Vlad C. Sandulache, MD, PhD
A 65-year-old African American man presented to an Otolaryngology Head and Neck Surgery clinic at a tertiary Veterans Health Administration (VHA) facility for evaluation. The patient recalled a past diagnosis of oropharyngeal cancer (OPC), possibly associated with the human papillomavirus (HPV). After receiving the diagnosis at another VHA facility, the patient opted to seek care at a local, non-VHA facility and received approximately 7 weeks of daily radiation and weekly infusions of chemotherapy.
Six years after his initial diagnosis and treatment, the patient said he had a persistent cough with any meaningful attempts to eat or drink. He also noted he lost at least 10 lbs in the last 3 months and had been hospitalized twice during the past winter. During his second hospitalization he spent 4 days on a ventilator in the intensive care unit.
On examination, the patient appeared frail and cachectic, with significant fibrosis of the neck skin and moderate trismus. His dentition was in poor health, and an in-clinic flexible endoscopy demonstrated clear silent aspiration of oral secretions. Given his failure to thrive, the patient was urgently admitted to the hospital. A modified barium swallow study performed by the head and neck Speech Pathology team demonstrated gross aspiration with all consistencies. After extensive counseling, the patient agreed to the placement of a gastrostomy tube. He was discharged in stable condition with adequate supplies and self-care training. He was advised to continue follow-up in the Head and Neck Cancer Survivorship clinic.
Two years later, in the early phase of the COVID-19 pandemic, the patient was admitted to the hospital with COVID pneumonia. Given the damage to his lungs over the previous decade from recurrent episodes of aspiration pneumonia, the patient succumbed.
An Unexpected, Unrelenting Epidemic
Shifting population dynamics and behaviors have led to an explosion in the incidence of cancers associated with infection by oncogenic subtypes of HPV, among which cancer of the oropharynx represents the most common malignancy.1,2 OPC now afflicts more than 30,000 new patients in the United States each year.3 Given current vaccination rates against oncogenic HPV, the overall trend of increasing incidence is not expected to stabilize until the 2040s.3 Traditional cancers of the head and neck region were previously fatal after 5 years in more than 60% of cases; however, today patients with HPV-associated OPC can expect a more than 80% chance of being alive 5 years after treatment.4-7 Combining the increasing incidence of OPC with a high chance of oncologic cure has led to an ever-expanding cohort of OPC survivors.
Enthusiasm about a high rate of survival after an HPV-associated OPC diagnosis is now partially dampened by an increasing realization that neither oncologists nor healthcare systems are remotely prepared for this rapidly expanding cohort of OPC survivors. Their unique needs and problems have yet to be objectively defined and quantified.
Relationship Between Survival and Long-Term Toxicity in HPV-Associated OPC
Survivorship care after OPC treatment is a growing challenge in terms of the number of patients affected, the negative impact on quality of life (QOL), and the potential burden on the healthcare system. The rapidly growing number of OPC survivors who are living long enough to develop delayed adverse effects related to their past OPC treatment1,2,8 includes many patients in whom toxicities can be truly debilitating,9,10 generating significant unmet needs.
Tumor and Treatment Toxicity
Although HPV-associated OPC demonstrates an excellent response to conventional chemoradiotherapy (CRT), this finding cannot be interpreted to mean that reducing treatment intensity is safe for patients with this disease. Prospective trials have now demonstrated that neither replacing or eliminating conventional chemotherapy, nor significantly reducing radiation doses, can be considered safe at this time.11-15 As a result, a patient with newly diagnosed HPV-associated OPC in 2025, and potentially even 2030, is likely to receive the same treatment as patients who were treated in the late 2010s.14
Three decades ago, the chronic effects of tumor and treatment were largely limited to a small cohort of survivors; however, today they affect more patients.1,2,7 Chronic xerostomia, dysphagia, trismus, radiation fibrosis, and osteoradionecrosis (ORN) now confront tens of thousands of OPC survivors; over the coming decades, these treatment effects have the potential to affect millions of patients.16-22
While most acute toxicities resolve within several months of completing CRT, late CRT sequelae tend to be dynamic and can progress silently over many years.16,23 Adverse effects vary widely, with many toxicities (eg, dysphagia, ORN) being particularly debilitating. Many of these effects occur in a radiation dose–dependent fashion, but radiation dose does not fully predict late toxicities, pointing to a role for other, yet unidentified contributing factors.24,25
Dysphagia in Survivors of OPC
About two-thirds of survivors of head and neck cancer (HNC) who seek follow-up care 5 years after treatment report dysphagia and at least partial dependence on a feeding tube.26 The incidence of dysphagia increases proportionately with higher radiation doses delivered to the pharyngeal constrictors and supraglottic larynx.18 Dysphagia can severely reduce QOL years after treatment, necessitating substantial changes in diet and social behavior among OPC survivors. Often, patients are forced to choose between chronic malnutrition or starvation and feeding tube dependence.27 Loss of a normal oral diet is frequently one of the most affected QOL measures for OPC survivors.28
In addition to effects on QOL, dysphagia can have life-threatening consequences. In a recent systematic review and meta-analysis, life-threatening aspiration occurred after > 24 months at a reported incidence ranging from 3% to nearly 35%. Although a reduction in radiation dose to the pharyngeal constrictors can reduce chronic dysphagia,27 whether this can be done safely in most OPC patients, particularly those with bulky primary tumors, remains unclear.
Osteoradionecrosis (ORN) in Survivors of OPC
ORN is one of the most potentially serious complications of CRT and may not manifest for years after treatment. Its median time of onset after radiotherapy is 8 years in patients with OPC.24 Bone injury and impaired healing of the alveolar mucosa are signs of ORN, which occurs in ~7% of patients receiving intensity-modulated radiation therapy for OPC.17 ORN is accompanied by pain, difficulties with chewing, exacerbation of concomitant dysphagia and, in the advanced stage—gross cosmetic deformity secondary to mandibular or maxillary fracture and/or decay.29 Despite the severity of this complication, we are just beginning to understand why ORN develops in a subset of patients. Although ORN is generally more common in patients with advanced-stage OPC who receive higher doses of radiation to a larger overall bone volume,17,19,24,30 comprehensive translational research efforts focused on ORN (as well as other late toxicities of OPC treatment) are still in their infancy.
Unmet Needs in Predicting and Evaluating Late Toxicities
Predicting which patients will experience long-term treatment toxicities or which types of late toxicities they may develop is not yet possible. Whereas increased data collection and prognostic models can help inform healthcare systems as to the expected frequencies of toxicity, they are unlikely to be prognostic at the individual patient level. As such, there is a critical need for individualized biomarker strategies that can predict one’s risk of toxicity and identify normal tissue shifts in biology and function early in the process to initiate interventions before significant deterioration. Adding to the complexity of predicting late toxicities is the lack of standardization in instruments used to categorize them. Examples of tools that may be used to categorize dysphagia include the Common Terminology Criteria for Adverse Events v4.0 grading scale, the Radiation Therapy Oncology Group grading system, and the European Organization for Research and Treatment of Cancer Performance Status Scale for Head and Neck Cancer.20 The MD Anderson Symptom Inventory for head and neck cancer may also be used to catalog dysphagia and other common symptoms of HNC, as well as treatment-related concerns.31 Magnetic resonance imaging-based techniques coupled with machine learning approaches represent emerging tools that may have a role in identifying early radiation-induced bone changes that can facilitate early detection of ORN.32,33 Although conventional and newer tools can be used to generate objective metrics of treatment-related toxicity, consistent and appropriate deployment across the entire cohort of OPC survivors in the United States remains a distant goal.
Calibrating Treatment Intensity to Disease Intensity
Given the risk of severe and potentially life-threatening consequences of radiation-based treatment, there is a large unmet need to better calibrate treatment intensity to the intensity of HPV-associated OPC.14,34 In light of the good prognosis of the disease in most patients, recent efforts have focused on identifying ways to de-escalate treatment intensity while preserving the good outcomes known to be possible for patients with HPV-associated OPC. Improving tolerability and limiting the risk of late effects of radiation-based treatment is especially important with the aging population of HPV-associated OPC survivors, who would also be expected to have unrelated comorbidities.1
Various modes of de-escalation have been studied, including adding surgery to CRT, reducing radiation dose, and modifying systemic therapy regimens. Most of these efforts have largely failed to identify a safe regimen for treatment de-escalation that applies to a majority or even a significant plurality of patients with OPC.14,35,36 Although CheckMate 141 and KEYNOTE-048 garnered excitement when immune checkpoint inhibitors (ICIs) significantly prolonged overall survival and had a more favorable safety profile than standard systemic therapy in recurrent and metastatic OPC,11,37,38 adding definitive frontline avelumab to CRT failed to prolong progression-free survival versus CRT alone in the phase 3 JAVELIN Head and Neck 100 trial.13 Combined with additional recent trial data, these findings make it unlikely that an ICI-based regimen will provide previously unavailable de-escalation options for patients with OPC in the near future.
Considering continued de-escalation efforts, it is important to remember that survival is not uniform among all patients with HPV-associated OPC. For example, patients with HPV-associated OPC and a history of current or prior heavy tobacco use have not experienced the same dramatic prolongation in overall survival as their nonsmoking counterparts.36 Patients with recurrent disease also face a dismal prognosis, with failure rates of about 70% with salvage treatment with surgery, re-irradiation, or systemic therapy.38-41 Therefore, de-escalation may not be appropriate in all patients, but identifying which patients are at risk of overtreatment is not straightforward. Better risk stratification of patients may provide part of the solution but will require rigorous testing and long-term follow-up to establish.
Discussion
There is an urgent need to carefully consider how to manage long-term survivors of HPV-associated OPC. With ever-increasing numbers of patients who are living years beyond their OPC treatment, continual reevaluation of treatment strategies in certain subsets of patients and making concerted efforts to identify and manage late toxicities early is paramount. Yet there remains a critical gap in knowledge due to insufficient metrics for both toxicity intensity and the frequency of debilitating, life-threatening toxicity. Unfortunately, the lack of tools available combined with the mismatch in disease intensity with treatment intensity likely results in excessive treatment-induced toxicity for many patients.
In the absence of clear evidence about which treatment strategy to use for individual patients, clinicians are tasked with making therapeutic choices without being fully able to predict outcomes. Patient preference is important to consider, but these conversations can be complicated. How does one talk to a patient about their willingness to risk a cancer recurrence and potentially risk late toxicities when the clinician does not know whether that individual patient will develop late toxicities, or know how severe they will be? It is a tradeoff between QOL (ie, possible feeding tube dependence) and survival—yet the magnitude of the effect on QOL remains impossible to predict at present for the individual patient.
Moreover, the needs of individual OPC survivors vary. A cross-sectional study performed at Princess Margaret Cancer Centre found that 61% of the 158 participants had unmet needs related to their cancer survivorship.42 Meeting the needs of survivors may require the development of better screening instruments that can manage various complications early and effectively. Continuing to follow OPC survivors with a multidisciplinary team would most certainly be beneficial and has been reported to improve QOL.43 Continual Speech Pathology management and therapy from the time of diagnosis into the survivorship phase of care has been suggested as one way to improve functional outcomes.44 Given that coordinating long-term care teams is logistically challenging, well-planned research is warranted to equip these teams to provide OPC survivors with the care they need. These efforts will be particularly important considering the large number of survivors who will need this type of care in the coming decades. The time to start is now well past.