The branched N-glycan of PD-L1 predicts immunotherapy responses in patients with recurrent/metastatic HNSCC.

Immunotherapy has revolutionized cancer treatment, but the lack of a reliable predictive biomarker for treatment response remains a challenge. Alpha-1,6-Mannosylglycoprotein 6-ß-N-Acetylglucosaminyltransferase 5 (MGAT5) is a key regulator of complex N-glycan synthesis, and its dysregulation is associated with cancer progression. The lectin Phaseolus vulgaris leukoagglutinin (PHA-L) specifically binds to mature MGAT5 products. Previous studies have indicated elevated PHA-L staining in head and neck squamous cell carcinoma (HNSCC), which implies increased activity of MGAT5. However, the specific role of MGAT5 in HNSCC remains unclear. In this study, we found significantly higher PHA-L staining and MGAT5 expression in HNSCC tumors compared to adjacent non-tumor tissues. Using a mass spectrometry (MS)-based glycoproteomic approach, we identified 163 potential protein substrates of MGAT5. Functional analysis revealed that protein substrates of MGAT5 regulated pathways related to T cell proliferation and activation. We further discovered that PD-L1 was among the protein substrates of MGAT5, and the expression of MGAT5 protected tumor cells from cytotoxic T lymphocyte (CTL) killing. Treatment of nivolumab alleviated the protective effects of MGAT5 on CTL activity. Consistently, patients with MGAT5-positive tumors showed improved responses to immunotherapy compared to those with MGAT5-negative tumors. Using purified PD-L1 from HNSCC cells and a glycoproteomic approach, we further deciphered that the N35 and N200 sites carry the majority of complex N-glycans on PD-L1. Our findings highlight the critical role of MGAT5-mediated branched N-glycans on PD-L1 in modulating the interaction with the immune checkpoint receptor PD-1. Consequently, we propose that MGAT5 could serve as a biomarker to predict patients' responses to anti-PD-1 therapy. Furthermore, targeting the branched N-glycans at N35 and N200 of PD-L1 may lead to the development of novel diagnostic and therapeutic approaches.

Association Between Antibodies That Bind Epstein-Barr Virus (EBV) gp350 and gH/gL and Shedding of EBV in Saliva From Nasopharyngeal Carcinoma Multiplex Family Members in Taiwan.

Elevated levels of Epstein-Barr virus (EBV) gp350 and gH/gL antibodies have been associated with a lower risk of developing nasopharyngeal carcinoma (NPC), although the evidence remains inconclusive and unexplained. We conducted a longitudinal study within a high-risk Taiwanese cohort, analyzing total immunoglobulin against EBV-gp350 and -gH/gL in blood and EBV DNA shedding in saliva. Contrary to our hypothesis-that elevated levels of antibodies previously shown to be associated with a lower NPC risk should result in a decrease in EBV shedding in saliva-higher anti-gp350 antibodies at baseline were significantly associated with detectable EBV DNA in saliva at follow-up (odds ratio [OR], 1.99 [95% confidence interval {CI}, 1.03-3.97]; P = .04). Higher anti-EBV-gH/gL antibodies at baseline were not significantly associated with risk of detectable EBV DNA at follow-up (OR, 0.69 [95% CI, .35-1.32]; P = .26). These findings underscore the complexity of virus-host interactions and emphasize the need for further investigations into their role in EBV-associated diseases.

Feasibility of Using Needle Rinse Fluid for Cobas Human Papillomavirus (HPV) Assay in Diagnosing HPV+ Oropharyngeal Cancer with Neck Lymph Node Aspiration.

