Immunohistochemical p16 overexpression and Rb loss correlate with high‐risk human papillomavirus infection in endocervical adenocarcinomas

p16 is a sensitive surrogate marker for transcriptionally active high‐risk human papillomavirus (HR‐HPV) infection in endocervical adenocarcinoma (ECA); however, its specificity is not perfect.


Introduction
2][3] The relative and real incidence of endocervical adenocarcinomas (ECAs) has been increasing worldwide, and up to 20-25% of all invasive cervical tumours belong to this category. 4he histological classification of ECAs had been based on invasiveness, cellular morphology and mucin, production rather than on HPV infection, in the former fourth World Health Organisation (WHO) classification of tumours of the female reproductive organs.For example, intestinal type (HPV-positive) and gastric type (HPV-negative) adenocarcinomas had been placed into the category of 'mucinous' carcinomas. 5However, the current fifth WHO classification has shifted the priority to the presence or absence of HR-HPV infection and set up the categories of 'HPV-associated' and 'HPV-independent' adenocarcinomas. 2 HPV-associated AIS can show morphological variations such as conventional type and mucinous type; the latter includes stratified mucin-producing intraepithelial lesion (SMILE).HPVassociated adenocarcinomaan invasive canceris roughly classified into usual and mucinous types, and histological patterns are further assigned for each type; e.g.usual type (conventional, micropapillary pattern and villoglandular variant) and mucinous type [not otherwise specified (NOS), intestinal adenocarcinoma, signet-ring cell adenocarcinoma and stratified mucinproducing carcinoma].HPV-independent AIS is rare.HPV-independent adenocarcinomas are further subclassified into gastric, clear cell, mesonephric and endometrioid types.[8] HPV-independent endometrioid adenocarcinomas are now considered extremely rare, accounting for no more than 1% of all primary ECAs. 4 This category should be distinguished from cervical invasion of endometrioid carcinoma of the endometrium origin. 2,9ccording to the recent consensus, serous carcinoma detected in the cervix represents metastasis or direct extension from adnexal or endometrial serous carcinoma; thus, the entity of 'serous carcinoma' is not listed as a primary tumour of the uterine cervix in the present fifth WHO classification. 216 immunohistochemistry (IHC) is an effective (but flawed) indirect test for HR-HPV infection.Approximately 95% of HPV-associated carcinomas showed p16 overexpression with diffuse block positivity or every-cell staining pattern. 4p16 is highly sensitive for HR-HPV infection; however, it is not an entirely specific surrogate marker.1][12] In addition, because serous carcinoma (endometrial or adnexal origin), including that invading the cervix, often overexpresses p16, p16-positivity should not readily diagnose HPVassociated adenocarcinoma of the cervix. 2,13,14t the molecular level, in HPV-infected cells the HPV E7 protein binds to the Rb protein and induces not only p16 protein overexpression but also Rb protein degradation through the ubiquitin-proteasome pathway. 15Previously, we found that HR-HPV infection was typically associated with the partial loss (mosaic) pattern of Rb expression by IHC, and we proposed that the combination of characteristic expression patterns of p16 and Rb could reliably predict HR-HPV infection in head and neck SCCs in the sinonasal tact, 16 oropharynx, 17 conjunctiva and lacrimal sac. 180][21] The exact molecular mechanism underlying the diffuse expression of p16 in HPV-independent adenocarcinoma of the uterine cervix and serous carcinoma of the female genital tract is unclear.According to a previous p16, Rb and HPV in endocervical adenocarcinoma 1179 report and public database, RB1 gene alteration is less common in these carcinomas. 22,23A previous study has reported that histone modification with activation of lysine demethylase 6B and demethylation of histone 3 lysine 27 might be responsible for p16 overexpression in serous carcinoma of the uterine corpus. 24his study aimed to determine the potential benefits of the combined use of p16-IHC and Rb-IHC to predict HR-HPV infection in ECAs and evaluate the prognostic significance of HR-HPV infection in ECAs.

