HPV-Related Oropharyngeal Cancer in Southern Thailand: Proportion Trend and Survival Outcome

Background: Persistent high-risk human papillomavirus (HPV) infection is one of the major etiologies of oropharyngeal squamous cell carcinoma (OPSCC). This study aimed to determine the proportion, temporal trend, and prognostic significance of HPV-related OPSCC in Thai patients. Methods: The study included patients with OPSCC who were treated at Songklanagarind Hospital (Songkhla, Southern Thailand) from 2009 to 2020. HPV status was screened by p16 expression using immunohistochemistry and confirmed by real-time polymerase chain reaction. Cox regression was used to determine prognostic significance. Results: The overall proportion of HPV+ OPSCC was 15.3% (95% confidence interval [CI]: 12.1–18.5) with a slightly increased proportion from 10.6% in 2009–2010 to 16.5% (2019–2020) (P for trend = 0.166). Among the HPV+ cases, HPV16 was detected in 65.3%, HPV18 in 34.7%, and other high-risk HPV types in 24%. Patients with P16+ or HPV+ OPSCC had significantly better overall survival (hazard ratio [HR]: 0.63, 95% CI: 0.45–0.90 and HR: 0.63, 95% CI: 0.45–0.88, respectively). Conclusion: Thai patients in the southern region have a low proportion of HPV-related OPSCC with an increasing trend. Both P16 expression and HPV DNA status are strong independent prognostic factors of OPSCC.


Introduction
Oropharyngeal cancer is an important cancer worldwide with a global age-standardized incidence rate (ASR) of 1.8 per 100,000 men [1].In Thailand, the nationwide ASR is 1.7 per 100,000 and it is 2.0 per or specificity, including DNA-or RNA-based methods are recommended in the case of an equivocal p16 expression [18].
A vast majority of studies regarding the prevalence of HPV-related OPSCC and its temporal trend are from Western countries [4,7] and only a few studies are from East Asian countries [19,6,8,20].Recently, studies from central and northeastern regions of Thailand reported an overall proportion of HPV-related OPSCC of 11.5%-17.7%[21][22][23].The current study presented the overall proportion of HPV-related OPSCC and its temporal trend over 12 years in a cohort of patients treated in a tertiary university hospital in Southern Thailand.Additionally, clinicopathological characteristics and the prognostic significance of HPV-related OPSCC were evaluated.p16 IHC was used as a screening marker for HPV infection and a real-time polymerase chain reaction (PCR) for HPV DNA detection as a confirmation method.

Patients and clinical data
The study included patients with primary OPSCC, treated at Songklanagarind Hospital from January 2009 to December 2020.This 1000-bedded tertiary university hospital in Songkhla province provides comprehensive care serving a population in the southern region.The majority (more than 80%) of cancer patients in the region who require radiation or chemotherapy or need complex surgeries are referred to this hospital.The oropharyngeal site was defined following the International Classification of Diseases version 10 (ICD-10), as a base of the tongue, tonsil, soft palate, uvula, pharyngeal wall, and overlapped area of the oropharynx.Only patients with available paraffin-embedded tissue blocks in the Department of Pathology were included, and tumor samples with limited tumor cells for IHC staining were excluded.
Demographic and clinical data, including age, sex, history of smoking, alcohol drinking, betel nut chewing status, tumor site, clinical stage, date of last follow-up, and status of last follow-up, were retrieved from electronic medical records.Pathological information was obtained from pathological reports.Clinical staging was based on the American Joint Committee on Cancer 7 th (2009-2017) and the 8 th edition cancer staging (starting in 2018).The date and cause of death were obtained from the database of the hospital cancer registry, which was updated through the National Civil Database bi-annually.The study was approved by the Human Research Ethics Committee of the Faculty of Medicine, Prince of Songkhla University (REC.63-241-5-1).

