Skip to main content
PMC home page
Translational Oncology logoLink to Translational Oncology
. 2019 May 16;12(7):959–967. doi: 10.1016/j.tranon.2019.04.012

Cytological Immunostaining of HMGA2, LRP1B, and TP63 as Potential Biomarkers for Triaging Human Papillomavirus-Positive Women1

Yunhui Jiang *,2, Chengyi Zhu †,2, Dan He ‡, Qinglei Gao §, Xun Tian , Xin Ma #, Jun Wu **, Bhudev C Das ††, Konstantin Severinov ‡‡, Inga Isabel Hitzeroth §§, Priya Ranjan Debata ¶¶, Rong Liu §, Liang Zou ##, Long Shi ##, Hua Xu ##, Kaixiu Wang ##, Yuxian Bao ***, Leung Ross Ka-Kit †††, Zeshan You ‡‡‡, Zifeng Cui §§§,⁎, Zheng Hu §§§,§,⁎
PMCID: PMC6525307  PMID: 31102921

Abstract

Background: Since human papillomavirus (HPV) DNA testing has been promoted as primary screening strategy, the triage method has also evolved from morphological testing to a molecular biomarker detection to improve screening efficiency. In this study, we investigated the performance of three HPV integration hot-spots, HMGA2, LRP1B, and TP63, as potential triage markers in HPV screening tests. Materials and Methods: This cross-sectional study was conducted from November 2016 to December 2017 in the First Affiliated Hospital of Sun Yat-sen University. Immunocytochemistry was carried out using residual cervical cell samples from 121 HPV-positive cases (23 normal, 24 cervical intraepithelial neoplasia (CIN) 1, and 74 CIN2+). Results: Of the 121 cases, 77 showed completely paired for the three biomarkers. In these 77 cases, receiver operating characteristic (ROC) analysis of HMGA2 showed the best potential for detecting CIN2+ among HPV+ cases (sensitivity 70%; specificity 91.89%; AUC 0.839). TP63 was second most effective biomarker (AUC 0.838; sensitivity 80%; specificity 81.08%). In contrast, LRP1B had the smallest AUC (0.801) among the three biomarkers but had the highest sensitivity (90%) and specificity (56.76%). To test the triage value of combining the three biomarkers, logistic regression was conducted followed by ROC comparison analysis. Promisingly, the combination of the three biomarkers gave the largest AUC of 0.951 with 92.5% sensitivity and 89.1% specificity (P < .0001 compared to liquid-based cytology test by Z-test). Conclusions: A combination of HMGA2, LRP1B, and TP63 as potential biomarkers may be useful for screening during triage of HPV-positive patients, particularly for detecting CIN2 + .

Introduction

Cervical cancer is the fourth most common cancer in females worldwide and corresponding screening methods have been evolving for nearly half a century. Currently [1], HPV and cytology co-testing are the main screening methods utilized in most developing countries. However, since 2015, developed countries have gradually implemented the HPV test alone as the primary screening vehicle with flexible cytology triage to provide stronger risk-based stratification effects [2], which attributes to high sensitivity of HPV test. Another benefit of this strategy is that a reproducible HPV test as primary method significantly decreases diagnostic failure caused by the subjective judgment of cytologists, contributing to low-quality screening outcomes [3], [4]. However, the high sensitivity also brings up adverse impact of overdiagnosis and overtreatment and further management strategies for HPV-positive patients must be developed, especially to young women. For triage options, cytology test alone was reported reduction in specificity [5] while the sensitivity was still weak [6]. Therefore, it is important to develop additional triage methods to further improve screening strategies.

According to our previous study, HPV integration plays an important role in the progression of cervical lesions and is incorporated at hot spots in the host genome [7]. Integration at hot spots may result in downregulation of tumor suppresser genes such as low-density lipoprotein receptor family (LRP1B) [7] and elevate the expression of oncogenes such as the high mobility group A2 (HMGA2) gene [7] and tumor protein 63 (TP63, isoform ΔNp63 confirmed as oncogene) [8], [9]. These genes may serve as potential risk-stratified predictors of cervical cancer.

In this study, we evaluated the levels of three biomarkers in HPV-positive patients by immunocytochemistry to determine if they can be used to inform triage.

Materials and Methods

Study Design

Selected subjects were women attending the outpatient clinic and cervical cancer screening center between November 2016 and December 2017 in the First Affiliated Hospital of Sun Yat-sen University (Figure 1). Patients were co-tested for HPV and liquid-based cytology test (LCT) or they had equivalent HPV test results from another medical institution. Women who were older than 65 years, had been tested repeatedly, or were administered treatment were excluded. Two hundred twenty-seven HPV-positive women were selected for this cross-sectional study and treated according to the 2016 American College of Obstetricians and Gynecologists guidelines. Among the patients who were referred for colposcopy, 121 cases with LCT results were included (23 normal, 24 cervical intraepithelial neoplasia (CIN) I, and 74 CIN2+), and cervical cell samples were collected by the pathology department in BD SurePath LBC preservative fluid (BD Diagnostics, Franklin Lakes, NJ, USA) for immunocytochemistry analysis. Additionally, 10 HPV + LCT- cases without colposcopy referral were used as controls for pretesting. This study was approved by the ethics committee of the First Affiliated Hospital of Sun Yat-sen University (No. [2016]177) and all the participants gave their consent for sampling.

Figure 1.

Figure 1

Open in a new tab

Flowchart of the study. Women with both HPV and LCT test performed at approximately at the same time were chosen. Coloscopy was performed in 121 of the 227 HPV+ patients, with 23 reported as normal, 24 as CIN1, and 74 as CIN2+. Ten cases of HPV+ but negative LCT results were selected as controls. Immunocytochemistry was conducted on 131 samples.