C A S E S E L E C T I O N
Biopsy specimens and surgically resected specimens of formalin-fixed paraffin-embedded (FFPE) tissues were retrospectively collected from 108 ECAs, which included five cases of HPV-associated AIS, 78 cases of HPV-associated adenocarcinoma and 25 cases of HPV-independent adenocarcinoma (Table 1).HPVassociated and HPV-independent adenocarcinomas were defined as invasive cancers according to the fifth WHO classification. 2In addition, 13 patients with adenocarcinomas of endometrial or equivocal origin were examined as the control group.Here, uterine cancer of equivocal origin represents a tumour involving the boundary of upper endocervix and lower endometrium.All 121 patients were histopathologically diagnosed based on biopsy findings and subsequently underwent surgery or received radiation at Kyushu University between 1998 and 2022.Patients' clinical data were obtained from their medical records.Invasive ECAs were staged using the 2018 International Federation of Gynaecology and Obstetrics (FIGO) staging system.For old cases (1998-2017), the evidence was re-evaluated according to the new 2018 FIGO staging system for uterine cervical adenocarcinomas (Supporting information, Table S1).The Institutional Review Board at Kyushu University approved this study (permission code: 22050-00).

H I S T O P A T H O L O G I C A L E V A L U A T I O N
Two investigators (N.Y. and H.Y.) independently reviewed and diagnosed the patients according to the fifth edition of the WHO classification of female genital tumours. 2 HPV-associated AISs were subclassified into conventional and mucinous types, including SMILE.HPV-associated adenocarcinomas were classified into usual and mucinous types, and the predominant histological pattern was also recorded for usual type (conventional, micropapillary pattern and villoglandular variant) and mucinous type (NOS, intestinal adenocarcinoma, signet-ring cell adenocarcinoma and stratified mucin-producing carcinoma). 2 HPV-independent adenocarcinomas were subclassified into gastric type, clear cell type, mesonephric type, endometrioid adenocarcinoma and NOS.In our series, no patients had HPV-independent AIS and adenocarcinoma, mesonephric type.Endometrioid carcinomas from the uterine corpus or lower uterine segment were carefully excluded.Consequently, in our series, endometrioid carcinomas of absolutely endocervical origin were not recorded.Instead, 13 cases of uterine cancer of endometrial or equivocal origin (six endometrioid carcinomas and seven serous carcinomas) were used as the control group.
For the IHC, 4-lm-thick FFPE tissue sections and primary monoclonal antibodies against p16 (E6H4, prediluted, CIN histology kit; Roche, Heidelberg, Germany) and Rb (G3-245, dilution 1/50; BD Pharmingen, Franklin Lakes, NJ, USA) were used.The IHC methods were as described previously. 16,17The p16 protein expression was considered positive (protein overexpression) if nearly all tumour cells showed strong and diffuse nuclear and cytoplasmic staining ('block' positivity). 25A previously reported scoring system was utilised to categorise Rb-IHC expression as complete loss, preserved expression or partial loss, as follows [16][17][18] : for complete loss, Rb expression was present in < 10% (disappeared in > 90%) of the tumour cells; for preserved expression, Rb expression was present in > 90% of the tumour cells; and for partial loss, Rb expression was spotty (mosaic), present in ≥ 10 to ≤ 90% of the tumour cells.For evaluation of Rb, we searched the most prevalent area (hot-spot) of immunoreactive tumour cells at low-power magnification (910 ocular and 94 objective), and then scored the percentages of Rbexpressing cells in one field at higher-power magnification (910 ocular and 920 objective). 17Only nuclear staining was evaluated for Rb expression.Intratumoural endothelial cells were used as an internal positive control, and positive staining for Rb in those cells were confirmed to avoid false negativity. 17wo pathologists (N.Y. and H.Y.) independently evaluated the Rb expression patterns without information regarding the p16 IHC and HPV ISH results.

I S H F O R H I G H -R I S K H P V M R N A
In-situ hybridisation (ISH) was performed to detect HPV RNA using an E6/E7 mRNA probe set (Advanced Cell Diagnostics, Newark, CA, USA) for 18 HR-HPV types (types 16, 18, 26, 31, 33, 35, 39,   45, 51, 52, 53, 56, 58, 59, 66, 68, 73 and 82) with an RNA scope 2.5HD detection kit (Advanced Cell Diagnostics), according to the manufacturer's protocol.The presence of any nuclear or cytoplasmic dots was considered positive. 26As a positive control for HPV-ISH, oropharyngeal SCC specimens were used. 16

S T A T I S T I C A L A N A L Y S I S
Statistical analysis was performed using Fisher's exact test for categorical variables.Univariate and multivariate regression analyses with Cox proportional hazards model were conducted to identify prognostic factors.A Kaplan-Meier analysis and the log-rank test were used to derive and compare survival curves.Overall survival (OS) was defined as the time from the first day of therapy until death or to the last follow-up date for censored cases.Analyses were performed with JMP16 (SAS Institute, Cary, NC, USA).P-values of < 0.05 were considered significant.p16, Rb and HPV in endocervical adenocarcinoma 1181

C L I N I C A L F I N D I N G S
The clinicopathological findings of the 108 ECAs are summarised in Supporting information, Table S1.The ages ranged from 28 to 83 (median = 45) years, and 68% of the patients were diagnosed with stage I.The OS ranged from 1 to 245 (median = 30) months.
For the control group (endometrioid carcinomas and serous carcinomas of endometrial or equivocal origin), all 13 cases were negative for HR-HPV and six of 13 (46%) were positive for p16; however, all showed preserved Rb expression pattern (Figures 4  and 5).In this group, all six cases with p16 positivity were serous carcinoma (see below result).

R E L I A B I L I T Y O F P 1 6 A N D R B E X P R E S S I O N P A T T E R N S T O P R E D I C T H P V I N F E C T I O N
The 108 primary ECAs were divided into p16 + /HPV + (n = 83, 77%), p16 + /HPV À (n = four, 4%) and p16 À / HPV À (n = 21, 19%) groups (Figure 4).Among the p16-positive cases (n = 87), tumours with partial-loss pattern of Rb were exclusively HPV-positive (n = 83) and those with Rb-preserved pattern were exclusively HPV-negative (n = four).All 21 cases with p16 negativity showed preserved Rb pattern and lacked HPV infection.Thirteen cases of the control group were divided into p16 + /HPV À (n = six, 46%) and p16 À / HPV À (n = seven, 54%) groups (Figure 4).All 13 cases showed preserved Rb patterns.
Compared with p16, the combination of p16 overexpression and Rb-partial loss showed equally excellent sensitivity (each 100%) and improved specificity (100 versus 84%) and positive predictive values (100 versus 95.4%) in ECAs (Table 2).Similarly, the combined use of p16 and Rb patterns revealed excellent sensitivity (each 100%) and improved specificity (100 versus 73.6%) and positive predictive values (100 versus 89.2%) in the ECA and control groups (Table 2).

H I S T O P A T H O L O G I C A L F I N D I N G S O F H P V -A S S O C I A T E D A I S A N D A D E N O C A R C I N O M A
Representative histopathological images of HPVassociated AIS, conventional type (n = three), and SMILE (n = two) are shown in Supporting information, Figure S1.
Among 78 HPV-associated adenocarcinomas, the conventional usual type (n = 60), micropapillary pattern (n = one), villoglandular variants (n = six), mucinous type NOS (n = five), intestinal adenocarcinomas (n = two), signet-ring cell adenocarcinomas (n = two) and stratified mucin-producing carcinomas (n = two) were predominant histopathological patterns (Figure 1, Supporting information, Table S2).These cases showed consistent histological appearances and immunohistochemical profiles.An exceptional case of usual type adenocarcinoma was positive for ER and PgR simulating endometrioid adenocarcinoma; otherwise, features such as 'floating mitosis', p16 positivity and Rb partial loss pattern were the same as conventional usual type HPV-associated adenocarcinomas (Supporting information, Figure S2).Adenocarcinomas with predominant micropapillary pattern superficially resembled serous carcinomas in terms of confluent growth with micropapillary tufting; however, the presence of HPV infection avoided the diagnosis of serous carcinomas (Figure 1).

H I S T O P A T H O L O G I C A L F I N D I N G S O F H P V -I N D E P E N D E N T A D E N O C A R C I N O M A S
Gastric type HPV-independent adenocarcinomas (n = 14) showed consistent histopathological features characterised by glandular proliferation of atypical epithelial cells with abundant clear or pale eosinophilic cytoplasm and distinct cell borders, 2 as well as immunopositivity for HIK1083 (Figure 2).Three of 14 (21%) cases were positive for p16.Rb expression was preserved in all 14 cases irrespective of the p16 status (Table 1).Clear cell type HPV-independent adenocarcinomas (n = 10) showed consistent histopathological features; i.e. the carcinoma cells with clear, eosinophilic or granular cytoplasm proliferate in tubulocystic, papillary and/ or solid architecture sometimes with hobnail pattern, 2 as well as immunopositivity for HNF1b (Figure 2).One of 10 (10%) cases was positive for p16, but still showed preserved Rb expression (Table 1).Only one patient had HPV-independent adenocarcinoma, NOS with peculiar morphology; i.e. the tumour cells with eosinophilic or pale cytoplasm were arranged in irregular glandular, cribriform pattern and solid pattern (Figure 3).Intracytoplasmic mucin was observed.The tumour existed in the uterine cervix and did not affect the endometrium.This case was negative for the markers often seen in other type ECAs, including ER, PgR, CDX2, HIK1083, HNF1b and GATA3, and showed scattered immunoreactivity for p53 consistent with wild-type expression pattern (Supporting information, Figure S3).

H I S T O P A T H O L O G I C
All six cases of endometrioid carcinomas were positive for ER and PgR but negative for p16 and HR-HPV (Figure 5).Rb expression was preserved.The majority (six of seven cases; 86%) of serous carcinomas were positive for p16; however, HR-HPV was negative, and Rb expression was preserved (Table 1).Aberrant expression of p53 was seen in all seven cases, including diffuse (n = four) or null (n = three) pattern.

C L I N I C O P A T H O L O G I C A L A N D P R O G N O S T I C A S S O C I A T I O N O F H P V I N F E C T I O N
HR-HPV infection was significantly associated with younger age, lower stage, lower rates of lymph node  metastasis and lower rates of recurrence (P < 0.0001, P = 0.0396, P = 0.0115, and P = 0.0450, respectively; Supporting information, Table S3).
In the Kaplan-Meier estimate using log-rank tests, HPV infection was significantly associated with longer OS among patients with invasive ECAs (P = 0.0254; Figure 6).When cases were limited to advanced stage ECAs, HPV-positive cases tended to have longer OS (P = 0.0534; Figure 6).HPV-associated adenocarcinomas of usual type had significantly better prognosis A, Primary endocervical adenocarcinomas.Among p16-positive cases, the partial loss pattern of Rb was absolutely associated with HR-HPV infection, whereas cases with preserved pattern of Rb were negative for HR-HPV.All p16-negative cases showed a preserved pattern of Rb and absence of HR-HPV infection.B, Adenocarcinomas of the control group (endometrioid carcinoma and serous carcinoma, and invasion from the endometrium or undetermined primary site).All cases, irrespective of p16 status, showed preserved expression of Rb and absence of HR-HPV infection.
Figure 5. Representative histology, immunohistochemistry for p16 and Rb and in-situ hybridisation (ISH) for high-risk human papillomavirus (HR-HPV) in invasive adenocarcinomas of the control group (endometrioid carcinoma and serous carcinoma, invasion from endometrial cancer or undetermined primary site).A, Endometrioid carcinomas.Mucin-depleted, atypical columnar cells with hyperchromatic nuclei proliferate in well-formed or fused glandular patterns, superficially mimicking the usual type of adenocarcinomas.Apical mitoses and apoptotic bodies were not observed.p16 expression can be focal but not 'block' positive, judged as negative result.Rb expression is preserved and HR-HPV is negative.Tumour cells are diffuse positive for PgR.B, Serous carcinoma of the endometrium.Tumour cells are arranged in micropapillary patterns, lined by cells with high-grade nuclear atypia, superficially mimicking the micropapillary pattern of the usual type of adenocarcinoma.p16 is diffusely expressed, judged as positive.Rb expression is preserved, and HR-HPV is negative.Tumour cells show aberrant, p53 overexpression.than HPV-independent adenocarcinomas of gastric and clear cell types (P = 0.0070; Supporting information, Figure S4).

Discussion
The incidence of HPV infection in ECAs is gaining more attention because the fifth WHO classification clearly divides ECAs into HPV-associated and HPV-independent tumours.In this study, HR-HPV infection was detected in 76% of the invasive ECAs, in agreement with previous studies worldwide (57-90%). 3,4,10ost ECAs are aetiologically associated with HPV clades A9 (typically HPV16) and A7 (typically HPV18).HPV viral oncoproteins E6 and E7 inactivate p53 and Rb, respectively.This inactivation is associated with the integration of the HPV genome into the host genome, leading to the genomic instability and accumulation of somatic mutations. 27The E7-binded Rb protein is degraded through the ubiquitin-proteasome pathway. 15ptimal HPV testing in the practical diagnosis of ECAs is not yet established because of the advantages and disadvantages in any testing methods.HR-HPV mRNA chromogenic ISH, recognising 18 HR-HPV types, is approximately as sensitive as p16 IHC and is more specific; 4 however, this test is not yet widely available for clinical use in several countries, including Japan.Although p16 IHC is a highly sensitive and inexpensive test for HR-HPV infection, standalone p16 is not a perfect surrogate marker. 4In this series, four of 25 (16%) HPV-independent adenocarcinomas showed a diffuse, 'block' positivity for p16 despite the consistent morphology of gastric or clear cell type adenocarcinomas.In this study, the combination of p16 overexpression and Rb partial loss pattern were closely associated with HR-HPV infection in ECAs and showed excellent sensitivity, specificity and positive predictive values.
The result indicates that the p16/Rb IHC panel may be a reliable method to predict HR-HPV infection in ECAs, especially in situations in which HPV-ISH is not available.A similar trend has been demonstrated in SCCs in the sinonasal tract, 16 oropharynx, 17 conjunctiva and lacrimal sac. 18In our previous study of oropharyngeal SCC we conducted blinded scoring of Rb to check the interobserver variability, and revealed high reproducibility of Rb interpretation. 17oreover, in our preliminary study, HR-HPV infection was closely associated with the partial loss (mosaic) pattern of Rb IHC in HPV-associated SCCs of the uterine cervix (unpublished data), suggesting that the combination of p16 and Rb expression patterns might represent a useful surrogate for HR-HPV infection among the histological types; however, further study is needed to reach a definitive conclusion.
For the differential diagnosis, the combined use of p16 and Rb may help in histological typing of ECAs, particularly in cases with equivocal morphology and/ or unusual result of phenotypical markers.Examples of diagnostic dilemmas are as follows: 1. HPV-independent adenocarcinomas of gastric type can be positive for p16 (as mentioned above).In addition, this type of cancer can possess focal overlapping morphology with HPVassociated adenocarcinomas of usual type. 2,6,282. HPV-associated adenocarcinomas of usual type can resemble endometrioid adenocarcinomas in terms of histomorphology and ER/PgR expression. 29Reportedly, ER and PgR were expressed in 5 and 20% of HPV-associated adenocarcinomas, usual type. 12It is well known that endometrioid adenocarcinoma of the endometrium is usually positive for ER and PgR, 30 but the data on endometrioid adenocarcinoma of true endocervical origin are extremely limited because of its rarity.According to a large series of ECAs which were strictly classified according to the HPV status, ER and PgR expressions were present in each 33% (one of three cases) of HPV-independent endometrioid adenocarcinomas of the uterine cervix. 123. HPV-associated adenocarcinomas with micropapillary pattern can share the features of p16 positivity and morphology with serous carcinomas. 2,4 is reported that aberrant expression of p53 was seen in most cases of serous carcinoma, 14 but only 3.6% of HPV-associated adenocarcinomas, usual type. 12When p16 staining is diffusely positive, p53 IHC could be helpful in the differential diagnosis in addition to Rb IHC. 4. HPV-associated adenocarcinoma, mucinous type can exhibit overlapping features with HPVindependent adenocarcinoma, gastric type concerning the presence of intracytoplasmic mucin and p16 positivity. 2,3,6In addition, caution is needed because p53 aberrant expression has been reported in 18% of HPV-associated adenocarcinomas, mucinous type and 52% of HPVindependent adenocarcinomas, gastric type. 12 this study, endometrioid adenocarcinomas and serous carcinomas of true endocervical origin did not exist, and the result was consistent with previous reports. 2Most cases of serous carcinomas of endometrial or equivocal origin were positive for p16; however, they were prone to have the features of HPVnegative and preserved Rb pattern.Thus, in addition to the aid of p16/Rb IHC, the recognition of the prevalence of each ECA subtype may help for differential diagnosis.
Only one case of HPV-independent adenocarcinoma, NOS, was noted in our series.This case showed peculiar histological appearance and uncharacteristic immunohistochemical profile (Figure 3 and Supporting information, Figure S3).Further studies are needed to characterise the clinicopathological features of rare subtypes of HPV-independent ECAs.
Regarding biological behaviour, HPV-independent tumours are generally more aggressive than their HPV-associated counterparts among oropharyngeal and sinonasal SCCs. 16,17,31,32The prognostic significance of HR-HPV infection remains controversial in ECAs.According to Stolnicu et al., a nearly significant association between HPV infection and favourable prognosis exists among patients with ECAs on multivariate analysis. 33Hodgson et al. also reported that HPV infection in ECAs correlated with worse diseasefree and disease-specific survival on the univariate analysis. 8Kojima et al. reported that only gastric type HPV-independent adenocarcinomas had a significantly higher risk for disease recurrence than other types of ECAs. 6In the present study, HPV infection correlated with better prognosis in the univariate analysis.In addition, HPV-associated adenocarcinomas of usual type had significantly better prognosis than HPV-independent adenocarcinomas, including gastric and clear cell types (P = 0.0070).The results support the rationale of the current fifth WHO classification emphasising HPV infection.
In conclusion, the findings of this study suggest that p16-IHC and Rb-IHC combination may be a useful surrogate for the detection of transcriptionally active HR-HPV in ECAs.Compared with p16 alone, this combination IHC may become a helpful adjunct for the more accurate stratification of patients with ECAs.In this context, we propose inclusion of Rb-IHC into the diagnostic algorithm for ECAs and adenocarcinomas of equivocal origin in the routine clinical practice.In addition, the subclassification of ECAs based on HPV infection might provide important information for prognostic prediction and therapeutic strategies.

Figure 1 .
Figure 1.Representative histology, immunohistochemistry for p16 and Rb and in-situ hybridisation (ISH) for high-risk human papillomavirus (HR-HPV) in invasive HPV-associated adenocarcinoma of the uterine cervix.A-G, All subtypes/variants show a diffuse nuclear and cytoplasmic expression ('block' positivity) of p16, partial loss of Rb and various degrees of positive signals for HR-HPV mRNA ISH.A, Usual type.Well-formed glandular proliferation of adenocarcinoma cells with conspicuous apical mitoses.B, Usual type, micropapillary pattern.Confluent growth of carcinoma cells with micropapillary tufting.C, Villoglandular variant.Exophytic papillary growth of mucin-depleted, tall columnar cells.D, Mucinous type, not otherwise specified, adenocarcinoma.Cribriform proliferation of the atypical mucinous epithelium with pale cytoplasm.E, Intestinal adenocarcinoma.Glandular proliferation of goblet cell-type atypical epithelium.Tumour cells show diffuse CDX2 expression (inset).F, Signet-ring cell adenocarcinoma.Discohesive proliferation of carcinoma cells with intracytoplasmic mucinous vacuole and peripherally located nuclei.G, Stratified mucin-producing carcinoma.Sheet-like proliferation of carcinoma cells with eosinophilic to pale cytoplasm.Mucicarmine staining revealed intracytoplasmic mucin (inset).

Figure 2 .
Figure 2. Representative histology, immunohistochemistry for p16 and Rb and in-situ hybridisation (ISH) for high-risk human papillomavirus (HR-HPV) in invasive HPV-independent adenocarcinoma of the uterine cervix.A-D, All subtypes show preserved Rb expression and are negative for HR-HPV mRNA ISH.A, Gastric type.Adenocarcinoma cells with pale cytoplasm proliferate in well-differentiated glandular structure.Typically, p16 is negative and HIK1083 is positive.B, Gastric type, p16-positive case.p16 is positive in a subset of gastric type adenocarcinomas; otherwise, features are the same as p16-negative cases.C, Clear cell type.Tumour cells are arranged in papillary, glandular and solid patterns, lined by cells with clear and eosinophilic cytoplasms.Typically, p16 is negative and HNF1b is positive.D, Clear cell type, p16-positive case.p16 is positive in a subset of clear cell type adenocarcinomas; otherwise, features are the same as p16-negative cases.

Figure 3 .
Figure 3.A case of invasive, HPV-independent, not otherwise specified, adenocarcinoma of the uterine cervix.A, On gross appearance, the protruded tumour is located in the uterine cervix.B, The cross-section demonstrates a tumour mainly involving the endocervix.C,D, The corresponding panoramic view of histology shows that the tumour (arrows) is apart from the endometrium (arrowheads).E, Tumour cells are arranged in irregular glandular and cribriform patterns, lined by cells with eosinophilic cytoplasm.Cytoplasmic mucin is observed.

Figure 4 .
Figure 4. Mutual relationship among p16 expression, Rb status and high-risk human papillomavirus (HR-HPV) infection in uterine cervical adenocarcinomas.A, Primary endocervical adenocarcinomas.Among p16-positive cases, the partial loss pattern of Rb was absolutely associated with HR-HPV infection, whereas cases with preserved pattern of Rb were negative for HR-HPV.All p16-negative cases showed a preserved pattern of Rb and absence of HR-HPV infection.B, Adenocarcinomas of the control group (endometrioid carcinoma and serous carcinoma, and invasion from the endometrium or undetermined primary site).All cases, irrespective of p16 status, showed preserved expression of Rb and absence of HR-HPV infection.

Figure 6 .
Figure 6.Prognostic analyses in invasive endocervical adenocarcinomas (ECAs) according to high-risk human papillomavirus (HR-HPV) infection.A, HR-HPV infection in invasive ECAs is significantly associated with longer overall survival (P = 0.0254).B, HR-HPV infection tends to have longer overall survival period in invasive ECAs of advanced clinical stage (> stage IB1; P = 0.0533).

Table 2 .
The reliability of immunohistochemical expression pattern of p16 and Rb to predict HR-HPV infection