BACKGROUND: Human papillomavirus (HPV) is a crucial prognostic factor in oropharyngeal cancer (OPC). p16 is a surrogate marker for diagnosing HPV+ OPC, however it is not direct evidence of HPV existence. OBJECTIVE: The purpose of our study was to evaluate an HPV DNA test-Cobas HPV assay-in diagnosing HPV+ OPC through neck lymph node aspiration. METHODS: Patients with suspected neck mass who received fine needle aspiration (FNA) or core needle biopsy (CNB) at the National Taiwan University Hospital between January 2018 and December 2022 were reviewed. Besides routine cytology and pathology study, needle rinse fluid was collected for the Cobas HPV assay to detect high-risk HPV. RESULTS: We analyzed 137 patients with suspected lymph nodes, 32 (23.4%) of whom were HPV+ OPC patients and 105 (76.6%) of whom had non-HPV-related disease. FNA was performed in 31 patients and CNB was performed in 106 patients, according to the size and necrosis status of the lymph nodes. For diagnosing HPV+ OPC, CNB combined with p16 immunohistochemistry staining showed sensitivity of 93.3%, specificity of 97.8%, positive predictive value (PPV) of 87.5%, negative predictive value (NPV) of 98.9%, and accuracy of 97.2%. On the other hand, for the needle rinse Roche Cobas HPV assay, the test showed sensitivity of 96.9%, specificity of 100%, PPV of 100%, NPV of 99.1%, and accuracy of 99.3%. Compared with p16 IHC staining, the Cobas HPV test showed better PPV with statistical significance (p = 0.04). CONCLUSION: The Cobas HPV assay is a US FDA-approved, highly automated, and readily used technique to directly detect the presence of high-risk HPV. We recommend utilizing the Cobas HPV assay in combination with routine cytology or histopathology examination in the work-up of neck lymphadenopathy.

Kinetics of EBV antibody-based NPC risk scores in Taiwan NPC multiplex families.

Nasopharyngeal carcinoma (NPC) risk prediction models based on Epstein-Barr virus (EBV)-antibody testing have shown potential for screening of NPC; however, the long-term stability is unclear. Here, we investigated the kinetics of two EBV-antibody NPC risk scores within the Taiwan NPC Multiplex Family Study. Among 545 participants with multiple blood samples, we evaluated the stability of a 2-marker enzyme-linked immunosorbent assay score and 13-marker multiplex serology score using the intra-class correlation coefficient (ICC) by fitting a linear mixed model that accounted for the clustering effect of multiple measurements per subject and age. We also estimated the clustering of positive tests using Fleiss's kappa statistic. Over an average 20-year follow-up, the 2-marker score showed high stability over time, whereas the 13-marker score was more variable (p < .05). Case-control status is associated with the kinetics of the antibody response, with higher ICCs among cases. Positive tests were more likely to cluster within the same individual for the 2-marker score than the 13-marker score (p < .05). The 2-marker score had an increase in specificity from ~90% for single measurement to ~96% with repeat testing. The 13-marker score had a specificity of ~73% for a single measurement that increased to ~92% with repeat testing. Among individuals who developed NPC, none experienced score reversion. Our findings suggest that repeated testing could improve the specificity of NPC screening in high-risk NPC multiplex families. Further studies are required to determine the impact on sensitivity, establish optimal screening intervals, and generalize these findings to general population settings in high-risk regions.

Cleaning and Harmonizing Medical Image Data for Reliable AI: Lessons Learned from Longitudinal Oral Cancer Natural History Study Data.

For deep learning-based machine learning, not only are large and sufficiently diverse data crucial but their good qualities are equally important. However, in real-world applications, it is very common that raw source data may contain incorrect, noisy, inconsistent, improperly formatted and sometimes missing elements, particularly, when the datasets are large and sourced from many sites. In this paper, we present our work towards preparing and making image data ready for the development of AI-driven approaches for studying various aspects of the natural history of oral cancer. Specifically, we focus on two aspects: 1) cleaning the image data; and 2) extracting the annotation information. Data cleaning includes removing duplicates, identifying missing data, correcting errors, standardizing data sets, and removing personal sensitive information, toward combining data sourced from different study sites. These steps are often collectively referred to as data harmonization. Annotation information extraction includes identifying crucial or valuable texts that are manually entered by clinical providers related to the image paths/names and standardizing of the texts of labels. Both are important for the successful deep learning algorithm development and data analyses. Specifically, we provide details on the data under consideration, describe the challenges and issues we observed that motivated our work, present specific approaches and methods that we used to clean and standardize the image data and extract labelling information. Further, we discuss the ways to increase efficiency of the process and the lessons learned. Research ideas on automating the process with ML-driven techniques are also presented and discussed. Our intent in reporting and discussing such work in detail is to help provide insights in automating or, minimally, increasing the efficiency of these critical yet often under-reported processes.

