The incidence of human papillomavirus-positive (HPV+) oropharyngeal squamous cell carcinoma (OPSCC) is rising rapidly and has exceeded cervical cancer to become the most common HPV-induced cancer in developed countries. Since patients with HPV + OPSCC respond very favorably to standard aggressive treatment, the emphasis has changed to reducing treatment intensity. However, recent multi-center clinical trials failed to show non-inferiority of de-escalation strategies on a population basis, highlighting the need to select low-risk patients likely to respond to de-intensified treatments. In contrast, there is a substantial proportion of patients who develop recurrent disease despite aggressive therapy. This supports that HPV + OPSCC is not a homogeneous disease, but comprises distinct subtypes with clinical and biological variations. The overall goal for this review is to identify biomarkers for HPV + OPSCC that may be relevant for patient stratification for personalized treatment. We discuss HPV + OPSCC as a heterogeneous disease from multifaceted perspectives including clinical behavior, tumor morphology, and molecular phenotype. Molecular profiling from bulk tumors as well as single-cell sequencing data are discussed as potential driving factors of heterogeneity between tumor subgroups. Finally, we evaluate key challenges that may impede in-depth investigations of HPV + OPSCC heterogeneity and outline potential future directions, including a section on racial and ethnic differences.
Oropharyngeal squamous cell carcinoma (OPSCC) represents a subset of head and neck squamous cell carcinomas (HNSCC) comprising malignancies in the tonsils, base of tongue, soft palate, and posterior pharyngeal wall. The global incidence of OPSCC has escalated in high-income and developed countries [1,2,3], which is primarily due to the rise in human papillomavirus (HPV) infection [3,4,5]. OPSCC has already overtaken cervical cancer to become the most common HPV-associated (HPV+) cancer in the United States and United Kingdom [1, 4, 6]. The 8th edition of the American Joint Committee on Cancer (AJCC-8) recently recognized HPV + OPSCC as a separate entity from classical, tobacco-associated HPV-negative [HPV(-)] OPSCC) [7, 8] with distinct demographics , clinical features [9, 10] and molecular profiles [11,12,13]. Importantly, patients with HPV + OPSCC have better survival due to favorable response to treatment [9, 10]. Similar to other HPV-related cancers, HPV E6 and E7 oncoproteins are key drivers for OPSCC largely due to inhibition of p53 and Rb, respectively [14, 15]. Clinically, p16 positivity by immunohistochemistry (IHC) is used to diagnose HPV-related malignancies with over 95% concordance to HPV DNA positivity . However, discordance between HPV and p16 occurs in some patients with OPSCC who show poorer outcomes than those with HPV+ and p16+ OPSCC . Additional confirmatory testing using in situ hybridization (ISH) or polymerase chain reaction (PCR) to detect HPV DNA or RNA is recommended .
Despite superior survival of HPV + OPSCC, standard treatment is associated with high toxicities and compromised quality-of-life [18, 19]. As a result, there is growing interest in de-escalating treatment for these patients. However, recently completed clinical trials to de-intensify chemoradiation in unselected populations failed to demonstrate non-inferiority [18, 20]. Moreover, a subset of HPV + OPSCC continues to progress despite standard aggressive treatment [10, 21]. These findings suggest heterogeneity among primary HPV + OPSCC that influences clinical outcomes. Here, we discuss evidence demonstrating dichotomy in primary HPV + OPSCC from clinical, morphological, and molecular perspectives. We also evaluate factors that may contribute to tumor heterogeneity, with the goal of summarizing biomarkers that could predict patient risk and optimize treatment selection.
From oral HPV infection to OPSCC development
More than 200 types of HPV have been identified; 15 are high-risk and associated with cancer. Among these, HPV16 induces 80–90% of all HPV + OPSCC [22, 23]. Transmission of HPV to the oropharynx is through sexual contact  and infection normally occurs in basal and mitotically active epithelial cells. The productive life cycle of HPV is closely related to keratinocyte differentiation in stratified epithelium  that likely triggers extensive viral genome amplification, via a poorly understood mechanism  (Fig. 1). Most HPV infections are transient and cleared within 2 years [26,27,28]; failure to eliminate high-risk oral HPV may trigger neoplastic progression and malignancy [26,27,28]. Detection of high-risk HPV in oral samples has been shown to significantly increase the risk of OPSCC , and HPV oral DNA has been detected periodically for up to 7 years before OPSCC diagnosis [27, 29]. In contrast, seropositivity of HPV E6 could be detected 20–30 years before OPSCC . Therefore, HPV infection likely occurs decades before clinical presentation. Within the oropharynx, HPV + OPSCC is highly distributed at the tonsillar crypt, which has reticulated epithelium that allows virions easy access to basal epithelial cells . The basement membrane is also porous to immune cells that directly contact the basal epithelium (Fig. 1) . Resident myeloid cells within the crypt may foster a permissive microenvironment that enables HPV to evade host immune surveillance [32, 33].
During persistent infection, E6 and E7 are highly expressed (Fig. 1), encouraging molecular alterations that shape the epithelial environment to initiate carcinogenesis through their key functions in suppressing p53 and Rb . Other oncogenic functions of E6 and E7 include compromising cellular DNA repair, enhancing genomic instability, and increasing immune escape (Fig. 1) . Increased E6 and E7 proteins may promote integration of HPV DNA into host genome, but only 50–70% of HPV+ OPSCCs harbor HPV integration [35,36,37]. It is likely that episomal HPV induces different carcinogenic mechanisms from integrated HPV . Integration breakpoints and subsequent viral genome linearization may occur at the open reading frame of E1, E2, or E5, leading to their deletion or truncation and concomitant loss of expression . As E2 is a transcriptional repressor for E6/E7, expression of E6 and E7 is enhanced . However, recent studies showed that disruption of the E2 gene does not necessarily follow viral integration [35, 39]. In OPSCC with integrated HPV but intact E2, methylation at the E2-binding sites in the upstream regulatory regions of E6 and E7 inhibited E2-mediated repression of E6/E7 transcription, resulting in E6/E7 overexpression despite the presence of E2 .
Since HPV16 dominates HPV + OPSCC cases, most studies focus on HPV16 + OPSCC and assume that non-HPV16 genotypes behave similarly. Consequently, the exact prognostic impact of distinct HPV genotypes is unclear; some studies suggest that patients with non-HPV16 variants have inferior survival outcomes [40, 41], while others show no or minimal effect on survival [42, 43]. HPV16 is also more likely to infect the tonsillar crypt than the HPV33 variant , although the underlying molecular basis is not yet understood.
Clinical management of HPV-associated OPSCC
Although HPV + OPSCC is associated with superior treatment outcomes, treatment recommendations for these patients remain the same as those for HPV(−) OPSCC. Early-stage patients receive surgery or radiation (RT), while patients with locally advanced cancer receive multimodal treatment, either concurrent chemotherapy and radiation (CRT) or surgery followed by adjuvant CRT  (Fig. 2). RT as definitive treatment is 70 Gy, while 60–66 Gy is recommended for adjuvant treatment . Platinum-based agents such as cisplatin are usually given as chemotherapy [44, 45].
In the definitive setting, standard-of-care treatment often entails substantial morbidity [18, 20] and clinical trials focused on strategies to de-intensify standard therapies. These include: (i) substitution of cisplatin with cetuximab and immunotherapy in the standard CRT regimen; (ii) RT dose reduction to definitive and/or elective regions; (iii) adaptation of RT based on mid-treatment response; (iv) induction chemotherapy followed by de-intensified RT or CRT, and (v) transoral surgery (TORS) with de-escalated and risk-based adjuvant therapy (Supplementary Table 1). However, currently, no large multi-center phase III trial has proven non-inferiority of de-intensification paradigms in the definitive setting. There is a compelling need to identify true low-risk HPV + OPSCC likely to respond to de-escalated regimens.
Despite the good prognosis of primary HPV + OPSCC, locoregional recurrences and distant metastasis are observed [10, 21, 46]. Current guidelines recommend periodic clinical and radiological examinations for over 5 years to detect recurrence  (Fig. 2). Patients with recurrent tumors receive salvage therapy with curative intent such as salvage surgery, re-irradiation, or systemic chemotherapy  (Fig. 2). Salvage therapy significantly reduces risk of death in recurrent OPSCC by ~50%, with p16+ responding better than p16- tumors [21, 48]. Recently, PD-L1/PD-1 immune checkpoint inhibitors were approved by the U.S. Food and Drug Administration (FDA) to treat recurrent and metastatic HNSCC, regardless of HPV or p16 status . Although p16+ HNSCC patients had superior responses to PD-L1/PD-1 inhibitors than their p16(−) counterparts, this was limited to recurrent and metastatic tumors [50,51,52,53]. In the definitive setting of primary and locally advanced HNSCC, PD-L1/PD-1 inhibitors did not provide benefit regardless of p16 status [54,55,56]. Since HPV E6 and E7 are constitutively expressed in HPV + OPSCC, they are attractive targets for therapeutic vaccines . These therapeutic vaccines have yielded promising results in several Phase I and II clinical trials in locally advanced or recurrent and metastatic HPV + HNSCC, especially in conjunction with PD-L1/PD-1 therapy [58,59,60]. Further validation from prospective Phase III trials is needed for implementation into clinical routine.
