Approximately 85% of adults acquire a human papillomavirus (HPV) infection in their life. Most HPV infections are transient, asymptomatic, and eliminated by the immune system [1
]. However, HPV viral infection can persist latently in a subset of the population. Individuals with persistent high-risk HPV (hr-HPV) infection may acquire epithelial cell abnormalities and subsequently develop cancers at the site of infection [3
]. Persistent hr-HPV infection is particularly associated with cervical, anogenital and oropharyngeal cancers [4
]. Hr-HPV positive oropharyngeal squamous cell carcinoma (OPSCC) is a distinct subtype of head and neck carcinoma. Risk factors that may prevent the natural clearance of oropharyngeal hr-HPV infection are genetic and lifestyle factors like smoking and alcohol consumption [6
]. The prevalence of cancer of the oropharynx due to hr-HPV infection has increased, particularly in North America and Europe [7
]. It differs from HPV-negative head and neck squamous cell carcinoma (HNSCC) by its risk factor profile, clinical behavior, and molecular biology. Compared to HPV-negative HNSCC, hr-HPV positive OPSCC better responds to treatment and has a significantly better prognosis [8
It is unclear whether hr-HPV persists in oropharyngeal tissues in patients with hr-HPV-positive OPSCC following cancer treatment and which consequences this might have. Zhang and coworkers collected blood at diagnosis and post-treatment in 64 patients with p16-positive OPSCC to test for serum antibodies to E6 and E7 proteins of HPV 16. At diagnosis, most patients were seropositive to HPV 16 E6 (85%). In the post therapeutic samples, HPV 16 antibody levels decreased slowly over time, but only three patients became seronegative [9
]. In another study, salivary and serum immunoglobulin G (IgG) antibodies targeting E2, E6, and E7 were measured in 44 patients with OPSCC at the beginning and 6–7 weeks following the completion of treatment. In this study, E7-directed antibodies were detected in saliva in most of the patients and were associated with the HPV status. The median of salivary E7 antibody levels decreased significantly post-treatment [10
Fakhry and coauthors used oral rinse samples for HPV detection in 396 patients with oral and oropharyngeal cancer, of which 51% were HPV-positive before therapy. After treatment the HPV prevalence decreased. In patients who received surgical resection, the HPV prevalence decreased from 69.2% to 13.7%. In a subset of patients who required postoperative radiotherapy, the HPV prevalence decreased from 70% to 38% after surgery and then to 1% after radiotherapy. HPV detection in oral rinses was performed several times for patients who received radiotherapy. The median time to clearance was 42 days (95%CI, 37–49 days). The only factor significantly associated with reduced clearance was current smoking. HPV-positivity with the same genotype was detected after treatment in 14.3% of initially HPV-positive patients and among these patients, the cumulative incidence of recurrence was 45.3%. HPV DNA detection after completion of therapy was significantly associated with increased risk of recurrence and death [11
In cervical dysplasia, hr-HPV infection can persist following treatment and promote disease recurrence. Söderlund-Strand et al. performed a long-term follow-up study obtaining cervical samples for HPV DNA testing and cytological analysis from 178 women with abnormal smears referred for conization. Three years after treatment 3.1% of women were persistently HPV-positive with the same HPV genotype as before treatment. Recurrent or residual cervical intraepithelial neoplasia (CIN) in histopathology was found among 9 (5.1%) women during follow-up. All these women had a type-specific HPV persistence. The authors concluded that only type-specific HPV persistence predicted recurrent or residual disease [12
In a previous publication, we reported that surface brushings from oropharyngeal cancer reliably detect HPV-DNA. In 53 patients with OPSCC, sensitivity and specificity of the brush test was 86% (95%CI: 70–95%) and 89% (95%CI: 65–99%) [13
]. Also, Broglie et al. reported liquid-based brush cytology specimens from oropharyngeal lesions to be a reliable method to identify patients with hr-HPV OPSCC. The authors collected brush cytology specimens prospectively from 50 patients with OPSCC. The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of brush cytology to identify hr-HPV-DNA-positive and p16-positive OPSCC samples were 88%, 83%, 94%, 95%, and 81%, respectively [14
In this study, we examined whether hr-HPV is still detectable in surface brushings after treatment in patients with initially hr-HPV-positive OPSCC. Moreover, we compared the course of disease in patients with and without post-treatment hr-HPV in oropharyngeal brushings.