Immunohistochemistry (IHC) for p16 expression and evaluation
The 3-μm-thick sections were deparaffinized with xylene and rehydrated in graded alcohol.An automated immunostainer (Leica BOND-MAX, Melbourne, Australia) was used for IHC for p16 expression.Antigens were retrieved in the Tris-EDTA buffer (Bond Epitope Retrieval Solution 2, Leica Biosystem, Newcastle Upon Tyne, UK), pH 9, in a pressure cooker at 95°C for 4 min.
Sections were first incubated with bond peroxidaseblocking reagent (Bond Polymer Refine Detection, Leica Biosystem, Newcastle Upon Tyne, UK) and then with primary antibodies against p16 at a dilution of 1:5 (clone E6H4, CINtec® p16 Histology; Roche, Tuscon AZ, USA).A bond polymer refine detection kit (Leica) was used to detect the antigen-antibody reaction, followed by color development using 3,3'-diaminobenzidine as a chromogen and Meyer's hematoxylin as a counterstain.
Immunostaining for the p16 was evaluated by the percentage of positively stained tumor cells.The intensity of staining was scored as strong (3+), moderate (2+), weak (1+), or negative (0).Moderate to strong intensities and diffuse nuclear and cytoplasmic staining in ≥70% of the tumor cells were considered positive for p16 expression.All sections were independently examined by a senior pathologist and a third-year resident.Discrepancies were resolved by a discussion on a multi-head microscope.

HPV DNA detection
HPV DNA detection was done for the purpose of this study.All p16+ tumors and 50 random p16− tumors were confirmed for the presence of HPV.DNA by real-time PCR.The QIAamp® DNA FFPE Tissue Kit (Qiagen GmbH, Hilden, Germany) was used to extract DNA from 5 to 10 5-micron (depending on tissue size) tissue cut from paraffin-embedded tissue blocks and stored at −20℃ until used.The 14 HR HPV with 16/18 Genotyping Real-time PCR Kit (HBRT-H14; Hybribio, Chaozhou, China) was used for real-time PCR.The kit has been designed to detect 14 HR HPV types, including HPV 16,18,31,33,35,39,45,51,52,56,58,59, 66, and 68, with specific detection of HPV16 and 18 genotypes.Cellular internal control was included for each sample to monitor the whole testing process, starting from DNA extraction to signal detection.Bio-Rad CFX Manager (C1000 Touch Thermal Cycler, Bio-Rad, Germany) was used to assess results following the manufacturer's instructions.Positive controls were valid when the threshold cycle (Ct) was ≤36, while negative controls were valid when undetected.Samples were re-run if either control was deemed invalid.The PCR results were interpreted as HPV type 16, HPV type 18, other HR types, or negative for detection.

Statistical analysis
Descriptive analysis of clinicopathological data was presented in percent, mean (standard deviation), and median (interquartile range [IQR]) as appropriate.The chi-square test or Fisher's exact test was used to test the comparison of clinicopathological variables of HPV+ versus HPV− OPSCC as appropriate.The proportion of p16+ OPSCC and HPV+ OPSCC by 2-year intervals were calculated along with a 95% confidence interval (CI).The chi-square test was used to test the significance of the trend of proportion.The Kaplan-Meier method was used to estimate the overall survival (OS) and the logrank test was used to test for differences between survival curves.Cox regression analysis was used to obtain independent associations of p16 and HPV status with OS.All variables were tested for proportional hazard assumption, and a stratified Cox regression model was applied if the variable DOI:10.31557/APJCP.2024.25

Association of clinicopathological characteristics with HPV status
Table 3 shows the clinicopathological characteristics among patients with HPV+ and HPV− OPSCC.Patients with HPV+ tumors were more likely to be younger, female patients, non-smokers, and non-betel chewers.HPV+ tumors were more likely to be moderately or poorly differentiated SCC (76%) compared to HPV tumors (59.4%).The distribution of tumor size and clinical stage were not different between the two groups.