Margin to depth of invasion ratio as an indicator for stratifying close margins in early-stage oral squamous cell carcinoma.

OBJECTIVES: In early-stage oral squamous cell carcinoma (OSCC) patients, whether the margin-to-depth-of-invasion ratio (MDR) can assist in stratifying the prognosis remains unclear. METHODS: Patients diagnosed with early stage OSCC at National Taiwan University Hospital between January 2007 and December 2021 were reviewed. Patients with margin > 1 mm were classified into two groups: MDR < 0.5 and MDR ≥ 0.5. RESULTS: We analyzed 911 pT1-2N0M0 OSCC patients, 723 (79.36 %) with MDR ≥ 0.5 and 188 (20.64 %) with MDR < 0.5. Patients in the MDR < 0.5 group displayed a significantly higher local recurrence rate (odds ratio 2.81, p = 0.002) compared with MDR ≥ 0.5 group. The 5-year disease-free survival were 80.8 % for clear margin, 76.3 % for close margin (MDR ≥ 0.5), and 65.2 % for close margin (MDR < 0.5). The overall survival displayed a similar pattern, with 5-year rates of 88.3 % for clear margin, 86.8 % for close margin (MDR ≥ 0.5), and 75.0 % for close margin (MDR < 0.5). There were no significant overall survival differences between the two MDR ≥ 0.5 groups, but both were significantly superior to patients with MDR < 0.5 (p = 0.001; p = 0.01). After multivariant cox analysis, MDR < 0.5 was a significant risk factor for disease-free survival (p < 0.001). CONCLUSION: For early stage OSCC patients without positive margin (≦1mm), the survival outcome between MDR ≥ 0.5 group and MDR < 0.5 group was significantly different. The MDR < 0.5 group had significantly higher risk of local recurrence that may warrant adjuvant treatment.

The expression of nicotinic acetylcholine receptor subunits and their associations with local immune cells and prognosis in oral squamous cell carcinoma.

BACKGROUND: Nicotinic acetylcholine receptors (nAChRs) are ligand-gated ion channels that may be responsible for cancer cell proliferation, epithelial-mesenchymal transition (EMT), and immune regulation. However, little is known about the associations of different nAChR subunits with tumor microenvironment in oral squamous cell carcinoma (OSCC). METHODS: We retrospectively reviewed pathology samples from 75 OSCC patients by immunohistochemistry. In addition, a cohort of 307 OSCC patients in The Cancer Genome Atlas was analyzed. RESULTS: Subunit α1 was specific to peri-OSCC skeletal muscle. Increased α1 was associated with increased CD44 (cancer stem cells), increased CD3 and 8 (T cells), increased CD56 and 16 (natural killer cells), a decreased T stage, and an increased N stage. Increased α3 was associated with increased CD56 and 16. Increased α5 was associated with decreased CD3, 8, and 56, a decreased T stage, an increased N stage, worse survival, and decreased epithelial features. Increased α7 was associated with increased CD3, 8, 56, and 16, decreased tumor/peritumor ratios of CD3, 8, and 56 immune cells, and increased epithelial features. Increased local immune cells were associated with a better prognosis. CONCLUSIONS: α5 is the only subunit associated with decreased local immune cells and worse survival, while α1, α3, and α7 are associated with increased local immune cells in OSCC. α5 and α7 are correlated with different EMT states to be mesenchymal-like and epithelial-like OSCC, respectively. Protein expression data of the nAChR subunits, complementary to gene expression data, could provide meaningful information regarding the EMT status of OSCC associated with immune responses and prognosis.

