Management of elderly patients with head and neck cancer Free

Abstract

There are no established guidelines for managing older patients with head and neck cancer. Most clinical trials that define current standard therapy included few elderly patients. On the other hand, there is great variability in patients’ comorbidities, physical functions, cognitive function, familial and financial background and values. The key point appears to be appropriate geriatric assessment, clarifying the patients’ outcomes and a multidisciplinary team approach, including the treatment decision-making policy. Although these processes should be scientific in nature, the evidence for the treatment of elderly head and neck patients is very limited.

This review summarizes the evidence available regarding the management of geriatric assessment, each treatment modality and the multidisciplinary team approach for older patients with head and neck cancers.

Introduction

There is no definition of the age at which old people should be treated as elderly or older people. The National Comprehensive Cancer Network (NCCN) Older Adult Guidelines note that this definition is derived from the Medicare classification and should be defined by functional status rather than chronologic age (1). On the other hand, in many previous works, old was defined as 65 or 70 years or above: for example, the European Medicines Agency, American Society of Clinical Oncology (ASCO) guidelines for geriatric oncology and the Japan Clinical Oncology Group (JCOG) policy for geriatric research used 65 years, whereas the European Organization for Research and Treatment of Cancer (EORTC) older patient task force took 70 years as the cut-off for old age (2–5). Regarding older patients with head and neck cancer (HNC), the elderly patient is defined as being 70 years or older in a recent review (6). It has been estimated that 24% of newly diagnosed HNCs are in patients older than 70 years according to the surveillance, epidemiology and end result (SEER) database in the USA (7,8). In Japan, ~47% of newly diagnosed HNC patients are older than 70 years according to the first visit case registry of Japanese Society of Head and Neck Cancer (9). Despite increasing numbers of elderly patients, most previous clinical trials of standard treatment for HNC included few elderly patients; we still know little about whether older people should receive standard treatment (10).

In Europe and North America, HNC that occurs in older patients often presents with a different profile than in younger patients: a high proportion of women, no history of smoking or drinking, primary lesion mostly of the oral cavity and a few with lymph node metastases (11–13). On the other hand, the situation is slightly different in Japan, the same proportion of sex, but a slightly higher proportion of oral and laryngeal cancers in patients 70 years old and over [Tables 1 and 2; (9)]. In both cases, we should pay attention to the potential increased risk of treatment toxicity, shorter life expectancy and patients’ values and goals, especially in elderly patients (11,14).

With limited evidence, the decision of how to treat HNC in elderly persons is difficult and complicated. One of the reasons is that the treatment of HNC causes considerable changes not only in appearance, but also in swallowing, speech, hearing, smell and taste (15). In particular, airway and swallowing impairment was common after laryngeal and oropharyngeal cancer treatment in older patients, which resulted in worse survival (16,17). In addition, the decision should include the patient’s preference, because older patients are sometimes interested in participating in the decision-making process. Priority in treatment selection differs depending on the patient’s values and beliefs. Furthermore, the mind and body of elderly persons are sometimes vulnerable, and post-treatment outcomes can be difficult to predict.

The importance of GA in predicting survival and treatment outcomes in many fields of cancer has been reported (18,19). Despite the fact that the NCCN Older Adult Guidelines, the ASCO guideline and International Society of Geriatric Oncology (SIOG) statements recommend GA or a screening tool for every older patient with cancer, it has rarely been performed in both actual practice and clinical trials for older patients with HNC (1,3,20,21). Especially in Japan, the treatment policy for HNC in elderly persons was often selected based on the criteria of the institution or the attending physician, in which the actual age was the most important criterion (22).

There are unique issues to consider when caring for an older adult cancer patient, because the biological characteristics of certain cancers and their responsiveness may differ in older patients compared with their younger counterparts (1). The NCCN guidelines for HNC recommend that all HNC patients need access to the full range of support services and specialists with expertise in their management, which can also be applied to elderly patients with HNC (23).

