genotyping information for six HPV types—16, 18, 31, 45, 51, and 52—while another eight HR-HPV types are reported in three distinct groups: P1 (33/58), P2 (56/59/66), and P3 (35/39/68). The Onclarity HPV Assay (BD) also detects the human β-globin gene as a control for both sample and process adequacy. Briefly, the sample was heated at 120°C for approximately 30 minutes in the Viper LT prewarm station, to lyse and homogenize the specimen, and then cooled to room temperature before being transferred onto the deck of the instrument, where it underwent automated sample processing and PCR amplification/detection.15 Cytology With or Without Prior Knowledge of HPV Status Original readings were performed with prior knowledge of HPV status, and women older than 24 years who showed a positive HR-HPV test and/or cytology class of ASC-US or worse were referred for colposcopic examination and biopsy, upon medical judgment. All histopathologic reports were reviewed and confirmed by two pathologists. In cases where there were discrepancies between the two readings, a third pathologist was consulted for a final consensus diagnosis. Six months later, the same samples were selected, previous pencil dots and original labels with patient IDs were removed, and slides were read by the same cytologists without prior knowledge of HPV DNA status and blinded to histology results. First and second readings of the slides were performed in the same laboratorial environment by the same group of cytotechnologists. Ethics The Ethics Committee of the Faculty of Medicine, University of São Paulo approved this study (No. 075/13), and all patients enrolled provided written informed consent. Statistical Analyses Frequency and estimates of sensitivity, specificity, negative predictive value, and positive predictive value for detection of CIN 2+ were obtained for both screening strategies: cytology with or without prior knowledge of HPV status. The confidence intervals (CIs) were calculated using the exact binomial distribution. Distribution of cytologic diagnosis according to the HPV status was performed using the test of a single proportion. Agreement between Pap readings (with or without prior knowledge of HPV status) was assessed using the κ statistic and the McNemar exact test for paired data. The odds ratio (OR) for matched samples was calculated to compare the proportion of nonnegative results between both Pap readings. All CIs and calculated statistical tests considered a level of significance of 5%. Analyses were performed using the Stata/IC 12.0 statistical software package (StataCorp, College Station, TX). Results HPV Genotype Distribution In total, 79.3% (1,885/2,376) of the study population comprised women aged 25 to 64 years, the target population for whom CC screening is recommended in Brazil. Among the 2,376 samples analyzed, 372 (15.7%) were HPV positive. Single-type infection was observed in 69.6% (259/372), while 22.6% (84/372) had coinfection by two HPV types, 4.8% (18/372) by three, 1.9% (7/372) by four, 0.8% (3/372) by 5, and 0.3% (1/372) by six. The most frequent HPV types in the series were P2 (HPVs 56/59/66) and P3 (HPVs 35/39/68) groups (4.3% each), followed by 16 (3.3%), 52 (2.5%), 31 (2.2%), P1 (HPVs 33/58) (2.1%), 51 (1.6%), 18 (1.1%), and 45 (0.9%). Cytology Four (0.2%) samples were classified as unsatisfactory for cytologic analysis and excluded. Overall, 239 (10.2%) smears had a different cytology classification in between both readings. Cytology readings with prior knowledge of HPV status revealed 10.5% of abnormal samples (ASC-US or worse), while without prior knowledge of HPV status, this rate dropped to 7.6% (P < .05). Of all the samples classified as abnormal, a significant increase was observed in the number of atypical squamous cells—cannot exclude high-grade squamous intraepithelial lesion (ASC-H) from one