In adoptive T-cell therapy, T cells extracted from a patient are engineered to target the tumor cells, expanded, and re-infused in the same patient. A recent Phase I clinical trial explored the effectiveness of T-cell receptor-engineered T cells targeting HPV E7 in 12 recurrent and metastatic HPV+ cancer patients . Tumor regression was observed in 6 of 12 patients, suggesting high curative potential . However, the patient cohort is small and comprises multiple cancer types, including cervical and anal cancers . More clinical trials are needed to determine the effectiveness of T-cell therapy in locoregional HPV + OPSCC.
Although currently available therapies are effective for HPV + OPSCC at various stages (Fig. 2), responses vary due to inter-patient heterogeneity. Since de-escalation trials largely focus on primary HPV + OPSCC (Supplementary Table 1), this review focuses on tumor heterogeneity in primary HPV + OPSCC. This will allow emphasis on intrinsic characteristics associated with tumors that recur after initial treatment.
Clinical heterogeneity of HPV-associated OPSCC
HPV status is a strong determinant of superior response to therapy, but prospective studies consistently demonstrated that a subset of patients (~25%) had disease progression within 2 years of treatment [9, 10]. This suggests that not all primary tumors respond favorably to standard aggressive therapy. To better understand differential clinical behaviors, Ang and colleagues performed recursive-partitioning analysis that classified all OPSCC into distinct risk groups . Patients with HPV + OPSCC and tobacco exposure have an intermediate-risk phenotype with poorer survival than non-smokers. In contrast, HPV(−) OPSCC was classified in the high-risk group . Several studies reported positive associations between tobacco exposure and survival of HPV + OPSCC, although controversies remain [62, 63]. Risk stratification analysis is essential for treatment de-escalation that aims to personalize therapeutic regimens. In two early clinical trials (RTOG 1016 and De-ESCALaTE), replacing cisplatin with cetuximab, a drug targeting EGFR, in the standard CRT regimen led to inferior survival in unselected HPV + OPSCC with no reduction in acute and long-term toxicities [18, 19]. In contrast, the FDA approval of TORS provides an alternative for radiation therapy de-escalation, as reflected by several de-intensification trials (Supplementary Table 1) but surgery has toxicity as well. Consequently, patient-reported outcomes have been numerically superior for definitive CRT strategies [64, 65]. Furthermore, some tumors still recur after TORS and survival rates after recurrence are low at around 60% . To predict patients with lowest risk of recurrence, several de-escalation trials are selecting patients based on clinical, biologic, and imaging markers such as smoking history, extranodal extension, diffusion MRI, and FDG-PET metrics, and genomic alterations (Supplementary Table 1). Results from these clinical trials are awaited.
In summary, HPV + OPSCC demonstrate large variations in clinical behavior and treatment response. Results from completed and ongoing treatment de-intensification studies support the notion that patients with HPV + OPSCC require personalized treatment strategies. However, even with the AJCC-8 classification that better stratifies HPV + OPSCC than AJCC-7, there is no significant difference in overall survival between Stages I and II, or between Stages II and III patients [67, 68]. This suggests that clinical variables not included in AJCC-8 may significantly influence prognosis. Some of the proposed factors include extranodal extension [69,70,71,72], perineural invasion [73, 74], and angiolymphatic invasion , but controversies regarding their prognostic impact remain. Although p16 IHC is used to diagnose presence of HPV + OPSCC, a study shows that its cellular localization dictates survival outcomes [75, 76]. Understanding biological factors that implicate clinical dichotomy will refine cancer staging and better inform optimal therapeutic strategy for specific patient subgroups.
Morphologic heterogeneity of HPV+OPSCC
Based on morphologic variations, there are three main subtypes: non-keratinizing, keratinizing, and non-keratinizing with maturation (NKM) . These differ in keratinization, stromal reaction, and nuclear and cytoplasmic features among others (Table 1). About 50% of OPSCC show non-keratinizing morphology and most (70–98%) are HPV+ [78,79,80]. In contrast, keratinizing morphology is mostly associated with HPV(-) OPSCC, although a small fraction (~3–15%) of HPV + OPSCC also present this phenotype [78, 79, 81]. This suggests that a subset of HPV + OPSCC resembles the more aggressive HPV(−) cancer. In some keratinizing OPSCC, no true keratin formation is observed, but a dense layer of cytoplasmic eosinophilia imparted by keratin intermediate filaments is observed . Together, these three subtypes comprise ~90% of OPSCC (Table 2), while the other 10% are rare variants such as basaloid, undifferentiated, and papillary carcinomas [80, 81].
Given that most non-keratinizing OPSCC are HPV+, it is associated with favorable survival [78, 82, 83]. However, in a cohort of 208 p16+ patients, the non-keratinizing phenotype yielded better survival only in p16- but not in p16+ OPSCC. Within the p16+ sub-population, keratinization was only prognostic in smokers and not in non-smokers . This suggests that morphologic subtype alone is an insufficient prognostic classifier for p16+ OPSCC.
While OPSCC can be grouped into distinct morphologic subtypes, it is important to discuss specific histologic characteristics that provide substantive prognostic information. One such feature is nuclear pleomorphism, such as anaplasia and multinucleation. Anaplasia is the presence of three or more cells with nuclei that have diameters of five or more lymphocyte nuclei, while multinucleation occurs when each tumor cell has three or more nuclei . Anaplasia and multinucleation (A/M) increase from non-keratinizing to NKM to keratinizing subtypes . The distribution of A/M also differs, with non-keratinizing OPSCC showing focal distribution, while the other subtypes likely present more diffuse distribution. A/M was significantly associated with worse disease-specific survival regardless of p16 status ; this was substantiated in multiple studies regardless of whether pathologists’ evaluation or computerized scoring was used [84,85,86,87]. As A/M reflects genomic instability, it likely highlights a subset of genetically complex OPSCC. However, the prognostic value of A/M in HPV + OPSCC has been challenged by other studies [88, 89], where A/M was not predictive of survival or recurrence. Therefore, A/M may be insufficient to predict prognosis of HPV + OPSCC.
With the rise of artificial intelligence, researchers are using machine-learning approaches to characterize histopathologic variables in HPV + OPSCC [84, 85, 90]. In one study, patients with poorer prognosis demonstrated greater inter- and intra-tumoral variations in nuclear morphologic parameters such as nuclear texture, granularity, and cytomorphology . The molecular and biologic basis behind these divergent features both within and between tumors remains unclear. Understanding the interplay between molecular and morphologic phenotypes will help us understand how morphologic variations point toward distinct biological pathways that may influence treatment choices for subsets of HPV + OPSCC.
Molecular heterogeneity of HPV+OPSCC
Heterogeneity in clinical and morphological profiles implies intrinsic heterogeneity that could be delineated by molecular profiling. Indeed, molecular characterization of HNSCC cohorts revealed distinct HPV+ subtypes (Table 2). Earlier molecular profiling studies such as The Cancer Genome Atlas (TCGA) did not distinguish OPSCC from non-OPSCCs [11, 91,92,93]; this is a major concern as OPSCC differs significantly from non-OPSCC. Consequently, non-OPSCC were overrepresented in the subtype with poorest prognosis. Nevertheless, these studies provide important clues into molecular heterogeneity of HPV + OPSCC, as most HPV + HNSCC are OPSCC. This could be validated by more recent studies that restrict analysis to HPV + OPSCC (Table 2).
Analyzing gene expression from key molecular characterization studies on HPV + HNSCC [11, 91,92,93], Qin et al.  noted substantial congruency and derived three main molecular subtypes of HPV + HNSCC. The first subtype is immune-enriched and associated with low HPV integration frequency and high mesenchymal differentiation, the second is highly keratinized and basal-like with high stromal content and metabolic signatures, and the third is a highly keratinized subtype with low stromal content and suppressed immune responses. Overall, molecular classification studies revealed variables associated with prognosis, including viral transcript expression and host genomic alterations. Importantly, a subgroup of HPV + OPSCC demonstrates aggressive clinical features and molecular profiles similar to HPV(-) counterparts, including poor clinical outcomes and treatment resistance (Table 2).
While HPV + OPSCC subtypes mentioned above reflect heterogeneity at inter-patient level, single-cell sequencing studies also revealed huge intratumoral heterogeneity within the same patient [95, 96]. Intratumoral heterogeneity is the phenomenon where different cell populations coexist in the same tumor; metastasis and treatment resistance may result when aggressive or persistor clones are selected over others. In HPV + OPSCC cancer cells, variations in chromosomal aberrations, pathway activation, and HPV viral gene transcript expression were reported between and within tumors with important prognostic implications . Moreover, HPV + OPSCC also showed differing intratumoral inflammation and fibroblast elastic differentiation plasticity . In the following sections, we discuss molecular variations that may contribute to inter- and intratumoral heterogeneity. This is important because molecular phenotyping illuminates more extensive information on tumoral heterogeneity than traditional histopathologic assessment, enabling robust stratification that more accurately reflects prognosis and treatment responses.