Association of p16 expression and HPV status with OS
The median follow-up time was 13.3 months (IQR: 5.6-31.1 months).The median survival time of the entire cohort was 14.7 months (95% CI: 12.9-16.9months).Kaplan-Meier analysis revealed significantly better OS in patients with p16+ and HPV+ tumors than those with p16− and HPV− tumors (p < 0.001) (Figure 2).The median survival time of patients with p16+ or HPV+ was 33.3 months compared to 13.6 months in patients with p16− or HPV− tumors.

Cox regression analysis for OS
Univariate Cox regression revealed a significant association in age, clinical stage, treatment, p16 expression, and HPV status with the increased risk of death (Table 4).Multivariable analysis revealed that the treatment did not meet the proportional hazard assumption, thus we used stratified Cox regression by fitting the model according to the strata of treatment.p16 expression and HPV status were separately evaluated in a multivariate model as they are highly correlated.Table 5 shows the final multivariate model.p16+ (HR: 0.63, 95% CI: 0.45-0.90)and HPV+ tumors (HR: 0.63, 95% CI: 0.45-0.875)were strongly associated with favorable survival outcomes.
did not meet the assumption criteria.A p value of < 0.05 was considered statistically significant.The R Program version 4.0.2(R Foundation for Statistical Computing, Vienna, Austria) was used for all analyses.

Patient characteristics
There was a total of 560 patients with OPSCC treated at our institute for a period of 12 years.Tissue blocks of 65 patients were not available and 5 blocks had inadequate tissue for IHC staining, leaving 490 cases for the final analysis.Table 1 shows the clinicopathological characteristics of the study cohort.The median age was 65 years.Most patients were male (93.9%) and had a history of smoking (88.5%) and alcoholic drinking (79.8%).The majority of patients were at stage III or IV at diagnosis (80.5%).