case to 34, low-grade squamous intraepithelial lesion (LSIL) from 64 to 82, and high-grade squamous intraepithelial lesion (HSIL) from 11 to 18 when HPV status was informed, whereas for the cytologic samples classified as ASC-US, no significant absolute difference was observed between cytology readings with or without prior knowledge of HPV status, respectively (n = 116 × 105) Table 1. As the absolute number of samples classified as ASC-US did not differ much between HPV results in previous knowledge groups, a superficial overview does not reveal the dual effect of HPV testing over ASC-US diagnostics; 59 of 105 smears were lowered to negative for intraepithelial lesion or malignancy (NILM) upon HPV knowledge, and of these, only 11 (18.6%) were HPV positive. On the other hand, 82 samples categorized as NILM were elevated to ASC-US when the HPV result was informed, with 44 (54%) being HPV DNA positive. Six (33.3%) of 18 samples previously classified as HSIL in the original readings changed to LSIL when the HPV result (all HPV positive) was concealed, and four (22.2%) of 18 cases changed to NILM cytology Table 2. McNemar test showed a P value less than .001, which is evidence that the changes when reading cytology with or without prior knowledge of HPV status are statistically significant (κ = 0.56). Table 1 Distribution of Cytologic Diagnosis According to the HPV Status Cytology With Prior Knowledge of HPV Status, No. (%) Without Prior Knowledge of HPV Status, No. (%) P Value 95% CI NILM 2,122 (89.5) 2,191 (92.4) <.05 91.3-93.4 ASC-US 116 (4.9) 105 (4.4) >.05 3.6-5.2 LSIL 82 (3.5) 64 (2.7) <.05 2.0-3.3 ASC-H 34 (1.4) 1 (0.01) <.05 0.0-0.1 HSIL 18 (0.8) 11 (0.5) <.05 0.2-0.7 Total 2,372 (100) 2,372 (100) Cytology With Prior Knowledge of HPV Status, No. (%) Without Prior Knowledge of HPV Status, No. (%) P Value 95% CI NILM 2,122 (89.5) 2,191 (92.4) <.05 91.3-93.4 ASC-US 116 (4.9) 105 (4.4) >.05 3.6-5.2 LSIL 82 (3.5) 64 (2.7) <.05 2.0-3.3 ASC-H 34 (1.4) 1 (0.01) <.05 0.0-0.1 HSIL 18 (0.8) 11 (0.5) <.05 0.2-0.7 Total 2,372 (100) 2,372 (100) ASC-H, atypical squamous cells—cannot exclude high-grade squamous intraepithelial lesion; ASC-US, atypical squamous cells of undetermined significance; CI, confidence interval; HPV, human papillomavirus; HSIL, high-grade squamous intraepithelial lesion; LSIL, low-grade squamous intraepithelial lesion; NILM, negative for intraepithelial lesion or malignancy. View Large Table 2 Changes in the Cytologic Classification According to Prior Knowledge of the HPV DNA Test Result Cytology With Prior Knowledge HPV Status Cytology Without Prior Knowledge HPV Status, No. NILM ASC-US LSIL ASC-H HSIL Total NILM 2,057 59 5 0 1 2,122 ASC-US 82 27 7 0 0 116 LSIL 25 14 40 0 3 82 ASC-H 23 3 6 0 2 34 HSIL 4 2 6 1 5 18 Total 2,191 105 64 1 11 2,372 Cytology With Prior Knowledge HPV Status Cytology Without Prior Knowledge HPV Status, No. NILM ASC-US LSIL ASC-H HSIL Total NILM 2,057 59 5 0 1 2,122 ASC-US 82 27 7 0 0 116 LSIL 25 14 40 0 3 82 ASC-H 23 3 6 0 2 34 HSIL 4 2 6 1 5 18 Total 2,191 105 64 1 11 2,372 ASC-H, atypical squamous cells—cannot exclude high-grade squamous intraepithelial lesion; ASC-US, atypical squamous cells of undetermined significance; HPV, human papillomavirus; HSIL, high-grade squamous intraepithelial lesion; LSIL, low-grade squamous intraepithelial lesion; NILM, negative for intraepithelial lesion or malignancy. View Large In general, knowledge of HPV presence led to an upgrade of the cytology class, whereas its absence led to a downgrade. Blinded reading identified only one case of ASC-H, while HPV-informed reading revealed 34 cases of ASC-H. This single ASC-H case was upgraded to HSIL (HPV positive), while the 34 cases of ASC-H on the first reading changed to 23 cases of NILM (four HPV positive and ultimately revealing CIN 2+ lesions), three cases of ASC-US (one HPV positive), six