Counseling the Patient with Potentially HPV-Related Newly Diagnosed Head and Neck Cancer

Abstract

The recent emergence of a clinically distinct subset of head and neck cancers (HNC) caused by infection with the human papillomavirus (HPV) necessitates critical reevaluation of the existing counseling paradigm for patients with newly diagnosed HNC. Herein we propose a structural framework for patient counseling in which HPV testing is incorporated and the impact of HPV-status is discussed in the context of multiple medical and psychosocial domains. We strive to maintain a balance between making recommendations based on the best available scientific evidence and acknowledgment of uncertainty for both patients and providers. We anticipate that both the standard-of-care diagnostic workup and treatment, and counseling guidelines for these patients will change rapidly in the years ahead, as data from ongoing and planned prospective clinical trials become available.

Introduction

This article is intended to serve as a guide for clinicians treating patients with newly diagnosed head and neck cancer (HNC) potentially related to the human papillomavirus (HPV). This patient population includes patients with tumors of the oropharynx, primarily involving the tonsil and base of tongue, or tumors of unknown primary site with spread to the cervical lymph nodes. In light of a large and ever-expanding body of literature suggesting that HPV-mediated Head and Neck Cancers (HPV-HNC) have a distinct etiopathogenesis and clinical behavior when compared to “classical” tobacco/alcohol associated HNC, viral testing is an emerging component of the diagnosis and staging of tumors arising in certain anatomic sub-locations within the head and neck. Although current staging [1] and treatment guidelines do not yet distinguish between viral and non-viral HNC, this will change as guidelines are updated to reflect the existing scientific evidence, and as the results of clinical trials in this patient population become available. Absent formal recommendations from advisory bodies to guide the integration of HPV-testing into clinical practice, we propose a counseling strategy that is patient-centered and evidence-based, and that also acknowledges uncertainty where it exists, particularly when uncertainty may have implications affecting prognosis and therapeutic decision making. As part of the patient counseling process, it is important for providers involved in the care of patients with potentially HPV-mediated HNC to familiarize themselves with the aspects of this disease that differentiate it from non HPV-mediated HNC.

Emergence of HPV-HNC as a Distinct Clinical Entity

Squamous cell carcinoma of the head and neck (HNSCC) arises from the mucosal surfaces of lip, oral and nasal cavities, as well as the nasopharynx, oropharynx, and larynx. As a group, squamous cell carcinomas are the fifth most common cancer subtype worldwide [2, 3]; however, incidence rates vary considerably by region, with foci of high incidence in Southeast Asia, and Eastern Europe [4]. Eight-five to ninety percent of head and neck cancers are squamous cell carcinomas [5], and classically HNSCC was associated with exposure to alcohol and tobacco products [6], risk factors that have been casually associated with the disease [7, 8]. In addition, exposure to both agents in combination may act synergistically to increase risk significantly over that associated with either agent in isolation [9]. In the past, tobacco and alcohol accounted for upwards of 75–90 % [10] of HNSCC cases. The remaining 10 % of cases lacking attributable exposure were believed to be due to other minor risk factors, including diet [11] and/or micronutrient deficiencies [12], as well as poor oral hygiene [13] or a lack of consistent access to dental care [14]. It has also been shown that family history of head and neck cancer contributes to disease risk [15], suggesting the existence of genetic risk-modifying factors. In addition, individuals with heritable cancer risk or cancer-susceptibility syndromes, such as Li-Fraumeni syndrome, Fanconi’s anemia and Bloom-Machacek syndrome, are known to be at increased risk [16, 17].

The incidence of head and neck squamous cell carcinoma within the United States, particularly of the oral cavity, hypopharynx and larynx, rose steadily before reaching a plateau in the late twentiethcentury— an observation attributed to changes in the prevalence of alcohol consumption [18] and tobacco-use behaviors [19]. Thereafter, a significant decline in the disease incidence at these subsites was observed [20, 21]. However, during the same 20-year period, incidence of oropharyngeal cancer— particularly within the pharyngeal tonsil and base-of-tongue—increased dramatically both in the United States [22] and Europe [23], in dramatic contrast to the decreased or stable incidence of HNSCC at other oral and oropharyngeal subsites [24].