Performance and Operational Feasibility of Epstein-Barr Virus-Based Screening for Detection of Nasopharyngeal Carcinoma: Direct Comparison of Two Alternative Approaches.

PURPOSE: Two Epstein-Barr virus (EBV)-based testing approaches have shown promise for early detection of nasopharyngeal carcinoma (NPC). Neither has been independently validated nor their performance compared. We compared their diagnostic performance in an independent population. METHODS: We tested blood samples from 819 incident Taiwanese NPC cases (213 early-stage, American Joint Committee on Cancer version 7 stages I and II) diagnosed from 2010 to 2014 and from 1,768 controls from the same region, frequency matched to cases on age and sex. We compared an EBV antibody score using immunoglobulin A antibodies measured by enzyme-linked immunosorbent assay (EBV antibody score) and plasma EBV DNA load measured by real-time PCR followed by next-generation sequencing (NGS) among EBV DNA-positive individuals (EBV DNA algorithm). RESULTS: EBV antibodies and DNA load were measured for 2,522 (802 cases; 1,720 controls) and 2,542 (797 cases; 1,745 controls) individuals, respectively. Of the 898 individuals positive for plasma EBV DNA and therefore eligible for NGS, we selected 442 (49%) for NGS testing. The EBV antibody score had a sensitivity of 88.4% (95% CI, 86.1 to 90.6) and a specificity of 94.9% (95% CI, 93.8 to 96.0) for NPC. The EBV DNA algorithm yielded significantly higher sensitivity (93.2%; 95% CI, 91.3 to 94.9; P = 1.33 × 10-4) and specificity (98.1%; 95% CI, 97.3 to 98.8; P = 3.53 × 10-7). For early-stage NPC, the sensitivities were 87.1% (95% CI, 82.7 to 92.4) for the EBV antibody score and 87.0% (95% CI, 81.9 to 91.5) for the EBV DNA algorithm (P = .514). For regions with a NPC incidence of 20-100/100,000 person-years (eg, residents in southern China and Hong Kong), these two approaches yielded similar numbers needed to screen (EBV antibody score: 5,656-1,131; EBV DNA algorithm: 5,365-1,073); positive predictive values ranged from 0.4% to 1.7% and 1.0% to 4.7%, respectively. CONCLUSION: We demonstrated high sensitivity and specificity of EBV antibody and plasma EBV DNA for NPC detection, with slightly inferior performance of the EBV antibody score. Cost-effectiveness studies are needed to guide screening implementation.

Is the N2A Category Still Necessary for Oral Cancer Patients After Extranodal Extension Upgrade?

BACKGROUND: The presence of single-node metastasis (Ns) sometimes could be encountered in patients with oral squamous cell carcinoma (OSCC). The survival outcome for different Ns should be worthy of discussion. METHODS: Patients diagnosed with OSCC at the National Taiwan University Hospital between January 2007 and December 2018 were reviewed. All patients with Ns were classified into two groups: with and without extranodal extension (ENE). RESULTS: We analyzed 311 OSCC patients with Ns: 77 (24.76%) with and 234 (75.24%) without ENE. Lymph node (LN) >3 cm was the only significant factor associated with ENE (odds ratio 17.21, p < 0.001). The 5-year, disease-free survival of N1/N2A and N3B patients was 60.5% and 49.4%, respectively (p = 0.04), and the 5-year overall survival was 63.1% and 33.6%, respectively (p = 0.0001). Four fifths of Ns patients with LN >3 cm were upgraded to N3B category as ENE+. Postoperative radiotherapy (PORT) could provide significant benefit in regional control for Ns patients with (p = 0.03) and without (p = 0.0004) other adverse features. After multivariant Cox analysis, ENE+ was a modest and significant risk factor for disease-free (p = 0.08) and overall survival (p = 0.001). By contrast, the LN>3cm and N2A category were not significant risk factors for disease-free and overall survival. CONCLUSIONS: For OSCC patients with Ns, the survival outcome between N3B category and N1/N2A category was significantly different. After ENE+ upgrades (>80%), there were fewer N2A patients, and these patients became more comparable to N1 patients. PORT could significantly improve regional control for Ns patients.