cases of LSIL (all HPV positive), and two cases of HSIL (HPV positive). There was moderate agreement (κ = 0.51) between the original readings and blinded cytology, and the McNemar test showed a P value less than .001, a statistically significant difference reinforced by an OR of 7.5 (95% CI, 4.1-15.0) when cytology was performed with prior knowledge of HPV-positive status. Figure 1 shows the results when comparing the two cytologic readings in HPV-positive and HPV-negative samples. Figure 1 View largeDownload slide A, Cytology assessments in human papillomavirus (HPV)–positive samples (n = 372). B, Cytology assessments in HPV-negative samples (n = 2,002). ASC-H, atypical squamous cells—cannot exclude high-grade squamous intraepithelial lesion; ASC-US, atypical squamous cells of undetermined significance; HSIL, high-grade squamous intraepithelial lesion; LSIL, low-grade squamous intraepithelial lesion; NILM, negative for intraepithelial lesion or malignancy. Figure 1 View largeDownload slide A, Cytology assessments in human papillomavirus (HPV)–positive samples (n = 372). B, Cytology assessments in HPV-negative samples (n = 2,002). ASC-H, atypical squamous cells—cannot exclude high-grade squamous intraepithelial lesion; ASC-US, atypical squamous cells of undetermined significance; HSIL, high-grade squamous intraepithelial lesion; LSIL, low-grade squamous intraepithelial lesion; NILM, negative for intraepithelial lesion or malignancy. Considering the samples read with prior knowledge of HPV-negative status, on NILM cytologic results (n = 2,002), we found 96.1% (1,926/2,002) vs 96.0% (1,919/2,002) in original readings and blinded cytology, respectively, with no difference between these two arms. There was moderate agreement (κ = 0.38) between the original readings and rereadings; no differences were observed (P = .69 [McNemar test]; OR, .91; 95% CI, 0.60-1.36) when cytology was read with prior knowledge of HPV-negative status. According to the protocol, 319 (13.4%) women were referred to triage by colposcopy due to HPV and/or altered cytology in the first reading, with 131 cases HPV positive, 22 cases ASC-US positive/HPV negative, and 166 cases positive for both tests. Of these, 143 (45%) had no abnormalities upon colposcopic inspection, and 70 were submitted for colposcopy-driven biopsies, revealing 20 cases of cervical intraepithelial neoplasia grade 1, 13 cases of CIN 2, and two cases of CIN 3, whereas 35 biopsy specimens were within normal limits. The remaining 106 (33%) women did not have a colposcopy visit. All 15 cases of CIN 2+ were HR-HPV positive, with HPV 16 the most frequent HR-HPV type (8/15, 53.3%), both in single and coinfections, followed by the P1 group (HPVs 33/58, 26.7%), with one of these being a single infection, as depicted in Table 3. Table 3 Profile of 15 CIN 2+ Cases Observed According to the Cytology and HPV Status Age, y HPV Type FPa Informed HPV Status Blind HPV Status Histology 62 P1 IC HSIL HSIL CIN 3 31 16, P1 1 ASC-H NILM CIN 3 25 16, 31, P1 1 ASC-H NILM CIN 2 35 16 4 NILM NILM CIN 2 59 P1 1 ASC-H HSIL CIN 2 35 P1 IC HSIL ASC-H CIN 2 33 52 5 ASC-H NILM CIN 2 63 31, 52, P3 1 HSIL HSIL CIN 2 22 16, P1, P2 2 ASC-H LSIL CIN 2 34 16, 31 3 LSIL HSIL CIN 2 35 31 2 NILM NILM CIN 2 21 16 1 HSIL ASC-US CIN 2 29 16 1 HSIL LSIL CIN 2 23 45, 52, P1, P2, P3 1 ASC-H LSIL CIN 2 20 16 3 ASC-H NILM CIN 2 Age, y HPV Type FPa Informed HPV Status Blind HPV Status Histology 62 P1 IC HSIL HSIL CIN 3 31 16, P1 1 ASC-H NILM CIN 3 25 16, 31, P1 1 ASC-H NILM CIN 2 35 16 4 NILM NILM CIN 2 59 P1 1 ASC-H HSIL CIN 2 35 P1 IC HSIL ASC-H CIN 2 33 52 5 ASC-H NILM CIN 2 63 31, 52, P3 1 HSIL HSIL CIN 2 22 16, P1, P2 2 ASC-H LSIL CIN 2 34 16, 31 3 LSIL HSIL CIN 2 35 31 2 NILM NILM CIN 2 21 16 1 HSIL ASC-US CIN 2 29 16 1 HSIL LSIL CIN 2 23 45, 52, P1, P2, P3 1 ASC-H LSIL CIN 2 20 16 3 ASC-H NILM CIN 2 