Demographics of this emergent population also differed substantially from those of patients with “classical” HNSCC, with less frequent history of alcohol and/or tobacco exposure [25], and patients were commonly white men of upper-mid social strata [26–28], presenting with lesions of the oropharynx an average of 3–5 years earlier [29] than the average “classical” head and neck cancer patient. An association between the human papillomaviruses and head and neck cancer was proposed based on unique epidemiologic and pathological features common to the patient cohort.

Critical to this association was the observation that within this new cohort of head and neck cancer patients, there was an increase in the prevalence of behavioral risk factors associated with transmission of sexually transmitted infections, including HPV [30]. This included an earlier age of sexual debut [31], increased numbers of total lifetime sexual partners, as well as a increase in the prevalence of sexual behaviors in which there was oro-genital contact [32]. Shortly thereafter, it was demonstrated that a history of oral HPV infection frequently preceeded a later diagnosis of head and neck squamous cell carcinomas [33, 34]. Data from a number of large population-based studies show that 20– 25 % of head and neck cancer patients show some evidence of HPV exposure [35]; as a group, these patients are significantly more likely to have been exposed to HPV than appropriately matched controls [36, 37]. Even more striking was the association between HPV infection and cancers of the oropharynx [25, 38], in which evidence of exposure is even more prevalent than sites elsewhere within the head and neck [9]. Evidence of HPV exposure is found in over 50–75 % of patients among this subset of of head and neck cancers [39].

Virology and Molecular Biology of Oncogenic HPV

HPV are a family of epitheliotrophic DNA viruses, of which there are greater than 120 known genotypes. Over 40 genotypes are are known to possess a trophism for mucosal surfaces [37], particularly within the human anogenital tract where they are nearly ubiquitous amongst sexually active adults [40]. The virus is known to be transmitted via direct contact with infected cutaneous or mucosal surfaces [41]. HPV is widely appreciated to be a sexually transmitted virus, with a large body of literature describing genital-genital, digital-genital, and oro-genital transmission. Cross-sectional data suggests that high-risk HPV genotypes are present in the saliva of 1–3 % [42] of adults, and yet precisely how HPV is introduced into the upper aerodigestive tract is a subject of some debate [4]; however, oro-genital [27] and even oro-oral transmission have been described [43]. It is also unclear what features of the oropharynx predispose it towards persistent viral infection resulting in transformation [37], and which features distinguish it from other sites within the head and neck in which HPV-mediated disease is rare. It is also widely appreciated that a subset of HPV are carcinogenic in humans, secondary to their causal association with a number of anogenital cancers in humans, including invasive cervical cancer, for which HPV is believed to be necessary for malignant transformation [38, 44].

Clinically relvant HPV genotypes are frequently subdivided into two groups based upon their capacity to induce transformation of of normal squamous epithelial cells [45]. Those that are most commonly associated with benign papillomatous disease form the low-risk subgroup. Conversely, those that are associated with human malignancies and are capable of inducing transformation in vitro are classified as high-risk—the group being typified by genotypes 16, 18, 31, 33, and 35 [46]. Genotypes 16 and 18 in particular are known to be capable of inducing transformation in both the genital, and upper aerodigestive tract [47]. In tumors bearing viral DNA, HPV16 is the most prevalent genotype [48], comprising upwards of 85–97 % of cases [26]. Genotype 18 is found both independently and in combination with HPV16 in approximately 2–5 % of cases—the other remaining high-risk genotypes are found in a minor fraction of cases [37].

HPV-HNC appear to have a distinct molecular phenotype, as defined by gene expression profiles [49], patterns of large-scale chromosomal abnormalities [50], copy number variation [51], and the absolute number of somatic abnormalities identified in a given tumor [52, 53]. More generally, HPV-positivity is inversely associated with molecular markers of poor prognosis [54, 55]. Establishing a mechanism whereby viral infection might concievably lead to cancer in humans was necessary to demonstrate causality, and in the setting of HPV-associated cancers, the viral early proteins E6 and E7 are believed to contribute both to tumor initiation and maintenance of the malignant phenotype in HPV-HNC [56–60].