Chemoprevention of oral cancer: A review and future perspectives.

BACKGROUND: Oral cancer causes significant morbidity and mortality. Chemoprevention utilizes medication or natural compounds to reverse oral premalignant lesions and to prevent second primary tumors. METHODS: A comprehensive PubMed database and Cochrane Library search from 1980 to 2021 was performed using the keywords "leukoplakia," "oral premalignant lesion," and "chemoprevention." RESULTS: Chemopreventive agents included retinoids, carotenoids, cyclooxygenase inhibitor, herbal extracts, bleomycin, tyrosine kinase inhibitors, metformin, and immune checkpoint inhibitors. Although some agents demonstrated effect in reducing premalignant lesions and preventing second primary tumors, the results among different studies were highly variable. CONCLUSIONS: The results of different trials, albeit inconsistent, provided substantial information for future studies. In the era of personalized medicine, future studies will focus on identifying specific biomarkers and molecular profile to monitor and to prevent malignant transformation. Larger trials are warranted to validate the effect of chemopreventive agents.

Spindle Cell Carcinoma of the Head and Neck: Clinical Characteristics and Molecular Signatures.

OBJECTIVE/HYPOTHESIS: Spindle cell carcinoma of the head and neck (HNSpCC) is a rare variant of head and neck squamous cell carcinoma (HNSCC). This study evaluated the clinical characteristics and molecular signatures of such tumors. STUDY DESIGN: Retrospective analysis. METHODS: Medical records of patients diagnosed with HNSpCC from 1996 to 2018 were reviewed. The clinicopathologic features, treatment modalities, and survival status were carefully recorded. Whole exome sequencing (WES) was performed to evaluate the genetic signatures of HNSpCC. RESULTS: We found that among all 71 patients included in this study, the majority of them were male, with tumors developing predominantly in the oral cavity. The 1-, 3-, and 5-year disease-specific survival (DSS) rates were 64.6%, 49.5%, and 43.9%, respectively. A high local recurrence (LR) and distant metastasis (DM) rate (47.9%-25.3%, respectively) were observed. A significant proportion (28.2%) of patients with the worst prognosis had history of previous head and neck cancer (HNC) and had been treated with radiotherapy (RT). WES revealed that those post-RT SpCC shared common mutations with their previous HNC (pre-RT SCC), but gained additional genetic traits, such as hypoxia and cell-ECM interaction that were favorable for survival in an irradiated microenvironment. Distinct genetic landscapes in primary and post-RT SpCC were also found. CONCLUSIONS: This study demonstrates that HNSpCC is a unique entity with more aggressive behavior than conventional HNSCC. HNSpCC arising from a previously irradiated field is a predictor of dismal survival. Both genetic and microenvironmental factors contribute to this highly invasive tumor. LEVEL OF EVIDENCE: 4 Laryngoscope, 133:2183-2191, 2023.

Proactive Swallowing Rehabilitation in Patients with Recurrent Oral Cancer Receiving Salvage Treatment: Long-Term Swallowing-Related Outcomes.