HPV integration and extrachromosomal DNA
Since HPV is an etiologic factor for many OPSCCs variations in HPV biology likely affect clinical responses. Integration of HPV into the host genome is common in HPV+ cancers and could indicate poor prognosis . RNA sequencing on HPV + HNSCC, 18 from University of Michigan and 66 from TCGA, of which 80% were OPSCC , identified two molecular subtypes, KRT and IMU. KRT has an increased keratinization signature, while IMU has a stronger immune response and mesenchymal signature . Importantly, KRT OPSCCs are more aggressive than IMU and have significantly higher HPV integration frequency. Locati and colleagues published similar findings based on gene expression analysis in eleven cohorts . They identified three distinct subtypes of HPV + HNSCC: Cl1, Cl2, and Cl3. The Cl1 corresponds to IMU with best survival and lowest HPV integration rates . Positive correlation between HPV integration and poor prognosis were also reported by others [36, 37, 97, 98]; HPV integration-positive HNSCC had unfavorable survival similar to HPV(-) cancer . Furthermore, HPV integration in HNSCC predicts aggressive clinical phenotypes such as large tumor size and perineural invasion . However, other studies reported conflicting findings [35, 98,99,100]. Inconsistent results for the prognostic significance of HPV integration could be due to variations in detection methods. While some methods such as genome-wide sequencing detect integration events from genomic DNA, others such as PCR and RNA sequencing determine integration events through RNA transcripts . These may affect the determination of HPV integration in tumor samples, as a small proportion of HPV-integrated tumors may lack active viral transcripts .
Various studies support that HPV integration in OPSCC is a nonrandom event, preferring genomic regions that harbor structural and copy number variants; this leads to dysregulated host gene expression near the integration site (35, 100, 101*, 102*). Many dysregulated genes, including PD-L1, SOX2, TP63, FGFR3 and MYC, are involved in processes related to cancer progression such as immune cell function, epithelial differentiation, and proliferation (35, 36, 100, 101*, 102*). It is possible that the integration site determines OPSCC progression and therapy response. For example, Walline et al. (2016) showed that therapy-responsive OPSCC had HPV integration mostly at intergenic chromosome regions; in contrast, recurrent OPSCC were more likely to harbor integration at cancer-associated genes (103*). However, findings from this study were limited by small sample size (20 OPSCC) (103*). In cervical cancer, patients with “productive” HPV integration (i.e., generation of actively transcribed viral-host fusion transcript) have higher E6/E7 expression, increased tumor aggressiveness, and immune evasion compared to those with “silent” integration (i.e., no viral-host fusion transcript) (104*). Similar studies on a large OPSCC cohort could determine the diverse consequences of HPV integration.
Genome-wide HPV integration drives pervasive instability that can affect OPSCC development and aggressiveness (35, 37, 100, 101*, 105*). Instability encourages host structural and copy number variations via genomic amplification, deletion, structural rearrangement, recombination, chromosomal translocations, and inversion (35, 37, 100, 101*, 105*). Using whole-genome sequencing in 105 OPSCC, Akagi et al. (2023) identified heterocateny as the predominant genomic structural alteration induced by HPV integration (102*). In heterocateny, numerous, diverse and repetitive virus and host DNA concatemers coexist within cancer cells leading to intratumoral heterogeneity. This promotes clonal evolution where tumor clones with genomic alterations promoting carcinogenesis are further amplified (102*).
Extrachromosomal DNA (ecDNA) is emerging as a critical driver of human cancers and could explain why some genes have high amplification rates (106*). Briefly, ecDNA is a circular DNA that exists outside chromosomes. Its circular structure and lack of centromeres allows ecDNA to be more accessible for transcription and promote unequal segregation into daughter cells, in turn driving oncogene transcription while enhancing intratumoral heterogeneity (107*). Fusion transcripts containing HPV E6/E7 genes and host oncogenes were found in ecDNAs in OPSCC cell lines and patient samples, likely due to integration that promotes a highly unstable genome and results in multiple excision and reintegration events (102*, 108*, 109*). Consequently, virus-host transcripts found within ecDNA have structural variations and higher expression compared to those in intrachromosomal regions (102*, 109*). Therefore, ecDNA is another possible link between HPV integration and OPSCC progression. However, the precise roles of HPV-induced ecDNA amplification in cancer is unconfirmed.
In summary, HPV integration likely promotes OPSCC progression through three driver events: (i) dysregulation of oncogenes or tumor suppressor genes important for tumorigenesis, (ii) genome-wide alterations in distinct cancer subclones to promote intratumoral heterogenenity, and (iii) ecDNA that promotes oncogene amplification and heterogeneity. Important questions remain: What are the biological and clinical implications of HPV integration-induced dysregulated gene expression and genome instability? When and how is HPV integration triggered? Can driver events that induce integration be counteracted or prevented? Development of HPV + OPSCC experimental models and accurate viral integration breakpoint prediction technologies will help evaluate the diagnostic, prognostic, and therapeutic significance of HPV integration in OPSCC. Furthermore, there is a need to determine whether tumors with HPV integration but minimal active viral transcript also share the same prognostic implications as those that harbor “silent” HPV integration .
Variations in HPV transcript expression and functions
E2 gene deletion and elevated E6/E7 transcript levels may portend inferior clinical outcomes (110*, 111*). Higher E6/E7 transcript and protein in serum of untreated patients also predicted higher risk of recurrence after treatment (112*, 113*). Other studies reported conflicting findings (114*, 115*). Comparing HPV gene expression in 84 patients with primary HPV16 + HNSCC, Zhang et al. (2016) observed no significant change in total E6/E7 between KRT and IMU subtypes . However, KRT with poorer clinical outcomes, has an increased ratio of alternatively spliced HPV16 E6 isoforms (E6*) relative to full-length expression . In a follow-up study, HPV + OPSCC with high E6* activity scores are larger, have higher HPV integration rates, and poorer overall survival (116*). Furthermore, E6* scores were positively correlated with mutational burden, suggesting roles in DNA damage and radiation response (116*). Therefore, E6* may be important for progression and therapeutic response of OPSCC. However, the functional roles of E6* in OPSCC are still poorly understood. This is despite multiple reports that the spliced variant, E6*I is more abundantly transcribed in OPSCC than full-length E6 (103*, 109*, 117*), while another study showed almost equal E6 and E6*I in TCGA HNSCC samples (118*) Perhaps, the best characterized function of E6* is that it does not inhibit p53, unlike its full-length isoform (119*, 120*). Some studies suggested that alternative splicing of E6 into E6*I promotes efficient translation of E7 proteins (121*, 122*), but conflicting findings have been reported (123*, 124*). A major caveat is that most studies that interrogate the link between E6* and p53 were performed on cell lines originating from cervical cancer or other non-HPV related cancers (119*), while only one study used an overexpression approach in a HPV-negative OPSCC cell line that harbors p53 mutations (125*). There is a dire need to interrogate and validate functional and mechanistic roles of E6* isoforms using appropriate experimental models.
Another distinguishing molecular feature is that KRT has lower E2/4/5 transcript levels than IMU . This is consistent with a TCGA study that demonstrated two subgroups of HPV16 + HNSCC; one with high integration rate and increased E6/E7 expression, and the other predominantly integration-negative with minimal E6/E7, but significant upregulation in E2/4/5 . A subsequent functional study suggested that E2/4/5 expression drives alternative carcinogenic mechanisms independent of E6/E7 (126*). Gleber-Netto et al. (2019) proposed two HPV16 + OPSCC subtypes that differed primarily by expression of 38 HPV16-correlated genes, and hence likely differed in HPV-modulated functions and biological pathways (114*). These two subtypes also vary significantly in survival, with HPV + C1 having poorer overall survival than HPV + C2. However, no significant differences in viral integration status or gene expression of HPV16 E2-E7 were observed between C1 and C2 (114*). In contrast, transcript expression of E1 and E1^E4 spliced isoforms was significantly higher in HPV + C2 than HPV + C1 (114*). In HPV16+ cell lines, elevated E1^E4 potentiated radiation sensitivity (114*). It is unclear if differential E1^E4 expression is due to HPV integration, as statistical association between integration and E1^E4 is weak (114*).
Puram et al. (2023), reported diverse intratumoral heterogeneity in HPV transcript expression in single-cell transcriptomic analysis of 12 HPV + OPSCC . In one patient, malignant epithelial cells identified by chromosomal aberrations were observed in the tumor margin originally interpreted as histologically negative; these cells expressed higher E5 transcript than those in the tumor bulk , suggesting that E5 drives tumor invasion. There was also a subset of malignant cells within HPV+ tumors that had undetectable HPV expression (defined as HPVoff cells), indicating that HPV transcripts were lost or reduced; such suppression may be driven by epigenetic regulation . Importantly, these HPVoff cells demonstrated phenotypes that are typically associated with HPV(-) OPSCC such as epithelial senescence, radiation resistance, and invasion . Together, these findings support previous molecular characterization in bulk HPV+ tumors, where a proportion of HPV + OPSCC demonstrated aggressive phenotypes, similar to HPV(−) tumors.
