Aptima HR-HPV Testing of Cytology Specimens Is an Effective Supplement for p16 Staining to Improve Diagnostic Accuracy for HPV-Related Oropharyngeal Squamous Cell Carcinoma

Head and neck squamous cell carcinomas (HNSCCs) represent a heterogeneous group of tumors arising from the squamous epithelium of the oral cavity, oropharynx, hypopharynx, and larynx. The pathogenesis of HNSCC has been correlated with exposure to tobacco and excessive alcohol consumption. In recent years, a subset of HNSCCs, especially those arising in the oropharynx, have been confirmed to be associated with high-risk human papillomavirus (HR-HPV) infection [1]. The World Health Organization (WHO) classification of head and neck tumors in 2017 separated oropharyngeal carcinoma from traditional oral cancers and divided oropharyngeal squamous cell carcinoma (OPSCC) into HPV-positive and HPV-negative types based on HPV status [2]. HPV-positive OPSCC has distinct pathological features, such as high nuclear/cytoplasm ratio and a nonkeratinizing and basaloid morphology. In addition, compared with similarly staged HPV-negative cancers, HPV-positive OPSCC patients have improved survival and a better response to both chemoradiotherapy and induction therapy [3, 4].

Determining the status of HR-HPV is critical for the diagnosis and management of OPSCC; however, there are variable sensitivity (SN) and specificity (SP) levels for different detection methods [5]. Immunohistochemistry (IHC) for p16 has been recommended as an HPV surrogate marker by several organizations, including the College of American Pathologists (CAP) and the American Joint Committee on Cancer (AJCC) [6, 7]. Although p16 IHC staining is a simple and practical approach, there is still a small percentage of patients who test positive for p16 but negative for HPV. Moreover, these patients have been reported to have a significantly reduced 5-year survival rate compared to that of patients positive for both p16 and HPV [8]. Regarding the different therapeutic decisions for HPV-positive and -negative OPSCC patients, an HPV-specific test may be necessary for p16 false-positive or false-negative patients in the near future.

HPV E6/E7 mRNA qRT–PCR has been regarded as a gold-standard method for testing for transcriptionally active HPV [6, 9, 10]. Although PCR requires DNA or RNA extraction from formalin-fixed paraffin-embedded (FFPE) samples, more specific techniques are needed to process fragmented or ruptured RNA after FFPE sample processing. In addition, the small size of OPSCC biopsy samples might be another obstacle to PCR-based techniques. The Aptima HPV assay, an in vitro nucleic acid amplification test for the qualitative detection of E6/E7 viral messenger RNA (mRNA) from 14 high-risk types of HPV, was approved by the US Food and Drug Administration (FDA) for cervical cancer screening. As a fully automated and reliable HPV detection platform, the Aptima HPV assay has the potential to become a routine diagnostic method for detecting HPV E6/E7 mRNA in OPSCC patients.

However, the development and routine implementation of Aptima HPV testing in OPSCC patients are still limited by the current clinical workflows. In clinical practice, lymph node enlargement is the initial presenting symptom in up to half of OPSCC patients, and fine-needle aspiration (FNA) cytology is typically the initial diagnostic investigation for an enlarged lymph node [7]. However, for the other half of patients without lymph node metastasis, FFPE tissue samples from primary tumors are usually the only specimens available. Aptima testing has been approved by the FDA for use on cytology specimens but not on biopsies, as FFPE tissue biopsies are not suitable for this assay. To perform this testing, it is essential to obtain effective cytology specimens [11].

In this paper, we presented a new clinical workflow at our institution and combined adjuvant brush cytology with biopsy in OPSCC patients for HPV detection. Aptima HPV testing was performed on both brush cytology of the primary tumor and FNA cytology of metastatic lymph nodes, while p16 IHC staining was performed on the corresponding biopsies. The two approaches were then compared and assessed. In addition to HPV detection, adjuvant brush cytology can provide additional data for diagnosis, which may be crucial for cases with unsatisfactory biopsies. We hope that Aptima HPV testing of cytology specimens from OPSCC patients will become a parallel testing approach to identify potentially problematic p16 IHC results in the near future.