While it is recognized that expression of viral E6/E7 are necessary criteria for transformation, it is unknown whether the molecular derrangements induced by the virus are sufficent to cause disease. Both proteins are frequently detected within the oropharynx of affected individuals [27] and within the body of HPV-associated tumors [41]. Considered “viral oncoproteins,” E6 and E7 are believed to inactivate p53 [61, 62] and the retinoblastoma (pRb) gene product [63], respectively, both of which then become targets for degradation [64, 65]. Loss of these critical regulatory proteins leads to dysfunction within multiple cellular pathways, including cell-cyle regulatory control, DNA damage repair, apoptosis [66], and reactivation of host telomerase [67].

HPV Testing in HNC: Rationale for Testing and Patient Counseling

As stated previously, published treatment guidelines do not yet distinguish between viral and non-viral HNC, and therefore do not consider HPV-testing a regular component of diagnosis, staging or treatment planning. However, it is the opinion of the authors that HPV-testing is sufficiently useful to justify routine use in high-probability cases, because of the significant impact of HPV status on patient prognosis. If one also considers the importance of stratifying patients by HPV-status when designing clinical trials, HPV testing is essential for patients to have the option of enrolling in existing or future trials. Thus, we consider HPV testing for potentially virally related HNSCC an emerging standard-of-care that is recommended for all patients presenting with HNSCC at a potentially HPV-related subsite (tonsil, tongue base, and—where feasible—lymph nodes from HNSCC of unknown primary). Depending on the institution, this usually involves identification of viral DNA by polymerase chain reaction (PCR) or in situ hybridization (ISH). While immunohistochemical staining for the cellular p16 protein is strongly upregulated in the majority of HPV-related oropharyngeal cancer (OPC), and can be used as an additional validation of HPV status, it is best used as an adjunct to, and not a replacement for, direct HPV testing (Table 1).

Table 1. Summary of available HPV testing modalities, their specific attributes and drawbacks for use in patient counseling.

HPV testing methods

HPV DNA in situ hybridization [106]	The current gold standard, along with PCR analysis. Requires a high degree of technical expertise.
HPV DNA PCR [107, 108]	Modern DNA PCR technique that is extremely sensitive, but can be affected by contamination. Can readily be used to distinguish among oncogenic HPV types. Along with in situ hybridization, a gold standard technique.
HPV E6/E7 mRNA rt-PCR [109]	Detects biologic activity/ongoing viral oncogene production required for maintenance of malignant phenotype. Currently a research technique.
Immunohistochemistry [106, 110]	Host p16(INK4a) expression highly correlated with HPV-positivity. Therefore, frequently used as surrogate for direct HPV-testing, particularly in research studies.

Despite the current lack of guidelines regarding pre-test and post-test counseling, once testing is undertaken, the potential emotional impact of the information provided—both in terms of impact on patient prognosis and identification of sexually transmitted etiology—suggests that counseling should be provided. While little literature addresses the need for counseling directly in the subset of patients with HPV-HNC, we may consider relevant examples from similarly affected populations (reviewed comprehensively in [68]). In patients with HPV-unrelated HNC, disease-related anxiety and emotional distress are both common and known to negatively affect quality of life [69, 70]. However, it appears that interventions designed to mitigate disease-associated distress are underutilized or ineffective, as patients frequently report worsening anxiety during the course of their treatment [71]. Consider also that women with HPV-associated cervical dysplasia or invasive carcinoma may suffer from feelings of confusion, fear, and guilt in response to their diagnosis [72, 73]. It has also been demonstrated that feelings that contribute to self-blame and isolation may negatively affect information-seeking behaviors during physician–patient interactions, and may impair patient adherence to treatment recommendations [74]. Finally, a questionnaire-based study of patients with HPV-HNC treated at a major cancer center revealed significant gaps in patient understanding of HPV’s contribution to their cancer, transmission of the virus, and potential precautions [75]. Over one-third of patients in this study specifically desired more information about HPV from caregivers. Thus, counseling should address the major areas of patient concern: HPV biology, HPV transmission and transmissibility, impact on prognosis and treatment planning and post-diagnosis harm reduction. It is important to acknowledge that our current knowledge base is unable to answer all of the questions that a patient may have, and to reiterate that this is a field of active and ongoing research. Counseling strategies must provide ample opportunities for the patient to identify areas of concern and to ask questions of the provider.