Overall, inconsistencies between association of HPV transcripts and biologic functions in OPSCC reveal our lack of understanding of the roles of HPV proteins in carcinogenesis. It is possible that HPV transcript expression and/or activity should be considered together with other molecular factors such as HPV integration to predict clinical outcomes and therapy responses. The finding of HPVoff cells in HPV + OPSCC also has potential clinical significance due to their reduced treatment responses and increased invasion in vitro. Future studies will need to understand how HPVoff cells emerge in HPV + OPSCC, and their functional implications.
HPV circulating tumor DNA
ctDNAs are short fragments of DNA released into systemic circulation during tumor cell apoptosis and/or necrosis. ctDNA detection represents a non-invasive approach to assess tumor burden dynamically (127*). HPV ctDNAs are detected in saliva and plasma in up to 95% of HPV + OPSCC using quantitative PCR, digital droplet PCR, and capture-based next-generation sequencing (127*, 128*). These assays commonly target high-risk HPV E6/E7 oncogenes due to abundance in cancer patients (127*); in contrast, ctDNAs are undetectable in individuals with HPV infection but no cancer (128*).
Varying levels of HPV ctDNAs are observed in primary and metastatic OPSCC before, during, and after treatment. These variations have important diagnostic and prognostic implications. Elevated pre-treatment ctDNA correlated with higher overall disease burden (129*, 130*), larger tumor size (131*), and increased nodal burden (132*, 133*, 134*, 135*). In 110 primary OPSCC, patients with detectable pre-treatment ctDNA had higher clinical nodal stage, larger lymph nodes, and were more likely to present lymphovascular invasion (135*). In contrast, in a multi-institutional study, patients with more aggressive tumors (higher tumor and nodal stages, increased HPV16 integration) had low or undetectable pre-treatment HPV16 ctDNA (136*). The reasons behind these conflicting findings are unclear but could be due to differences in how HPV positivity was determined and whether recurrent tumors were included in the analysis. Low pre-treatment ctDNA may also predict recurrence after primary CRT, although this remains to be validated (133*, 136*). Together, these preliminary studies suggest that varying levels of pre-treatment HPV ctDNAs are related to stages of OPSCC progression including tumor growth, nodal metastasis, lymphovascular invasion, and HPV genomic integration. Future studies could explore if pre-treatment HPV ctDNA is a marker of aggressive tumors to inform therapy selection.
During treatment, HPV ctDNA levels change drastically due to increased tumor cell death (133*). Preliminary studies suggested that ctDNA levels peak from weeks 1–3 but clear by weeks 4–7 of standard CRT in primary OPSCC (131*, 133*, 136*). Kinetics of rise of HPV ctDNA and clearance correlate with therapeutic response (133*, 136*). In 103 non-metastatic p16+ OPSCC, patients who had >95% clearance of HPV ctDNA at Week 4 of CRT were likely to remain disease-free after treatment (136*). In 34 advanced p16+ OPSCC, an increase in HPV ctDNA levels at week 2 of CRT was associated with lower risk of disease progression but levels and clearance at weeks 4 or 7 did not predict progression (133*). Similar results were observed in patients with recurrent and metastatic tumors, where longitudinal changes in HPV ctDNA correlated with treatment response (137*, 138*). Successful remission should be reflected by complete clearance of HPV ctDNA in both primary and metastatic disease (131*, 136*). However, HPV ctDNA persists in some patients even after treatment, indicating residual disease (139*, 140*); elevation post-therapy is predictive of recurrence (140*).
Collectively, there is wide inter-patient variability in HPV ctDNA levels and kinetics before, during, and post-treatment. It is unclear if these variations are reflected in HPV+ molecular subtypes, such as KRT and IMU. Nevertheless, ease and convenience to obtain and analyze HPV ctDNA makes it an attractive candidate for real-time disease monitoring. Moreover, changes likely occur even before anatomic or radiologic response (131*, 140*), enabling timelier intervention (i.e., adaptation of ongoing treatment and initiation of salvage therapy) to improve treatment outcomes.
Variations in other biologic processes
Based on microarray and RNA sequencing studies (Table 2), distinct subgroups of HPV + OPSCC frequently exhibit differential enrichment in keratinization, metabolism, and immune responses. As mentioned in section “Morphologic heterogeneity of HPV+ OPSCC”, HPV + OPSCC mostly presents a non-keratinizing morphology, although a small proportion demonstrates focal or diffuse keratinization. It is unclear whether keratinization in HPV + OPSCC truly portends worse survival. However, elevated expression of keratin 17 has been correlated with poor prognosis in OPSCC regardless of HPV status (141*). Moreover, several omics analyses proposed that elevated expression of keratinization genes predicts inferior survival (92, 142*). Whether keratin gene expression truly reflects histopathology and aggressiveness requires further investigation.
Using whole exome sequencing to compare the mutational landscape, Harbison and colleagues reported that primary HPV + OPSCC that recurred after treatment had mutations in genes involved in metabolic and oxidative stress responses compared to non-recurrent tumors (143*). Expression of an oxidative metabolic gene signature comprising 200 genes negatively correlated with overall survival in three separate multi-institutional HPV + OPSCC cohorts (144*). Based on the TCGA cohort and preliminary experiments on HPV + OPSCC cell lines, increasing levels of full-length E6, but not E6*, mitigated mitochondrial anti-oxidant capacity and promoted cisplatin sensitivity (144*). Conversely, previous studies reported an important role of HPV16 E6*, not full-length E6, in inducing oxidative stress and modulating mitochondrial function in cervical and oropharyngeal cancers (116*, 145*, 146*). Together, these studies support an essential role of oxidative stress and mitochondrial metabolism in driving HPV + OPSCC progression. Other processes such as glucose metabolism and hypoxia are also associated with prognosis (147*, 148*, 149*). Glucose- and hypoxia-based imaging markers predict survival and recurrence in HPV + OPSCC (148*, 150*), although controversy remains (151*, 152*).
Molecular profiling of HPV + OPSCC consistently demonstrated that subtypes with better prognosis were enriched in immune-related processes (92, 93, 142*, 153*, 154*). Tumors that do not harbor HPV integration have higher lymphocytic infiltration and stronger immune signatures characterized by enrichment in T (CD3 + , CD4 + , CD8 + ), B, NK, and CD34+ cells . In agreement, tumor-infiltrating lymphocytes (TILs) have been associated with improved prognosis in HPV + OPSCC (155*, 156*, 157*, 158*). T-cell immunity may promote survival in HPV + OPSCC by inducing a TH1 anti-tumor response (155*, 159*). CD163+ dendritic cells are key players that drive tumor-specific T cell infiltration and Th1 polarization (159*). Moreover, HPV proteins such as E2/E5/E6/E7 promote immunoevasion that contributes to malignant transformation and therapy resistance through various mechanisms; these include suppression of MHC class molecules, inhibition of stimulator of interferon genes (STING) activation, modulation of CD4+ and regulatory T-cell function, and regulation of the PD-1/PD-L1 immune checkpoint (160*). Therefore, targeting of HPV proteins such as E5, E6 and E7 is a potential therapeutic strategy to improve sensitivity to anti–PD-1/PD-L1 immunotherapy (161*).
While most studies focus on the role of T-cell immunity in HPV + OPSCC, HPV-specific B cells also provide prognostic information (162*, 163*, 164*, 165*, 166*). In a cohort of 72 OPSCC, higher infiltration of CD20 + B cells and B cell/CD8 + T cell interactions predicted superior prognosis, surpassing HPV status and T-cell infiltration (162*). In another study, increased abundance of peri-tumoral B cells in the lymph nodes correlates with better survival (167*). Additionally, cancer-associated fibroblasts with elastic differentiation negatively predict overall survival . Analysis of immune genes in the TCGA HNSCC cohort revealed that SYNGR3, a neuronal gene expressed in T and B cells, was overexpressed in HPV + HNSCC compared to HPV(-) HNSCC. This was validated in multiple cohorts. Enhanced SYNGR3 expression correlates with improved survival in HPV + HNSCC (168*). Together, the immune environment clearly modulates HPV + OPSCC progression; however, whether and how each immune component regulates HPV-driven tumor development in the oropharynx requires further investigation. In summary, heterogeneity in processes such as immune activation and hypoxia reflects the biology of HPV + OPSCC. Variations in biologic processes provide valuable clues to biomarkers that may refine selection for de-intensification clinical trials. In a Phase 2 trial, patients who responded well to de-intensified 30 Gy compared to the original 70 Gy had either no pre-treatment hypoxia or resolution of hypoxia during treatment (148*). Concurrently, Zeng et al. (2022) proposed a classification where immune-rich patients classified by a three-gene immune classifier had excellent survival outcomes and were more likely to remain disease-free even after receiving de-escalated RT (30 Gy) (153*). Other studies identified rapid alterations in immune cell composition and transcriptomes during early treatment stage (169*, 170*), suggesting that molecular markers can predict mid-treatment response to assist precise adaptive treatment selection. Immune biomarkers have the potential to predict responses to immunotherapies as well. Currently, there is evidence showing that high PD-L1 expression correlates with improved response to PD-L1/PD-1 inhibitor in recurrent and metastatic patients . In locally advanced patients, clinical benefit for PD-L1/PD-1 inhibitor is uncertain due to high variability in response among high-PD-L1 expressors .