Immunohistochemistry

Tissue microarrays CIN481, CIN1201, CR501, and BC10025 were obtained from Alenabio Biotechnology (Xi'an, China) and Fanpu Biotech (Guilin, China) (Table 1). First, the tissue microarray slides were baked at 65 °C for 1 h, and then deparaffinized in xylene and passed through graded alcohol followed by antigen retrieval with 1 mM EDTA, pH 8.0 (Invitrogen, Carlsbad, CA, USA) in a steam pressure cooker at over 120 °C for 2 min. The slides were treated with 3% H2O2 to block endogenous peroxidase for 30 min and washed carefully in phosphate-buffered saline with 10% Tween. The slides were first incubated with diluted rabbit anti-LRP1B/TP63 [8] /HMGA2 [10] antibody (LRP1B, ATLAS: HP 1069094, 1:100; TP63, GeneTex, Irvine, CA, USA: GTX102425, 1:350; HMGA2, Protein Tech, Rocky Hill, NJ, USA: 20795-I-AP, 1:250) overnight at 4 °C. After carefully washing the slide with PBS, they were treated with Pika general antibody (Gene Tech, GK500705, Shanghai, China; horseradish peroxidase-conjugated rabbit/mouse antibody) for 30 min, and 3, 3′ diaminobenzidine chromogen was used to detect expression within 3 min. Finally, the slides were counterstained with hematoxylin, dehydrated, and covered.

Table 1.

Histology information of tissue microarrays CIN481, CIN1201, CR501, and BC10025

Tissue microarray CIN481 CIN1201 CR501 BC10025
Normal or inflammation 4 10 4 4
 CIN1 12 51 - -
 CIN2 11 16 - -
 CIN3 9 20 - -
Squamous cell carcinoma 10 4 40 30
Adenocarcinoma 2 1 2 -
Adenosquamous carcinoma - - 4 10
Cancer adjacent tissue - - - 10
Total 48 102 50 54
Open in a new tab

The histological score (H-score) was used to evaluate the performance of tissue staining. Specifically, immuno-intensity was defined as negative (0), weak (1+), moderate (2+), and strong (3+). The H-score consisted of a sum of the combination of immunoreactivity for intensity multiplied by percentages. The possible range of the H-score was 0–300.

Immunocytochemistry

To perform immunocytochemistry analysis of the 121 cervical cell samples, similar procedures and dilution ratios of the three biomarkers were used for immunohistochemistry. Briefly cytological slides of liquid-based cytology were prepared by a ThinPrep 2000 processor (Cytyc Corporation, Boxborough, MA, USA). All slides were used for immunocytochemistry on the same day as they were prepared without baking and deparaffinization. Additionally, 0.5% Triton X-100 (Solarbio, Beijing, China) was added to slides, and the slides were incubated for 30 min before blocking with peroxidase.

The same microscope (Axio Imager Z1, 2010A731, Zeiss, Oberkochen, Germany) with a digital camera was used for all cytological staining evaluation. For each slide, the area with the largest number of cells and highly-stained signal was selected for semi-quantitative analysis under 100× magnification. Image-Pro Plus 6.0 software (Media Cybernetics, Rockville, MD, USA) was used to calculate the integral optical density (IOD) of selected area. The Log-transformed sum IOD value (Log (IOD[SUM])) of each slide was used for further statistical analysis.

Statistical Analysis

Statistically significant differences (P < .05) were evaluated by Kruskal-Wallis H-test and Bonferroni comparisons using GraphPad Prism software version 7.0 (GraphPad, Inc., San Diego, CA, USA) as well as graphing. Significant degrees were marked as * P < .05, **P < .01, ***P < .001, and ****P < .0001. For each biomarker, a receiver operating characteristic (ROC) was generated with CIN1+ and CIN2+ as the diagnostic results. To examine the combination performance for detecting CIN, stepwise multiple logistic regression was used to construct a model. ROC and logistic regression analyses were performed with MedCalc software (version 15.2.2, Ostend, Belgium).

The optimal cutoff value of each biomarker was determined by the maximal Youden index. The area under the ROC curve (AUC) was used to determine the diagnostic effect with 95% confidence intervals of the proportions calculated. The Z test was applied to compare ROC models (P < .05 was considered as significant). For the Pap test, ASCUS+ was used as the cutoff value.

Results

The Three Biomarkers Were Differentially Expressed in Cervical Lesion Tissues

Three slides of CIN tissue microarrays were used for immunohistochemistry analysis to analyze the expression of LRP1B, HMGA2, and TP63. Generally, the expression levels of HMGA2 and TP63 were elevated along with the progression of cervical lesions. Compared to HMGA2, TP63 showed stronger expression which was useful for predicting cervical pathologic grades (P < .0001 and 0.0036 respectively) (Figure 2). However, HMGA2 showed a strong ability to distinguish between CIN1 and CIN2+. Specifically, the significance values (P-value) of the stratification effect for HMGA2 and TP63 for detecting CIN2+ were 0.0126 and 0.0221, respectively (Figure 2). Notably, HMGA2 is also expressed in cervical adenocarcinoma (Figure 2). To support this result, we also immunohistochemically evaluated HMGA2 and TP63 in the tissue microarrays BC10025 and CR501, respectively. The intensity score revealed that HMGA2 showed stronger staining in cervical adenocarcinoma and squamous carcinoma compared to in tissue adjacent to cervical carcinoma and normal cervical tissues (S1 Fig). Additionally, TP63 was significantly overexpressed mainly in cervical squamous carcinoma. However, both HMGA2 and TP63 showed weak non-specific expression in the normal cervix, which may contribute to the low sensitivity of this test (Figure 2 and Table 2).

