Palliative radiotherapy in head and neck cancers: Evidence based review
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0973-1075.30244
Source of Support: None, Conflict of Interest: None
Squamous cell carcinoma of head and neck (SCCHN) is one of the commonest cancers seen in India, constituting up to 25% of their overall cancer burden. Advanced SCCHN is a bad disease with a poor prognosis and patients usually die of uncontrolled loco-regional disease. Curative intent management of loco-regionally advanced SCCHN has become more evidence-based with active clinical research in the form of large prospective randomized controlled trials and meta-analyses. However, little has been written about palliative radiotherapy (PRT) in head and neck cancers. It is widely recognized that PRT provides effective palliation and improved quality-of-life in advanced incurable malignancies. It is in this context that this study proposes to review the existing literature on palliative radiotherapy in advanced incurable SCCHN to help formulate consensus guidelines and recommendations.
Keywords: Head neck cancer, palliation, quality-of-life, radiotherapy
Squamous cell carcinoma of the head and neck (SCCHN) is one of the commonest cancers seen in developing countries, including India, constituting up to 25% of the overall cancer burden. Registry estimates suggest that 675,000 people are diagnosed annually to have head and neck cancers of which 60% occur in low to middle income countries. The vast majority of them present with loco-regionally advanced disease, with only 25-30% presenting in early stages amenable to cure. It is indeed unfortunate that less than 10% of the peer-reviewed published literature on SCCHN comes from the developing world where it is largely prevalent. Advanced SCCHN carries a poor prognosis and patients usually die of uncontrolled loco-regional disease. The five-year survival even with aggressive treatment is less than 20%, with a median survival of around 12 months.
Head and neck cancers represent a significant treatment challenge because the disease is within an anatomic environment that contains several critical tissues, such as the spinal cord, salivary glands, mandible, nerves, major blood vessels and the organs of speech, swallowing, hearing and respiration. Common symptoms include pain, dysphagia, odynophagia, otalgia, hoarseness, cough and respiratory distress. There can be significant overlap between symptoms and treatment toxicity. Curative intent management of loco-regionally advanced SCCHN has become more evidence-based with data from large prospective randomized controlled trials and meta-analyses, resulting in a large body of high-quality evidence for formulating consensus guidelines and recommendations., It is widely recognized that palliative radiotherapy (PRT) provides effective palliation and improved quality-of-life (QOL) in advanced incurable malignancies,, and accounts for a significant portion of cancer care across the world. However, little has been written about PRT in head and neck cancers. Poor compliance to therapy, limited enrolment in prospective trials and high attrition rates render outcome assessment difficult and challenging in this population. Furthermore, limitation in resources both in terms of personnel and radiation equipment in developing countries has led to a situation where timely delivery of PRT to patients with short life-expectancy is compromised. An expert panel had earlier concluded that insufficient information precludes estimations of the frequency, degree of or duration of symptomatic relief from PRT of head and neck cancer. There is a paucity of guidelines in current literature regarding the optimal choice of palliative regimens for these patients with inadequate information on time, dose and fractionation; toxicity of such palliative regimens; and QOL issues pertinent to them. Though there are QOL questionnaires and scales addressing the symptoms in head and neck cancers,,, the majority of them pertain to radical treatment for SCCHN. There is still no proper assessment tool for PRT for head and neck cancers specifically incorporating and focusing on symptom indices. It is in this context that this article proposes to review the existing literature on palliative radiotherapy in advanced incurable SCCHN to help formulate consensus guidelines and recommendations.
Literature search strategy
Published data for this review were identified by systematically searching the PubMed-MedLine restricted to the English language from 1980 till date with the following Me dical S ubject H eading ( MeSH ) terms: "head neck cancer"; "radiotherapy"; "palliation" and "quality-of-life" using BOOLEAN search algorithms. Studies reporting exclusively on re-irradiation for loco-regionally recurrent disease were not included. Studies involving hyperthermia were also not considered. All pertinent articles were retrieved and selected studies considered for this review. Relevant cross references from selected studies were also considered. The primary outcome of interest were symptom relief and QOL. Secondary endpoints were loco-regional control, disease-free survival and toxicity.
How should we select patients for treatment with palliative intent?
In the absence of reliable and robust prognostic factors, it is common to recommend radical loco-regional treatment even for advanced SCCHN with the intent of maximizing loco-regional control and achieving a potential cure. It is often difficult to identify subsets of patients with advanced disease best suited for palliative therapy alone as compared to those in whom radical intent treatment could still be considered. The factors that should guide the treating oncologist in choosing patients for palliative intent treatment alone are i) inoperable, fixed and unresectable disease; ii) very advanced loco-regional disease not amenable to cure; iii) poor physical condition and medical co-morbidities; iv) widely metastatic disease; v) achievable symptomatic relief; and vi) short life-expectancy.
Does palliative radiotherapy improve outcome?
Best supportive care alone is associated with a median survival of three to six months in advanced SCCHN., In the largest study on the natural history of untreated head and neck cancer, 808 patients were followed-up until their death. All patients were given best supportive care, but specific oncological treatment was not pursued due to advanced tumor stage, poor performance status or patient refusal. The median overall survival was 100 days (range 1 day to 53.8 months). Performance status was the only predictor of survival ( P <0.001) on multivariate analysis. Approximately 50% of untreated patients died within 4 months of diagnosis, but a small subset of patients with low tumor burden and good performance status survived upto 4 years.
It has been argued that PRT neither improves survival, nor positively impacts on QOL of patients with SCCHN. Burns et al treated 76 consecutive patients with advanced cancers of the upper aero-digestive tract with radical or palliative intent. Clinical stage, tumor site and performance status were important prognostic factors. Overall mean survival was 15 months with a 2-year disease-free survival of 16%. Patients treated with radical intent had a mean survival of 19.4 months with a 2-year disease-free survival of 29%. On the other hand, palliative intent treatment was associated with a mean survival of 8.4 months and virtually no long-term survivors. Reasonable palliation was achieved in 25% of patients with return to normal function, but the remaining continued to have significant distressing symptoms related to speech and swallowing. The outcome of palliative intent treatment was no better than best supportive care leading the authors to conclude that there is little benefit associated with palliative chemotherapy or radiotherapy in SCCHN.
There is no level I evidence regarding the use of palliative head and neck radiotherapy, but several retrospective series,,, case-control studies,, single arm prospective trials,,,, and one small randomized controlled trial affirm that its use is associated with an improvement in outcome.
In a retrospective study, forty patients with advanced neck disease from an unknown primary were treated with either 30 Gy/10 fractions/2 weeks or 20 Gy/2 fractions with a 1-week inter-fraction interval. There was a good 1-year response rate (77% and 48% respectively), with a similar symptomatic response rate of 68% and 38% respectively.
In a series of 331 elderly patients (age >70 years), Lusinchi et al planned 54 patients with external radiation alone (30 Gy/15 fractions/3 weeks) with palliative intent. Good responders went on to receive further radiation to curative doses with or without interruptions. Irradiation had to be discontinued in 18/54 (33%) patients even before the planned 30 Gy due to poor performance status, low tolerance to PRT, progression and logistical reasons. Local control at three years was 19%. The two year and five year overall survival for patients treated with PRT was 16% and 5% respectively. Since this was a retrospective study, QOL improvement and extent of palliation could not be ascertained from the reported data.
In a series of 160 patients with N3a neck disease, radiotherapy was given to 54 patients with palliative intent. Eleven of these 54 were lost to follow-up and the mean survival of the remaining patients was only 6 months. Almost a third of the patients failed to complete planned treatment due to progression or poor performance status. Two patients survived beyond 2 years with no evidence of disease.
Cyclical accelerated split-course radiotherapy with simultaneous chemotherapy was given to 34 patients with advanced unresectable SCCHN. Radiotherapy consisted of 23.4 Gy/9 days divided into 1.8 Gy, twice daily (from day 3-11) for a total tumor dose of 70.2 Gy/51 days in 3 cycles. The two year local control and survival was 81% and 58% respectively. Overall toxicity was acceptable with excellent palliation of symptoms. This was comparable to historical controls treated with palliative intent radiotherapy.
Carvalho et al compared patients with advanced head and neck cancer who received treatment with those who remained untreated until death having the same demographic and clinical characteristics. They found a significant difference between the survival rates of the untreated group and those of the treated groups that was independent of the type of treatment performed ( P < 0.00001) or the tumor response to treatment ( P < 0.0001).
Paris et al reported on a phase I/II study of aovel fractionation involving 370 cGy/fraction given twice-a-day for 4 fractions over 2 days. This was repeated every 3-4 weeks giving a total dose of 44 Gy over 9 weeks. Good palliation was achieved in 33 (84.6%) of 39 lesions in 37 patients with minimal acute toxicity and no long-term complications. The mean survival for the entire cohort was 4.5 months (range 2 weeks to 31 months)
Minatel et al used a higher dose regimen for palliation in inoperable head and neck cancer patients. They treated 58 patients with split-course radiotherapy, 50 Gy/20 fractions with a 2-week break after the first 25 Gy with concurrent bleomycin. This regimen was associated with a local control rate of 69% with median response of seven months. Symptomatic improvement was seen in 81% of patients, but there was 79% grade 3 toxicity.
Twenty five patients with advanced SCCHN were treated with a short course of PRT (30 Gy/10 fractions/2 weeks). Baseline symptoms were assessed with an 11-point numerical scale for pain, dysphagia, cough, insomnia and dyspnea. At 1-month post treatment, all 22 patients with pain and >90% patients with dysphagia, dyspnea and insomnia experienced >50% symptom relief. Cough was relieved in 60% patients. The median duration of response was 3 months. No patients experienced grade 3 or worse toxicity.
Mohanti et al treated 505 patients with stage IV SCCHN with a uniform dose of 20 Gy/5 fractions over one week. Seventy percent of patients presented with two or more distressing symptoms. At 1-month assessment 189 (37%) achieved a partial response and had ambulatory physical state suited for further curative-dose radiotherapy. Good symptom relief (50% or more) was found in 57% for pain, 53% for dysphagia, 57% for hoarseness, 47% for otalgia, 76 % for respiratory distress and 59% for cough. The main acute toxicity of PRT was patchy oro-pharyngeal mucositis and dermatitis. Median overall survival with palliative radiotherapy was 200 days. The 153 patients who went on to receive further curative-dose radiotherapy had a significantly better overall survival (400 days).
The QUAD SHOT consisted of 14 Gy/4 fractions, twice daily at least 6 hours apart for 2 consecutive days. This regimen was repeated at 4-weekly interval for a further 2 courses, if no tumor progression was observed. It was based on the concept of delivering short cyclical courses of treatment, with each cycle designed to give a biologically equivalent dose below the threshold for producing mucositis and a maximum cumulative dose of 42 Gy/12 fractions imposed by consideration of delayed radiation effects. Thirty patients had at least one treatment and sixteen patients completed all 3 cycles. Sixteen patients (53%) had an objective response and a further seven had stable disease. Median overall survival was 5.7 months (range 0.6-26.7 months), with a median progression-free survival of 3.1 months (range 0.6-11.4 months). No patient experienced grade 3 or worse toxicity. Patients were asked to rate their overall quality of life on a scale of 1-7 (very poor to excellent) and to indicate whether they thought radiation treatment had been worthwhile. In 67% patients, the performance status stabilized or improved. The treatment was well tolerated, with overall improved QOL in 11 of 25 (44%) evaluable patients. Treatment was felt to be worthwhile by 43%, 58% and 63% of patients after first, second and third courses of cyclical radiation respectively.
The only prospective controlled trial of 64 patients with inoperable SCCHN randomized patients to either conventional radiotherapy (60-70 Gy/30-35 fractions/6-7 weeks) or short-course high fractional dose radiotherapy (400 cGy per fraction, 40-48 Gy/10-12 fraction/2-3 weeks, 4 fractions per week). The palliative benefits of radiation were comparable in both the arms. Complete regression of tumor was seen in the majority of patients in both the groups and there was no reported difference in acute or late toxicity. The overall survival was poor in both arms, reflecting a selection bias towards advanced incurable disease.
What is the optimum dose-fractionation schedule?
In the last decade or so, clinical trials and consensus guidelines utilizing short-course PRT have evolved for several incurable solid tumors such as bone metastases, brain metastases and lung cancer. No such large prospective randomized controlled trial has been done for PRT in advanced incurable SCCHN. It has been argued that a higher total dose is needed for growth restraint and sustained palliation in head-neck cancers. Various dose-fractionation schedules [Table - 1] that have been used in the aforementioned sites have been extrapolated for use in palliative head-neck radiotherapy. Although the quality of evidence is not very robust, the weight of evidence favors a short-course fractionated regimen (20 Gy/5 fractions or 30 Gy/10 fractions) or cyclical treatment (QUAD-SHOT) as compared to single fraction or protracted courses of radiotherapy.
How should we assess QOL in patients having head and neck cancer who are treated with palliative intent?
Performance status scale for head and neck cancer; European organization of research for treatment in cancer quality-of-life questionnaire with head-neck module (EORTC QLQ C-30 and HN 35) and functional assessment of cancer therapy and head-neck module  are a few of the tools addressing general and head-neck specific issues which have been used traditionally for QOL assessment. Most of these have been formulated for definitive and radical radiotherapy. There is an urgent need to develop and validate robust tools to accurately assess QOL in the palliative setting having proper symptom, psychosocial and medication indices incorporated into them. Although there have been several studies addressing the importance of medications in palliation of head and neck cancers, hardly any progress has been made in formulating a proper medication index.,, When selecting a disease-specific QOL instrument for head and neck cancer patients, careful consideration must be given to disease sub-site, treatment, psychosocial issues timing of assessment, clinical setting, local culture, study purpose and research question.
Can we select a subset of patients for dose escalation using biomarkers?
Good responders to PRT have traditionally been treated further to curative doses with modest benefits using response as a surrogate for favorable biology. In this context biomarkers can further qualify and quantify resistant or aggressive disease and help select good responders for dose escalation with a potential to improve outcome. Many molecular markers such as the over-expression of epidermal growth factor receptor, cyclin D-1, cyclin A; p53 and Rb inactivation; p16 deletion; E-cadherin downregulation; and gene expression profiling have been correlated with outcome in SCCHN, a measure of the complexity of the process of carcinogenicity and progression. It is highly unlikely that a single biomarker can accurately predict prognosis due to the complex interplay of factors and crosstalk between pathways, but a panel of these could establish distinct molecular subsets and help select patients for dose escalation. There is a growing need to carry out biologic correlative studies and validate these leads in the clinic for palliative radiotherapy.
It is recommended to have a proper efficacy judgement criteria for palliative head and neck radiotherapy where the primary endpoints are symptom relief and not survival. International consensus on PRT endpoints for future clinical trials in bone metastases have addressed issues which are equally applicable to head and neck palliation. These include analgesic assessments; primary endpoint definition; timing, frequency and duration of follow-up assessment; progression and duration of response. More with advanced incurable SCCHN should be accrued onto prospective trials of palliative treatment with appropriate endpoints and QOL tools.
[Table - 1]