As mentioned in section “Clinical Management of HPV-associated OPSCC”, HPV E6/E7 vaccines and T-cell therapy are emerging immunotherapeutics against HPV + OPSCC. Therefore, biomarkers that predict response to these therapies have not been discovered. Nevertheless, selecting HPV + OPSCC patients suitable for immunotherapy remains a top priority. Future studies should aim to uncover novel markers that can predict response to immunotherapies especially in the upfront definitive setting to aid selection for treatment de-escalation.
Genomic and epigenomic alterations
Various genomic and epigenomic alterations have been identified among HPV+ subtypes. E2F3 amplification is higher in classical than inflamed/mesenchymal subtypes, and coupled with overexpression of cell cycle genes including MCM2, MCM10, CDKN2A, E2F2, and RPA2 . Zhang et al. (2016), observed that KRT has increased frequency of PIK3CA activating mutations and copy number gains, as well as amplifications at chr3q . IMU harbored frequent copy number loss on chr16q, which was absent in HPV-KRT and HPV(-) samples . Lee et al. (2018) observed differential PIK3CA and NOTCH mutations between subtypes (171*). PIK3CA is the most commonly mutated gene; ~20–30% of HPV + OPSCC carry this mutation (11, 117*, 172*). Mutations of phosphoionositide 3-kinase (PI3K) components and downstream mediators have been associated with prognosis and treatment response in HPV + OPSCC (173*). Patients with mutant rather than wild-type PIK3CA are more likely to develop recurrence three years after completing de-intensified CRT (174*). Therefore, patients with wild-type PIK3CA would be better candidates for de-escalated therapy in primary HPV + OPSCC.
The TCGA head and neck cancer cohorts were classified into different subgroups based on DNA methylation profile . HPV + HNSCC samples were distributed across all subgroups, supporting heterogeneity in global DNA methylation . The subgroup that is hypomethylated is associated with higher non-synonymous mutations, reduced EMT signature, and superior overall survival . In another study on a TCGA HNSCC cohort, tumors with HPV integration had similar methylation patterns as HPV(-) HNSCC and normal tissues, whereas HNSCC with episomal HPV (integration-negative) had different methylation profiles from the other three groups. These findings substantiate the proposition that HPV-integrated HNSCC has distinct carcinogenic mechanisms from those with episomal HPV . It is also of interest to explore if methylation patterns differ based on HPV+ molecular subtypes. Using the University of Michigan HNSCC cohort, Liu and colleagues (175*) characterized 5-hydroxymethylation (5hmC) profiles in the KRT and IMU subtypes identified previously . Using 5hmC profiles only, less aggressive IMU could be easily distinguished from KRT, which had 5hmC profiles similar to those of HPV(-) HNSCC. Closer annotation of 5hmC profiles revealed hyper-5hmC and increased expression of cell migration genes in IMU. In contrast, KRT had hyper-5hmC enrichment related to keratinization and cell junctions, with accompanied elevation in gene expression.
The apolipoprotein B mRNA editing enzyme catalytic polypeptide-like (APOBEC) cytidine deaminases is a family of enzymes that convert cytosine to uracil in nucleotides. HPV + OPSCC is highly enriched with a APOBEC mutagenesis signature that strongly correlates with mutational burden (117*, 176*, 177*). Overexpression of APOBEC3A (A3A), A3B, A3F and A3H were detected in HPV-infected keratinocytes and OPSCC compared to their HPV(-) counterparts, due to dysregulation by E6 and E7 oncoproteins (178*, 179*, 180*, 181*). Mutations induced by APOBEC can occur simultaneously in viral and host genomes (182*); key somatic mutations include driver mutations in the PIK3CA helical domain and other immune genes (176*, 183*). In another study, A3A promoted HPV16 E2 hypermutation and associated with increased viral genome integration in OPSCC (184*). While these findings suggest that APOBEC mutagenesis signature portends poorer prognosis, another study contradicts by demonstrating that A3 enzymes improve survival and response to cisplatin in HPV + HNSCC (185*). Future studies will need to confirm the prognostic impact of APOBEC mutagenesis signature in HPV + OPSCC. It will also be interesting to investigate if distinct molecular subtypes of HPV + OPSCC express differential APOBEC-induced mutational pattern and load.
Overall, molecular characterization of HPV + OPSCC revealed diverse inter- and intra-tumoral levels; some studies proposed molecular subtypes but it is unclear how these subtypes affect treatment selection. This could be due to lack of prospective studies that associate treatment outcomes with molecular subtypes in HPV + OPSCC. A web-based tool that can differentiate the molecular subtypes of HPV + OPSCC will be useful for researchers to classify their samples and associate them with important clinical variables.
Challenges and future directions
Despite the suggestion of several biomarkers from patient and/or experimental studies, none is currently available to classify HPV + OPSCC into high and low risk groups for de-escalated therapy. One key reason could be lack of substantial validation, due to conflicting results, possibly from small population size or inconsistencies in selection criteria for clinical trials. Future studies that include several hundred patients (up to thousands) with corresponding data from molecular assays, imaging, and clinical outcomes could help to identify and validate key biomarkers with substantial clinical implications. Another major caveat is that many biomarkers are identified from bulk tumor populations whereas spatial and temporal changes related to tumor progression may yield more predictive results. Delineating the main factors that drive heterogeneity in HPV + OPSCC will help to identify candidates for de-intensified therapy and personalized treatment strategies.
To better understand the biology of HPV + OPSCC, in vitro and in vivo models that can recapitulate the heterogeneity in HPV + OPSCC are needed. However, HPV + HNSCC cell lines harbor multiple genetic artifacts, which are misrepresented in patient specimens. These artifacts include overrepresentation of EGFR amplification and underrepresentation of canonical PIK3CA-activating mutations and TRAF3 (186*). Furthermore, many HPV + HNSCC cell lines are poorly engrafted in immune-deficient mice (186*, 187*) and only a few HPV+ oropharyngeal cell lines exist (186*). Patient-derived xenografts or organoids could overcome these shortcomings to improve capture of heterogeneity and architecture (187*, 188*). However, success rates for engrafting HPV + OPSCC into xenograft or organoids are low (187*), and most do not include the tumor microenvironment, which is important for progression (189*). Novel ex vivo tumor organoids that incorporate tumor microenvironment characteristics are under development and require validation (190*). There is a pressing need for models that recapitulate genetic variation, tumor morphologies, and architectures within the tumor microenvironment in distinct HPV + OPSCC subgroups.
Recent studies revealed that race/ethnicity-specific differences may impact incidence, survival, and morphology of cancers (191*). In HNSCC, patients with European ancestry (White) significantly outnumber those of African ancestry (Black) (10, 85, 192*, 193*). White patients are more likely to present HPV + HNSCC (192*, 193*). In contrast, Black patients are usually diagnosed at a younger age and have poorer survival, even after accounting for HPV status, cancer stage, and healthcare access (192*, 194*). The reason behind this racial disparity is unclear, but may be due to differing environmental factors, diet, and genetic variations (192*, 194*). A multi-institutional study comprising 744 HPV + OPSCC suggested that Black patients have a higher density of multinucleated tumor cells than White patients, independent of AJCC-8 stage . This suggests that certain prognostic biomarkers for HPV + OPSCC should be population-specific. Other studies revealed variations in immune, mutational and transcriptomic profiles among HNSCC in White versus Black patients (192*, 195*, 196*). However, whether these findings translate to HPV + OPSCC remains unknown. Unraveling the characteristics of HPV + OPSCC in different racial and ethnic subgroups may provide insights into biologic factors underlying their prognostic variations.
Recognition of HPV + OPSCC with its favorable clinical outcomes sparked interest in de-escalating treatment to reduce side effects. However, inter- and intratumoral heterogeneity are a monumental challenge. Biomarkers are important for assigning patients to accurate risk subgroups. In this review, we discussed potential biomarkers based on clinical, histopathologic, and molecular variations (Fig. 3). It is likely that molecular features correspond to clinical and morphologic profiles but their links remain largely unexplored. There is a need to integrate clinical, histopathologic, and molecular variations to develop a robust and clinically actionable paradigm that groups HPV + OPSCC into subtypes. Multidisciplinary collaboration is needed to accurately integrate the plethora of markers and determine the most robust and reliable biomarker combination for routine diagnosis. A more detailed understanding of tumor development, evolution and resistance will provide clinically-relevant information to improve outcomes for all HPV + OPSCC; this can be achieved through appropriate experimental models that better represent heterogeneity in HPV + OPSCC.