The present study aimed to evaluate the impact of proactive swallowing rehabilitation on swallowing function and quality of life in patients with recurrent oral cancer in the first 2 years after salvage treatment. Consecutive adult patients with recurrent oral cancer who received salvage surgery and free flap reconstruction were recruited prospectively, to whom proactive swallowing rehabilitation was provided. Body weight (BW); fiberoptic endoscopic evaluation of swallowing (FEES), functional oral intake scale (FOIS), and diet level; 10-item eating assessment tool (EAT-10), and MD Anderson Dysphagia Inventory (MDADI); and adherence at baseline, 1, 3, 6, 12, 18 and 24 months were evaluated. A total of 50 patients were included during May 2018 to July 2020. Compared to the baseline, significant deterioration in BW, FOIS, and MDADI was noted at one month. However, a trend of recovery was observed in BW and FOIS from one month, and in MDADI from three months. All patients were free of tube feeding at 18-24 months and tolerated diet with special preparations or compensation. Safe swallowing could be achieved in approximately 80% participants after 12 months of diet modification or compensatory maneuvers. Proactive swallowing therapy was feasible in patients with recurrent oral cancer receiving salvage treatment. Although this patient population might have pre-existing dysphagia from previous treatments, rehabilitation could facilitate safe per oral intake and maintain adequate nutrition with adaptive maneuvers or compensatory strategies. Patients who underwent proactive swallowing rehabilitation had better recovery in the functional oral intake level.

Radiation-induced sarcoma of head and neck: Clinical characteristics and molecular signatures.

BACKGROUND: Radiation-induced sarcoma of the head and neck (RISHN) is a rare yet devastating potential complication of radiotherapy treatment. We aimed to evaluate the clinicopathological characteristics and molecular signatures of RISHN in patients who underwent radiotherapy for head and neck cancer (HNC) to identify high-risk patients and enable earlier cancer detection. METHODS: This study retrospectively evaluated 24 sarcoma patients who received radiotherapy for HNC between 1994 and 2019. Patients were divided into two groups based on RISHN latency period. Patient demographics, initial tumor staging, risk factors, and survival between groups were analyzed, and whole-exome sequencing (WES) of selected samples was performed. RESULTS: The median age at diagnosis of RISHN was 54 years, and the male-to-female ratio was 2:1. The latency period ranged from 0.8 to 64.4 years (median 6.5 years), with a median survival of 21.5 months. Primary cancer in the oral cavity, treatment with alkylating agents, alcohol consumption, betel nut chewing, and smoking were identified as risk factors for short (<5 years) latency periods. The majority of RISHN cases occurred in the oral cavity (58.3%). WES analysis showed that tumor necrosis factor and cell cycle checkpoint pathways were differentially involved in both patient groups. CONCLUSIONS: Although case numbers were small, our cohort represents the largest case series of RISHN from a single institution to date. Clinicians must be aware of factors affecting RISHN development and latency, and risk factor identification may lead to earlier detection and prevention in the future.

Extraction of Ruler Markings For Estimating Physical Size of Oral Lesions.

Small ruler tapes are commonly placed on the surface of the human body as a simple and efficient reference for capturing on images the physical size of a lesion. In this paper, we describe our proposed approach for automatically extracting the measurement information from a ruler in oral cavity images which are taken during oral cancer screening and follow up. The images were taken during a study that aims to investigate the natural history of histologically defined oral cancer precursor lesions and identify epidemiologic factors and molecular markers associated with disease progression. Compared to similar work in the literature proposed for other applications where images are captured with greater consistency and in more controlled situations, we address additional challenges that our application faces in real world use and with analysis of retrospectively collected data. Our approach considers several conditions with respect to ruler style, ruler visibility completeness, and image quality. Further, we provide multiple ways of extracting ruler markings and measurement calculation based on specific conditions. We evaluated the proposed method on two datasets obtained from different sources and examined cross-dataset performance.

Automatic Detection of Oral Lesion Measurement Ruler Toward Computer-Aided Image-Based Oral Cancer Screening.

Intelligent computer-aided algorithms analyzing photographs of various mouth regions can help in reducing the high subjectivity in human assessment of oral lesions. Very often, in the images, a ruler is placed near a suspected lesion to indicate its location and as a physical size reference. In this paper, we compared two deep-learning networks: ResNeSt and ViT, to automatically identify ruler images. Even though the ImageN et 1K dataset contains a "ruler" class label, the pre-trained models showed low sensitivity. After fine-tuning with our data, the two networks achieved high performance on our test set as well as a hold-out test set from a different provider. Heatmaps generated using three saliency methods: GradCam and XRAI for ResNeSt model, and Attention Rollout for ViT model, demonstrate the effectiveness of our technique. Clinical Relevance- This is a pre-processing step in automated visual evaluation for oral cancer screening.