Figure 2.

Figure 2

Open in a new tab

Immunohistochemistry performance of three biomarkers. A, Representative immunohistochemistry results of HMGA2, TP63, and LRP1B in tissue microarrays CIN 481 and CIN 1201. The expression of HMGA2 and TP63 increased along with the lesion degree, while LRP1B exhibited a tumor suppressor pattern. B, H-Score method evaluated the performance of the three biomarkers semi-quantitively on corresponding CIN TMA. Kruskal-Wallis H-test with Bonferroni comparisons was used for analysis Expression of HMGA2 and TP63 differed between CIN1 and CIN2+ (P-value .0126 and .0221, respectively), while the general effect of TP63 was stronger than that of HMGA2 (P-value <.0001 and .0036, respectively).

Table 2.

Performance comparison of three biomarkers and LCT for CIN1+ and CIN2+ detection in two different sample groups

Detection CIN1+ (partially paired samples)
Strategy AUC (95% CI) Maximum Youden Index, % Sensitivity, % Specificity, % n
HMGA2 0.754 (0.656–0.836) 0.5730 61.64 95.65 96
TP63 0.799 (0.703–0.875) 0.6087 69.57 91.30 92
LRP1B 0.786 (0.695–0.861) 0.5397 88.75 65.22 103
LCT 0.719 (0.629–0.799) 0.439 61.29 81.61 121



Detection CIN2+ (partially paired samples)
Strategy AUC (95% CI) Maximum Youden Index, % Sensitivity, % Specificity, % n
HMGA2 0.855 (0.769–0.919) 0.6479 74.55 90.24 96
TP63 0.850 (0.761–0.916) 0.638 78.43 85.37 92
LRP1B 0.782 (0.690–0.857) 0.495 88.14 61.36 103
LCT 0.705 (0.613–0.786) 0.4094 67.61 71.33 121



Detection CIN1+ (completely paired samples n = 77)
Strategy AUC (95% CI) Maximum Youden Index, % Sensitivity, % Specificity, % Z-test significance (compared to LCT)
HMGA2 0.745 (0.633–0.837) 0.5306 57.41 95.64 0.9642
TP63 0.780 (0.671–0.867) 0.5982 68.52 91.30 0.6305
LRP1B 0.767 (0.656–0.855) 0.5040 85.19 65.22 0.7203
LCT 0.740 (0.628–0.834) 0.4807 61.11 86.96 -
Logistic Combination 0.882 (0.788–0.944) 0.6651 92.59 73.91 0.0332



Detection CIN2+ (completely paired samples n = 77)
Strategy AUC (95% CI) Maximum Youden Index, % Sensitivity, % Specificity, % Z-test significance (compared to LCT)
HMGA2 0.839 (0.738–0.913) 0.6189 70 91.89 0.0196
TP63 0.838 (0.736–0.912) 0.6108 80 81.08 0.0289
LRP1B 0.801 (0.695–0.884) 0.4676 90 56.76 0.0374
LCT 0.664 (0.547–0.768) 0.3277 62.5 72.27 -
Logistic Combination 0.951 (0.877–0.987) 0.8169 92.5 89.1 <0.0001
Open in a new tab

In contrast, LRP1B expression was observed, although staining signal was not strong in the normal cervix, it showed relatively lower levels in CIN. However, the staining of LRP1B showed no significant association with the progression to cervical lesions (Figure 2). Based on this result, we predicted that downregulation of LRP1B occurs in the early stages of cervical morphological changes during the progression from normal cells to CIN. Additionally, LRP1B may serve as a sensitive biomarker for cervical lesion detection. However, its downregulation may be difficult to detect by immunohistochemistry methods.

The immunohistochemistry results of the tissue microarrays indicated that HMGA2 and TP63 can serve as histological diagnostic biomarkers for cervical lesions.

Combining Immunocytochemistry of LRP1B, HMGA2, and TP63 Showed a Stronger Triage Effect than LCT for CIN2+ Detection

To explore the dilution ratio of the three biomarkers, we first tested the immunocytochemistry method using Siha, HeLa, 10 HPV+ LCT- cervical cell samples and HEK293T cell lines. Consistent with the histological results, LRP1B, HMGA2, and TP63 showed the same expression patterns. We then conducted immunochemistry analysis of cervical cell samples using the same dilution ratio. For the 121 HPV+ cases with histology results, we analyzed the slides semi-quantitatively. Consistent with the immunohistochemistry results, HMGA2 and TP63 were significantly overexpressed in CIN2+ compared to in CIN1 or normal cell samples, indicating their potential triage value. Unexpectedly, LRP1B performed better in HPV+ cytology than in cervical lesion tissues, with P < .0001 and .0105 for distinguishing normal tissue from CIN2+ and CIN1 from CIN2 respectively (Figure 3).

Figure 3.

Figure 3

Open in a new tab

Immunocytochemistry performance of three biomarkers. A, Representative immunocytochemistry of three biomarkers on residual cervical cell samples from normal to CIN2+. Consistent with the immunohistochemistry results, HMGA2 and TP63 acted as oncogenic biomarkers and LRP1B showed a decreased expression pattern along with progression of cervical lesions. B, Image-Pro Plus (IPP) semi-quantitative results of three biomarkers. Log-transformed Sum IOD value (Log (IOD[SUM])) of each slide was used for Kruskal-Wallis H-test with Bonferroni comparisons. Expression of HMGA2 and TP63 significantly differed between CIN1 and CIN2+, as well as normal and CIN2+ (P < .0001). LRP1B showed a lower ability to distinguish between CIN1 and CIN2+ (P-value .0105).