Mahal BA, Catalano PJ, Haddad RI, Hanna GJ, Kass JI, Schoenfeld JD, et al. Incidence and demographic burden of HPV-associated oropharyngeal head and neck cancers in the United States. Cancer Epidemiol Biomark Prev. 2019;28:1660–7.
Lorenzoni V, Chaturvedi AK, Vignat J, Laversanne M, Bray F, Vaccarella S. The current burden of oropharyngeal cancer: a global assessment based on GLOBOCAN 2020. Cancer Epidemiol Biomark Prev. 2022;31:2054–62.
Chaturvedi AK, Engels EA, Pfeiffer RM, Hernandez BY, Xiao W, Kim E, et al. Human papillomavirus and rising oropharyngeal cancer incidence in the United States. J Clin Oncol. 2011;29:4294–301.
Faraji F, Rettig EM, Tsai H-L, El Asmar M, Fung N, Eisele DW, et al. The prevalence of human papillomavirus in oropharyngeal cancer is increasing regardless of sex or race, and the influence of sex and race on survival is modified by human papillomavirus tumor status. Cancer. 2019;125:761–9.
Jemal A, Simard EP, Dorell C, Noone A-M, Markowitz LE, Kohler B, et al. Annual report to the nation on the status of cancer, 1975–2009, featuring the burden and trends in human papillomavirus (HPV)–associated cancers and HPV vaccination coverage levels. JNCI J Natl Cancer Inst. 2013;105:175–201.
Lechner M, Liu J, Masterson L, Fenton TR. HPV-associated oropharyngeal cancer: epidemiology, molecular biology and clinical management. Nat Rev Clin Oncol. 2022;19:306–27.
Lydiatt W, O’Sullivan B, Patel S. Major changes in head and neck staging for 2018. Am Soc Clin Oncol Educ Book. 2018;38:505–14.
O’Sullivan B, Huang SH, Su J, Garden AS, Sturgis EM, Dahlstrom K, et al. Development and validation of a staging system for HPV-related oropharyngeal cancer by the International Collaboration on Oropharyngeal cancer Network for Staging (ICON-S): a multicentre cohort study. Lancet Oncol. 2016;17:440–51.
Fakhry C, Westra WH, Li S, Cmelak A, Ridge JA, Pinto H, et al. Improved survival of patients with human papillomavirus–positive head and neck squamous cell carcinoma in a prospective clinical trial. JNCI: J Natl Cancer Inst. 2008;100:261–9.
Ang KK, Harris J, Wheeler R, Weber R, Rosenthal DI, Nguyen-Tân PF, et al. Human papillomavirus and survival of patients with oropharyngeal cancer. N Engl J Med. 2010;363:24–35.
Lawrence MS, Sougnez C, Lichtenstein L, Cibulskis K, Lander E, Gabriel SB, et al. Comprehensive genomic characterization of head and neck squamous cell carcinomas. Nature. 2015;517:576–82.
Seiwert TY, Zuo Z, Keck MK, Khattri A, Pedamallu CS, Stricker T, et al. Integrative and comparative genomic analysis of HPV-positive and HPV-negative head and neck squamous cell carcinomas. Clin Cancer Res. 2015;21:632–41.
Sartor MA, Dolinoy DC, Jones TR, Colacino JA, Prince MEP, Carey TE, et al. Genome-wide methylation and expression differences in HPV(+) and HPV(-) squamous cell carcinoma cell lines are consistent with divergent mechanisms of carcinogenesis. Epigenetics. 2011;6:777–87.
Mantovani F, Banks L. The Human Papillomavirus E6 protein and its contribution to malignant progression. Oncogene. 2001;20:7874–87.
Giarrè M, Caldeira S, Malanchi I, Ciccolini F, Leão MJ, Tommasino M. Induction of pRb degradation by the human papillomavirus type 16 E7 protein is essential to efficiently overcome p16INK4a-Imposed G1 cell cycle arrest. J Virol. 2001;75:4705–12.
Fakhry C, Lacchetti C, Rooper LM, Jordan RC, Rischin D, Sturgis EM, et al. Human papillomavirus testing in head and neck carcinomas: ASCO clinical practice guideline endorsement of the College of American Pathologists Guideline. J Clin Oncol. 2018;36:3152–61.
Mehanna H, Taberna M, von Buchwald C, Tous S, Brooks J, Mena M, et al. Prognostic implications of p16 and HPV discordance in oropharyngeal cancer (HNCIG-EPIC-OPC): a multicentre, multinational, individual patient data analysis. Lancet Oncol. 2023;24:239–51.
Gillison ML, Trotti AM, Harris J, Eisbruch A, Harari PM, Adelstein DJ, et al. Radiotherapy plus cetuximab or cisplatin in human papillomavirus-positive oropharyngeal cancer (NRG Oncology RTOG 1016): a randomised, multicentre, non-inferiority trial. Lancet. 2019;393:40–50.
Mehanna H, Robinson M, Hartley A, Kong A, Foran B, Fulton-Lieuw T, et al. Radiotherapy plus cisplatin or cetuximab in low-risk human papillomavirus-positive oropharyngeal cancer (De-ESCALaTE HPV): an open-label randomised controlled phase 3 trial. Lancet. 2019;393:51–60.
Yom SS, Torres-Saavedra P, Caudell JJ, Waldron JN, Gillison ML, Xia P, et al. Reduced-dose radiation therapy for HPV-associated oropharyngeal carcinoma (NRG Oncology HN002). J Clin Oncol. 2021;39:956–65.
Fakhry C, Zhang Q, Nguyen-Tan PF, Rosenthal D, El-Naggar A, Garden AS, et al. Human papillomavirus and overall survival after progression of oropharyngeal squamous cell carcinoma. J Clin Oncol. 2014;32:3365–73.
Castellsagué X, Alemany L, Quer M, Halec G, Quirós B, Tous S, et al. HPV Involvement in head and neck cancers: comprehensive assessment of biomarkers in 3680 patients. JNCI. 2016;108:407–20.
Ndiaye C, Mena M, Alemany L, Arbyn M, Castellsagué X, Laporte L, et al. HPV DNA, E6/E7 mRNA, and p16INK4a detection in head and neck cancers: a systematic review and meta-analysis. Lancet Oncol. 2014;15:1319–31.
Wierzbicka M, San Giorgi MRM, Dikkers FG. Transmission and clearance of human papillomavirus infection in the oral cavity and its role in oropharyngeal carcinoma—a review. Rev Med Virol. 2022:e2337.
White EA. Manipulation of epithelial differentiation by HPV oncoproteins. Viruses. 2019;11:369.
Kreimer AR, Pierce Campbell CM, Lin H-Y, Fulp W, Papenfuss MR, Abrahamsen M, et al. Incidence and clearance of oral human papillomavirus infection in men: the HIM cohort study. Lancet. 2013;382:877–87.
D’Souza G, Clemens G, Strickler HD, Wiley DJ, Troy T, Struijk L, et al. Long-term persistence of oral HPV over 7 years of follow-up. JNCI Cancer Spectr. 2020;4:pkaa047.
Bettampadi D, Sirak BA, Abrahamsen ME, Reich RR, Villa LL, Ponce EL, et al. Factors associated with persistence and clearance of high-risk oral human papillomavirus (HPV) among participants in the HPV infection in men (HIM) study. Clin Infect Dis. 2020;73:e3227–e34.
Agalliu I, Gapstur S, Chen Z, Wang T, Anderson RL, Teras L, et al. Associations of oral α-, β-, and γ-human papillomavirus types with risk of incident head and neck cancer. JAMA Oncol. 2016;2:599–606.
Kreimer AR, Ferreiro-Iglesias A, Nygard M, Bender N, Schroeder L, Hildesheim A, et al. Timing of HPV16-E6 antibody seroconversion before OPSCC: findings from the HPVC3 consortium. Ann Oncol. 2019;30:1335–43.
Ferris RL, Westra W. Oropharyngeal carcinoma with a special focus on HPV-related squamous cell carcinoma. Annu Rev Pathol Mech Dis. 2023;18:515–35.
Mattox AK, Roelands J, Saal TM, Cheng Y, Rinchai D, Hendrickx W, et al. Myeloid cells are enriched in tonsillar crypts, providing insight into the viral tropism of human papillomavirus. Am J Pathol. 2021;191:1774–86.
Galliverti G, Wullschleger S, Tichet M, Murugan D, Zangger N, Horton W, et al. Myeloid cells orchestrate systemic immunosuppression, impairing the efficacy of immunotherapy against HPV+ cancers. Cancer Immunol Res. 2020;8:131–45.
Della Fera AN, Warburton A, Coursey TL, Khurana S, McBride AA. Persistent human papillomavirus infection. Viruses. 2021;13:321.
Parfenov M, Pedamallu CS, Gehlenborg N, Freeman SS, Danilova L, Bristow CA, et al. Characterization of HPV and host genome interactions in primary head and neck cancers. Proc Natl Acad Sci 2014;111:15544–9.