The present study was based on two different patient cohorts. One cohort included 30 patients who were suspected of having oropharyngeal malignancies and underwent endoscopy examination. The other cohort included 30 patients with HNSCC who were suspected of having lymph node metastasis and underwent FNA examination for enlarged lymph nodes. Figure 1 shows the study design and the case selection. All patients underwent examinations at the National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, between October 2020 and November 2021. The study protocol was reviewed and approved by the Ethics Committee of the National Cancer Center/Cancer Hospital.

After a detailed history was collected, the patients underwent endoscopy examination. A cytological brush was introduced through a separate channel in the endoscope after routine biopsy. The disposable endoscopic brush (Endoscopic Cytobrush, Micro-Tech, Nanjing, China) was extended into the target site and then advanced up to the lesions via up-and-down or rotary motions. The trapped exfoliated cells were rinsed in PreservCyt solution (Hologic, Marlborough, MA, USA) for HPV testing and cytological diagnosis.

Palpable lesions were aspirated free-hand by cytopathologists, and nonpalpable lesions were aspirated by experienced radiologists under real-time ultrasound guidance. The aspirates were smeared onto one slide, and the residual specimens in the needle were rinsed in PreservCyt solution (Hologic) for HPV testing and cytological diagnosis.

After routine cytology preparation, residual cells were tested for E6/E7 mRNA of 14 HR-HPV genotypes (i.e., 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68) with an automatic instrument, the Panther System (Hologic, San Diego, CA, USA). Three main steps were carried out in order in this system: target capture, target amplification by transcription-mediated amplification, and detection of the amplification products (amplicon) by the hybridization protection assay.

All cytological diagnoses were made by two of the authors (Dr. Guo and Dr. Zhao). The diagnosis of OPSCC by brush cytology is a new field for us; thus, all brush cytology of all cases was reviewed by the two cytopathologists simultaneously to reach a consensus diagnosis. The cytological diagnoses of brush cytology were classified according to the Bethesda System for Reporting Cervical Cytology, and the categories were modified appropriately. Five categories were introduced: negative for intraepithelial lesion or malignancy, atypical squamous cells, low-grade squamous intraepithelial lesion (shown in Fig. 2a), high-grade squamous intraepithelial lesion (HSIL) (shown in Fig. 2b), SCC (shown in Fig. 2c), and other malignant tumors (shown in Fig. 2d, e) [12].

FFPE tissues from 44 primary tumors and 16 metastatic lymph nodes were sectioned, and p16 IHC (monoclonal antibody, clone H18040, Roche) was then performed routinely. A moderate-to-strong staining intensity for the nuclei and cytoplasm in no less than 70% of tumor cells was interpreted as p16 IHC positive [13]. The p16 IHC results were evaluated by two pathologists.

RNAscope is a newly developed technique for RNA in situ hybridization detection, and it is valuable for resolving the issue of mRNA degradation in FFPE samples [14]. RNAscope has been considered a gold-standard method to be performed for cases with discordant results from Aptima HPV testing and p16 IHC. RNAscope (Advanced Cell Diagnostics, SV, CA) was used according to the manufacturer’s instructions by using a cocktail to recognize E6/E7 mRNA of 18 HR genotypes (i.e., 16, 18, 26, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, 68, 73, and 82). Both positive and negative control probes were used for individual cases.

When the results of Aptima HPV testing and p16 IHC were consistent (HPV+/p16+, HPV−/p16−), these results were the standard; when the results were discordant (HPV−/p16+, HPV+/p16−), RNAscope detection was used, and the result of RNAscope was considered the standard. The results of Aptima HPV testing and p16 IHC were compared with the standard results, and the SN, SP, positive predictive value, and negative predictive value of each test were calculated with 95% Wilson confidence intervals.

A total of 60 HNSCC patients with simultaneous Aptima HPV testing and p16 IHC were enrolled. The patient cohort included 50 males (83.3%) and 10 females (16.7%), and the mean age was 58.4 years, ranging from 34 to 84 years. There were 39 OPSCC and 21 non-OPSCC patients. The primary sites of HNSCC were the oropharynx (39/60, 65%), tongue (3/60, 5%), oral cavity (6/60, 10%), laryngeal pharynx (9/60, 15%), larynx (2/60, 3.3%), and esophagus (1/60, 1.7%). Among 60 histologic specimens, 44 (73.3%) were from primary tumors, and 16 (26.7%) were from metastatic lymph nodes. For cytological specimens, 30 were collected by brushing from the primary site, and the other 30 were collected by FNA from metastatic lymph nodes (Table 1).