ASC-H, atypical squamous cells—cannot exclude high-grade squamous intraepithelial lesion; ASC-US, atypical squamous cells of undetermined significance; CIN 2, cervical intraepithelial neoplasia grade 2; CIN 3, cervical intraepithelial neoplasia grade 3; FP, FocalPoint; HPV, human papillomavirus; HSIL, high-grade squamous intraepithelial lesion; IC, not processed; LSIL, low-grade squamous intraepithelial lesion; NILM, negative for intraepithelial lesion or malignancy. aThe software ranked each case into one of five categories: quintiles 1 (high risk) through 5 (low risk). View Large Two (13%) of 15 women with CIN 2+ had NILM in the original readings (with HPV knowledge). Considering the readings without prior knowledge of HPV status, six (40%) of 15 cases of CIN 2+ were missed by cytology. FocalPoint quintiles classified one of the two cases of CIN 3 in quintile 1 and 70% of CIN 2 (9/13) in quintiles 1 and 2. FocalPoint analyses were not possible in two samples (one case of CIN 2 and one case of CIN 3) due to artifacts or bubbled coverslips. The distribution of CIN 2+ cases according to the cytologic diagnosis and HPV types is shown in Table 3. As shown in Table 4, the sensitivity of computerized-guided cytology with prior knowledge of HPV status to detect CIN 2+ was higher compared with blinded cytology: 86.7% vs 60.0%, respectively. A slight loss in specificity was observed: 89.9% vs 92.7%, respectively. However, no significant differences were observed in positive and negative predictive values. HPV test alone detected more cases of CIN 2+ than either of the cytology strategies. It detected 100% of the cases of CIN 2+, with a decrease in the specificity (84.9%) and slight loss in the positive predictive value, without changing the negative predictive value. Table 4 Sensitivity and Specificity of HPV vs Cytology With and Without Prior Knowledge of HPV Status to Detect CIN 2+ Cases Characteristic HPV, No./Total No. (%) 95% CI, % Cytology With Prior Knowledge HPV Status, No./Total No. (%) 95% CI, % Cytology Without Prior Knowledge HPV Status, No./Total No. (%) 95% CI, % Sensitivity 15/15 (100.0) 100.0-100.0 13/15 (86.7) 69.5-100.0 9/15 (60.0) 35.2-84.8 Specificity 2,004/2,361 (84.9) 83.5-86.4 2,120/2,357 (89.9) 88.8-91.2 2,189/2,361 (92.7) 91.7-93.8 NPV 2,004/2,004 (100.0) 100.0-100.0 2,120/2,122 (99.9) 99.8-100 2,189/2,195 (99.7) 99.5-99.9 PPV 15/372 (4.0) 2.0-6.0 13/250 (5.2) 2.5-8.0 9/181 (5.0) 1.8-8.1 Characteristic HPV, No./Total No. (%) 95% CI, % Cytology With Prior Knowledge HPV Status, No./Total No. (%) 95% CI, % Cytology Without Prior Knowledge HPV Status, No./Total No. (%) 95% CI, % Sensitivity 15/15 (100.0) 100.0-100.0 13/15 (86.7) 69.5-100.0 9/15 (60.0) 35.2-84.8 Specificity 2,004/2,361 (84.9) 83.5-86.4 2,120/2,357 (89.9) 88.8-91.2 2,189/2,361 (92.7) 91.7-93.8 NPV 2,004/2,004 (100.0) 100.0-100.0 2,120/2,122 (99.9) 99.8-100 2,189/2,195 (99.7) 99.5-99.9 PPV 15/372 (4.0) 2.0-6.0 13/250 (5.2) 2.5-8.0 9/181 (5.0) 1.8-8.1 CI, confidence interval; HPV, human papillomavirus; NPV, negative predictive value; PPV, positive predictive value. View Large Discussion Numerous studies have confirmed that HPV testing is more sensitive than cytology, and thus it offers a higher cancer prevention benefit.16 Cervical cytology is recognized as being laborious and subjective, as well as having large variations in the cytologic classification.17,18 Reasons for this lack of reproducibility include differences between cytotechnician and pathologist skills, technical difficulties for accurate visualization of the cellular alterations, inadequate sampling of the squamous-columnar junction, borderline cytologic abnormalities, high workload demands, and so forth.19,20 Overall, cytologic screening with prior knowledge of HPV status resulted in an increased number of abnormal smears greater than ASC-US (P < .05; OR, 2.1; κ = 0.56), particularly