Koneva LA, Zhang Y, Virani S, Hall PB, McHugh JB, Chepeha DB, et al. HPV integration in HNSCC correlates with survival outcomes, immune response signatures, and candidate drivers. Mol Cancer Res. 2018;16:90–102.
Labarge B, Hennessy M, Zhang L, Goldrich D, Chartrand S, Purnell C, et al. Human papillomavirus integration strictly correlates with global genome instability in head and neck cancer. Mol Cancer Res. 2022;20:1420–8.
Balaji H, Demers I, Wuerdemann N, Schrijnder J, Kremer B, Klussmann JP, et al. Causes and consequences of HPV integration in head and neck squamous cell carcinomas: state of the art. Cancers 2021;13:4089.
Reuschenbach M, Huebbers CU, Prigge E-S, Bermejo JL, Kalteis MS, Preuss SF, et al. Methylation status of HPV16 E2-binding sites classifies subtypes of HPV-associated oropharyngeal cancers. Cancer 2015;121:1966–76.
Bratman SV, Bruce JP, O’Sullivan B, Pugh TJ, Xu W, Yip KW, et al. Human papillomavirus genotype association with survival in head and neck squamous cell carcinoma. JAMA Oncol. 2016;2:823–6.
Chatfield-Reed K, Gui S, O’Neill WQ, Teknos TN, Pan Q. HPV33+ HNSCC is associated with poor prognosis and has unique genomic and immunologic landscapes. Oral Oncol. 2020;100:104488.
Garset-Zamani M, Carlander AF, Jakobsen KK, Friborg J, Kiss K, Marvig RL, et al. Impact of specific high-risk human papillomavirus genotypes on survival in oropharyngeal cancer. Int J Cancer. 2022;150:1174–83.
Varier I, Keeley BR, Krupar R, Patsias A, Dong J, Gupta N, et al. Clinical characteristics and outcomes of oropharyngeal carcinoma related to high-risk non–human papillomavirus16 viral subtypes. Head Neck. 2016;38:1330–7.
Sher DJ, Adelstein DJ, Bajaj GK, Brizel DM, Cohen EEW, Halthore A, et al. Radiation therapy for oropharyngeal squamous cell carcinoma: executive summary of an ASTRO evidence-based clinical practice guideline. Pract Radiat Oncol. 2017;7:246–53.
Bourhis J, Sire C, Graff P, Grégoire V, Maingon P, Calais G, et al. Concomitant chemoradiotherapy versus acceleration of radiotherapy with or without concomitant chemotherapy in locally advanced head and neck carcinoma (GORTEC 99-02): an open-label phase 3 randomised trial. Lancet Oncol. 2012;13:145–53.
Huang SH, Perez-Ordonez B, Weinreb I, Hope A, Massey C, Waldron JN, et al. Natural course of distant metastases following radiotherapy or chemoradiotherapy in HPV-related oropharyngeal cancer. Oral Oncol. 2013;49:79–85.
Network NCC. NCCN guidelines for patients- oropharyngeal cancer. Head and neck cancers series. 2020. Available from: https://www.nccn.org/patients/guidelines/content/PDF/hn-oropharyngeal-patient.pdf.
Guo T, Qualliotine JR, Ha PK, Califano JA, Kim Y, Saunders JR, et al. Surgical salvage improves overall survival for patients with HPV-positive and HPV-negative recurrent locoregional and distant metastatic oropharyngeal cancer. Cancer. 2015;121:1977–84.
Guo T, Kang SY, Cohen EEW. Current perspectives on recurrent HPV-mediated oropharyngeal cancer. Front Oncol. 2022;12:966899.
Seiwert TY, Burtness B, Mehra R, Weiss J, Berger R, Eder JP, et al. Safety and clinical activity of pembrolizumab for treatment of recurrent or metastatic squamous cell carcinoma of the head and neck (KEYNOTE-012): an open-label, multicentre, phase 1b trial. Lancet Oncol. 2016;17:956–65.
Chow LQM, Haddad R, Gupta S, Mahipal A, Mehra R, Tahara M, et al. Antitumor activity of pembrolizumab in biomarker-unselected patients with recurrent and/or metastatic head and neck squamous cell carcinoma: results from the phase Ib KEYNOTE-012 expansion cohort. J Clin Oncol. 2016;34:3838–45.
Ferris RL, Blumenschein G, Fayette J, Guigay J, Colevas AD, Licitra L, et al. Nivolumab for recurrent squamous-cell carcinoma of the head and neck. N Engl J Med. 2016;375:1856–67.
Ferris RL, Blumenschein G, Fayette J, Guigay J, Colevas AD, Licitra L, et al. Nivolumab vs investigator’s choice in recurrent or metastatic squamous cell carcinoma of the head and neck: 2-year long-term survival update of CheckMate 141 with analyses by tumor PD-L1 expression. Oral Oncol. 2018;81:45–51.
Lee NY, Ferris RL, Psyrri A, Haddad RI, Tahara M, Bourhis J, et al. Avelumab plus standard-of-care chemoradiotherapy versus chemoradiotherapy alone in patients with locally advanced squamous cell carcinoma of the head and neck: a randomised, double-blind, placebo-controlled, multicentre, phase 3 trial. Lancet Oncol. 2021;22:450–62.
Mell LK, Torres-Saavedra P, Wong S, Chang S, Kish JA, Minn AJ, et al. Radiotherapy with durvalumab vs. cetuximab in patients with locoregionally advanced head and neck cancer and a contraindication to cisplatin: phase II results of NRG-HN004. Int J Radiat Oncol Biol Phys. 2022;114:1058.
Machiels JP, Tao Y, Burtness B, Tahara M, Rischin D, Alves GV, et al. LBA5 Primary results of the phase III KEYNOTE-412 study: Pembrolizumab (pembro) with chemoradiation therapy (CRT) vs placebo plus CRT for locally advanced (LA) head and neck squamous cell carcinoma (HNSCC). Ann Oncol. 2022;33:S1399.
Roof L, Yilmaz E. Immunotherapy in HPV-related oropharyngeal cancers. Curr Treat Options Oncol. 2023;24:170–83.
Aggarwal C, Cohen RB, Morrow MP, Kraynyak KA, Sylvester AJ, Knoblock DM, et al. Immunotherapy targeting HPV16/18 generates potent immune responses in HPV-associated head and neck cancer. Clin Cancer Res. 2019;25:110–24.
Aggarwal C, Saba NF, Algazi AP, Sukari A, Seiwert T, Haigentz M, et al. 916MO Safety and efficacy of MEDI0457 plus durvalumab in patients (pts) with human papillomavirus-associated recurrent/metastatic head and neck squamous cell carcinoma (HPV+ R/M HNSCC). Ann Oncol. 2020;31:S661–S2.
Raguz J, Schmidt S, Kleisner T, Zerbs M, Bekic G, Feher S, et al. Abstract 4058: TheraT—a highly versatile arenavirus based vector platform for intravenous and intratumoral cancer immunotherapy. Cancer Res. 2020;80:4058.
Nagarsheth NB, Norberg SM, Sinkoe AL, Adhikary S, Meyer TJ, Lack JB, et al. TCR-engineered T cells targeting E7 for patients with metastatic HPV-associated epithelial cancers. Nat Med. 2021;27:419–25.
Ma SJ, Yu H, Yu B, Waldman O, Khan M, Chatterjee U, et al. Association of pack-years of cigarette smoking with survival and tumor progression among patients treated with chemoradiation for head and neck cancer. JAMA Netw Open. 2022;5:e2245818-e.
O’Sullivan B, Huang SH, Siu LL, Waldron J, Zhao H, Perez-Ordonez B, et al. Deintensification candidate subgroups in human papillomavirus–related oropharyngeal cancer according to minimal risk of distant metastasis. J Clin Oncol. 2013;31:543–50.
Nichols AC, Theurer J, Prisman E, Read N, Berthelet E, Tran E, et al. Radiotherapy versus transoral robotic surgery and neck dissection for oropharyngeal squamous cell carcinoma (ORATOR): an open-label, phase 2, randomised trial. Lancet Oncol. 2019;20:1349–59.
Palma DA, Prisman E, Berthelet E, Tran E, Hamilton S, Wu J, et al. Assessment of toxic effects and survival in treatment deescalation with radiotherapy vs transoral surgery for HPV-associated oropharyngeal squamous cell carcinoma: the ORATOR2 Phase 2 randomized clinical trial. JAMA Oncol. 2022;8:845–51.
Brody RM, Shimunov D, Cohen RB, Lin A, Lukens JN, Hartner L, et al. A benchmark for oncologic outcomes and model for lethal recurrence risk after transoral robotic resection of HPV-related oropharyngeal cancers. Oral Oncol. 2022;127:105798.
Vijayvargiya P, Trivedi S, Rupji M, Song H, Liu Y, Jiang R, et al. Comparison of the Seventh and Eighth Edition of American Joint Committee on Cancer (AJCC) staging for selected and nonselected oropharyngeal squamous cell carcinomas. Oncologist. 2022;27:48–56.