Importantly, not all of the 121 cases with LCT results were suitable for evaluating the three markers mainly because of the poor condition of the slides. Therefore, we analyzed the ROC in two groups: i) partially paired samples (n = 121) and ii) completely paired samples for three biomarkers (n = 77). For the second group, logistic regression analysis was performed to examine the performance when combining the three biomarkers; the results were compared to those of the Z-test (Table 2). The ROC curves were plotted for LRP1B, HMGA2, TP63, combination of the three biomarkers, and LCT results for detecting CIN1+ and CIN2+ (Figure 4).

Figure 4.

Figure 4

Open in a new tab

Receiver operating characteristic curve of three biomarkers and LCT for detection of CIN1+ (A) and CIN2+ (B).

Notably, the triage effect of LCT in the first group (n = 121) was consistent with that reported previously [5], achieving 61.29% sensitivity and 81.61% specificity for detecting CIN1+ and 67.71% and 71.33%, respectively, for CIN2+. Additionally, the AUC of LCT was larger for detecting CIN1+ than for CIN2+. In contrast, either single or combined biomarker patterns performed better for detecting CIN2+ than for CIN1+. For detecting CIN1+, although the three biomarkers alone and logistic regression of TP63 and LRP1B resulted in a larger AUC than that obtained for LCT, the results of the Z-test were not significant. For CIN2+, all three markers and logistic regression of the three biomarkers showed a stronger triage effect than LCT and the logistic regression combination was significant (P < .0001) (Table 2).

Specifically, in the group of 77 cases used to detect CIN2+ among the HPV+ cases, the AUC of LCT was 0.664 with a 62.5% sensitivity and 72.27% specificity. For a single biomarker, ROC analysis of HMGA2 showed the best potential (sensitivity 70%; specificity 91.89%; AUC 0.839). Moreover, TP63 was the second-best biomarker with an AUC of 0.838, sensitivity of 80%, and specificity of 81.08%. LRP1B showed the smallest AUC of 0.801 among the three biomarkers but had the highest sensitivity of 90% and specificity of 56.76%. Therefore, HMGA2 and TP63 shared high specificity and relatively low sensitivity, while LRP1B showed high sensitivity and low specificity. Together, the logistic regression of three biomarkers had a comprehensive effect compared to any of three biomarkers for detecting CIN2+ among HPV+ cases (Table 2).

The logistic combination model for detecting CIN1+ involved only TP63 and LRP1B evaluated by stepwise multivariate regression with the variate of P > .1 excluded (Logit(p) = −1.40418 LRP1B+ 2.42022 TP63–1.4989, P-value of likelihood ratio test <.0001). The AUC, sensitivity, and specificity of this model were 0.882, 92.59%, and 73.91%, respectively, while the logistic model for detecting CIN2+ that included all three biomarkers with AUC, sensitivity, and specificity increased to 0.951, 92.5%, and 89.1%, respectively (Logit(p) = 1.76981 × HMGA2–2.07416 × LRP1B+ 2.11305 × TP63–4.9213, P-value of likelihood ratio test <.0001).

In conclusion, a combination of HMGA2, TP63, and LRP1B may be useful as HPV+ triage biomarkers for CIN2+ (Figure 4).

Discussion

Cervical cells collected for screening are a convenient and reliable source of material. Several clinical researches have been conducted to improve cervical cancer screening strategies. Because HPV is necessary but not the only determining factor in the progression to cervical cancer, an increasing number of screening methods have begun to focus on molecular alterations during tumorigenesis [11], [12], [13], [14], [15]. For instance, p16/Ki67 and microRNAs have been reported as promising for improving the sensitivity and specificity of HPV-positive triage effects [14], [16]. Relative changes in these molecular biomarkers may occur before morphological transformation, which can serve as a better indicator than the Pap test for the triage of patients who are HPV-positive. In this study, we selected our previously identified hot spot integration genes, HMGA2, TP63, and LRP1B, to assess their triage effects by immunocytochemistry and semi-quantitative evaluation. Generally, ROC analysis demonstrated that the semi-quantitative immunostaining values of the three biomarkers gave greater AUC values than the LCT test for distinguishing between both CIN1+ and CIN2+ among HPV+ cases.

HPV integration is a functional step of related carcinogenesis and HPV DNA may randomly scatter in the human genome during the early phases. However, the integration hot spots provide advantages for long-term virus survival and selection, making them predominant in invasive cervical cancers. The carcinogenetic mechanism of these viral-benefit integration hot spots results from downregulation of tumor suppressors or upregulation of oncogenes. Therefore, genes such as HMGA2, TP63, and LRP1B can be used to predict HPV transforming infection, and our results support the application of immunostaining cytology as triage method. Additional molecular tests for identifying HPV integration hot spots should be conducted.

HMGA2 plays an important role in regulating DNA activities in embryogenesis but is not detectable in adult cells [17]. Based on our previous study, HMGA2 was found at 7.8% frequency as a new HPV integration hot spot accompanied by significant upregulation in cervical cancer tissues [7]. A recent study reported dramatically increased expression of HMGA2 in CIN and cervical cancer cells and that it had a stronger diagnostic effect than the HPV 16 copy number. The underlying mechanism is linked to inhibition of cell apoptosis [18]. Additionally, high HMGA2 expression was found to contribute to the metastasis and poor prognosis of various cancers, including gastrointestinal, breast, oral cavity, lung, and ovarian cancers [19], [20], [21], [22], [23]. We found that HMGA2 expression slightly different between CIN1 and normal cells in both tissue and cytology analyses (Figure 2, Figure 3), corresponding to its decreased triage sensitivity (57.41% in the second group in Table 2). However, its significant difference between CIN2+ and CIN1 indicates its potential as a diagnostic tool for detecting CIN2+ among HPV+ cases with 70% sensitivity and 91.89% specificity.

LRP1B is a tumor suppresser gene located in the region 2q22.1–22.2 with 91 exons and a length of more than 500 kb [24]. As previously reported, HPV integrates into the LRP1B intron with a frequency of 5.8%, leading to down-regulation of its expression in cervical cancer [7]. A study suggested that LRP1B inhibits tumor cell growth by increasing Stat3 phosphorylation and p21C1P1 levels as well as by inactivating the phosphorylation of SRC [25]. Inactivation of LRP1B contributed to tumorigenesis of lung, gastrointestinal, thyroid oral, urothelial, ovarian, and colon cancer [26], [27], [28], [29], [30]. In our study, LRP1B did not show a strongly decreased expression pattern from normal to CIN cervical tissues (Figure 2). However, it performed much better in HPV+ cytology by immunostaining as a triage biomarker, indicating a close relationship between HPV infection and LRP1B downregulation. As observed in the larger sample group, the triage effect of LRP1B for CIN1+ and CIN2+ was the same. Additionally, the AUC of LRP1B was not stronger than that of TP63 for detecting CIN1+ and ranked last for detecting CIN2+. Regardless, as a tumor suppresser, its high sensitivity and low specificity for detecting cervical precancers among HPV+ cases may compensate for the limitations of the other two biomarkers, which shared low sensitivity and high specificity.

TP63 is closely associated with cancer progression but its two isoforms, ΔNp63 and Tap63, have different effects. The former is considered as an oncogene that inhibits the tumor suppressive activities of the latter [31]. Given the similarities between TP63 and p53 gene, a previous study showed that Tap63 functions as a tumor suppressor gene by activating major apoptosis pathways such as p53 [32]. In contrast, upregulation of ΔNp63 can induce tumorigenesis by disrupting Tap63 and p53 transactivation function and is related to various types of squamous cell cancer [33], [34]. Moreover, TP63 and its position at 3q28 were observed as frequent HPV integration spots in both cervical cancers and oropharyngeal squamous cell carcinomas [9], [35], [36]. Over-expression of ΔNp63 was reported in cervical cancer and is useful for differentiating cervical lesion severity [15], [37]. Based on these findings, we chose ΔNp63 as a potential biomarker. We found that this protein was the primary biomarker for detecting CIN1+ and achieved nearly the same AUC as HMGA2 for CIN2+ prediction, indicating its powerful triage effect.

In this study, we observed expression changes in three biomarkers, HMG2, LRP1B, TP63, immunocytochemically in different degrees of HPV + cervical lesions to evaluate their triage potential. The three biomarkers were all HPV integration hot spots with related expression changes observed in previous studies. Our results demonstrated that combining the three biomarkers for evaluation by logistic regression improved the sensitivity to 92.5% and specificity to 89.1%, which is suitable for triage HPV screening.

However, there were some limitations to our study. Immunocytochemistry analysis lacks objectivity compared to other quantitative molecular tests of DNA, mRNA, protein, and methylation. Additionally, our sample size was small, which could impact our results.

Importantly, the combination of immunocytochemistry and morphological methods may improve the current triage methods for HPV screening. Our study provides an alternative method to traditional cytological tests by taking advantage of the expression of HPV integration hot spots.

Acknowledgments

Acknowledgements

Not applicable.

Declarations of Interest

None.

Footnotes

1

Funding: This research was supported by the National Science and Technology Major Project of the Ministry of science and technology of China [grant number 2018ZX10301402]; Nature and Science Foundation of China [grant numbers 81761148025,81630060, 81572570, 81772787, 81403166); National Key Research & Development Program of China [grant number 2016YFC0902901]; Guangzhou Science and Technology Programme [grant number 201704020093 to Zheng Hu]; Fundamental Research Funds for the Central Universities [grant number 17ykzd15]; Three Big Constructions—Supercomputing Application Cultivation Projects sponsored by National Supercomputer Center In Guangzhou; and Jingmen Key Science and Technology Planning Project [grant number YFZD2016039].

Contributor Information

Yunhui Jiang, Email: jessayhome@163.com.

Chengyi Zhu, Email: zhuxuyang0304@163.com.

Dan He, Email: hedan4103@163.com.

Qinglei Gao, Email: qlgao@tjh.tjmu.edu.cn.

Xun Tian, Email: 9941563@qq.com.

Xin Ma, Email: urologist@foxmail.com.

Jun Wu, Email: junwu1@mit.edu.

Bhudev C. Das, Email: bcdas@amity.edu.

Konstantin Severinov, Email: K.Severinov@skoltech.ru.

Inga Isabel Hitzeroth, Email: Inga.Hitzeroth@uct.ac.za.

Priya Ranjan Debata, Email: prdebata@gmail.com.

Rong Liu, Email: liurongdoctor@163.com.

Liang Zou, Email: ZL_886@163.com.

Long Shi, Email: shilong-zhou@hotmail.com.

Hua Xu, Email: xuhua2018jmey@163.com.

Kaixiu Wang, Email: jmeywkx@163.com.

Yuxian Bao, Email: yuxianbao112@163.com.

Leung Ross Ka-Kit, Email: rossleung@163.com.

Zeshan You, Email: youzeshan@163.com.

Zifeng Cui, Email: cuizf@163.com.

Zheng Hu, Email: huzheng1998@163.com.

References

  • 1.Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(6):394–424. doi: 10.3322/caac.21492. Epub 2018/09/13. (PubMed PMID: 30207593) [DOI] [PubMed] [Google Scholar]
  • 2.Huh WK, Ault KA, Chelmow D, Davey DD, Goulart RA, Garcia FA, Kinney WK, Massad LS, Mayeaux EJ, Saslow D. Use of primary high-risk human papillomavirus testing for cervical cancer screening: interim clinical guidance. Obstet Gynecol. 2015;125(2):330–337. doi: 10.1097/AOG.0000000000000669. Epub 2015/01/09. (PubMed PMID: 25569009) [DOI] [PubMed] [Google Scholar]
  • 3.Sankaranarayanan R, Nene BM, Shastri SS, Jayant K, Muwonge R, Budukh AM, Hingmire S, Malvi SG, Thorat R, Kothari A. HPV screening for cervical cancer in rural India. N Engl J Med. 2009;360(14):1385–1394. doi: 10.1056/NEJMoa0808516. (PubMed PMID: 19339719) [DOI] [PubMed] [Google Scholar]
  • 4.Wright TC, Stoler MH, Behrens CM, Sharma A, Zhang G, Wright TL. Primary cervical cancer screening with human papillomavirus: end of study results from the ATHENA study using HPV as the first-line screening test. Gynecol Oncol. 2015;136(2):189–197. doi: 10.1016/j.ygyno.2014.11.076. (PubMed PMID: 25579108) [DOI] [PubMed] [Google Scholar]
  • 5.Richardson LA, El-Zein M, Ramanakumar AV, Ratnam S, Sangwa-Lugoma G, Longatto-Filho A, Cardoso MA, Coutlee F, Franco EL, Consortium PS. HPV DNA testing with cytology triage in cervical cancer screening: Influence of revealing HPV infection status. Cancer Cytopathol. 2015;123(12):745–754. doi: 10.1002/cncy.21596. Epub 2015/08/01. (PubMed PMID: 26230283) [DOI] [PubMed] [Google Scholar]
  • 6.Isidean SD, Mayrand MH, Ramanakumar AV, Rodrigues I, Ferenczy A, Ratnam S, Coutlee F, Franco EL, Group CCS Comparison of Triage Strategies for HPV-Positive Women: Canadian Cervical Cancer Screening Trial Results. Cancer Epidemiol Biomark Prev. 2017;26(6):923–929. doi: 10.1158/1055-9965.EPI-16-0705. Epub 2017/01/18. (PubMed PMID: 28096198) [DOI] [PubMed] [Google Scholar]
  • 7.Hu Z, Zhu D, Wang W, Li W, Jia W, Zeng X, Ding W, Yu L, Wang X, Wang L. Genome-wide profiling of HPV integration in cervical cancer identifies clustered genomic hot spots and a potential microhomology-mediated integration mechanism. Nat Genet. 2015;47(2):158–163. doi: 10.1038/ng.3178. Epub 2015/01/13. (PubMed PMID: 25581428) [DOI] [PubMed] [Google Scholar]
  • 8.Zhu D, Jiang XH, Jiang YH, Ding WC, Zhang CL, Shen H, Wang XL, Ma D, Hu Z, Wang H. Amplification and overexpression of TP63 and MYC as biomarkers for transition of cervical intraepithelial neoplasia to cervical cancer. Int J Gynecol Cancer. 2014;24(4):643–648. doi: 10.1097/IGC.0000000000000122. Epub 2014/03/26. (PubMed PMID: 24662128) [DOI] [PubMed] [Google Scholar]
  • 9.Liu Y, Lu Z, Xu R, Ke Y. Comprehensive mapping of the human papillomavirus (HPV) DNA integration sites in cervical carcinomas by HPV capture technology. Oncotarget. 2016;7(5):5852–5864. doi: 10.18632/oncotarget.6809. 10.18632/oncotarget.6809 Epub 2016/01/07. (PubMed PMID: 26735580; PubMed Central PMCID: PMCPMC4868726) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Nikolic MZ, Caritg O, Jeng Q, Johnson JA, Sun D, Howell KJ, Brady JL, Laresgoiti U, Allen G, Butler R. Human embryonic lung epithelial tips are multipotent progenitors that can be expanded in vitro as long-term self-renewing organoids. elife. 2017;6 doi: 10.7554/eLife.26575. Epub 2017/07/01. (PubMed PMID: 28665271; PubMed Central PMCID: PMCPMC5555721) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Fujii T, Saito M, Hasegawa T, Iwata T, Kuramoto H, Kubushiro K, Ohmura M, Ochiai K, Arai H, Sakamoto M. Performance of p16INK4a/Ki-67 immunocytochemistry for identifying CIN2+ in atypical squamous cells of undetermined significance and low-grade squamous intraepithelial lesion specimens: a Japanese Gynecologic Oncology Group study. Int J Clin Oncol. 2015;20(1):134–142. doi: 10.1007/s10147-014-0688-0. Epub 2014/04/20. (PubMed PMID: 24744261) [DOI] [PubMed] [Google Scholar]
  • 12.Liu P, Iden M, Fye S, Huang YW, Hopp E, Chu C, Lu Y, Rader JS. Targeted, Deep Sequencing Reveals Full Methylation Profiles of Multiple HPV Types and Potential Biomarkers for Cervical Cancer Progression. Cancer Epidemiol Biomark Prev. 2017;26(4):642–650. doi: 10.1158/1055-9965.EPI-16-0368. (PubMed PMID: 28069683) [DOI] [PubMed] [Google Scholar]
  • 13.Wilting SM, Steenbergen RDM. Molecular events leading to HPV-induced high grade neoplasia. Papillomavirus Research. 2016;2:85–88. doi: 10.1016/j.pvr.2016.04.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Tian Q, Li Y, Wang F, Li Y, Xu J, Shen Y, Ye F, Wang X, Cheng X, Chen Y. MicroRNA detection in cervical exfoliated cells as a triage for human papillomavirus-positive women. J Natl Cancer Inst. 2014;106(9) doi: 10.1093/jnci/dju241. (PubMed PMID: 25190727; PubMed Central PMCID: PMC4188123) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Varga N, Mozes J, Keegan H, White C, Kelly L, Pilkington L, Benczik M, Zsuzsanna S, Sobel G, Koiss R. The Value of a Novel Panel of Cervical Cancer Biomarkers for Triage of HPV Positive Patients and for Detecting Disease Progression. Pathol Oncol Res. 2017;23(2):295–305. doi: 10.1007/s12253-016-0094-1. Epub 2016/08/09. (PubMed PMID: 27497597) [DOI] [PubMed] [Google Scholar]
  • 16.Wentzensen N, Fetterman B, Castle PE, Schiffman M, Wood SN, Stiemerling E, Tokugawa D, Bodelon C, Poitras N, Lorey T. p16/Ki-67 Dual Stain Cytology for Detection of Cervical Precancer in HPV-Positive Women. J Natl Cancer Inst. 2015;107(12):djv257. doi: 10.1093/jnci/djv257. Epub 2015/09/18. (PubMed PMID: 26376685; PubMed Central PMCID: PMCPMC4675094) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Cleynen I, Van de Ven WJ. The HMGA proteins: a myriad of functions (Review) Int J Oncol. 2008;32(2):289–305. (Epub 2008/01/19. PubMed PMID: 18202751) [PubMed] [Google Scholar]
  • 18.Wang L, Shen H, Zhu D, Feng B, Yu L, Tian X, Ren C, Gao C, Li X, Ma D. Increased high mobility group A 2 expression promotes transition of cervical intraepithelial neoplasm into cervical cancer. Oncotarget. 2018;9(8):7891–7901. doi: 10.18632/oncotarget.24080. 10.18632/oncotarget.24080 Epub 2018/03/01. (PubMed PMID: 29487700; PubMed Central PMCID: PMCPMC5814267) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Morishita A, Zaidi MR, Mitoro A, Sankarasharma D, Szabolcs M, Okada Y, D'Armiento J, Chada K. HMGA2 is a driver of tumor metastasis. Cancer Res. 2013;73(14):4289–4299. doi: 10.1158/0008-5472.can-14-3394. Epub 2015/08/05. (10.1158/0008-5472.can-12-3848. PubMed PMID: 23722545; PubMed Central PMCID: PMCPmc3715567) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Kong D, Su G, Zha L, Zhang H, Xiang J, Xu W, Tang Y, Wang Z. Coexpression of HMGA2 and Oct4 predicts an unfavorable prognosis in human gastric cancer. Med Oncol. 2014;31(8):130. doi: 10.1007/s12032-014-0130-5. Epub 2014/07/20. (PubMed PMID: 25037576) [DOI] [PubMed] [Google Scholar]
  • 21.Chang KP, Lin SJ, Liu SC, Yi JS, Chien KY, Chi LM, Kao HK, Liang Y, Lin YT, Chang YS. Low-molecular-mass secretome profiling identifies HMGA2 and MIF as prognostic biomarkers for oral cavity squamous cell carcinoma. Sci Rep. 2015;5:11689. doi: 10.1038/srep11689. Epub 2015/07/04. (PubMed PMID: 26138061; PubMed Central PMCID: PMCPmc4650660) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Eide HA, Halvorsen AR, Bjaanaes MM, Piri H, Holm R, Solberg S, Jorgensen L, Brustugun OT, Kiserud CE, Helland A. The MYCN-HMGA2-CDKN2A pathway in non-small cell lung carcinoma--differences in histological subtypes. BMC Cancer. 2016;16:71. doi: 10.1186/s12885-016-2104-9. Epub 2016/02/10. (PubMed PMID: 26858029; PubMed Central PMCID: PMCPmc4746877) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Wu J, Wei JJ. HMGA2 and high-grade serous ovarian carcinoma. J Mol Med (Berl) 2013;91(10):1155–1165. doi: 10.1007/s00109-013-1055-8. Epub 2013/05/21. (PubMed PMID: 23686260) [DOI] [PubMed] [Google Scholar]
  • 24.Liu CX, Musco S, Lisitsina NM, Yaklichkin SY, Lisitsyn NA. Genomic organization of a new candidate tumor suppressor gene, LRP1B. Genomics. 2000;69(2):271–274. doi: 10.3748/wjg.v17.i24.2909. Epub 2011/07/08 2000/10/14. (10.1006/geno.2000.6331. PubMed PMID: 11031110) [DOI] [PubMed] [Google Scholar]