Discussion
This study revealed a low proportion of HPV+ OPSCC (15.4%); however, an increasing trend of this proportion is evident.The proportion of HPV+ OPSCC increased from 13% to 16.5% in 12 years.Consistent with most previous studies, our study supported the evidence of favorable survival outcomes in patients with HPV-related OPSCC.
A high prevalence of HPV-related OPSCC has been reported in various countries, especially those in North America and Europe [5,9].In our study, we revealed a low prevalence of HPV+ tumors among OPSCC (15.4%).
As OPSCC patients need complex treatments, they are majorly referred to our institute, therefore, our results, more or less, represent the figure of HPV-related OPSCC in the southern Thai population.Our study showed almost DOI:10.31557/APJCP.2024.25  exactly similar to the previous three studies from other regions of Thailand and other studies from Southeast Asia.These included two reports from the central region of Thailand [22,23] and one report from the northeastern region [21] which revealed the proportion of HPV-related OPSCC of 14.5%, 15.6%, and 17.7% from a total sample size of 110, 64 and 96, respectively.These three studies used PCR-based methods for HPV detection, thus the results are comparable.A study from Malaysia also revealed a low proportion of HPV-related OPSCC (16.7%) [24].All of this evidence may indicate a low proportion of HPV-related OPSCC in Southeast Asian populations.However, a higher prevalence of HPV-related OPSCC (28%-38%) was reported in other Asian countries, including Japan, China, and Taiwan [19,6,8].Reports from Middle East Asia also documented a high prevalence (up to 80%) of HPV+ OPSCC [25,26].This is probably due to the more Westernized lifestyle of certain population groups as well as other factors in these countries compared to the Southeast Asian populations.
We found an increasing trend of HPV+ OPSCC proportion from 10.56% in (2009-2010) to 16.5% (2019-2020), but with no statistical significance, probably due to the small number of cases in each time interval.Nevertheless, our results, more or less, represent the figure of HPV-related OPSCC in the southern Thai population.The increasing trend of HPV+ OPSCC has also been reported by the study from the northeastern region [21].This study reported a significant increase in HPV prevalence by 2% annually from 16% in 2012 to 26% in 2017.The study used the same HPV PCR detection kits as our study, but their result may be more solid as they performed HPV DNA analysis in all tumors while our study confirmed the presence of HPV DNA only in p16+ tumors and a random set of p16− cases.Other East Asian countries, including Taiwan and Korea, also reported an increasing trend of HPV+ OPSCC.A large study from Taiwan revealed an increasing trend for 18 years, but with no statistical significance [19].Another study from South Korea demonstrated a significantly increased HPV+ OPSCC proportion from 33.3% in 2008-2009 to 83.3% in 2020 [27].However, this study used p16 expression to define HPV status which might lead to overestimated HPV+ rate.All this evidence indicates an increasing trend of HPV-related OPSCC in the Asian population which is similar to Western countries, although the rate of increase is controversial.
HPV16 is consistently reported as the major genotype (>85%) in HPV-related OPSCC in the Western population [28][29][30].This information in the Asian population is scarce.One large study from Taiwan [19] and from Thailand [21] reported a high frequency of HPV16 up to 83% and 82%, respectively.However, the representative HPV type in the latter study may be limited due to the small number of HPV+ cases (n = 17).Interestingly, our study revealed a different result of a considerably lower proportion of HPV16 (62.67%).Additionally, our cohort had a higher proportion of multiple infections (21.3%) compared to <10% in the aforementioned studies [28,19].The difference in the frequency of specific HPV genotypes might have a clinical impact.A recent systematic review [31] revealed a significant impact on survival in three of six studies, of which two studies revealed a better survival among HPV16 cases compared to other HR HPV genotypes while the other one revealed the reverse results.Therefore, the determination of specific HPV genotypes in HPV-related OPSCC may be important for patient management; however, further meta-analysis or future trials are needed.
The association of clinicopathological characteristics with HPV status in OPSCC appears to be similar to previous studies in Western countries and Thailand [32,21,23].Patients with HPV+OPSCC are younger, non-smokers.The tumor occurs more frequently at the tonsil and has poorly differentiated histology.The results regarding prognostic HPV status are also consistent with previous studies [32,28].Patients with HPV+ tumors (HR: 0.63, 95% CI: 0.45-0.88)had better OS.The prognostic value of p16 expression was exactly equivalent to that of HPV DNA status (HR: 0.63, 95% CI: 0.45-0.89).This supports the clinical utility of p16 expression evaluated by IHC in clinical practice.
Our study has some limitations.We could not evaluate a portion of patients treated in our hospital (about 12%) due to unavailable tissue blocks.In addition, not all tumor samples were tested for HPV DNA analysis.We performed DNA analysis in p16+ samples and selected p16− samples; therefore, the HPV prevalence in this study may be underestimated.Additionally, this is a hospital-based study, thus the results may not represent the incidence and trend of HPV-related OPSCC in the population.However, this study is a large series of its kind and is the largest study regarding HPV-related OPSCC in Thailand.
In conclusion, the present study reports a potentially increasing proportion of HPV-related OPSCC in the Southern Thailand population, although the overall proportion is low.HPV-related OPSCC, evaluated by either p16 IHC or HPV PCR analysis, was confirmed to be associated with favorable survival outcomes.

Table 2 .
Mutually Exclusive Distribution of HPV Type among the HPV-Positive Cases.

Table 3 .
Clinicopathological Characteristics of Patients Classified by HPV DNA Status.

Table 4 .
.1.57HPV-Related Oropharyngeal Cancer in Thailand Univariate Cox Regression Analysis for OverAll Survival of Patients with Oropharyngeal Cancer.

Table 5 .
Multivariate Cox Regression Analysis for Overall Survival of Patients with Oropharyngeal Cancer.†, adjusted by strata of treatment; ‡, HPV status and p16 expression