Among all 60 HNSCC patients, 28 were p16+/HPV+, 29 were p16−/HPV−, 2 were p16+/HPV−, and 1 was p16−/HPV+. The overall concordance rate between Aptima HPV testing and p16 IHC was 95.0%. Three cases with discordant results for p16 and HPV were reexamined by the RNAscope method, and all were confirmed to be HPV negative (shown in Fig. 3a–i).

The sensitivity and negative predictive value of Aptima HPV testing and p16 IHC were consistent at 100%, while the SP and positive predictive value of Aptima HPV testing versus p16 IHC were 96.9% and 96.6% versus 93.8% and 93.3%, respectively (Table 2). Among the 60 HNSCC patients, the prevalence rates of HPV infection in OPSCC and non-OPSCC was 61.5% (24/39) and 19.0% (4/21), respectively.

The percentage of tumor cells with moderate-to-strong intensity staining of the nuclei and cytoplasm was evaluated in individual cases, and the results are presented in Figure 4. Twenty-four cases showed negative staining, 5 showed no more than 10% tumor cell positivity, and 30 showed more than 80% tumor cell positivity. The remaining case showed an unusual staining pattern, with 100% strong nuclear staining but no certain cytoplasmic staining (shown in Fig. 3e). According to the 70% cutoff value, 50% of cases (30/60) were p16 positive; however, all but the case with unique nuclear staining could easily be interpreted as p16 IHC negative or positive. The Aptima HPV test result for this special case was HPV negative, and the RNAscope assay further confirmed this negative result. This emphasized that a positive HPV result can be reported only when >70% of tumor cells achieve not only nuclear but also cytoplasmic staining [15].

We divided 30 p16-positive patients into OPSCC and non-OPSCC cohorts. Among 25 OPSCC patients, 7 had a keratinizing morphology, and 18 had a nonkeratinizing morphology, while among 5 non-OPSCC patients, 1 had a keratinizing morphology, and 4 had a nonkeratinizing morphology. Regarding the 7 p16-positive keratinizing OPSCC patients, 6 were HPV-positive, and 1 was HPV-negative, while all the p16-positive nonkeratinizing OPSCC patients were HPV-positive (Table 3). HPV-positive OPSCC has distinct pathological features of a nonkeratinizing morphology. However, we also observed 7 p16-positive patients showing a keratinizing morphology. Subsequent Aptima HPV testing revealed that 1 keratinizing p16-positive patient was HPV negative, and this result was further confirmed by the RNAscope method. Our observation suggests that further HPV tests are necessary to determine HPV status in p16-positive OPSCC patients presenting a nonclassical keratinizing histologic morphology.

As shown in Table 4, 35 patients with oropharyngeal tumors simultaneously underwent examinations by brush cytology and biopsy pathology. Regarding the final diagnosis, 30 cases were OPSCC, 1 was adenoid cystic carcinoma, 1 was diffuse large B-cell lymphoma, 1 was papilloma, and 2 were HSIL. Of particular note was that 6 biopsies of the primary mass for the 30 OPSCCs were originally diagnosed as HSIL by histopathology but finally confirmed to be SCC. Nevertheless, all 6 OPSCCs were accurately diagnosed as SCC by brush cytology. These six cases could be histopathologically underdiagnosed; thus, additional brush cytology accompanied by endoscopic biopsy improved the diagnostic accuracy for oropharyngeal malignancies.

The development of approaches to detect HPV status in HNSCC patients for the purpose of diagnosis has become a research hotspot worldwide. p16 IHC staining is the most widely used method in routine clinical work and is simple and cost-effective. Although p16 IHC staining shows good concordance with HPV DNA/RNA detection, a small proportion of patients display inconsistent results. Nauta et al. [16] analyzed the prognosis of a group of oropharyngeal cancer patients with p16+/HPV DNA-results and found that the 5-year overall survival rate of this group was significantly worse than that of the p16+/HPV DNA + group but similar to that of the p16−/HPV DNA-group, further indicating the possibility of false-positive staining in the p16 IHC test. Therefore, HPV-specific testing should be carried out considering the differential prognosis. An editorial also raised the question of whether it is necessary to test oropharyngeal cancer by both p16 and HPV simultaneously [17]. The Aptima HPV assay has the potential to be a standard diagnostic approach for identifying HPV E6/E7 mRNA in OPSCC patients since it is a completely automated, affordable, and reliable HPV detection platform. However, as this assay is only suitable for use on cytology specimens, it can only be used in OPSCC patients with metastatic lymph nodes through FNA under the current workflow [18‒20]. Aiming to overcome this limitation, our study first proposed performing Aptima HPV detection on brush cytology of primary tumors under endoscopy to determine the HPV status of OPSCC.