  • 25.Beer AG, Zenzmaier C, Schreinlechner M, Haas J, Dietrich MF, Herz J, Marschang P. Expression of a recombinant full-length LRP1B receptor in human non-small cell lung cancer cells confirms the postulated growth-suppressing function of this large LDL receptor family member. Oncotarget. 2016;7(42):68721–68733. doi: 10.18632/oncotarget.11897. Epub 2016/09/15. (PubMed PMID: 27626682; PubMed Central PMCID: PMCPmc5356585) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Xiao D, Li F. Integrative analysis of genomic sequencing data reveals higher prevalence of LRP1B mutations in lung adenocarcinoma patients with COPD. 2017;7(1):2121. doi: 10.1038/s41598-017-02405-9. (PubMed PMID: 28522810) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Prazeres H, Torres J, Rodrigues F, Pinto M, Pastoriza MC, Gomes D, Cameselle-Teijeiro J, Vidal A, Martins TC, Sobrinho-Simoes M. Chromosomal, epigenetic and microRNA-mediated inactivation of LRP1B, a modulator of the extracellular environment of thyroid cancer cells. Oncogene. 2017;36(1):146. doi: 10.1038/onc.2016.143. Epub 2016/08/09. (PubMed PMID: 27499094) [DOI] [PubMed] [Google Scholar]
  • 28.Langbein S, Szakacs O, Wilhelm M, Sukosd F, Weber S, Jauch A, Lopez Beltran A, Alken P, Kalble T, Kovacs G. Alteration of the LRP1B gene region is associated with high grade of urothelial cancer. Lab Investig. 2002;82(5):639–643. doi: 10.1038/labinvest.3780458. (Epub 2002/05/11. PubMed PMID: 12004004) [DOI] [PubMed] [Google Scholar]
  • 29.Maru Y, Tanaka N, Ohira M, Itami M, Hippo Y, Nagase H. Identification of novel mutations in Japanese ovarian clear cell carcinoma patients using optimized targeted NGS for clinical diagnosis. Gynecol Oncol. 2017;144(2):377–383. doi: 10.1016/j.ygyno.2016.11.045. Epub 2016/12/13. (PubMed PMID: 27939411) [DOI] [PubMed] [Google Scholar]
  • 30.Wang Z, Sun P, Gao C, Chen J, Li J, Chen Z, Xu M, Shao J, Zhang Y, Xie J. Down-regulation of LRP1B in colon cancer promoted the growth and migration of cancer cells. Exp Cell Res. 2017;357(1):1–8. doi: 10.1016/j.yexcr.2017.04.010. Epub 2017/04/15. (PubMed PMID: 28408316) [DOI] [PubMed] [Google Scholar]
  • 31.Nekulova M, Holcakova J, Coates P, Vojtesek B. The role of p63 in cancer, stem cells and cancer stem cells. Cell Mol Biol Lett. 2011;16(2):296–327. doi: 10.2478/s11658-011-0009-9. Epub 2011/03/29. (PubMed PMID: 21442444) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Gressner O, Schilling T, Lorenz K, Schulze Schleithoff E, Koch A, Schulze-Bergkamen H, Lena AM, Candi E, Terrinoni A, Catani MV. TAp63alpha induces apoptosis by activating signaling via death receptors and mitochondria. EMBO J. 2005;24(13):2458–2471. doi: 10.1038/sj.emboj.7600708. (PubMed PMID: 15944736; PubMed Central PMCID: PMCPMC1173149) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Pallier K, Cazes A, El Khattabi L, Lecchi C, Desroches M, Danel C, Riquet M, Fabre-Guillevin E, Laurent-Puig P, Blons H. DeltaN TP63 reactivation, epithelial phenotype maintenance, and survival in lung squamous cell carcinoma. Tumour Biol. 2012;33(1):41–51. doi: 10.1007/s13277-011-0239-5. Epub 2011/10/12. (PubMed PMID: 21986963) [DOI] [PubMed] [Google Scholar]
  • 34.Gaya JM, Lopez-Martinez JM, Karni-Schmidt O, Bonal DM, Algaba F, Palou J, Villavicencio H, Benson MC, Cordon-Cardo C, Castillo-Martin M. DeltaNp63 expression is a protective factor of progression in clinical high grade T1 bladder cancer. J Urol. 2015;193(4):1144–1150. doi: 10.1016/j.juro.2014.10.098. Epub 2014/12/03. (PubMed PMID: 25444981) [DOI] [PubMed] [Google Scholar]
  • 35.Schmitz M, Driesch C, Jansen L, Runnebaum IB, Durst M. Non-random integration of the HPV genome in cervical cancer. PLoS One. 2012;7(6):e39632. doi: 10.1371/journal.pone.0039632. Epub 2012/07/05. (PubMed PMID: 22761851; PubMed Central PMCID: PMCPMC3384597) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Walline HM, Komarck CM, McHugh JB, Bellile EL, Brenner JC, Prince ME, McKean EL, Chepeha DB, Wolf GT, Worden FP. Genomic Integration of High-Risk HPV Alters Gene Expression in Oropharyngeal Squamous Cell Carcinoma. Mol Cancer Res. 2016;14(10):941–952. doi: 10.1158/1541-7786.mcr-16-0105. Epub 2016/07/17. (PubMed PMID: 27422711; PubMed Central PMCID: PMCPmc5065754) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Canham M, Charsou C, Stewart J, Moncur S, Hoodless L, Bhatia R, Cong D, Cubie H, Busby-Earle C, Williams A. Increased cycling cell numbers and stem cell associated proteins as potential biomarkers for high grade human papillomavirus+ve pre-neoplastic cervical disease. PLoS One. 2014;9(12):e115379. doi: 10.1371/journal.pone.0115379. Epub 2014/12/23. (PubMed PMID: 25531390; PubMed Central PMCID: PMCPmc4274002) [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Translational Oncology are provided here courtesy of Neoplasia Press

RESOURCES