Unfortunately, there have been limited reports about the selection of older HNC patients for either standard treatment or reduced intensity treatment. Because most of them are retrospective studies, the conclusions have weak evidence. For this review, previous reports that focused on the management of elderly patients with HNC for each treatment modality and phase were reviewed.

Definitive radiotherapy with and without systemic therapy

Concomitant chemoradiotherapy (CCRT) with cisplatin is regarded as the world standard treatment for patients with advanced HNC (24). This treatment is indicated for not only those with unresectable disease, but also for patients who do not wish to undergo surgery even if their disease is resectable. The tolerability of this treatment has been confirmed in the usual Japanese population as well (25,26). This treatment was chosen based on a meta-analysis by Pignon et al. in which CCRT improved overall survival (OS) of patients with unresectable HNC compared to radiotherapy alone. However, no survival-prolonging effect was observed in the subgroup 71 years or older (24). A similar study by Machtay et al. who analyzed three Radiation Therapy Oncology Group (RTOG) chemoradiation trials, showed that the occurrence of severe toxicity was significantly higher in the older age group (27). In addition, although p16-positive head and neck squamous cell carcinoma (SqCC) is considered to be highly sensitive to chemotherapy and radiation therapy, it is not recommended at this time to change the treatment policy based on human papilloma virus (HPV) status (28), but this also applies to elderly people. In the subgroup analysis of the RTOG1601 trial, the hazard ratio between CCRT and radiotherapy (RT) plus cetuximab showed no significant difference also in patients 65 years old and over (29). Haehl et al. examined the additional effect of chemotherapy on OS compared to radiation therapy, with a threshold of 75 years; an additional effect was seen in patients 65–74 years old, but not in those 75 years old and over (30).

Radiation monotherapy is one of the potential curative treatments for elderly persons in whom drug treatment should be restricted. Radiation causes mucositis, pain, xerostomia and dysgeusia in the acute phase, whereas caries, trismus and osteoradionecrosis occur in the long-term phase. Intensity-modulated radiation therapy (IMRT) has significantly reduced these risks (31). Radical RT using IMRT or other highly conformal techniques to reduce acute and late toxicities is appropriate in elderly patients without severe comorbidities (32). Sparing the parotid glands with IMRT significantly reduces the incidence of xerostomia, which is involved in swallowing function, and sparing the carotid arteries reduces late cerebrovascular events (33,34).

Adjuvant therapy

In the adjuvant therapy setting following HNC resection, the RTOG 9501 trial and the EORTC 22931 trial showed the survival benefit of postoperative chemoradiotherapy over radiotherapy alone (35–37). Because only 6% of the patients were 70 years or over in the prior trial, while the latter excluded them, there had been skepticism about the additional concomitant chemotherapy in elderly patients. In a sub-analysis of our JCOG 1008 trial, which showed that a weekly cisplatin regimen was non-inferior to a tri-weekly cisplatin regimen as postoperative adjuvant chemoradiotherapy for high-risk locally advanced HNC, there was no significant difference in survival between patients 65 years and over and others (38). Woody et al. performed a SEER database analysis of high-risk HNC patients over 70 years old after surgical resection, in which chemoradiotherapy was still associated with an improvement in OS after their inverse propensity score-matched analysis (39). This result suggests that there is a group of patients who may benefit from chemoradiotherapy as postoperative adjuvant treatment, even in elderly HNC patients over 70 years of age. For p16-positive oropharyngeal cancer, which Skillington et al. analyzed in their cohort study, the addition of chemotherapy to the adjuvant radiotherapy for surgically managed p16-positive oropharyngeal SqCC may be unnecessary regardless of age (40).