In the current study, Aptima HPV testing showed a similarly high sensitivity and SP to p16 IHC in detecting the HPV status of OPSCC patients, and the overall concordance rate between Aptima HPV testing and p16 IHC was 95.0%. A previous study demonstrated a lower sensitivity of 87.5% and a similarly high SP of 100% for Aptima HPV testing compared to p16 IHC on FNA samples of OPSCCs, along with a concordance rate of 92% [11]. Considering the possibility of false positivity or negativity in these two methods, we performed a third detection method (RNAscope method) to validate the diagnosis when the results were discordant. Although comparisons between the methods of p16 IHC and Aptima HPV were performed in a previous study, there would likely be an evaluation deviation due to lack of an objective gold standard. Considering its high cost and difficulty of operation, the RNAscope method is not widely available in clinical practice. Thus, we only used RNAscope testing as a third approach to reexamine discordant cases in this study.

In addition, the concordance rate in the current study was higher than that (77.3%) in previous reports of HPV DNA and p16 IHC [21]. Compared with HPV DNA testing, the Aptima HPV assay targets HPV E6/E7 mRNA, whose transcripts are considered to be the most relevant marker to assert the pathogenetic link between HPV infection and cancer. We have tried HPV testing in oral rinse samples, but this method was abandoned since the quantity of the obtained cells was unsatisfactory and may have further affected the sensitivity of the experiment. Moreover, previous findings evaluating the utility and prevalence of Aptima HPV testing in oral rinse samples indicate that this method has a moderate-to-poor sensitivity for HPV-related oropharyngeal carcinoma [22].

In our study, the false-positive rate of p16 IHC was 6.7%, within the range of 5–20% reported in the literature [23, 24]. It has been proposed that p16 protein levels can be elevated by non-virus-related factors, resulting in a false-positive rate of up to 20% [25]. Additional Aptima HPV testing on cytology specimens combined with routine p16 IHC can help identify false-positive cases detected by p16 IHC. Two cases showed p16 +/HPV− results and were further confirmed to be HPV negative through RNAscope detection. Therefore, the combination of these two methods can improve the accuracy of HPV status judgment and support accurate diagnosis and treatment for OPSCC patients. In addition, 28% of p16-positive OPSCC patients (7/25) showed a predominantly keratinizing morphology on histology, and of these 7 patients, 6 were HPV positive and 1 was HPV negative. According to the fifth edition of WHO classification of HNSCC, p16-positive keratinizing OPSCC requires further HPV etiological testing, and this could be achieved by Aptima HPV testing [26].

There is some evidence that a few cases showing 50–70% moderate-to-strong staining or more than 70% diffuse and weak cytoplasmic and nuclear staining in tumor cells could be HR-HPV related [27, 28], and thus, the criteria currently used in p16 IHC may contribute to false-negative results. The staining pattern of p16 IHC in this study was easily interpreted except for one case presenting a distinctive staining pattern, with 100% intensive nuclear staining and no visible cytoplasmic staining (shown in Fig. 4). Although this case should be interpreted as p16 IHC negative according to the recommended location of cytoplasmic and nuclear staining, the strong, diffuse staining of the nucleus is still very puzzling. Fortunately, subsequent Aptima HPV and RNAscope detection revealed HPV negativity. To the best of our knowledge, there have been no reports of such strong nuclear staining of p16 since single cytoplasm staining has been reported [29, 30]. Viral elimination is a normal response of the human immune system during the evolution of the disease, and protein expression decreases along with the reduction in pathogenic genes, resulting in the remaining protein being confined to the nucleus but not the cytoplasm. In another way, p16-related proteins are produced in the cytoplasm and then transported to the nucleus through cytoplasm-to-nucleus shuttling [31]. Occasionally, p16 overexpression occurs in the nucleus, but it has no or weak expression in the cytoplasm when almost all proteins in the cytoplasm are transferred to the nucleus. Given the value of HPV status in guiding the treatment of OPSCC patients, we emphasize that HPV-positive OPSCC can only be diagnosed when tumor cells show both positive cytoplasmic and nuclear staining, and we recommend performing HPV RNA/DNA testing for further confirmation if the results of p16 IHC are uncertain.