Wuerdemann N, Wittekindt C, Sharma SJ, Prigge ES, Reuschenbach M, Gattenlöhner S, et al. Risk factors for overall survival outcome in surgically treated human papillomavirus-negative and positive patients with oropharyngeal cancer. Oncol Res Treat. 2017;40:320–7.
Bauer E, Mazul A, Chernock R, Rich J, Jackson RS, Paniello R, et al. Extranodal extension is a strong prognosticator in HPV-positive oropharyngeal squamous cell carcinoma. Laryngoscope. 2020;130:939–45.
An Y, Park HS, Kelly JR, Stahl JM, Yarbrough WG, Burtness BA, et al. The prognostic value of extranodal extension in human papillomavirus-associated oropharyngeal squamous cell carcinoma. Cancer. 2017;123:2762–72.
Zhan KY, Eskander A, Kang SY, Old MO, Ozer E, Agrawal AA, et al. Appraisal of the AJCC 8th edition pathologic staging modifications for HPV−positive oropharyngeal cancer, a study of the National Cancer Data Base. Oral Oncol. 2017;73:152–9.
Kharytaniuk N, Molony P, Boyle S, O’Leary G, Werner R, Heffron C, et al. Association of extracapsular spread with survival according to human papillomavirus status in oropharynx squamous cell carcinoma and carcinoma of unknown primary site. JAMA Otolaryngol–Head Neck Surg. 2016;142:683–90.
Albergotti WG, Schwarzbach HL, Abberbock S, Ferris RL, Johnson JT, Duvvuri U, et al. Defining the Prevalence and Prognostic Value of Perineural Invasion and Angiolymphatic Invasion in Human Papillomavirus–Positive Oropharyngeal Carcinoma. JAMA Otolaryngol Head Neck Surg. 2017;143:1236–43.
Schmitd LB, Scanlon CS, D’Silva NJ. Perineural invasion in head and neck cancer. J Dent Res. 2018;97:742–50.
Zhao N, Ang MK, Yin XY, Patel MR, Fritchie K, Thorne L, et al. Different cellular p16INK4a localisation may signal different survival outcomes in head and neck cancer. Br J Cancer. 2012;107:482–90.
Lai S, Wenaas AE, Sandulache VC, Hartman C, Chiao E, Kramer J, et al. Prognostic significance of p16 cellular localization in oropharyngeal squamous cell carcinoma. Ann Clin Lab Sci. 2016;46:132–9.
Lewis JS. Morphologic diversity in human papillomavirus-related oropharyngeal squamous cell carcinoma: catch me if you can! Mod Pathol. 2017;30:S44–S53.
Chernock RD, El-Mofty SK, Thorstad WL, Parvin CA, Lewis JS. HPV-related nonkeratinizing squamous cell carcinoma of the oropharynx: utility of microscopic features in predicting patient outcome. Head Neck Pathol. 2009;3:186–94.
Shinn JR, Davis SJ, Lang-Kuhs KA, Rohde S, Wang X, Liu P, et al. Oropharyngeal squamous cell carcinoma with discordant p16 and HPV mRNA results: incidence and characterization in a large, contemporary United States cohort. Am J Surg Pathol. 2021;45:951–61.
Gondim DD, Haynes W, Wang X, Chernock RD, El-Mofty SK, Lewis JS,Jr. Histologic typing in oropharyngeal squamous cell carcinoma: a 4-year prospective practice study with p16 and high-risk HPV mRNA testing correlation. Am J Surg Pathol. 2016;40:1117–24.
Lewis JS, Mirabello L, Liu P, Wang X, Dupont WD, Plummer WD, et al. Oropharyngeal squamous cell carcinoma morphology and subtypes by human papillomavirus type and by 16 lineages and sublineages. Head Neck Pathol. 2021;15:1089–98.
Lewis JSJ, Scantlebury JB, Luo J, Thorstad WL. Tumor cell anaplasia and multinucleation are predictors of disease recurrence in oropharyngeal squamous cell carcinoma, including among just the human papillomavirus-related cancers. Am J Surg Pathol. 2012;36:1036–46.
Cooper T, Biron VL, Adam B, Klimowicz AC, Puttagunta L, Seikaly H. Association of keratinization with 5-year disease-specific survival in oropharyngeal squamous cell carcinoma. JAMA Otolaryngol Head Neck Surg. 2015;141:250–6.
Koyuncu CF, Lu C, Bera K, Zhang Z, Xu J, Toro P, et al.Computerized tumor multinucleation index (MuNI) is prognostic in p16+ oropharyngeal carcinoma.J Clin Investig. 2021;131:e145488.
Koyuncu CF, Nag R, Lu C, Corredor G, Viswanathan VS, Sandulache VC, et al. Image analysis reveals differences in tumor multinucleations in Black and White patients with human papillomavirus-associated oropharyngeal squamous cell carcinoma. Cancer. 2022;128:3831–42.
Alexiev BA, Obeidin F, Johnson DN, Finkelman BS, Prince R, Somani SN, et al. Oropharyngeal carcinoma: a single institution study of 338 primaries with special reference to high-risk human papillomavirus-mediated carcinoma with aggressive behavior. Pathol Res Pract. 2020;216:153243.
Wilde DC, Castro PD, Bera K, Lai S, Madabhushi A, Corredor G, et al. Oropharyngeal cancer outcomes correlate with p16 status, multinucleation and immune infiltration. Mod Pathol. 2022;35:1045–54.
Molony P, Werner R, Martin C, Callanan D, Sheahan P, Heffron C, et al. Tumour cell anaplasia and multinucleation as prognosticators in oropharyngeal squamous cell carcinoma. Head Neck Pathol. 2020;14:606–15.
Thompson LDR, Burchette R, Iganej S, Bhattasali O. Oropharyngeal squamous cell carcinoma in 390 patients: analysis of clinical and histological criteria which significantly impact outcome. Head Neck Pathol. 2020;14:666–88.
Hue J, Valinciute Z, Thavaraj S, Veschini L. Multifactorial estimation of clinical outcome in HPV-associated oropharyngeal squamous cell carcinoma via automated image analysis of routine diagnostic H&E slides and neural network modelling. Oral Oncol. 2023;141:106399.
Keck MK, Zuo Z, Khattri A, Stricker TP, Brown CD, Imanguli M, et al. Integrative analysis of head and neck cancer identifies two biologically distinct HPV and three non-HPV subtypes. Clin Cancer Res. 2015;21:870–81.
Zhang Y, Koneva LA, Virani S, Arthur AE, Virani A, Hall PB, et al. Subtypes of HPV-positive head and neck cancers are associated with HPV characteristics, copy number alterations, PIK3CA mutation, and pathway signatures. Clin Cancer Res. 2016;22:4735–45.
Locati LD, Serafini MS, Iannò MF, Carenzo A, Orlandi E, Resteghini C, et al. Mining of self-organizing map gene-expression portraits reveals prognostic stratification of HPV-positive head and neck squamous cell carcinoma. Cancers. 2019;11:1057.
Qin T, Li S, Henry LE, Liu S, Sartor MA. Molecular tumor subtypes of HPV-positive head and neck cancers: biological characteristics and implications for clinical outcomes. Cancers. 2021;13:2721.
Puram SV, Mints M, Pal A, Qi Z, Reeb A, Gelev K, et al. Cellular states are coupled to genomic and viral heterogeneity in HPV-related oropharyngeal carcinoma. Nat Genet. 2023;55:640–50.
Kürten CHL, Kulkarni A, Cillo AR, Santos PM, Roble AK, Onkar S, et al. Investigating immune and non-immune cell interactions in head and neck tumors by single-cell RNA sequencing. Nat Commun. 2021;12:7338.
Nulton TJ, Kim N-K, DiNardo LJ, Morgan IM, Windle B. Patients with integrated HPV16 in head and neck cancer show poor survival. Oral Oncol. 2018;80:52–5.
Pinatti LM, Sinha HN, Brummel CV, Goudsmit CM, Geddes TJ, Wilson GD, et al. Association of human papillomavirus integration with better patient outcomes in oropharyngeal squamous cell carcinoma. Head Neck. 2021;43:544–57.
Lim MY, Dahlstrom KR, Sturgis EM, Li G. Human papillomavirus integration pattern and demographic, clinical, and survival characteristics of patients with oropharyngeal squamous cell carcinoma. Head Neck. 2016;38:1139–44.
Symer DE, Akagi K, Geiger HM, Song Y, Li G, Emde A-K, et al. Diverse tumorigenic consequences of human papillomavirus integration in primary oropharyngeal cancers. Genome Res. 2022;32:55–70.
This work is supported by grants from NIH/NIDCR DE027551 (NJD) and NIH/NCI F058486 (NJD/MAS).
The authors declare no competing interests.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Lim, Y.X., Mierzwa, M.L., Sartor, M.A. et al. Clinical, morphologic and molecular heterogeneity of HPV-associated oropharyngeal cancer. Oncogene 42, 2939–2955 (2023). https://doi.org/10.1038/s41388-023-02819-y