in ASC-H (from one case to 34). These findings corroborate the results recently published by Wright et al11 showing a greater impact on LSIL and HSIL categories when cytology is read with prior knowledge of HPV status, with only 66.0% and 69.6%, respectively, of these diagnoses remaining the same after HPV informed readings. The same study revealed that 33.7% of ASC-US cases were reclassified as NILM after unblinded review, and 8.7% were upgraded to ASC-H or HSIL. In this study, the absolute number of samples classified as ASC-US did not differ much between HPV results in previous knowledge groups (116 × 105), essentially due to 59 samples that were downgraded to NILM, with most being HPV negative, and 82 that were upgraded to ASC-US, with most being HPV positive. Similarly, the ASC-US + ASC-H/LSIL + HSIL ratio remained at 1.4 on both readings because there was an increase in all altered categories—notably, ASC-H but also in LSIL and HSIL when HPV was known. In contrast, Benoy et al6 enrolled 2,905 women undergoing routine screening in Belgium and found a statistically significant difference only for LSIL (67 vs 46), an increase of 31% when HPV status was made available. Endorsing the findings described above, this influence is strongly observed in our subset of HPV-positive samples (OR, 7.5), with NILM cytology-blinded HPV results showing a decrease of 20.5% in samples previously positive in unblinded readings. This shift is probably attributable to a greater caution when slides from women known to be HPV positive are read by the cytotechnologists and pathologists, who may be influenced by this knowledge when interpreting cell abnormalities10,12 or increasing the importance of findings that would be otherwise ignored. In contrast, for the samples performed with prior knowledge of HPV-negative results, we found reasonable agreement (κ = 0.38), and no differences were observed between the readings (P = .69; OR, 0.91). Cytology is a subjective method. In this study, different assessments may have affected the results, and knowing that women were HPV negative in samples with borderline dysplastic changes possibly raised the confidence of cytotechnicians and pathologists toward NILM results. This was corroborated by the strong agreement found on NILM cases (96.1% vs 96.0% in both readings), since most smears were HPV negative. Richardson and colleagues10 compared the influence of the HPV status separately for three different settings; a total of 1,767 cervical cytology specimens were reread, and when women were HPV negative, the readings tended to be concordant between the original readings and rereadings. Certainly, the most important finding of this study was the demonstration that prior knowledge of the presence of HPV increased cytologic detection of CIN 2 lesions or worse compared with cytology screening alone, with a slight decrease in specificity but without reductions of positive and negative predictive values. Compared with blinded cytology, cytology with prior knowledge of HPV status revealed an increase of 26.7% in sensitivity and a slight decrease in specificity, similar to the gain of 17.4% in the sensitivity of cytology for CIN 2+ described by Benoy et al6 when performing cytology guided by HPV positive results, which also corroborates the recent study from Bergeron and colleagues12 that showed higher sensitivity for detection of precancer when cytology is evaluated with knowledge of HPV status. Currently, the modern cytology laboratory is moving beyond the Pap test read manually, attracted by the recently available technologies of computer-guided prescreening and the probability of abnormality based on quintile scores. The review by cytotechnologists and pathologists offers an additional instance for the recognition of cellular abnormalities. Laboratories with problems such as a high rate