Oral Cavity Anatomical Site Image Classification and Analysis.

Oral cavity cancer is a common cancer that can result in breathing, swallowing, drinking, eating problems as well as speech impairment, and there is high mortality for the advanced stage. Its diagnosis is confirmed through histopathology. It is of critical importance to determine the need for biopsy and identify the correct location. Deep learning has demonstrated great promise/success in several image-based medical screening/diagnostic applications. However, automated visual evaluation of oral cavity lesions has received limited attention in the literature. Since the disease can occur in different parts of the oral cavity, a first step is to identify the images of different anatomical sites. We automatically generate labels for six sites which will help in lesion detection in a subsequent analytical module. We apply a recently proposed network called ResNeSt that incorporates channel-wise attention with multi-path representation and demonstrate high performance on the test set. The average F1-score for all classes and accuracy are both 0.96. Moreover, we provide a detailed discussion on class activation maps obtained from both correct and incorrect predictions to analyze algorithm behavior. The highlighted regions in the class activation maps generally correlate considerably well with the region of interest perceived and expected by expert human observers. The insights and knowledge gained from the analysis are helpful in not only algorithm improvement, but also aiding the development of the other key components in the process of computer assisted oral cancer screening.

Rising incidence of HPV positive oropharyngeal cancer in Taiwan between 1999 and 2014 where betel nut chewing is common.

BACKGROUND: The incidence of human papillomavirus (HPV) positive oropharyngeal cancer (OPC) is rising but HPV negative OPC is decreasing in Western countries. In Taiwan, the incidence of HPV negative OPC is common but the incidence of HPV positive OPC remains unknown. The objective of this study is to estimate the incidence trend and the survival of HPV positive OPC in Taiwan. METHODS: Between 1999 and 2014, primary tumor tissues from 425 incident OPCs were obtained from 5 medical centers in Taiwan. 408 OPCs were evaluated by the EasyChip HPV genotyping (King-Car, I-Lan, Taiwan) and 369 OPCs by p16 staining. The clinical data were retrospectively obtained from the medical records. RESULTS: In our study, 29% of OPCs were HPV positive. The percentage of HPV positive OPC was stable from 1999 to 2014 (25% (1999-2002), 30% (2003-2006), 30% (2007-2010), 29% (2011-2014)). The estimated crude incidence rate of HPV positive OPC increased significantly from 0.62 (1999-2002), 1.06 (2003-2006), 1.52 (2007-2010) to 1.74 (2011-2014) per 100,000 person-year. The sensitivity and specificity of p16 staining for positive HPV infection were 92% and 91%, respectively. The 5-year overall survival rates for patients with HPV positive OPC and with HPV negative OPC were 67.8% and 49.0%, respectively (HR = 0.52 (0.35-0.76), p = 0.0005). Patients with HPV positive OPC but no betel nut/cigarette exposure had the best overall survival (5-year: 88.2%, p < 0.0001). Patients with HPV negative OPC and betel nut/cigarette exposure had the worst overall survival (5-year: 46.6%, p < 0.0001). Patients with HPV positive OPC but also with betel nut/cigarette exposure had poorer 5-year overall survival (48.3%, p < 0.01). CONCLUSION: The incidence of HPV positive OPC is increasing along with HPV negative OPC, which leads to stably low percentage of HPV positive OPC in Taiwan. HPV positive OPC may become an important head and neck cancer when the incidence of HPV negative OPC declines in the near future. P16 is a useful surrogate marker for HPV infection in OPC and a good prognostic indicator for treatment outcome of OPC. Patients with HPV positive OPC but no betel nut/cigarette exposure has an excellent prognosis. Betel nut/cigarette exposure significantly worsens the prognosis of HPV positive OPC.