Palliative radiotherapy in the head and neck

Introduction

Head and neck cancers represent the sixth most common cancer worldwide, with approximately 630,000 new patients diagnosed annually. While many advances within the field of head and neck cancer have focused on the improvement of curative-intent treatment, there remains a lack of universal guidelines for the palliative treatment of head and neck cancers. Disease and health-related factors associated with the delivery of palliative-intent treatment include the presence of distant metastatic disease at the time of initial presentation, very advanced locoregional disease, prior extensive surgery or radiotherapy (RT) to the head and neck, patient co-morbidities and poor performance-status precluding tolerance of curative-intent therapy, and patient choice. In such patients, definitive, curative-intent treatment with cancer-directed therapy is thought to be associated with significant treatment related toxicity, including the burden of daily treatments over many weeks. Such toxicity negatively impacts the health and quality of life (QoL) of these patients and outweighs the benefit of potentially curative therapy.

Supportive care measures used in head and neck cancer patients focus on managing the disease-associated symptoms, including pain, dysphagia, dyspnea, bleeding, and ulceration. Given the critical importance of the tissues of the head and neck to basic life function, such symptoms significantly impact QoL. Pain management with medications, anticholinergic blockade of secretions, tracheostomy for airway protection, and feeding tube placement for nutritional supplementation represent some of the common forms of care provided. For more symptomatic patients, palliative surgery or embolization of bleeding vessels can also be performed. ^, Multidisciplinary management of the supportive care these patients receive remains essential. Best supportive care is associated with a median survival of between 3 and 5 months, and even shorter for patients who have received prior therapy.

A subset of patients not eligible for radical curative-intent therapy may still be candidates for palliative cancer-directed therapy with chemotherapy, RT, or a combination of these. The goal of a palliative treatment is to alleviate symptoms secondary to the cancer with minimal treatment toxicity and side effects. Patients treated with palliative systemic therapy have improved survival (median survival of 10.1 months with the EXTREME regimen [cisplatin/carboplatin, 5-fluorouracil, and cetuximab] or 14.9 months with pembrolizumab over supportive care alone). ^,

Palliative radiation therapy

There have been many studies on the use of RT to mitigate disease-related symptomatology, and to improve patients’ speech, swallowing, breathing, and pain as well as to mitigate ongoing bleeding and tumor ulceration. While such studies advocate for the use of RT in the palliative setting, there is little high-level evidence to direct the selection of an optimal palliative radiation regimen.

Curative-intent RT is delivered with doses from 6000 to 7000 cGy divided into 180 to 200 cGy fractions, often delivered with concurrent chemotherapy. Such regimens, while curative, are often highly toxic. Common side effects of treatment are secondary to radiation damage to normal soft tissues and include mucositis, dysphagia, xerostomia, dysgeusia, and radiation dermatitis. Many palliative regimens aim to reduce the dose of radiation administered to below the threshold for producing severe side effects to optimize the balance between risk and benefit when cure is not the goal. Balancing the delicate divide between providing sufficient dose to provide durable local control, while not causing significant morbidity, is fundamental in the selection of each treatment regimen.

Overview of palliative regimens

The treatment of head and neck cancers with palliative RT has been studied in retrospective institutional studies and small phase II trials, as seen in Table 12.1 . The treatment schedules used in these studies range from short courses with simple techniques to stereotactic radiation to conventionally fractionated, standard-length courses. Treatment durations range from 1 to 7 weeks, with many regimens including predetermined breaks between cycles of radiation. Some regimens include the use of chemotherapy. Treatment response rates, changes in QoL, and toxicity profiles vary significantly based on treatment regimen. We reviewed the literature to identify evidence to guide the development of a patient-centric framework for selection of an optimal regimen for the palliative treatment of head and neck cancers. PubMed, EMBASE, MEDLINE, and Cochrane Library databases were searched for published studies on palliative radiation for head and neck cancers from 1993 to 2018.

QUAD shot

One popular approach in the palliative treatment of head and neck cancers in North America is the “QUAD shot” regimen. This approach was introduced in the palliation of pelvic malignancies in RTOG 8502. The regimen consists of 370 cGy delivered twice daily for two consecutive days (total of 1480 cGy/cycle), repeated every 3 to 4 weeks for a total of 4400 cGy in 3 cycles. For patients who tolerate the cycles well and who continue to respond, more than 3 cycles can be delivered. The cyclic nature of the regimen allows for disease response assessment following each course, at which time a decision is made regarding whether to proceed with another course. The biologically equivalent dose administered in QUAD shot is below the threshold for producing mucositis, and the separation between subsequent cycles of radiation allows for depleted mucosal stem cells to repopulate prior to the next cycle. In addition, this regimen allows patients to complete all radiation for 1 cycle within 2 days, limiting the travel burden and time required in a healthcare facility for palliative treatment.

The regimen was first studied in the head and neck setting by Paris et al. in 37 patients with head and neck cancer. Patients included were medically inoperable, failed chemotherapy, and/or had metastatic disease. Twenty-one patients completed all 3 cycles of radiation. An objective response rate of 77% was seen, with an improvement observed in those who completed more cycles. Average survival following therapy was 4.5 months. The majority of patients (85%) endorsed subjective improvement in presenting symptoms. No major toxicities were seen, and expected side effects, including skin changes, dysphagia, and xerostomia peaked 12 days following completion of each cycle. No late complications were observed.

A similar QUAD shot approach was used by Corry et al. In this study of 30 patients, 1400 cGy was administered over 2 days in twice daily fractions. Patients received up to two additional cycles if they demonstrated no tumor progression at the time of follow-up. All but one patient had metastatic disease, and two-thirds of the patients had an Eastern Cooperative Oncology Group (ECOG) performance status of 2 to 3 at baseline. Fifty-three percent of patients completed all 3 cycles, and 50% of all patients had an objective response. Median overall survival was 5.7 months and 44% of patients endorsed improved QoL following completion of therapy. Patients also endorsed improvements in dysphagia (33%) and pain (56%). No patients developed grade 3 toxicities, and the most common grade 1 and 2 toxicities were dermatitis (52%) and mucositis (67%). Similarly, low rates of grade 3 toxicities using the QUAD shot approach have been shown in other retrospective analyses. ^,

The significance of completing more cycles of the QUAD shot regimen was reported by Lok et al. in the largest retrospective study using the QUAD shot regimen. The study included 75 patients with pain, dysphagia, and bleeding secondary to head and neck cancer, 40% of whom had prior radiation at the site of palliation. Thirty-seven percent of patients completed at least 3 cycles of 1480 cGy, and some patients who tolerated and continued to respond to therapy received 5 cycles. Palliative response was significantly correlated with the increasing number of cycles, but not prior radiation, chemotherapy, surgery, or disease stage. Palliative response was seen in 65% of patients, and median overall survival was 5.7 months. Grade 3 dermatitis and functional mucositis were reported in four patients.

The properties of protons allow for significant dose sparing of neighboring tissues and organs at risk and has been studied in cases of head and neck re-irradiation. Proton therapy has been used to deliver the QUAD shot regimen in a cohort of 26 patients, as described by Ma et al. Eighty-eight percent of these patients had prior head and neck radiation. Overall palliative response was 73%, and grade 1 acute toxicity rate was 58%.

Addition of concurrent chemotherapy to QUAD shot

There have been efforts to formally determine the response of radiosensitizing chemotherapy combined with the QUAD shot regimen. Gamez et al. similarly demonstrated the tolerance of concurrent chemotherapy in a retrospective study of 21 patients with metastatic or recurrent head and neck cancer. The two most common indications for treatment in this cohort were pain and dysphagia. All patients received at least 1 cycle of QUAD shot RT with carboplatin (AUC = 2) or cetuximab 250 mg/m ² prior to each radiation cycle. Seventy-six percent of patients completed all 3 cycles of radiation. Objective response was high (85.7%), and the majority of complete responders (80%) completed all 3 cycles of radiation. Patients who completed all 3 cycles of radiation also had a 100% rate of presenting symptom resolution. Treatment was well tolerated with no grade 3 acute toxicities. All patients had at least a grade 1 toxicity, distributed amongst dermatitis, mucositis, and xerostomia. Thirty-five percent of patients had acute grade 2 toxicities, split between mucositis and xerostomia. Two cases of late grade 2 xerostomia and dysgeusia were reported. Carrascosa et al. similarly showed a high response rate (95%) in seven head and neck cancer patients treated with QUAD shot and concurrent paclitaxel 60 mg/m ² .

30 to 32 Gy/5 to 8 fractions

The “Hypo Trial” conducted by Porceddu et al. was a prospective multi-center trial of a hypofractionated palliative regimen for head and neck cancers in which patients were given 3000 cGy in 5 fractions, with 2 fractions administered at least 3 days apart weekly. Of the 37 patients enrolled in the study, the majority (51%) had a World Health Organization (WHO) performance status of 1. The most common presenting symptom was dysphagia, seen in 18/39 patients. Overall, 80% of patients had an objective response following treatment. Median survival was 6.1 months. Grade 3 mucositis was observed in 26% of patients, and 11% of patients experienced grade 3 dermatitis and dysphagia. Two patients had grade 4 dysphagia. Late grade 3 toxicities of skin, mucosa, fibrosis, and xerostomia (all n = 1) were also observed in this cohort.

Murthy and colleagues studied a regimen with similar biological dose equivalency and biweekly administration. Their regimen of 3200 cGy with 400 cGy fractions given twice a week had a treatment completion rate of 74% in their prospective study of 126 patients. Forty-two percent of patients had a response at the primary site of disease, and three-quarters of patients endorsed improvement in pain from baseline. Only one case of grade 3 mucositis was reported, and no long-term follow-up was reported due to significant travel burden for the majority of patients.

24 Gy/3 fractions

Higher dose per fraction hypofractionated regimens have been studied in larger retrospective studies. The “0-7-21” regimen from Canada delivered 24 Gy in three weekly fractions and was studied in 110 patients, the majority of whom had stage IV disease. This approach yielded high (82% and 81%) response rates for symptoms and tumor size, respectively; however, toxicities from this regimen were not reported.

20 to 25 Gy/5 fractions

Short course daily radiation regimens have been proposed and described in studies, including a phase II trial by Fortin et al. In their study of 32 patients with advanced head and neck cancer, patients received 2500 cGy in 5 fractions over the course of 1 week. Patients were treated with IMRT and the majority of patients expressed improvement in global QoL, physical functioning, swallowing, and pain compared to baseline. Thirteen percent of patients experienced a grade 3 toxicity.

Mohanti et al. described a similar weeklong treatment in a large retrospective study of 505 patients. Patients were treated with 2000 cGy in 5 fractions. Compared to Fortin et al., this regimen showed a lower objective response rate at 37% and symptom response rate of 50%. Additionally, all patients in this cohort experienced patchy mucositis when examined 1-month post-treatment. Paliwal et al. used this same regimen of 2000 cGy in 5 fractions in a 50-patient retrospective study.

Although systemic chemotherapy combined with RT is traditionally thought to be too aggressive to be used for palliative care due to combined toxicity, another recent phase II clinical trial done by Borg et al. showed promising outcomes with the treatment consisted of 2000 cGy in 5 fractions followed by 4 cycles of chemotherapy (FOLFOX regimen). Out of the 29 enrolled patients, 23 were in the per-protocol (PP) population (treated with at least 4 fractions of RT and 2 cycles of chemotherapy) and the median time to improvement was 2.0 months and the median duration of improvement was 12.2 months. They also showed that in their PP population, the endoscopic response rate was 78% with 22% complete responders, and the metabolic response rate of the primary tumor was 61% with 30% complete responders with the median overall survival at 16.0 months. Nevertheless, the most frequent grade 3 to 4 adverse events were neutropenia (32%), infection (25%), pain (14%), esophagitis (11%) and anorexia (11%), which can be effectively managed without compromising overall quality of care. Therefore, they suggested such palliative short-course RT followed by chemotherapy is a promising treatment strategy that can provide long-lasting relief of dysphagia.

30 Gy/10 fractions

Ghoshal and colleagues have studied a more conventional palliative regimen of 3000 cGy in 10 fractions. Seventy-two percent of the 25 patients in this study had metastatic disease, with pain and dysphagia as the main indications for delivery of palliative-intent RT. Ninety percent of patients experienced greater than 50% symptom relief following treatment. All patients had either grade 1 or 2 mucositis following treatment.

Ali et al. reported similar outcomes with the 3000 cGy in 10 fraction regimen in a prospective study of 30 patients with symptomatic disease. All patients had greater than 50% pain relief and improvement in baseline dyspnea. A similar toxicity profile was seen, with all patients having grade 1 or 2 mucositis. Twenty-seven percent of patients in the cohort went on to receive further radiation in 200 cGy fractions to a cumulative biological equivalent dose (BED) of 6600 cGy.

40 Gy/10 fractions

Two radiation regimens of 4000 cGy delivered over 10 fractions have been studied, each with slight variations in the design of treatment delivery. The first, studied by Kancherla and colleagues, administered 2000 cGy over 5 days, followed by a 2-week break, and then a second cycle of 2000 cGy in 1 week. There was a 72% objective response rate and 79% symptom response rate at 4 to 6 weeks after treatment completion amongst the 33 patients in this cohort. One- and 2-year overall survival were 42% and 35%. Grade 3 skin toxicity, mucositis, and esophagitis were seen in 3%, 6%, and 9% of patients, respectively, and 20% of all patients required hospitalization for toxicity management.

Das et al. prospectively studied 36 patients with stage IV or recurrent head and neck cancer. Patients received 4000 cGy in 10 fractions, with 2 fractions given each week. Sixty percent of patients had an improvement in performance status after treatments, and 88% of patients endorsed greater than 50% pain relief. No significant changes were noted in head and neck specific QoL scores following treatment. Similar to the toxicity profile seen by Kancherla et al., 21% of patients experienced grade 3 mucositis or dermatitis. Forty-eight percent of patients were recommended a nasogastric tube during treatment due to oropharyngeal odynophagia. To reduce such early mucosal toxicity, groups have studied the use of smaller fractions with similar cumulative doses.