of false-negative Pap tests are benefiting from such technologies, improving their sensitivity.3 Our findings demonstrate that the FocalPoint system classified most CIN 2+ cases in quintiles 1 and 2, which favor the use of computer-assisted screening in the daily routine to avoid false-negative results. All 15 CIN 2+ cases were positive for the HR-HPV assay, and a slight decrease in the specificity (84.9%) was observed when comparing unblinded and blinded cytology. It is noteworthy that in this study, only women older than 24 years who showed a positive HR-HPV test and/or cytology ASC-US+ were referred to colposcopy and biopsy. A limitation that must be acknowledged was the high rate of nonattendance to the colposcopic assessment (33%). Although this rate is worrisome, it does reflect the reality of CC screening and triage in Brazil. Several reasons account for this loss: difficulties in making the colposcopic appointment, lack of understanding, noncompliance, and so on. However, it should be emphasized that HPV positivity and cytologic categories did not differ between women who attended colposcopy and nonattendants, which suggests that results may be applicable to the whole study population. The main argument against primary HPV testing is that despite the very high negative predictive value for CIN 3+, it has a lower specificity compared with cytology, leading to higher referral rates for colposcopy and likely some unnecessary treatments. Due to its high specificity, triage of HPV-positive women by cytology has been widely advocated, allowing HPV-positive/NILM women to be followed by HPV/cytology in shorter intervals but precluding their referral to colposcopy, which has been shown to be a safe strategy by long-term follow-up studies.21 By this approach, it is expected that cytology has a better performance for triage of HPV-positive women compared with primary screening,22 leading to an earlier diagnosis of persistent lesions and enhanced efficacy without increasing the biopsy rate.5 However, in the current study, two HPV-positive smears were classified as NILM on both readings with FocalPoint scores of 2 and 4, in which corresponding biopsy specimens were shown to harbor CIN 2 lesions. In such cases, p16/Ki-67 immunohistochemistry may clarify the oncogenic potential of the lesions, eventually representing limitations of the employed algorithm. The best strategy for women who are HPV positive with a subsequent negative triage cytology remains controversial. Despite having a low risk of developing CIN 3+, it is possibly not low enough to wait for their next screening round (eg, 3-6 years) as preconized for those found HPV negative. The CIN 3+ risk of women found HPV positive with negative cytology after 2 years was 5% in women aged 30 to 60 years in the Netherlands, a high risk to return in a routine screening interval, which can differ according to the country. An additional cytology test after 6 to 12 months decreased the risk to less than 1%, which seemed an acceptable strategy when considering the risk of CIN 3+.23 In the United States, a positive result for either HPV 16 or 18 was found to bear a CIN 3+ risk high enough to justify direct referral to colposcopy, skipping cytology triaging.24 In conclusion, our data suggest that cytology interpreted with prior knowledge of the HPV status is more sensitive than blinded cytology. Prior knowledge of HPV status could help cytotechnicians and pathologists to be more careful about minor HPV-associated cellular changes and more accurately classify cytologic changes in a CC routine screening laboratory. HPV vaccination promotes a decrease in the incidence of premalignant lesions,25 which is expected to affect cytologic screening since the finding of abnormal cells will become a rare event and thus less reproducible and of limited effectiveness.26 In addition, several countries are progressively implementing HPV primary screening programs in replacement of cytology. These two significant improvements in the prevention of cervical cancer are pressing for a repositioning of cervical cytopathology. In these fast-evolving times for medicine, the cytology laboratory should not be refractory to radical changes.3 Experts and national programs are converging to cytology-based triage of HR-HPV–positive women. An optimal integrated screening and triage marker strategy may be achieved by, for instance, dual staining of abnormal cervical cells for p16INK4a and Ki-67 antigens (eg, providing a high positive predictive value that may identify women who need immediate interventions, counterbalancing the low positive predictive value of the presence of HPV-DNA).22. Acknowledgments: We thank Becton-Dickinson (BD) for technical support, equipment, and reagents for the study, as well as cytologists and pathologists of the FOSP Laboratory. We also thank the Instituto Nacional de Ciência e Tecnologia das Doenças do Papilomavírus Humano INCT-HPV (National Institute of Science and Technology of the Diseases Associated to the Papillomavirus) coordinated by Luisa Lina Villa, São Paulo, Brazil, for financial support through CNPq 573799/2008-3 and FAPESP 2008/57889-1 grants. References 1. Poljak M, Ginocchio CC. Editorial: recent advances in molecular detection of human papillomavirus and cervical cancer screening. J Clin Virol . 2016; 76( suppl 1): S1- S2. Google Scholar CrossRef Search ADS PubMed 2. Castle PE, Cremer M. Human papillomavirus testing in cervical cancer screening. Obstet Gynecol Clin North Am . 2013; 40: 377- 390. Google Scholar CrossRef Search ADS PubMed 3. Joste N, Gober-Wilcox J. The modern cytology laboratory: moving beyond the Pap test. Obstet Gynecol Clin North Am . 2013; 40: 199- 210. Google Scholar CrossRef Search ADS PubMed 4. Kitchener HC, Almonte M, Thomson C et al. HPV testing in combination with liquid-based cytology in primary cervical screening (ARTISTIC): a randomised controlled trial. Lancet Oncol . 2009; 10: 672- 682. Google Scholar CrossRef Search ADS PubMed 5. Ronco G, Dillner J, Elfström KM et al. ; International HPV Screening Working Group. Efficacy of HPV-based screening for prevention of invasive cervical cancer: follow-up of four European randomised controlled trials. Lancet . 2014; 383: 524- 532. Google Scholar CrossRef Search ADS PubMed 6. Benoy IH, Vanden Broeck D, Ruymbeke MJ et al. Prior knowledge of HPV status improves detection of CIN2+ by cytology screening. Am J Obstet Gynecol . 2011; 205: 569.e1- 569.e7. Google Scholar CrossRef Search ADS 7. Herbert A. Primary HPV testing: a proposal for co-testing in initial rounds of screening to optimise sensitivity of cervical cancer screening. Cytopathology . 2017; 28: 9- 15. Google Scholar CrossRef Search ADS PubMed 8. Passamonti B, Bulletti S, Camilli M et al. Evaluation of the FocalPoint GS system performance in an Italian population-based screening of cervical abnormalities. Acta Cytol . 2007; 51: 865- 871. Google Scholar CrossRef Search ADS PubMed 9. INCA. Instituto Nacional do Câncer. Ministério da Saúde. 2016. http://www.inca.gov.br. Accessed January 26, 2016. 10. Richardson LA, El-Zein M, Ramanakumar AV et al. ; PEACHS (Pap Efficacy After Cervical HPV Status) Study Consortium. HPV DNA testing with cytology triage in cervical cancer screening: influence of revealing HPV infection status. Cancer Cytopathol . 2015; 123: 745- 754. Google Scholar CrossRef Search ADS PubMed 11. Wright TCJr, Stoler MH, Aslam S et al. Knowledge of patients’ human papillomavirus status at the time of cytologic review significantly affects the performance of cervical cytology in the ATHENA study. Am J Clin Pathol . 