HPV-negative carcinoma of the uterine cervix: a distinct type of cervical cancer with poor prognosis

Transcription

1 DOI: / Gynaecological oncology HPV-negative carcinoma of the uterine cervix: a distinct type of cervical cancer with poor prognosis L Rodrıguez-Carunchio, a I Soveral, b RDM Steenbergen, c A Torne, b S Martinez, b P Fuste, b J Pahisa, b L Marimon, a J Ordi, a M del Pino b a Department of Pathology, Centre de Recerca en Salut Internacional de Barcelona (CRESIB), Hospital Clınic, University of Barcelona, Faculty of Medicine, Barcelona, Spain b Institute of Gynaecology, Obstetrics and Neonatology, Institut d 0 Investigacions Biomediques August Pi I Sunyer (IDIBAPS), Hospital Clınic, University of Barcelona, Faculty of Medicine, Barcelona, Spain c Unit of Molecular Pathology, Department of Pathology, VU University Medical Center, Amsterdam, the Netherlands Correspondence: Prof. J Ordi, Department of Pathology, CRESIB, Hospital Clınic, University of Barcelona, Faculty of Medicine, C/Villarroel 170, Barcelona, Spain. jordi@clinic.ub.es Accepted 2 July Published Online 17 September Objective Using highly sensitive polymerase chain reaction (PCR) techniques, we reanalysed all cervical carcinomas (CCs) found to be human papillomavirus (HPV)-negative by Hybrid Capture 2 (HC2) to determine the prevalence of true HPV-negativity. We also evaluated the characteristics of the patients with tumours with confirmed HPV-negativity. Design Observational study. Setting Barcelona, Spain. Population A cohort of 136 women with CC (32 adenocarcinomas, 104 squamous cell carcinomas) who had pre-treatment HC2 testing. Methods All negative cases were reanalysed and genotyped for HPV using three PCR assays (SPF10, GP5+/6+ and E7-specific assay). Main outcome measures Percentage of confirmed HPV-negative and HPV-positive tumours. Clinicopathological features and disease-free survival (DFS) and overall survival (OS) of both groups. Results Fourteen of 136 women (10.2%) were negative for HPV by HC2. After reanalysis by PCR-based techniques only 8/136 (5.8%) tumours were confirmed as HPV-negative, whereas in six cases different HPVs were identified [HPV 11, 16 (two tumours), 18, 45, and 68]. Confirmed HPV-negativity was more frequent in adenocarcinomas than in squamous cell carcinomas (5/32, 15.6% versus 3/104, 2.9%, respectively; P = 0.017). Patients with CCs with confirmed HPV-negativity had significantly worse DFS than women with HPV-positive tumours [51.9 months (95% CI months) versus months (95% CI months); P = 0.010]. In the multivariate analysis HPV-negativity and International Federation of Gynecology and Obstetrics (FIGO) staging were associated with increased risk of progression and mortality. Conclusions An HC2-negative result is an uncommon finding in women with CC, but in almost half of these cases HPVs are identified by more sensitive techniques. CCs with confirmed HPV-negativity are more frequently adenocarcinomas, and seem to be associated with worse DFS. Keywords Cervical cancer, human papillomavirus, p16, p53, prognosis, uterine cervix. Linked article This article is commented on by N Sethi, p. 128 in this issue. To view this mini commentary visit / Please cite this paper as: Rodrıguez-Carunchio L, Soveral I, Steenbergen RDM, Torne A, Martinez S, Fuste P, Pahisa J, Marimon L, Ordi J, del Pino M. HPV-negative carcinoma of the uterine cervix: a distinct type of cervical cancer with poor prognosis. BJOG 2015;122: Introduction Human papillomavirus (HPV) has been identified as the key causative agent for the development of cancer of the JO and MdP are both senior authors of the article. uterine cervix (CC). 1 The prevalence of HPV infection in CC has been estimated to be as high as 99%; 1,2 however, in almost all studies using highly sensitive HPV tests a small proportion of CCs are negative for HPV. 3,4 It has been suggested that HPV-negative CC may represent a biologically distinct subset of tumours carrying a poorer prognosis than 119

2 Rodrıguez-Carunchio et al. HPV-positive CCs, 5 7 but these findings have not been supported by other studies. 5 Thus, the significance of HPV-negativity in CC remains unclear. 3 Molecular techniques for HPV detection have a very high sensitivity for high-grade squamous intraepithelial neoplasia (HSIL) and CC. Evidence that sensitivity and negative predictive value are higher for HPV testing, than for the Pap test, 8 10 has resulted in the recent introduction of these techniques in the current guidelines of screening for CC prevention Hybrid Capture 2 (HC2), the first US Food and Drug Administration (FDA)-approved test, detects the presence of 13 high-risk HPV types and is one of the most frequently used methods for HPV screening in clinical practice; however, a small proportion of cases of HSIL and CC show a negative result with this test. 15 In a previous study by our group we analysed HSIL lesions negative for HC2 and found that, after HPV genotyping, common HPV types included in the HC2 probes were identified in over half of these lesions. 