The global annual HPV infection rate of OPSCC is 63% [32], with the HPV16 subtype accounting for 95% of these cases [33]. According to the International Agency for Research on Cancer (IARC), the incidence of HPV-associated OPSCC is variable in different regions of the world, with values of more than 60% in North America and less than 20% in Southern Europe [34]. By 2020, HPV-positive OPSCC might be a major epidemic tumor in Western countries [35]. In China, the HPV infection rate of OPSCC also varies across different regions and reports, with values of 20.8% in the south, 11.7% in the east, and 5.51% in the northeast [36‒38]. Nevertheless, the prevalence of HPV in OPSCC in our study was 61.5% (24/39), which was substantially higher than the 16.7% detected by HPV DNA testing between January 1999 and December 2009 in our center [39]. This inconsistency might have been caused by different HPV detection methods or the annually increasing incidence. Large samples, multicenter data, and effective detection methods are needed to reveal the true prevalence of HPV in OPSCC in the Chinese population. The infection rate of HPV is generally low in HNSCC at other sites, with values of 22.1% for the larynx and 24.2% for the oral cavity [23]. In the current study, the infection rate of non-OPSCCs was 19.0% (4/21), which is in line with that reported in previous studies [40, 41].

According to Table 4, 6 cases were underdiagnosed as HSIL by biopsy histopathology but accurately diagnosed as SCC by brush cytology. Stromal infiltration is crucial for paraffin specimens to distinguish HSIL from invasive carcinoma. Sometimes superficial biopsy coupled with a discontinuous normal basement membrane makes it difficult for histopathologists to diagnose invasive carcinoma. For cytology specimens, the diagnosis of invasive carcinoma is mainly based on several cytological features, such as marked pleomorphism of cell size and shape, an irregular chromatin distribution and prominent nucleoli. These cytological features have been shown to be effective in distinguishing cervical squamous cell carcinoma from HSIL [12]. In conclusion, additional brush cytology along with endoscopy biopsy improved diagnostic accuracy for oropharyngeal malignancies.

One limitation of the present study is that the sample size of our experimental patients was small. Compared with that in Europe and the USA, the incidence of oropharyngeal cancer is considerably lower in China [42], thus posing a challenge for the collection of experimental data. In the future, we will continue to recruit patients and conduct research with a larger database to improve the dependability of the study.

Based on our findings, Aptima HPV testing performed well and enabled accurate diagnosis during FNA and brush cytology sample testing; therefore, we propose a clinical workflow for adjuvant brush cytology along with biopsy in OPSCC patients. In this workflow, Aptima HPV testing and p16 IHC are recommended as a combined examination for HPV status in OPSCC patients to identify false-positive cases of p16 IHC and to achieve an accurate judgment of HPV infection status. At the same time, brush cytology can be a supplemental method for biopsy diagnosis and can improve the morphological diagnosis of malignant oropharyngeal tumors.

We thank the patients and investigators who participated in this study.

This study protocol was reviewed and approved by the National Cancer Center/Cancer Hospital, the Chinese Academy of Medical Sciences, and the Independent Ethics Committee, approval number [NCC2021C-301]. The written informed consent was obtained from participants (or their parent/legal guardian/next of kin) to participate in the study.

The authors have declared that no conflict of interest exists.

This work was supported by grants from the National Natural Science Foundation of China (Grant No. 81972804).

Xin Yang and Chunfang Hu contributed equally to this work. Xin Yang carried out the data analysis, interpreted the percentage of positive tumor cells in p16 IHC sections, and wrote the manuscript. Chunfang Hu reviewed the interpretation of p16 immunohistochemical sections and assisted in writing the manuscript. Huiqin Guo obtained financial support and designed and conducted the study. Huiqin Guo and Huan Zhao both made the cytological diagnoses. Zhihui Zhang and Xiaoguang Ni helped to brush the cytological specimens under endoscopy. Linlin Zhao and Jing Yu performed HPV testing. All the authors approved the final version of the manuscript.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.