Specific Aspects of Patient Counseling: Biology, Transmission and Transmissibility

Before HPV-testing is performed, the provider should offer a brief discussion of the potential mechanisms of carcinogenesis (viral vs. environmental), emphasizing that while HPV-related cancer has a generally better prognosis, both types of cancer are treatable. It is also important that before HPV testing is performed, patients have a clear understanding of what information will be provided and its potential implications, in order to decide who they are comfortable sharing this information with. If a patient’s test results are consistent with HPV-HNC, the provider should engage in a frank discussion of the most common modes of transmission and the natural history of the virus. While the information about biology of HPV presented above is likely more detailed than necessary for most patients, many patients may be quite unfamiliar with the mechanisms by which viruses can cause cancer, or even what a virus is. The clinician should assess both the patient’s initial knowledge base and desire for detailed information to determine in what depth to explain the topic. It is also important for the clinician to understand the patient’s starting assumptions about HPV and HPV-related cancer, since these topics are often sensationalized or misrepresented by the lay media.

With regard to viral transmission, the patient should be made aware that the virus is very common [42], and is known to be transmitted via contact with infected body areas [41]. This includes vaginal, anal and oral sexual behaviors; oro-oral contact, which is taken to include deep “French-style” kissing has also been implicated [43]. Most infections are asymptomatic, and therefore an individual in the community may not know that he/she is infected, and therefore capable of transmitting the virus to others. Epidemiologic evidence suggests that there is a significant “latent period,” between infection and the onset of HPV-associated cancers, which makes it impossible to conclusively define when, where, and from whom any individual acquired a high-risk HPV. The vast majority of HPV infections are transient, and most healthy individuals will effectively clear the virus. However, there may be significant person-to-person variability in the time required complete viral clearance.

Despite the fact that HPV-HNC is causally related to an acquired infectious agent, it is important to communicate clearly that the patient’s cancer is not transmissible by any known mechanism. The patient should be made aware that his or her partner is potentially at increased risk of subsequently developing HPV-HNC or other HPV-related cancers; however, the exact magnitude and even the certainty of this risk is unknown. There is currently no evidence to suggest behavioral modification to decrease the risk of transmission, and there are no recommendations or regulations requiring partner notification in documented cases. However, in some instances, the patient or his/her partner may inquire about means to further decrease the risk of viral transmission, including the utility of existing vaccines. Since most partners of patients with head and neck cancer have already been exposed to HPV in the past, there is currently no recommendation for use of prophylactic HPV vaccines in partners who do not meet conventional CDC guidelines [76–78]. However, while not supported by specific clinical studies, in the absence of other preventative measures, the partners of patients with HPV-HNC may find it reassuring to undergo periodic surveillance exams of the oral cavity and oropharynx, and should be encouraged to follow any age-appropriate and gender-appropriate established screening guidelines (such as for cervical cancer) where applicable.

Specific Aspects of Patient Counseling: Prognosis and Treatment Planning

Historically, the prognosis of patients with newly diagnosed head and neck cancer has been dependent upon the following criteria: age at the time of diagnosis [25]; response to treatment (for those patients treated with nonsurgical therapy); and for patients treated with surgery, radiologic and surgical-pathologic critera used to asses disease stage at the time of treatment and the completeness of surgical excision [44], including: tumor size, extent of nodal metastases within the lymphadenectomy specimen, and extent of locoregional and neurovascular invasion. However, it has become increasingly clear that HPV-positivity is also an independent prognostic factor associated with significantly improved prognosis in patients with newly diagnosed HPV-HNSC [79, 80]—an observation that is relevant regardless of the anatomic location of the tumor within the oropharynx [81], histologic grade of the primary lesion [35], or clinical stage at the time of diagnosis [82, 83]. Specifically, it appears that HPV-postivity is associated with significantly reduced risk of recurrence [84] and improved disease-free and progression-free survival [26, 35, 39]. HPV-positivity is also associated with a significant reduction in disease-specific mortality [25, 39], and all-cause mortality [26]. All-told, HPV-positivity—as determined by direct measures, or inferred via p16-status—is independently associated with a 40–50 % reduction in the risk of death from all causes when compared with HPV-unrelated OPC [85–88]. Improved prognosis within this cohort of patients appears to be attributable to enhanced treatment responsiveness, although decreased risk of second primary cancers may also play a role [89]. An expanding body of literature suggests that HPV-HNC, relative to HPV-unrelated disease, is more responsive to primary surgical therapy [90], as well as several adjuvant radiotherapy [10, 88] and/or chemoradiotherapy protocols [91, 92]. A number of hypotheses have been suggested as to why HPV-HNC appears to respond favorably to treatment, but at this time none have been confirmed [29, 93].