In this study, we questioned whether patients with hr-HPV DNA-positive OPSCC remain hr-HPV DNA-positive after treatment and if post-treatment hr-HPV DNA at the initial tumor site is associated with the rate of disease persistence or recurrence. Before and after treatment, brushings were taken from the oropharynx, including the surface of the previous tumor site and tested for hr-HPV-DNA. Post-treatment brushings were available in 62 patients. Overall, 88.7% of hr-HPV-positive patients were hr-HPV negative at follow-up. In seven patients, hr-HPV after treatment was detected, and all patients hr-HPV-positive for the same genotype developed a recurrence or tumor persistence. Detection of hr-HPV at follow-up was associated with a substantially increased risk for persistent or recurrent disease (OR 244.2; 95% CI: 10.4 to 5757.7).
Post-treatment hr-HPV positivity and persistent or recurrent disease are rare events in hr-HPV-related oropharyngeal carcinoma. Accordingly, the results on potential influencing factors are based on a low number of patients and should be considered with caution. However, our results are in line with previous data.
Also, Hanna and coworkers described a significant decrease in post-treatment E7 antibody levels in the salivary glands of patients with OPSCC [10
]. Rettig and coworkers investigated hr-HPV DNA in oral rinses in 157 patients with OPSCC. At initial diagnosis, HPV type 16 was detected in 67/124 patients. After therapy, oral HPV 16 DNA was detected in six patients (9%). All five patients with persistent oral HPV 16 DNA developed a recurrent disease. Of these patients, three died. Persistent HPV 16 DNA detection in oral rinses was associated with a greater than 20-fold increased risk of recurrence (hazard ratio [HR], 29.7 [95% CI, 9.0–98.2]) and death (HR, 23.5 [95% CI, 4.7–116.9]) [15
]. In a similarly designed study on 93 patients with OPSCC and HPV 16-positive cancer of unknown primary, pre- and post-treatment serum or saliva samples were taken to detect HPV 16 E6. The authors reported hr-HPV-positive post-treatment saliva to be associated with higher risk of recurrence (hazard ratio [HR], 10.7; 95% CI, 2.36–48.50) and reduced overall survival (HR, 25.9; 95% CI: 3.23–208.00). The combined saliva and plasma post-treatment HPV 16 DNA status was 90.7% specific and 69.5% sensitive in predicting recurrence within three years [16
]. Fakhry and coauthors used oral rinse samples for HPV detection and described a significant decrease in HPV DNA after therapy, about 14.3% patients remained HPV-positive compared to 11.3% in our study. Also in this study, the authors described a significantly lower two-year overall survival among HPV-positive patients with persistent HPV detection for the same genotype (tumor-type) after therapy than among those without detectable tumor-type DNA after therapy (68% vs. 95%; adjusted HR, 6.61; 95% CI, 1.86–23.44; p
= 0.003), as was recurrence-free survival (55% vs. 88%; adjusted HR, 3.72; 95% CI, 1.71–8.09; p
< 0.001) [11
Although, only few studies analyzed HPV-positivity after treatment in OPSCC patients and although the number of patients in our study is low, post-treatment HPV-positivity seems to be a strong predictor for overall survival. An inclusion of this observation in the clinical management in patients diagnosed with hr-HPV-positive OPSCC should be discussed. Post-treatment HPV detection in the early follow-up period can be a possible new standard to evaluate and predict the clinical course of those patients. As the brush test does not require biopsies and is easy to perform in awake patients without anesthesia, this is a very suitable and simple test method for HPV-DNA detection. Post-treatment hr-HPV-positive patients may require a very close clinical and radiological follow-up, these patients can be at high risk for a poor overall survival. One of our patients with post-treatment HPV-positivity received an HPV vaccination with Gardasil®9 (Recombinant Human Papillomavirus 9-valent Vaccine, Merck Sharp & Dohme BV, Haarlem, Netherlands) during full remission, hoping to reduce virus activity. However, he developed a locoregional recurrence. Interestingly, 2/7 patients were HPV-positive after therapy with a different hr-HPV genotype, these two patients were in full remission and experienced no recurrence. It seems that a new hr-HPV infection occurred and the risk in these patients for recurrence after therapy is as low as in post-treatment HPV-negative patients. However, higher numbers of patient are needed for further evaluation.