2016; 146: 391- 398. Google Scholar CrossRef Search ADS PubMed 12. Bergeron C, Giorgi-Rossi P, Cas F et al. Informed cytology for triaging HPV-positive women: substudy nested in the NTCC randomized controlled trial. J Natl Cancer Inst . 2015; 107: dju423. Google Scholar CrossRef Search ADS PubMed 13. Lorenzi AT, Syrjänen KJ, Longatto-Filho A. Human papillomavirus (HPV) screening and cervical cancer burden: a Brazilian perspective. Virol J . 2015; 12: 112. Google Scholar CrossRef Search ADS PubMed 14. Nayar R, Wilbur DC. The Pap test and Bethesda 2014. Cancer Cytopathol . 2015; 123: 271- 281. Google Scholar CrossRef Search ADS PubMed 15. Ejegod D, Bottari F, Pedersen H et al. The BD Onclarity HPV assay on samples collected in Surepath medium meets the international guidelines for human papillomavirus test requirements for cervical screening. J Clin Microbiol . 2016; 54: 2267- 2272. Google Scholar CrossRef Search ADS PubMed 16. Whitlock EP, Vesco KK, Eder M et al. Liquid-based cytology and human papillomavirus testing to screen for cervical cancer: a systematic review for the U.S. Preventive Services Task Force. Ann Intern Med . 2011; 155: 687- 697, W214. Google Scholar CrossRef Search ADS PubMed 17. Stoler MH, Schiffman M; Atypical Squamous Cells of Undetermined Significance-Low-Grade Squamous Intraepithelial Lesion Triage Study (ALTS) Group. Interobserver reproducibility of cervical cytologic and histologic interpretations: realistic estimates from the ASCUS-LSIL triage study. JAMA . 2001; 285: 1500- 1505. Google Scholar CrossRef Search ADS PubMed 18. Wright TCJr, Stoler MH, Behrens CM et al. Interlaboratory variation in the performance of liquid-based cytology: insights from the ATHENA trial. Int J Cancer . 2014; 134: 1835- 1843. Google Scholar CrossRef Search ADS PubMed 19. Castle PE. Understanding the limitations of cytology for screening. HPV Today—Newsletter on Human Papillomavirus. 2015. http://www.hpvtoday.com. Accessed January 26, 2016. 20. Branca M, Longatto-Filho A. Recommendations on quality control and quality assurance in cervical cytology. Acta Cytol . 2015; 59: 361- 369. Google Scholar CrossRef Search ADS PubMed 21. Dillner J, Rebolj M, Birembaut P et al. ; Joint European Cohort Study. Long term predictive values of cytology and human papillomavirus testing in cervical cancer screening: Joint European Cohort Study. BMJ . 2008; 337: a1754. Google Scholar CrossRef Search ADS PubMed 22. Wentzensen N, Schiffman M, Palmer T et al. Triage of HPV positive women in cervical cancer screening. J Clin Virol . 2016; 76( suppl 1): S49- S55. Google Scholar CrossRef Search ADS PubMed 23. Rijkaart DC, Berkhof J, Rozendaal L et al. Human papillomavirus testing for the detection of high-grade cervical intraepithelial neoplasia and cancer: final results of the POBASCAM randomised controlled trial. Lancet Oncol . 2012; 13: 78- 88. Google Scholar CrossRef Search ADS PubMed 24. Wright TCJr, Stoler MH, Sharma A et al. ; ATHENA (Addressing THE Need for Advanced HPV Diagnostics) Study Group. Evaluation of HPV-16 and HPV-18 genotyping for the triage of women with high-risk HPV+ cytology-negative results. Am J Clin Pathol . 2011; 136: 578- 586. Google Scholar CrossRef Search ADS PubMed 25. Garland SM, Kjaer SK, Muñoz N et al. Impact and effectiveness of the quadrivalent human papillomavirus vaccine: a systematic review of 10 years of real-world experience. Clin Infect Dis . 2016; 63: 519- 527. Google Scholar CrossRef Search ADS PubMed 26. El-Zein M, Richardson L, Franco EL. Cervical cancer screening of HPV vaccinated populations: cytology, molecular testing, both or none. J Clin Virol . 2016; 76 ( suppl 1): S62- S68. Google Scholar CrossRef Search ADS PubMed © American Society for Clinical Pathology, 2018. All rights reserved. For permissions, please e-mail: firstname.lastname@example.org
American Journal of Clinical Pathology – Oxford University Press
Published: Apr 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera