Introduction
Non-small cell lung cancer (NSCLC) accounts for ~85% of all lung cancer cases, and can be classified into three main clinical groups that determine the treatment approach according to the level of the disease: operable disease, locally advanced disease, and distant metastatic disease.Reference Halperin, Perez and Brady 1 The operable disease group accounts for only 20% of all NSCLC patients, and comprises stage I, II and selected stage III patients. A symptomatic disease, >3 cm tumour size, non-squamous cell histology, metastasis of numerous lymph nodes, and vascular invasion are among the factors that adversely affect prognosis.
Currently, postoperative radiotherapy (PORT) is indicated for disease with a close or positive surgical margin or for resected N2 (mediastinal lymph node involvement) disease. PORT following adjuvant chemotherapy is recommended for disease with a negative surgical margin that only includes mediastinal lymph node involvement. In these patients, the treatment volume includes the involved lymph node region, as well as the hilum and subcarinal lymph nodes of the same side, depending on the localisation of the primary tumour. In resected N2 disease, the PORT dose is 50 Gy in total. 60–66 Gy PORT is administered only to the surgical margin area in disease with a close or positive surgical margin and no lymph node involvement.
PORT is contraindicated in completely resected stage I disease, 2 and while its role in N2 disease has become gradually more supported based on new data, its role in N1 (pulmonary lymph node involvement) disease is not yet certain.Reference Chung, Stryker, O’Neill and DeMuth 3 – Reference Le Péchoux, Dunant and Pignon 7 Furthermore, randomised prospective studies have found out that PORT increases local control of N2 disease. 8 , 9 Although its survival effect was not shown in those studies, a Surveillance Epidemiology and End Results analysis reported that radiotherapy decreases survival in N0 (without lymph node involvement) disease, but provides a survival advantage in N1 and N2 disease.Reference Lally, Zelterman, Colasanto, Haffty, Detterbeck and Wilson 6
In the current study, we performed a retrospective assessment of the clinical treatment characteristics, local control rates, overall survival, and prognostic factors of 61 patients diagnosed with NSCLC and treated with PORT by a linear accelerator (LINAC).
Methods
Patient characteristics
A total of 61 patients who were diagnosed with NSCLC and received PORT by a LINAC between the years of 2000 and 2011 were included in the analysis. A summary of patient characteristics can be found in Table 1, and a summary of tumour characteristics and surgery data can be found in Table 2. All patients were treated with a Philips SLI-Plus 25® (Elekta, West Sussex, UK) LINAC between 2000 and 2008. After 2008, patients were treated with a Philips SLI-Plus 25® or an Elekta Synergy Platform® (Elekta) LINAC. PORT was administered to 15 patients (24·5%) for a positive surgical margin, 33 patients (54%) for mediastinal lymph node involvement, and 13 patients (21·5%) for both mediastinal lymph node involvement and a positive surgical margin.
Table 1 Patient characteristics
Abbreviation: KPS, Karnofsky performance status.
Table 2 Tumour characteristics and surgery data
A total of 33 patients received 2D treatment, while 3D treatment was planned for 28 of the patients; 50–66 Gy conformal radiotherapy with a fractional dose of 1·8–2 Gy was administered to the bronchial stump+ipsilateral hilum+mediastinal region for 58 patients and to the bronchial stump+ipsilateral hilum for three patients.
During the first 2 years following radiotherapy, the patients were evaluated using complete blood counts, routine biochemical examinations, physical examination and thoracic computerised tomography (CT) on a quarterly basis. Evaluations between the 2nd and the 5th years were performed semi-annually.
There were no selection criteria for the patients of this study. All the patients who received PORT during years 2000–2011 were included.
Statistical analysis
Statistical analysis was performed with SPSSTM 16 software (SPSS Inc., Chicago, IL, USA). Disease-free survival, overall survival, and local–regional control were calculated by the Kaplan–Meier method using the period from the date of surgery to the date of death or last evaluation. The variables which were significant (p≤0·05) in univariate analyses were included in multivariate analyses, which utilised Cox’s regression model to assess the prognostic factors.
Results
In the median 17-month follow-up period, 23 patients were disease-free, six had local relapse, three had distant metastasis, two had both local relapse and distant metastasis, and 27 died due to disease-related reasons. In their first evaluation in the 3rd month after PORT, all of the patients were examined by thoracic CT or positron emission tomography. Upon examination, we observed a complete response in 93·4% of the patients, local–regional and systemic progression in 4·9%, and a partial response in 1·6%.
Median survival was 25 months, while distant metastasis-free survival was 19 months. The 2- and 5-year overall survival rates were 49 and 32%, respectively, while the 2-year distant metastasis-free survival rate was 61%.
Treatment modality was a factor for overall survival. Out of 33 2D patients 10 survived (six disease-free survival), while out of 28 3D patients 24 survived (17 disease-free survival) which corresponded to a p-value of 0·021. Tumour size >4 cm (p=0·004), a positive surgical margin (p=0·01), and left lung localisation (p<0·05) of the tumour were also significant prognostic factors in terms of overall survival. Factors such as lower Karnofsky performance status (KPS), advanced age, history of smoking, weight loss, comorbidity, type of surgery, total radiotherapy (RT) dose, stage and chemotherapy did not have a statistically significant effect on overall survival.
The median for local–regional control was 20 months. Local–regional progression was determined in 10 of the 61 patients. None of the prognostic factors had any statistically significant effect on local–regional control.
Distant metastasis developed in 19 patients. The 2-year distant metastasis-free survival rate was 61%. Treatment modality (2D/3D) (p=0·005) and lymph node positivity (p=0·013) were significant prognostic factors on the distant metastasis-free survival. The distant metastasis-free survival rate in patients with negative lymph node status was 8·4 times higher (95% CI 1·57–45·45).
Discussion
In the present study, we found the 5-year overall survival rate and the local recurrence rate to be 32 and 16%, respectively, which are comparable with the retrospective studies listed in Table 3.
Table 3 Retrospective studies of PORT
Abbreviations: PORT, postoperative radiotherapy; NR, not reported.
In their study on stage IIIA-N2 patients, Dai et al. found that, with PORT, local recurrence-free, distant metastasis-free, disease-free, and overall survival rates were significantly higher in the 1st, 3rd and 5th years.Reference Dai, Hui and Ji 10 There are also studies that report a significant increase in 5-year overall survival for N1 and N2 patients who received PORT.Reference Liu, Zhao and Wang 11 , Reference Corso, Rutter, Wilson, Kim, Decker and Husain 12 Furthermore, Shen et al. observed higher overall survival rates among stage IIIA pN2 patients with two or more positive mediastinal lymph nodes that were treated with PORT and chemotherapy.Reference Shen, Ji and Zuo 13 In our study, the effect of the N stage on survival was not statistically significant. We determined that lymph node positivity was a prognostic factor for distant metastasis-free survival. Distant metastasis-free survival rates in N1 or N2 patients were eight times higher than in N0 patients (p=0·013).
Many prognostic factors that affect survival in lung cancer are reported in the literature.Reference Sawyer, Bonner and Gould 14 – Reference Caro, Salas, Ward and Goss 16 A significant increase in the rates of local recurrence-free survival (p<0·001), recurrence-free survival (p=0·013), and overall survival (p=0·002) following PORT was reported in a study by Moretti et al. on 83 patients with pathological N2 disease. The number of involved lymph nodes or stations, presence of extracapsular invasion, and a positive surgical margin were considered to be significant prognostic factors in overall survival.Reference Moretti, Yu and Chen 17 Mantovani et al. found out that that the number of positive mediastinal lymph nodes is a prognostic factor for local control in pN2 patients.Reference Mantovani, Levra and Filippi 18 In our study, the presence of a positive surgical margin did not have a significant effect on local–regional control or distant metastasis-free survival, but it was a significant prognostic factor for overall survival.
50 Gy RT and two to six cures of CT (median four cures) were given to the patients in a study by Zou et al. on stage III-N2 patients.Reference Zou, Xu and Li 19 Five-year overall survival rate was 30·5% in the patients who received postoperative chemoradiotherapy and 14·4% in those who did not (p=0·007). Higher disease-free survival rates were found with postoperative chemoradiotherapy (p=0·003). PORT and having received at least three cures of CT were found to be prognostic factors in terms of overall survival and disease-free survival. In our study, CT did not have a significant contribution to survival. However, CT information of many of the patients could not be obtained from their files.
In a study by Karakoyun et al., RT dose, KPS, age, left lateralisation of tumour and pneumonectomy were prognostic factors for overall survival.Reference Karakoyun-Çelik, Yalman, Bolukbasi, Cakan, Cok and Ozkok 20 There was a higher mortality rate in patients who received >54 Gy RT to left-sided tumours which was reported to be potentially due to cardiac toxicity. In our study, the overall survival rates were also low in the case of left-lung tumours irrespective of the total RT dose applied (p=0·03).
Billiet et al. studied PORT patients in three groups: patients who received RT via Co-60, patients receiving RT via LINAC, and patients receiving RT with both Co-60 and LINAC simultaneously. The overall survival rates of the LINAC group were 13% higher and local recurrence was 10% lower than the mean.Reference Billiet, Decaluwé and Peeters 21 It is important to note that all of the patients included in our study were treated with a LINAC.
The modality of the treatment of PORT in the clinic has been upgraded to field-in-field and intensity-modulated radiation therapy (IMRT) starting from 2012. Until then, all the patients received 2D or 3D conformal radiotherapy. These patients received doses ranging from 50 to 66 Gy and even the fraction dose was not constant (1·8 or 2 Gy). These differences in the dose scheme may have had negative impact on the study. Another limitation was the lack of chemotherapy information. Only 32 out of 61 patients’ info was reachable. Also higher total number of patients would have yielded more decisive results.
Conclusions
Wider treatment regions were exposed with Co-60 in many of the previously published PORT studies on patients diagnosed with and operated on for NSCLC. In our study, the results on overall survival and local control are compatible with other LINAC-based studies. The study shows better overall survival results with 3D-conformal modality with respect to 2D treatment. Today 3D-conformal PORT is accepted as the standard treatment by many clinicians for operable stage II-III disease. In the light of all the related studies, prospective randomised studies assessing adjuvant RT are needed based on modern radiotherapy standards such as field-in-field and IMRT.
Acknowledgements
None.
Financial support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Confilicts of Interest
None.