Surgery

Previous studies of surgical treatment for elderly HNC patients are limited and sometimes controversial. There have been several reports that elderly persons have higher risks in surgery for HNC. Examining free-flap reconstruction surgery, Spyropoulou et al. reported a higher mortality in HNC patients older than 70 years (41), and Bhama et al. reported that there was a significantly higher rate of death within 30 days and longer stay in the intensive care unit in patients older than 80 years (42). On the other hand, Goldstein et al. reported frailty as a predictor of perioperative complications in patients undergoing HNC surgery. Although many retrospective studies showed that age itself was not predictive of complications or mortality (43–45), the age threshold was set at 65 or 70 years (46). Peters et al. examined perioperative complications by classifying the patients into 10-year age groups in their retrospective study. A higher rate of preexisting comorbidities in patients aged 80 years or older, advanced tumor stage and prolonged surgery time were associated with surgical complications (47). They also examined predictors of complications in patients treated with free-flap reconstruction surgery, which identified comorbidity as the only significant predictor of complications, not chronological age (48). Similarly, Nakayama et al. reported in their retrospective study of free-flap reconstructive surgery for advanced HNC in elderly patients that poor geriatric-8 (G8) score and smoking habit were correlated with perioperative complications, whereas chronological age 80 years and over was not (49). A review of outcomes of free-flap reconstructive surgery in elderly persons, including studies published between 2000 and 2014, showed no differences in terms of free-flap success, surgical complications or mortality rate between older and younger patients (50). Thus, the following points are gradually becoming clearer: elderly patients with low comorbidity scores may be offered free-flap reconstruction with less reservation, and careful preoperative assessment and postoperative monitoring are mandatory (45,50).

The choice of treatment for advanced laryngeal cancer in the elderly is particularly controversial, because there are large individual differences in tolerance and beliefs related to aphonia when undergoing laryngectomy, and the expected functional decline in swallowing, vocalization and breathing when undergoing larynx-preserving therapy. Woodard et al. reported that long-term functional and quality of life (QoL) outcomes after total laryngectomy ranged from intermediate to high categories, especially in those older than 65 years at diagnosis, presence of no >2 comorbidities and no history of previous chemoradiation therapy (51). Examining QoL of patients following total laryngectomy vs chemoradiation, the overall QoL scores of both groups seemed similar, except for more sensory disturbances in the surgery group, with more dry mouth in the chemoradiation group (52). This indicates that total laryngectomy may be appropriate also for elderly patients who cannot tolerate chemoradiation, because more functional decline is expected in elderly persons when undergoing larynx-preserving therapy. There were some reports of total laryngectomy or total pharyngolaryngectomy in elderly persons, which resulted in good clinical outcomes, despite a lower level of recovery of speech function (53,54). Achim et al. showed that the predictors for wound complications in total laryngectomy, including pharyngeal cutaneous fistula, were sarcopenia and malnutrition, rather than chronological age (55). In recent years, there have been some clinical trials of transoral and endoscopic surgery combined with risk-adapted neoadjuvant and/or adjuvant therapy for locally advanced HNC, which aimed for better disease control and less functional decline (56–59). These new trials may also benefit older patients who meet certain conditions.

Systemic therapy for recurrence or metastasis

In systemic therapy in palliative settings, Vermorken et al. reported in their EXTREME trial the survival benefit of cetuximab added to a combination regimen of platinum and 5-fluorouracil, and that the benefit was not significant in 65 years or older, recurrent or metastatic head and neck cancer (RM HNC) patients, who made up only 18% of eligible patients (60). In contrast, Agriris et al. showed that palliative chemotherapy for RM HNC elderly patients (>70 years) had comparable survival benefit and response rate with those for younger patients (61). However, toxicity rates and chemotherapy-related deaths were significantly higher in elderly patients. In recent years, immune checkpoint inhibitors have been increasingly used for the systemic therapy of RM HNC. In the CheckMate 141 trial, nivolumab showed survival and QoL benefits compared with other chemotherapeutic agents also in RM HNC patients over 65 years of age (62,63). In the post-hoc analysis of this trial, OS and tumor response benefits with nivolumab versus single-agent chemotherapy were maintained regardless of age. In addition, the nivolumab arm had a lower rate of treatment-related adverse events regardless of age (64). The KEYNOTE-048 trial, which demonstrated that pembrolizumab (plus platinum and 5-fluorouracil) is an appropriate first-line treatment for recurrent or metastatic HNC, the OS benefit with this regimen versus cetuximab chemotherapy was maintained in elderly patients 65 years and over (65). Presley et al. showed in their review of immunotherapy in older adults with cancer that immune checkpoint inhibitor (ICI) efficacy and toxicity are comparable between younger adults and fit older adults included in clinical trials. Toxicity management in older adults must consider comorbidities and ideally include primary care teams and caregivers (66).

Palliative radiotherapy

With radiotherapy, one may be able to plan treatment of appropriate intensity for more older or oldest patients with HNC, especially in palliative settings. Bonomo et al. performed an observational, prospective study of hypofractionated radiation therapy for advanced HNC patients deemed unsuitable for CCRT or high-dose radiotherapy. It resulted in clinical benefit with low toxicity in frail, elderly patients (67). An ideal palliative radiation regimen should alleviate symptoms secondary to the cancer with minimal treatment toxicity and side effects while improving a patient’s QoL (68) There were wide variations in dose fractionation regimens between studies. Fraction sizes ranged from 1.8 to 8 Gy, and total doses ranged from 12 Gy to 54 Gy. The most common dose fraction regimen used was 30 Gy in 10 fractions, and this dose fraction has been commonly used also in Japan (69). Another popular approach in the palliative treatment of HNCs in North America is the ‘QUAD shot’ approach. This approach was first introduced in the palliation of pelvic malignancies in the RTOG 8502 trial (70). It consists of 14–14.8 Gy over 2 days in twice-daily fractions, and the patients who tolerate the cycle without tumor progression can receive two additional cycles (71). Corry et al. demonstrated that 53% of patients completed all 3 cycles, and 50% of patients had an objective response. Median OS was 5.7 months with confirmed improvement of QoL, swallowing difficulties and pain (71). This QUAD shot approach has been delivered more often in Japan recently. Ferro et al. investigated short-course accelerated radiotherapy for patients 80 years or older with advanced HNC. Of 17 patients who received a total dose of 20 Gy (2 cycles if possible), with no Grade 3 toxicity, 3-month symptom-free survival was over 80%, and the palliative response rate was 88%. Repeated short-course accelerated radiotherapy in a palliative setting of HNC is safe and well tolerated in older adults (72).

Geriatric assessments

As mentioned above, some treatment options for HNC can be expected to perform as well as in young adults, even in older age. However, it should be noted that the elderly patients selected in these clinical trials were an elite group with less comorbidities, more family support and a stable mental state. In other words, we can divide the elderly into two groups: the ‘fit’ group who can receive the same treatment as young adults, and the ‘unfit’ group who cannot. Aging is a process that consists of a gradual loss of physiological reserve that leads to impaired functions (73). Traditional tools to assess functional status, such as the Eastern Cooperative Group-Performance Status or Karnofsky Performance Scale, have often been shown to be inadequate for representing the functional status of this population (74,75). Though Comprehensive Geriatric Assessment (CGA) has been the gold standard for sorting out vulnerable and unfit patients, it consumes much time and resources (76–80). Therefore, several screening tools have been developed, such as the G8 [Table 3; (81)], Vulnerable Elders Survey-13 (82) and the Flemish version of the Triage Risk Screening Tool (83). According to the SIOG statements, GA provides an overview of the general health status of older individuals and determines the functional and physiological age, which is more important than chronological age (21,83). In addition, GA helps oncologists to understand the overall health status and problem, to predict life expectancy and tolerance of treatments and to influence treatment choices and interventions that can improve treatment compliance (18,79,83–85). There have been many studies of GA or GA screening tools for older patients with various cancers to clarify these aims (21,86), but there have been few studies of HNC patients (87,88). Neve et al. reported that the G8 tool identified twice the number of patients as vulnerable compared with the multidisciplinary team [MDT; (89)]. We also reported that the G8 tool predicted life expectancy and complications of treatment in older patients with HNC in a prospective, observational study (22). Brugel et al. are now expecting to show, in their randomized, controlled trial, the ‘EGeSOR study’, that CGA may have direct clinical benefits in the follow-up management of elderly patients with HNC (90). Geriatrician interventions are also important to make decisions regarding treatment policy and the need for geriatric interventions (60). Because of the heterogeneity of the aging process, chronological age sometimes differs from biological or functional age (74). Recently, the ELderly heAd and Neck cancer-Oncology eValuation (ELAN-ONCOVAL) study showed that geriatric assessment performed by a geriatrician identified a significantly higher number of patients who require multidisciplinary interventions (91). However, since there are not many geriatric oncologists at present, cooperation with them is an issue for the future (92). These trials may lead to modifications of the management model for elderly patients with HNC using GA.