8 In the present study, we aimed to analyse CC cases that were negative for HPV by HC2 with highly sensitive polymerase chain reaction (PCR) techniques, and to determine the clinicopathological characteristics of the patients with tumours with confirmed HPV-negativity. The main objective of the study was to provide insight into the differential clinical, pathological, and prognostic characteristics of the unusual group of HPV-negative CCs. Methods Study design and case selection From July 1999 to December women were diagnosed with CC at the Gynaecological Oncology Unit. Age, histological diagnosis (squamous cell, adenocarcinoma), and International Federation of Gynecology and Obstetrics (FIGO) staging were retrieved from the clinical records of all of the patients. From this group we identified all women who had HPV testing performed by HC2, either simultaneously or within the 6 months prior to the histological diagnosis. Of the 406 women diagnosed, 136 were identified for inclusion (33.5%). Patients tested for HPV by HC2 were younger than those not tested ( years versus years; P = 0.017), but no differences in FIGO stage (P = 0.166) or histological diagnosis (P = 0.444) were observed between the overall group and the subset of patients tested for HPV. Nodal status, smoking habit, immunological status and follow-up data were retrieved from the clinical records from all cases tested for HPV by HC2. Study protocol According to the protocol of the Gynaecological Oncology Unit, upon referral evaluation all women routinely undergo a complete study, including clinical history, biopsy of the tumour for histological confirmation, clinical examination for FIGO staging, and pre-treatment imaging evaluation (abdominopelvic magnetic resonance imaging and/or computed tomography scan). In all women the primary location of the tumour in the uterine cervix was confirmed in the imaging evaluation, which was carefully reviewed by an expert radiologist, and patients with suspected endometrial or extrauterine tumours were excluded. HC2 testing HC2 analyses were performed in the cervical specimens collected using the Digene cervical sampler kit (Qiagen, Gaithersburg, MD, USA). Samples were stored at 20 C until further processing. All the samples were analysed for the presence of high-risk HPV types only (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68). The recommended cut-off of 1.0 pg/ml relative light units (RLU) was used to classify a specimen as positive or negative. 15 Clinical management and treatment Women with CC at FIGO staging Ia2 and Ib1 underwent a laparoscopy and an intraoperative sentinel node evaluation. Patients with negative sentinel nodes were treated either with a radical vaginal hysterectomy assisted by laparoscopy or a radical trachelectomy. Women with a positive sentinel node underwent complete pelvic and paraaortic lymphadenectomy without radical hysterectomy, followed by concurrent chemoradiotherapy treatment. Patients with locally advanced cervical cancer (FIGO staging Ib2 IIIb) underwent a complete paraaortic lymphdenectomy, with selective pelvic lymphadenectomy in the case of suspicious pelvic lymph nodes, followed by concurrent chemoradiotherapy. Chemo-radiotherapy consisted of six weekly single doses of 40 mg/m 2 cisplatin, administered simultaneously with radiation therapy. All patients received external beam radiation to the pelvic region followed by brachytherapy (six fractions of 4 5 Gy, with two or three fractions per week). A parametrial boost was performed in cases with neoplastic involvement (mean dose of 12 Gy). In addition to the pelvic radiation therapy, patients with histologically confirmed metastatic paraaortic lymph nodes received the same radiation scheme extending the external beam radiotherapy field to the aortic area, up to the level of the T12 vertebra. Histological evaluation and immunohistochemical detection of p16 INK4a All histological samples were fixed in 10% formalin and embedded in paraffin. All the original slides were reviewed by a gynaecopathologist (JO). The revision was based on haematoxylin and eosin (H&E) morphological criteria, plus immunohistochemistry (IHC) when necessary to 120

3 HPV-negative carcinoma of the uterine cervix exclude involvement by extracervical neoplasms (estrogen receptor, vimentin, CEA, CK20). IHC was performed with the Autostainer Link 48 â (Dako Co., Carpinteria, CA, USA), using the EnVision system (Dako). All tumours were stained with p16 INK4a (CINtec Histology Kit, clone E6H4; Roche-Mtm-Laboratories, Heidelberg, Germany) and p53 (monoclonal antibody DO 7; Dako). 16 For p16 only cases with diffuse and strong staining were considered as positive for analytical purposes, whereas cases with irregular or focal staining were considered negative. 17 For p53 only cases with nuclear staining in more than 25% of viable cells were considered positive. 18 Tissue preparation and nucleic-acid isolation DNA extraction and HPV genotyping were performed on formalin-fixed paraffin-embedded (FFPE) tissue in tumour tissue from the surgical specimen or from pre-treatment biopsy. The samples were serially sectioned on a microtome. The first and last sections (3 lm) were stained with H&E for histological confirmation of the diagnosis. In-between sections were collected in RNAase-free reaction tubes for DNA isolation (sandwich cutting technique). Sectioning and sample preparation were carried out with the highest measures to avoid contamination and cross-contamination. Paraffin blocks lacking tissue were cut in between the patient samples as controls to ensure the lack of contamination. None of these control samples were positive in the HPV PCR assay, indicating the adequate avoidance of contamination. DNA was extracted from the formalin-fixed paraffin-embedded samples by overnight incubation in 250 ll of proteinase K solution (1 mg/ml) at 70 C. Subsequently, proteinase K was heat-inactivated by incubation of the sections at 95 C for 10 minutes, and samples were spun down and cooled down at 20 C for 1 2 minutes. DNA was isolated using a commercially available kit (QIAamp DNA minikit; Qiagen, Hilden, Germany), according to the manufacturer s instructions, and 10 ll of isolated DNA was used for PCR amplification. DNA yields were quantified spectrophotometrically using the NanoDrop ND 1000 (Thermo Scientific NanoDrop, Wilmington, DE, USA). DNA detection and genotyping HPV genotyping was performed in all samples testing negative for HC2. To assess the quality of the DNA, b globin PCR analysis was performed using the primers BGPCO3 and BGPCO5, as described elsewhere. 19 PCR products were analysed by electrophoresis on a 1.5% agarose gel. All samples were b globin PCR-positive, indicating adequate DNA quality. Mucosal HPV DNA was detected by L1 region-based, GP5+/6+ PCR, as described previously. 20 HPV DNA was typed by reverse-line blot, as described previously. 21 This system allows the typing of both high-risk (16, 18, 26, 30, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68, 70,73, 82i, 82m) and low-risk HPV types (6, 11, 34, 40, 42, 43, 44, 54, 55, 57,61, 64, 71, 72, 81, CP6108). In addition, HPV typing was performed using the SPF10-LiPA system (Innogenetics Diagnostica Iberia SL, Barcelona, Spain), as described previously. 22,23 The test includes the same genotypes identified by the GP5+/6+ PCR plus the low-risk types HPV53 and 74. Finally, HC2-negative CCs were tested for the presence of high-risk HPV E7 DNA using a sensitive E7 multiplex PCR, enabling the detection of 14 high-risk HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68). 24 HC2-negative CCs that were negative for SPF10, GP5+/ 6+, and HPV E7 PCR were considered tumours with confirmed HPV-negativity. HC2-negative CCs that were positive for SPF10, GP5+/6+, and/or HPV E7 PCR were considered HPV-positive samples, and were analysed together with HC2-positive CCs. Data analysis SPSS 18.0 (SPSS, Chicago, IL, USA) was used for statistical analysis. The results for categorical variables are expressed as absolute numbers and percentages, and the chi-square or Fisher s exact tests were used to compare the variables. Disease-free survival (DFS) was defined as the time from diagnosis to the first local recurrence or metastasis. Overall survival (OS) was defined as the time from the date of registration to the date of death or to the last date of follow-up. Death without documented progression was censored at the date of death. Survival data were analysed with the Kaplan Meier method, and survival curves were compared using the log-rank test. For data analysis women with locally advanced and advanced FIGO staging were considered as one group. Multivariate logistic regression models were used to analyse the factors studied as prognostic factors using the risk estimation as hazard ratio (HR) with 95% confidence intervals (95% CIs). Two different models were analysed. The first model included HPV negativity, FIGO staging (advanced versus early stage), and histology (adenocarcinoma versus squamous cell carcinoma) as co-variables. The second model included HC2-negativity, FIGO stage (advanced versus early stage) and histology (adenocarcinoma versus squamous cell carcinoma) as co-variables. Results Fourteen out of 136 women with CC (10.2%) had a negative HC2 test. Table 1 shows the histological type, the viral load by HC2 in RLU, the results of HPV detection by PCR using the three different techniques, and the p16 INK4a and p53 immunostaining of the 14 HC2-negative tumours. 121

4 Rodrıguez-Carunchio et al. Table 1. Histological type, viral load by HC2 testing, SPF10 PCR, GP5+/6+ PCR, and E7-specific PCR results and immunohistochemical staining for p16 INK4a and p53 in HPV-negative cervical cancers Patient Histological type HC2 (RLU) PCR HPV SPF10 PCR HPV GP5+/6+ PCR HPV E7 p16 INK4a p53 1 ADC, mucinous Positive + 2 ADC, mucinous Positive + 3 SCC, keratinising Positive + 4 SCC, keratinising Positive + 5 SCC, non-keratinising 0.85 Negative 68 Positive + 6 SCC, non-keratinising Negative Negative* ADC, mucinous 0.18 Negative Negative Negative (+ focal) + 8 ADC, mucinous 0.15 Negative Negative Negative + 9 ADC, mucinous 0.23 Negative Negative Negative + 10 ADC, mucinous 0.14 Negative Negative Negative (+ focal) + 11 SCC, keratinising 0.50 Negative Negative Negative + 12 SCC, non-keratinising 0.13 Negative Negative Negative + 13 SCC, non-keratinising 0.29 Negative Negative Negative Adenosquamous carcinoma 0.27 Negative Negative Negative (+ focal) ADC, adenocarcinoma; ANED, alive with no evidence of disease; AWD, alive with disease; DOD, died of disease; SCC, squamous cell carcinoma. *The probe did not include the oligonucleotides for HPV11. After the re-evaluation with PCR techniques only eight (5.8%) tumours were found to be CCs with confirmed HPV-negativity. HPV DNA was detected by PCR-based techniques in six out of the 14 tumours testing negative for HC2 (42.8%). The results obtained with SPF10 and GP5+/ 6+ were identical in terms of positivity or negativity, and with respect to HPV genotyping in four out of the six cases. One case was positive for HPV11 using SPF10 and negative using GP5+/6+, and the second case was positive for HPV68 using GP5+/6+ and negative using SPF10. All cases excepting one of the PCR-positive HPV cases harboured HPV types included in the HC2 analysis (16, 18, 45, and 68). All samples positive for high-risk HPV by PCR were positive for HPV E7 PCR, confirming the presence of the virus. The eight cases negative for HPV as well as the case positive for HPV11 by PCR were negative for E7. Viral load, as determined by RLU, was significantly higher in HC2-negative tumours positive for HPV by PCR than in CCs with confirmed HPV-negativity (0.236 versus 0.535, respectively; P = 0.019). Four of the five women with HPV types included in the HC2 test had RLU levels above 0.40, whereas only one out of the nine women with tumours negative for HPV or with HPV types not included in the test (one case with HPV11) had RLU values higher than 0.40 (P = 0.022). p16 INK4a immunostaining was positive in five of the six (83.3%) HPV-positive tumours and in two of the eight (25%) tumours with confirmed HPV-negativity (P = 0.10). Figure S1 shows the histological characteristics and the p16 INK4a staining of three HPV-negative tumours. p53 staining was positive in two of the six HPV-positive (33.3%) and in six of the eight HPV-negative (75%) tumours (P = 0.277). The immunohistochemical panel performed in all adenocarcinomas with a negative HC2 test showed positivity for CEA in all cases and negative staining for vimentin and cytokeratin 20. Estrogen receptors were positive in two tumours (cases 7 and case 10, HPV-negative). Table 2 shows the clinical characteristics of the 14 patients with HC2-negative CCs. Only three of them were diagnosed at early FIGO staging. Lymph node metastases were identified in four out of 14 (28.6%) of the cases. All patients with positive lymph nodes had a locally advanced CC. Table 3 shows the comparison of smoking habits, immunological status, histological type, FIGO staging (early versus advanced staging), and lymph node status between women with CC with confirmed HPV-negativity (n = 8) and HPV-positive CC (n = 128). Adenocarcinomas were more frequently negative for HPV than squamous cell carcinomas [5/32 (15.6%) versus 3/104 (2.9%); P = 0.007]. Although the difference did not reach the level of significance, there was a trend towards a more advanced stage in HPV-negative CC [7/8 (87.5%) versus 67/128 (52.3%); P = 0.053]. Figure 1 shows the DFS (2A) and the OS (2B) of women with CC according to HPV status. Patients with CC with confirmed HPV-negativity had significantly worse DFS than women with HPV-positive CC [51.9 months (95% CI months) versus months (95% CI months); P = 0.010]. Patients with CC with confirmed HPV-negativity also had lower OS than women 122

5 HPV-negative carcinoma of the uterine cervix Table 2. Age at diagnosis, FIGO staging, lymph node status, relapse, months of first relapse, final status and follow up in months of women with HPV-negative cervical cancer Patient Age FIGO staging Lymph node metastases Relapse Months Status Months at last follow-up 1 58 IVa No No ANED IIb Yes No ANED IIb No No ANED IIb No No ANED Ib1 No No ANED IIIa Unknown No AWD Ib1 No No ANED Ib2 Yes No ANED IIIb Yes Yes 22.7 AWD IIb No Yes 11.3 DOD IIb Yes No ANED IIb No Yes 7.5 DOD IIIb No Yes 6.5 DOD IIb No No ANED ANED, alive, no evidence of disease; AWD, alive with disease; DOD, died of disease. Table 3. Smoking habits, immunological status, histological type, FIGO staging, and lymph node status according to HPV status HPV-negative CC (n = 8) HPV-positive CC (n = 128) P n % n % Smoking habit Immunosupression Histological type Squamous cell carcinoma Adenocarcinoma FIGO staging Early stage (Ia Ib1) Advanced stage (Ib2 IV) Lymph node metastasis HPV-negative cervical carcinomas (CC) were tumours found to be negative for HPV both by Hybrid Capture 2 (HC2) and after reanalysis by PCR (SPF10, GP5+/6+, and RT-PCR for E7). HPV-positive CCs were tumours that were either HC2-positive or positive in reanalysis by PCR. Data are presented as total numbers and percentages. with HPV-positive tumours, but the differences did not reach statistical significance [67.7 months (95% CI months) versus months (95% CI months), respectively; P = 0.225]. No differences were observed in terms of DFS or OS on grouping patients according to HC2 results [78.2 months for HC2-negative (95% CI months) versus months for HC2-positive (95% CI ), for DFS; P = 0.308; 84.5 months (95% CI months) versus months (95% CI months) for OS; P = 0.688]. No differences were observed in terms of DFS or OS on grouping patients according to histologic type (105.3 months in squamous cell carcinoma [95% CI months) versus 97.4 months (95% CI months) in adenocarcinomas for DFS; P = 0.576; months (95% CI ) versus months (95% CI months) for OS; P = 0.092]. Table 4 shows the multivariate Cox models for disease progression and mortality. An HPV-negative status was associated with an increased risk of progression and mortality. An advanced FIGO stage was also associated with progression and increased mortality. 123

6 Rodrıguez-Carunchio et al. (A) (B) Figure 1. Kaplan Meier curves for disease-free survival (A) and overall survival (B), stratified by human papillomavirus (HPV) status. HPV-negative cervical carcinomas (CCs) were tumours that scored negative for HPV both by Hybrid Capture 2 (HC2) and after reanalysis by PCR (SPF10, GP5+/6+, and RT-PCR for E7). HPV-positive CCs were tumours that were either HC2-positive or positive in reanalysis by PCR. Discussion Main findings A small number of CCs showed a negative result with the HC2 screening technique (10.2%). This proportion dropped to almost half (5.8%) after re-evaluating the negative cases with three highly sensitive PCR techniques. A significant proportion (six out of 14) of the HC2-negative CCs harboured HPV types included in the HC2 test (16, 18, 45, and 68). HPV-negativity was more frequently observed in adenocarcinomas. The most relevant finding of our study was that women with confirmed HPV-negative tumours were more frequently diagnosed with adenocarcinomas, presented at more advanced stages, and had a worse DFS, with a trend, albeit not statistically significant, to a worse OS. Interestingly, the association between HPV status and DFS persisted in the multivariate analysis. No differences were observed in terms of DFS or OS on grouping patients according to either the HC2 results or the histologic type (squamous cell carcinoma versus adenocarcinomas). Strengths and limitations The main strength of our study is that it is the first comprehensive report of the characteristics of CCs testing negative for the widely used HPV screening test HC2, and it extensively analyses the negative cases with a series of highly sensitive PCR tests and evaluates the prognostic impact of HPV negativity. The main limitation of this study is the size of the sample, particularly the small number of tumours with confirmed HPV-negativity; however, because of the low frequency of HPV-negative CCs, even large referral centres are only able to recruit a very small number of women with these neoplasms. Thus, large multicentre studies are necessary to confirm the characteristics of this small subset of CCs. An additional limitation is that HPV testing was only performed in a limited subset (33.5%) of the patients with CCs diagnosed at our unit, as this procedure is not routinely performed in patients with CC; however, no differences in FIGO staging or in histological diagnoses were identified between the patients tested and those not tested for HPV. Further studies including a larger number of patients are warranted in order to increase the power to detect small differences and to exclude any biases in the results. Interpretation The HC2-negative rate observed in our series (10.2%) is similar to that observed in previous studies. 25 In addition, the lower percentage of HPV-negative CCs (5.8%) observed after re-evaluating the negative cases with more sensitive techniques is in keeping with the rates observed in other studies using highly sensitive PCR techniques. 3 PCR tests have a higher analytical sensitivity than HC2, 26 detecting very small quantities of HPV DNA. 27 Thus, the findings of our study strongly support the possible development of CC independently of HPV infection. After an extensive search for the presence of HPV in all HC2-negative cases using 124

7 HPV-negative carcinoma of the uterine cervix Table 4. Multivariate Cox models for disease progression and mortality Variable Mortality Disease progression HR (95% CI) P HR (95% CI) P Model 1 HPV negativity* 4.72 ( ) ( ) FIGO staging (advanced stages) 3.70 ( ) ( ) Histology (adenocarcinoma) 0.23 ( ) ( ) Model 2 HC2 negativity* 1.22 ( ) ( ) FIGO staging (advanced stages) 4.06 ( ) ( ) Histology (adenocarcinoma) 0.42 ( ) ( ) *Includes only tumours negative for all HPV tests. different techniques for HPV detection, a small proportion of CC proved to be negative by all techniques. Interestingly, within the group of women with HC2-negative samples, the mean RLU in women with positive HPV detected by PCR was higher than in women with tumours with confirmed HPV-negativity. Although it has been widely shown that 1.0 RLU has the best balance between sensitivity/specificity, and should be considered the standard for CC screening, some studies have pointed out that despite being considered as negative for HPV, HC2 test values between 0.4 and 1.0 RLU indicate the presence of a low viral load. 28 Indeed, all but one of the women with HC2-negative CCs harbouring HPVs by PCR test had RLUs above 0.40, whereas only one of the CCs with confirmed HPV-negativity had a viral load above this threshold. As observed by our group with HSIL lesions, 8 our study confirms that a significant proportion of the HC2-negative CCs (57.2%) harboured HPV types included in the HC2 test (16, 18, 45, and 68). Only one of the HC2-negative CCs was positive for a low-risk HPV type (HPV11). Low-risk HPVs are common in the population, but are rarely identified in CC. 3,29 Although it is unknown in which conditions these HPV types can induce cancer, there is strong evidence that these low-risk HPV types are causally involved in a very small percentage of anogenital carcinomas. 3,4,29 It is widely accepted that HPV is a necessary cause of CC; however, several studies have shown that a proportion of CCs, ranging from 4 to 52%, are HPV-negative. 3 5,30 32 The low sensitivity of some HPV-testing methods when applied to formalin-fixed paraffin-embedded tissues is likely to be the cause of the low HPV prevalence in CCs observed in some studies. 5 Poor-quality DNA from tumour specimens arising from suboptimal histological processing and error in histological classification by the inclusion of endometrial neoplasms as cervical adenocarcinomas have been proposed as possible causes of false HPV-negative results in CC. In the present study, the quality of DNA was assessed by b globin PCR in all of the samples, and was found to be acceptable and the diagnosis was confirmed by careful imaging and histological review with immunohistological staining to exclude extracervical tumours. Finally, a low HPV DNA content in some CCs has been considered as a possible cause of false-negative HPV-testing results, 3 which is in keeping with the results observed in our study. Another possible explanation for HPV-negativity could be the de-differentiation and subsequent loss of HPV within the tumour. Nevertheless, the design of the present study does not allow for determining whether the HPV-negative CCs arose independently of HPV infection or were initially HPV-positive and later became independent from the virus. In keeping with the results of previous reports, HPV-negativity was more frequently found in adenocarcinomas. 3,4,31,33 35 It has been shown that some subtypes of adenocarcinomas are commonly HPV-negative (minimal deviation adenocarcinoma, and gastric-type, clear-cell, serous, and mesonephric adenocarcinomas); however, all adenocarcinomas included in our study were of the conventional type. 36 A recent series analysing 760 cervical adenocarcinomas has reported that over 80% of these tumours are of the conventional or mucinous type, whereas non-mucinous variants are rare. Although non-mucinous variants (endometrioid, serous, clear cell, and minimal deviation) are more frequently HPV-negative (70 80%), almost 30% of these conventional mucinous tumours are HPV-negative. 37 The impaired prognosis of HPV-negative CC has already been reported in a few previous series. 6,7 This poorer prognosis for HPV-negative tumours has been observed in other locations in which HPV-associated carcinomas are 125

8 Rodrıguez-Carunchio et al. found. Indeed, in the head and neck region, HPV-positive tumours have consistently shown a better prognosis than HPV-negative neoplasms, and this phenomenon has been related to a better response to chemotherapy and radiation therapy. 38 Similarly, HPV-positive carcinomas of the vagina have shown a better prognosis than HPV-negative tumours. 39 In our series of CCs, tumours with confirmed HPV-negativity were frequently found in an advanced FIGO stage, and showed a higher rate of lymph node metastasis. We cannot exclude a better response to chemotherapy and radiation therapy for HPV-positive CCs, as observed in the head and neck region. Conclusion A negative result in HC2 testing is an uncommon finding in women with CCs, and about half of these tumours prove to harbour high-risk HPV types included in the HC2 test when tested with more sensitive PCR-based techniques. The unusual CCs with confirmed HPV-negativity are more frequently adenocarcinomas and are associated with a poorer DFS. Disclosure of interests The authors declare that they have no conflicts of interest. Contribution to authorship LRC, IS, MP, and JO designed the study, analysed and interpreted the data, and wrote the article. RS, AT, SM, PF, JP, and LM participated in the analysis of the data. All the authors approved the final article. Details of ethics approval Approved by the Ethics Committee of Clinical Research of the hospital clinic on 2 July 2013 (registry 2013/8486). Funding Funded by Instituto de Salud Carlos III (ICSIII)-Fondos de Investigacion Sanitaria and ERDF one way to Europe (PS09/1084, PI12/1165, PI12/1231). Acknowledgements We thank Donna Pringle for the English revision of the article. Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. Human papillomavirus (HPV)-negative carcinomas of the uterine cervix. (A, A 0 ) Mucinous adenocarcinoma (case 7); (B, B 0 ) mucinous adenocarcinoma (case 8); (C, C 0 ) keratinising squamous cell carcinoma (case 11). (A, B, C) haematoxylin and eosin stain; (A 0,B 0,C 0 ) p16 INK4a immunohistochemical staining. (A 0 ) Focal irregular staining. (B 0 ) Diffuse positive staining. (C 0 ) Negative staining. & References 1 Walboomers JM, Jacobs MV, Manos MM, Bosch FX, Kummer JA, Shah KV, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999;189: Monsonego J. Global challenges of cervical cancer prevention. Eur J Gynaecol Oncol 2000;21: de Sanjose S, Quint WG, Alemany L, Geraets DT, Klaustermeier JE, Lloveras B, et al. Human papillomavirus genotype attribution in invasive cervical cancer: a retrospective cross-sectional worldwide study. Lancet Oncol 2010;11: Li N, Franceschi S, Howell-Jones R, Snijders PJ, Clifford GM. Human papillomavirus type distribution in 30,848 invasive cervical cancers worldwide: variation by geographical region, histological type and year of publication. Int J Cancer 2011;128: Barreto CL, Martins DB, De Lima Filho JL, Magalhaes V. Detection of Human Papillomavirus in biopsies of patients with cervical cancer, and its association with prognosis. Arch Gynecol Obstet 2013;288: Higgins GD, Davy M, Roder D, Uzelin DM, Phillips GE, Burrell CJ. Increased age and mortality associated with cervical carcinomas negative for human Papillomavirus RNA. Lancet 1991;338: Riou G, Favre M, Jeannel D, Bourhis J, Le Doussal V, Orth G. Association between poor prognosis in early-stage invasive cervical carcinomas and non-detection of HPV DNA. Lancet 1990;335: del Pino M, Rodriguez-Carunchio L, Alonso I, Torne A, Rodriguez A, Fuste P, et al. Clinical, colposcopic and pathological characteristics of cervical and vaginal high-grade lesions negative for HPV by Hybrid Capture 2. Gynecol Oncol 2011;122: Clavel C, Cucherousset J, Lorenzato M, Caudroy S, Nou JM, Nazeyrollas P, et al. Negative human papillomavirus testing in normal smears selects a population at low risk for developing high-grade cervical lesions. Br J Cancer 2004;90: Cuzick J, Clavel C, Petry KU, Meijer CJ, Hoyer H, Ratnam S, et al. Overview of the European and North American studies on HPV testing in primary cervical cancer screening. Int J Cancer 2006;119: Saslow D, Runowicz CD, Solomon D, Moscicki AB, Smith RA, Eyre HJ, et al. American Cancer Society guideline for the early detection of cervical neoplasia and cancer. CA Cancer J Clin 2002;52: Meijer CJ, Berkhof J, Castle PE, Hesselink AT, Franco EL, Ronco G, et al. Guidelines for human papillomavirus DNA test requirements for primary cervical cancer screening in women 30 years and older. Int J Cancer 2009;124: Wright TC Jr, Massad LS, Dunton CJ, Spitzer M, Wilkinson EJ, Solomon D consensus guidelines for the management of women with abnormal cervical cancer screening tests. Am J Obstet Gynecol 2007;197: Saslow D, Solomon D, Lawson HW, Killackey M, Kulasingam SL, Cain J, et al. American Cancer Society, American Society for Colposcopy and Cervical Pathology, and American Society for Clinical Pathology screening guidelines for the prevention and early detection of cervical cancer. CA Cancer J Clin 2012;62: Ordi J, Alonso I, Torne A, Esteve R, Sierra E, Campo E, et al. Human papillomavirus load in Hybrid Capture II assay: does increasing the cutoff improve the test? Gynecol Oncol 2005;99:

9 HPV-negative carcinoma of the uterine cervix 16 del Pino M, Garcia S, Fuste V, Alonso I, Fuste P, Torne A, et al. Value of p16(ink4a) as a marker of progression/regression in cervical intraepithelial neoplasia grade 1. Am J Obstet Gynecol 2009;201: 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: Santos M, Landolfi S, Olivella A, Lloveras B, Klaustermeier J, Suarez H, et al. p16 overexpression identifies HPV-positive vulvar squamous cell carcinomas. Am J Surg Pathol 2006;30: de Roda Husman AM, Walboomers JM, van den Brule AJ, Meijer CJ, Snijders PJ. The use of general primers GP5 and GP6 elongated at their 3 ends with adjacent highly conserved sequences improves human papillomavirus detection by PCR. J Gen Virol 1995;4: Jacobs MV, Snijders PJ, van den Brule AJ, Helmerhorst TJ, Meijer CJ, Walboomers JM. A general primer GP5+/GP6(+)-mediated PCR-enzyme immunoassay method for rapid detection of 14 high-risk and 6 low-risk human papillomavirus genotypes in cervical scrapings. J Clin Microbiol 1997;35: van den Brule AJ, Pol R, Fransen-Daalmeijer N, Schouls LM, Meijer CJ, Snijders PJ. GP5+/6+ PCR followed by reverse line blot analysis enables rapid and high-throughput identification of human papillomavirus genotypes. J Clin Microbiol 2002;40: Kleter B, van Doorn LJ, ter Schegget J, Schrauwen L, van Krimpen K, Burger M, et al. Novel short-fragment PCR assay for highly sensitive broad-spectrum detection of anogenital human papillomaviruses. Am J Pathol 1998;153: Kleter B, van Doorn LJ, Schrauwen L, Molijn A, Sastrowijoto S, ter Schegget J, et al. Development and clinical evaluation of a highly sensitive PCR-reverse hybridization line probe assay for detection and identification of anogenital human papillomavirus. J Clin Microbiol 1999;37: Hesselink A, Berkhof J, van der Salm ML, van Splunter AP, Geelen TH, Van Kemenade FJ, et al. Clinical validation of the HPV-Risk assay: a novel, real-time PCR assay for the detection of high-risk human papillomavirus DNA by targeting the E7 region. J Clin Microbiol 2014;52: Li Z, Austin RM, Guo M, Zhao C. Screening test results associated with cancer diagnoses in 287 women with cervical squamous cell carcinoma. Arch Pathol Lab Med 2012;136: Szostek S, Klimek M, Zawilinska B, Rys J, Kope J, Daszkiewic E. Detection of human papillomavirus in cervical cell specimens by hybrid capture and PCR with different primers. Acta Biochim Pol 2006;53: Zaravinos A, Mammas IN, Sourvinos G, Spandidos DA. Molecular detection methods of human papillomavirus (HPV). Int J Biol Markers 2009;24: Jastania R, Geddie WR, Chapman W, Boerner S. Characteristics of apparently false-negative digene hybrid capture 2 high-risk HPV DNA testing. Am J Clin Pathol 2006;125: Guimera N, Lloveras B, Alemany L, Iljazovic E, Shin HR, Jung-Il S, et al. Laser capture microdissection shows HPV11 as both a causal and a coincidental infection in cervical cancer specimens with multiple HPV types. Histopathology 2013;63: Guan P, Howell-Jones R, Li N, Bruni L, de Sanjose S, Franceschi S, et al. Human papillomavirus types in 115,789 HPV-positive women: a meta-analysis from cervical infection to cancer. Int J Cancer 2012;131: Igidbashian S, Schettino MT, Boveri S, Barberis M, Sandri MT, Carinelli S, et al. Tissue genotyping of 37 in situ and invasive cervical cancer with a concomitant negative HC2 HPV DNA test. J Low Genit Tract Dis 2014;18: Lai CH, Huang HJ, Hsueh S, Chao A, Lin CT, Huang SL, et al. Human papillomavirus genotype in cervical cancer: a population-based study. Int J Cancer 2007;120: Andersson S, Larson B, Hjerpe A, Silfversward C, Sallstrom J, Wilander E, et al. Adenocarcinoma of the uterine cervix: the presence of human papillomavirus and the method of detection. Acta Obstet Gynecol Scand 2003;82: Clifford G, Franceschi S. Members of the human papillomavirus type 18 family (alpha-7 species) share a common association with adenocarcinoma of the cervix. Int J Cancer 2008;122: Tornesello ML, Losito S, Benincasa G, Fulciniti F, Botti G, Greggi S, et al. Human papillomavirus (HPV) genotypes and HPV16 variants and risk of adenocarcinoma and squamous cell carcinoma of the cervix. Gynecol Oncol 2011;121: Park KJ, Kiyokawa T, Soslow RA, Lamb CA, Oliva E, Zivanovic O, et al. Unusual endocervical adenocarcinomas: an immunohistochemical analysis with molecular detection of human papillomavirus. Am J Surg Pathol 2011;35: Pirog EC, Lloveras B, Molijn A, Tous S, Guimera N, Alejo M, et al. HPV prevalence and genotypes in different histological subtypes of cervical adenocarcinoma, a worldwide analysis of 760 cases. Mod Pathol 2014; doi: /modpathol [Epub ahead of print]. 38 Alos L, Moyano S, Nadal A, Alobid I, Blanch JL, Ayala E, et al. Human papillomaviruses are identified in a subgroup of sinonasal squamous cell carcinomas with favorable outcome. Cancer 2009;115: Alonso I, Felix A, Torne A, Fuste V, del Pino M, Castillo P, et al. Human papillomavirus as a favorable prognostic biomarker in squamous cell carcinomas of the vagina. Gynecol Oncol 2012;125: