A comparative study of different dose fractionations schedule of thoracic radiotherapy for pain palliation and health-related quality of life in metastatic NSCLC.

INTRODUCTION: To investigate the effect of different hypo fractionated thoracic radiotherapy schedules in relation to thoracic pain relief, overall survival and post radiotherapy HRQOL in metastatic NSCLC.
MATERIAL AND METHODS: Stage IV NSCLC and had intra-thoracic symptoms, included in the study. Patients were randomly assigned to three treatments arms. (i) 17 Gy in 2 fractions in one week (ii) 20 Gy in five fractions in one week. (iii) 30 Gy in 10 fractions in two weeks. BPI module was used to assess pain score before and after the thoracic radiotherapy. Functional assessment of cancer therapy-G (FACT-G) used to investigate changes in HRQOL. Clinicians' assessment of symptom improvement were recorded at 2(nd), 6(th) and 12(th) weeks after completion of TRT.
RESULTS: Pain relief, HRQOL and OS were equivalent in all the three arms. The median OS were 6 months, 5 months, 6 months in arm A, B and arm C, respectively.
CONCLUSION: Protracted palliative thoracic radiotherapy renders no added advantage of relief of symptoms, HRQOL and overall survival compared to short course palliative TRT in metastatic NSCLC.

RCT Entities:   Population Interventions Outcomes

INTRODUCTION

Metastatic non-small cell lung carcinoma (NSCLC) usually present with symptoms from the primary intra-thoracic tumor including dyspnea, chest pain, cough, and hemoptysis.[1] Thoracic palliative radiotherapy is an effective modality of treatment in relieving symptoms.[23] Patients with metastatic NSCLC are not suitable for curative treatment and yield a poor prognosis with limited survival. Consequently, a limited treatment period is usually advocated, if adequate palliative effect is attained. Hypo fractioned thoracic radiotherapy (TRT) may fulfill this criterion. Multiple randomized studies favor hypo fractionated treatment policy[4567] and others advise against hypo fractionated treatment because of increased toxicity.[891011]

MATERIALS AND METHODS

Patients

182 eligible patients were included in this trial between July 2007 and March 2012. Patients were recruited from the department of radiation oncology, Medical College and Hospitals, Kolkata and the department of radiation oncology, Burdwan Medical College and Hospital, Burdwan.

Inclusion criteria of the eligible patients were:

Histologicaly or cytologically confirmed NSCLC

Aged over 18 years

Stage IV

Performance status (PS) ≤ Eastern Cooperative Oncology Group (ECOG) 3

Expected survival more than 3 months

Patients had intra-thoracic symptoms

No history of previous TRT.

Written informed consent was taken from all the patients before random assignment to palliative TRT.

Eligible patients were assigned randomly to one of the three treatment arms with the help of blind envelope method. Twelve patients were ineligible because of incorrect stage and patient's refusal/terminated their palliative radiotherapy prematurely. Significantly, higher number of patients in arm C terminated their planned palliative TRT. All the patients were assessed on an intention to treat basis. Baseline patients’ characteristics are presented in Table 1. Fourteen patients died within 3 months of post radiotherapy. All these patients were excluded from statistical analysis.

Table 1

Baseline characteristic of all eligible patients according to assigned treatment arms

All the patients were subjected to the following diagnostic work up: Clinical examination, radiological examination (chest X-rays, CECT thorax, and abdomen), and blood chemistry profile. ECOG PS was recorded for every patient before thoracic palliative radiotherapy. CECT or magnetic resonance imaging (MRI) brain and whole body bone scan were performed whenever indicated by symptoms.

Thoracic radiotherapy

TRT administered with parallel oppose AP/PA fields encompassing the primary tumor with 1.5-2 cm margin using Co60 beams. Mediastinal and supraclavicular region were not routinely included in the portals unless enlarged node caused symptoms. Radiotherapy fractionated scheme are as follows:

Arm A: 17 Gy in 2 fractions in one week

Arm B: 20 Gy in 5 fractions in one week

Arm C: 30 Gy in 10 fractions in two weeks.

To prevent possible side effects, patients were prophylactically treated with prednisolone 50 mg once daily as indicated.

After completion of TRT, significant percentage of patients received palliative systemic therapy according to their PS and histopathology. Only six patients in arm A and two patients in arm B were re-irradiated to the chest because of bone metastasis. Forty-five patients received palliative radiotherapy outside the chest because of bone and brain metastasis.

Health-related quality of life

Consecutive patients requiring palliative TRT for metastatic NSCLC enrolled in longitudinal assessment of health-related quality of life (HRQoL) study at their first visit in the department of radiation oncology in Medical College and Hospitals, Kolkata and Burdwan Medical College and Hospital, Burdwan. Patients were followed up and HRQoL was assessed with the FACT-G questionnaire (FACT-G, Bengali format) before the radiotherapy, 2nd, 6th, and 12th week after completion of palliative TRT.[12]

FACT-G, a 27-item general module and is grouped into the four domains: Physical, functional, social well-being (each with seven items), emotional well-being (six items).[13]

The response scales are Likert type, with scoring ranges from 0 to 4. Domain score were obtained by summing the score of their item, ranging from 0 to 28 in the physical, functional, and social well-being and from 0 to 24 in the emotional well-being domains. All the domains score were added, so that a higher score indicates better HRQoL. The FACT-G is reliable and validated instrument for measuring HRQoL in cancer patients.

Pain relief evaluation

Brief pain inventory (BPI) is a self-administered assessment tool used in pain management. The basic concept of BPI is to provide the information on the intensity of pain and the degree to which it interferes with the daily activity. The BPI is a brief and easy to use tool for assessment of pain in clinical research. The long version of the BPI has the front and back body diagrams, the four pain severity items and seven pain interference items rated on 0-10 scale and the question about percentage of pain relief by analgesic. The long version uses 7 days recall period. The interference items are presented with 0-10 scale, with ‘0’ = no interference and ‘10’ = interference completely. The BPI measures how much pain has interfered with the seven daily activities including general activity, walking, work, mood, enjoyment of life, relation with others, and sleep. The BPI pain interference typically scored as the mean of the seven interference items. We chose Bengali translation of validated BPI for the interference items for statistical analysis.[14] Clinically significant change was defined as at least one-step change on the symptom scale.

Clinical follow up after 12th week was optional according to the individual needs. Subsequent palliative treatment allowed, such as chemotherapy, targeted therapy or best supportive care, were according to the requirements and PS. All the patients were followed up every 3 months thereafter.

Statistical analysis

The primary objective of the study was the patient-reported pain relief. Overall survival (OS) and the HRQoL among the three treatment arms were considered as the secondary objectives.

Categorical variables were analyzed using the Pearson's test or Chi-square test. Continuous variables were analyzed using Spearman's test and Kruskal-Wallis test. The log-rank was applied for comparison of OS.

RESULTS

There were no differences among the treatment arms with respect to the relief of chest pain. A total of 83.3% in arm A, 79.6% in arm B, and 87% in arm C (P > 0.05) had reported significant relief of chest pain within 2 weeks from the start of palliative TRT [Figure 1, Table 2]. Furthermore, the palliative effect of pain relief lasted throughout the primary follow up periods.

Figure 1

Pain relief (a) and health-related quality of life score (week 0-12) for FACT - G physical (b), functional (c), social (d), emotional (e)

Table 2

Mean scores for BPI and all FACT-G items module in arm A, arm B, and arm C at baseline and at 2nd, 6th, and 12th week after of palliative thoracic radiotherapy

HRQoL analysis

Mean scores of all the domains of FACT-G questionnaires at baseline and during follow up are listed in Table 2. At 2nd week of follow up, patients in all the three arms reported reduced physical and functional well-being scores compared with baseline scores, thereafter increased and reached its highest value at 6th week of post-TRT. Patients in arm A reported reduced social and emotional well-being scores compared with baseline scores, whereas patients in arm B and C reported increased mean social and emotional well-being scores during post palliative TRT [Figure 1].

Overall survival

The median OS was 5 months for the entire cohort as a whole. Different hypo fractionated palliative TRT schedule did not per se contribute to any significant difference in median OS [Figure 2]. The median OS were 6, 5, and 6 months in arm A, B and arm C, respectively [Table 3]. Statistically significant association (P < 0.005) was observed between the PS and the median OS of the patients. Median OS being 7 months in patients with good PS (ECOG0-1) as compared with 5 months in those with poor PS (ECOG 2-3) [Figure 3]. Subgroup of patients (arm C) with good PS and those treated with more fractionated (30 Gy in 10 fractions) palliative RT had better median OS compared with patients with poor PS [Figure 4]. Tumor histology did not per se contribute to any significant difference in median OS (P = 0.911).

Figure 2

Overall survival of patients according to different dose fractionation schedule

Table 3

Median overall survival of the patients analyzed with respect to clinical variables

Figure 3

Overall survival of the patients according to performance status

Figure 4

(a) Overall survival of patients (17 Gy per 2 F) according to performance status; (b) overall survival of patients (20 Gy per 5 F) according to performance status; (c) overall survival of patients (30 Gy per 10 F) according to performance status

On univariate analysis, variables such as gender, religion, histo-pathological type, radiation dose and fractionations and treatment portal size had no significant effect on median OS. However, PS had shown significant difference in median OS on univariate analysis [Table 4]. On multivariate analysis using Cox regression model, only PS had significant influence on OS [Table 5].

Table 4

Univariate analysis: Correlation between overall survival and clinical or demographic variables

Table 5

Multivariate analysis by Cox regression model for overall survival and clinical or demographic variables

DISCUSSION

Metastatic NSCLC is one of the leading causes of cancer-related deaths in India. Although treatment outcomes in advanced disease remain modest, with the paradigm shifts in the approach to treatments, they are steadily improving. Customized treatments based on the tumor histology and genetic profiles have becomes the standard of care.

Systemic chemotherapy with the platinum-based doublets is the standard of care in metastatic or advanced NSCLC.[151617] This leads to modest survival advantage and improve QoL. TRT remains an important treatment modality for patients with symptom from intra-thoracic disease. There is no consensus which fractionation scheme should be used for palliative TRT. Different palliative TRT fractionation schedules, ranging from 10 Gy in single fractionation to 30 Gy in 10 fractions are used in clinical practice considering the patients preference, PS, and institutional protocol. In 2002, Cochrane meta-analysis of 10 randomized palliative radiotherapy trials in NSCLC was published.[3] The palliative effect on disease-related symptoms were equivalent in all the fractionation schedules regardless of total radiotherapy dose.

Considering the poor prognosis for our patients, the primary objective of the study was the pain relief. This result was consistent with the previous studies in metastatic NSCLC.[457] HRQoL data did not reveal any significant difference among the study groups except social well-being. There is slight decrease in physical and functional well-being mean score of FACT G at 2nd week post-TRT in all the arms. It indicates a slow deterioration of HRQoL in the physical and functional domains just after palliative hypo fractionated TRT. However, social well-being mean score increased over time up to 12 weeks post treatment except in the arm A (17 Gy in 2 fractions). Whereas, social well-being score decreased just after post radiotherapy and continued to decrease throughout the follow up periods.

The TRT side effects were mild and temporary. Dysphasia, a symptom related to radiation oesophagitis, increased during and shortly after the radiotherapy compared with the pretreatment level. It appears early in arms A (17 Gy in 2 fractions) and B (20 Gy in 5 fractions) than arm C (30 Gy in 10 fractions) and subsides within few weeks after completion of radiotherapy.

PS is the most important prognostic factor for median OS. In our study, we observed longer survival period in patients with good PS, as expected. However, there was no significant difference detected in OS among the three different palliative radiotherapy arms. In addition, subgroup analyses of patients with poor PS, the hypo fractionated (17 Gy in 2 fractions) TRT was equally effective as more fractionated high dose TRT (30 Gy in 10 fractions). We observed a non-significant increase in median survival in patients with good PS and those treated with more fractionated high dose TRT (30 Gy in 10 fractions). Bezjak et a l.[11] reported a study (10 Gy in fractions vs. 20 Gy in 5 fractions) and found a significantly superior median survival in favor of higher dose TRT (6.0 vs. 4.2 months, P = 0.03). This difference was restricted to the good PS, which is similar to our results.

CONCLUSION

The efficacy of hypo fractionated TRT with respect to symptom palliation and OS in metastatic NSCLC is equivalent to that of more protracted or higher dose TRT. However, stage IV NSCLC with good PS may be benefited from more protracted TRT schedule. The side effect of hypo fractionated TRT was tolerable and comparable with protracted and higher dose TRT. Short-term hypo fractionated treatment is convenient for patients with limited expected survival.