MDT approach

An MDT approach was developed first in the middle 2000s to improve the quality of cancer care by preventing and diminishing treatment side effects (93–95). Today, MDTs for HNCs have been successfully implemented in many countries and are considered standard of care management for HNC patients (96). MDTs for HNC consist of head and neck surgeons, radiation oncologists and medical oncologists, at minimum. In addition, the team should consist of dentists, dietitians, SLPs, specialized nurses, social workers, psycho-oncologists and onco-geriatricians, if possible.

The dentist is necessary for HNC patients at each stage of the treatment process, because potential complications can occur in the oral cavity during and after each treatment: surgery, radiotherapy and chemotherapy (97,98). Postoperative complications such as oral and pharyngeal wound infection and aspiration pneumonia occur with poor oral hygiene. During and after (chemo)radiation therapy, the patients sometimes develop xerostomia, dysphagia, aspiration pneumonia and osteoradionecrosis (31,99). Any dental treatment should be preventive and continuous before, during and after the treatment (100).

Specialized dietitians are crucial, because early nutritional assessment and intervention can contribute to minimizing acute toxicity and treatment interruptions, and improving survival (95,101,102). Over 40% of HNC patients are already malnourished at diagnosis, which leads to definitive deterioration after treatment (103). In elderly HNC patients, sarcopenia has been correlated with an increased rate of surgical complications and decreased OS in those treated with (chemo)radiotherapy (55,104,105).

The effectiveness of swallowing exercises for HNC patients who received (chemo)radiotherapy has been reported repeatedly (106–108). However, Hansen et al. examined the characteristics of multidisciplinary care in the treatment of HNC in elderly patients using SEER-Medicare linked data, and they found rare consultations with an SLP before definitive therapy. Since the situation in Japan is the same, the number of studies of swallowing function, patient-reported outcomes and swallowing exercises has been very limited in Japan (109–111).

Psychological interventions are also important, because HNCs and the treatments are more emotionally traumatic and cause more anxiety than any other type of cancer in terms of physical impairment or change in appearance (60,112–114). It has been reported that HNC patients given a diagnosis of depression were more likely to die of cancer (115). In addition, an educational intervention based on self-management skills favored QoL and reduced anxiety and depression (115,116). Elderly cancer patients have a high prevalence of anxiety or depression; therefore, it is important for elderly HNC patients and their families to be offered the opportunity to meet others in a similar situation and not to isolate them (117).

The important points of the approach are shared decision-making and early comprehensive supportive services (60,118). It has currently become accepted that the MDT approach works in the care of various cancer patients (119). The same should be true for older patients with HNC. Therefore, more evidence is needed to link MDT-guided care and outcomes.

Conclusion

There is little evidence for the treatment and management of elderly patients with HNC, because past randomized, controlled trials excluded most of this population. The treatment decision requires a more accurate biological or functional age rather than the chronological age. CGA is a key tool for deriving it, and we must accumulate evidence to better predict prognosis and treatment adverse events in older patients with HNC. In addition, because survival may not be the most important outcome in elderly patients, the MDT approach, which aims to maintain QoL that can be reduced by treatment, as well as the cancer itself, will become even more important in the future.

Acknowledgments

The authors would like to express their appreciation to the Japan Clinical Oncology Group (JCOG) Head and Neck Cancer Study Group.

Funding

This work received no support.

Conflict of interest statement

The authors have no conflict of interest to declare.

References