Due to the anatomic complexity of the head and neck region and the aggressive nature of HNSCC, multi-modality therapy is often required. Combination therapies, such as chemoradiation, are associated with risk of serious morbidity, including short-term and long-term complications due to sacrifice or dysfunction of structures critical for normal speech, deglution, and maintainence of the upper airway. As the association between HPV and head and neck cancers has only recently gained wide acceptance, there are as yet few prospective therapeutic clinical trials that have stratified patients based upon HPV-status. In the absence of these data, treatment guidelines do not yet distinguish between HPV-mediated and unrelated disease. However, this does not mean that the diagnosis of HPV-HNC has no bearing on treatment decisions, which are made by consensus after discussion between the patient and caregivers. While providers should not advocate for treatment de-intensification on an ad hoc basis in the absence of clinical trials supporting this approach, a patient who is knowledgeable about the uniquely good treatment response and survival associated with HPV-HNC will bring this understanding to the assessment of risks and benefits involved in choosing between different treatment options [94]. This highlights the importance of ensuring HPV-HNC patients have a thorough and realistic understanding of the unique epidemiologic and clinical characteristics of their disease. Conversely, patients with HPV-unrelated HNSCC should be reassured that their cancer is treatable, and be given a realistic assessment of their prognosis.

Specific Aspects of Patient Counseling: Harm Reduction

It is clear that behavioral risk factors play an important role in the development of head and neck cancer. Perhaps most important amongst the known risk factors, especially with regards to HPV-HNC, is tobacco use. It is clear that historical and ongoing tobacco use affects clinical outcomes in patients with HPV-HNC. A clear dose-response relationship exists between the intensity and duration of tobacco exposure and individual risk of developing HNC. Additionally, tobacco-use behaviors at the time of diagnosis, or tobacco use that continues during treatment in patients with HPV-HNC, are independently associated with worse prognosis [^{95, 96}], particulary in response to adjuvant radiotherapy [⁹⁷]. Counseling strategies should therefore incorpoate an assessment of tobacco exposure at the time of diagnosis, and ongoing environmental factors contributing to maintainence of this behavior [⁹⁸]. It is well established that brief, intensive, counseling sessions delivered by a physician can be effective in reducing tobacco use [^{99, 100}]. They should explain the risk associated with continued tobacco use, with regards to disease-specific and all-cause mortality [¹⁰¹]. Relapse events are common, and repeat approaches by the provider are often required. The patient may also require additional support, and the provider must be prepared to discuss existing behavioral, pharmacologic, and combinatorial treatment regimens, or make appropriate referrals [^{102, 103}].

Patients should be counseled that post-therapy follow-up care is an essential part of their treatment, and that all patients are at risk of recurrence or second primary cancers, despite favorable HPV status. Additional interventions, while not necessarily supported by clinical evidence, may promote patient wellbeing and a sense of control. Promotion of a sound diet, patient-appropriate exercise regimen, and stress reduction is a common-sense recommendation for all cancer patients. Hard evidence regarding participation in HNC-specific patient support groups is limited by the retrospective nature of existing studies and small sample sizes [^{104, 105}]; however, anecdotal experience suggests that many patients find them helpful.

Conclusion

By 2020, the yearly incidence of HPV-HNC is expected to pass that of invasive cervical cancer; concurrently, HPV-HNC is expected to become the most common form of HNC [²²]. Head and neck cancer care providers eagerly await the results of high-quality, prospective, randomized clinical trials that take into account the unique aspects of HPV-HNC which distinguish it from classical tobacco/alcohol-induced head and neck cancer. Absent this data, however, current treatment guidelines do not distinguish between patients with HPV-mediated and unrelated disease, and therapeutic de-escalation should not be performed on an ad hoc basis. Therefore, providers who care for HPV-HNC patients should develop a concise, evidence-based, and patient-centered counseling strategy in order to insure that required information is effectively communicated to patients in a way that allows them to effectively navigate a complex and evolving treatment paradigm.