Also, in patients with cervical cancer, persistent HPV infection is associated with an increased risk for recurrence [17
]. In a study on 72 women with CIN, persistence or clearance of hr-HPV DNA was described as an early valid prognostic marker of failure or cure after treatment, more accurate than cytology or section margin status at the time of conisation. The absence of hr-HPV DNA had a 100% negative predictive value [18
]. Söderlund-Strand et al. described a type-specific HPV-persistence in women with a residual CIN. No recurrent or residual disease was detected in women with any other patterns of HPV positivity, e.g., type change or fluctuating positivity [12
In our study HPV 16 was the most common genotype before therapy, other genotypes like HPV 18, 33, and 35 were rare. Two patients were pre- and post-treatment positive for comparatively rare HPV genotypes 18 and 33. As reinfection with these rare types is unlikely, we assume tumor virus persistence rather than new HPV infection. Our findings about pretreatment HPV genotypes are in line with other studies. Chatfield-Reed and coauthors reported that of 99 hr-HPV positive HNSCC patients, 75.6% were positive for HPV 16 and 3% for HPV 18. In this study, 16.2% were positive for the genotype HPV 33, which we detected in only 1.9% [19
]. Fossum and coworkers reported in 166 OPSCC patients (77% hr-HPV positive), HPV 16 to be the predominating genotype (65%), followed by HPV 33 (17%), HPV 18 (2%), and HPV 31/35/56/59 in one patient each [20
In this study, post-treatment hr-HPV positivity was associated with clinical T-stage at diagnosis and tumor EGFR expression. An association of post-treatment hr-HPV positivity and primary tumor EGFR expression has not yet been reported. However, this observation is based on very few patients. EGFR protein over-expression has been reported in 70–100% of HNSCC, but 46–72% of OPSCC [21
]. Although the reason why HPV-positive tumors express less EGFR expression is currently unknown, smoking has been hypothesized to be a contributory factor [24
We did not observe an association of smoking and post-treatment hr-HPV positivity. This might be due to the low number of patients. Active smoking was admitted by 8/45 patients. Kero et al. reported a correlation between persistent oral HPV infection and smoking in 131 men who were sampled by serial oral scrapings. Genotype-specific HPV persistence was detected in 18/129 men. The mean persistence time ranged from 6 months to 30.7 months. The authors concluded that most of the persisting oral infections in males were caused by HPV 16, and smoking increased the oral hr-HPV persistence [25
]. Another assumed risk factor for HPV persistence is immunosuppression. In a cohort of 97 HIV/AIDS patients, a genotype-specific oral and oropharyngeal HPV persistence was described in 33.3% of patients, of which 13.3% were hr-HPV positive [26
However, the most significant observations of this study are that the majority of patients are hr-HPV-negative after therapy and that post-treatment hr-HPV positive patients seem to have an increased risk of tumor persistence or recurrence.
Limitations of the study are that not all OPSCC patients were enrolled consecutively in the study. From 12 patients, no post-treatment brushings were available because they missed the follow-up or received follow-up in another hospital. This includes also patients who may have had a relapse. Moreover, several otorhinolaryngologists were involved in sampling, but all were instructed how to collect samples. In this study, the brushings were obtained in some patients under general anesthesia during restaging panendoscopy and in some patients awake during clinical follow-up.
The brush test is a simple test method and does not require tumor biopsies for HPV-DNA detection, future clinical usefulness of this test includes an oropharyngeal HPV screening for populations at risk, e.g., immunosuppressed elderly population. Partners of HPV-positive OPSCC patients, who are often concerned about oral HPV transmission, can also be tested by oropharyngeal brushings.