Background
Surgery is the cornerstone of treatment for early-stage non–small cell lung cancer (NSCLC). Due to high recurrence rates following tumor resection, including radical surgery, studies have been carried out on the efficacy of adjuvant therapy. Several randomized trials and meta-analyses have confirmed that platinum-based adjuvant chemotherapy can improve overall survival (OS) in patients with stage II and IIIA NSCLC[
1‐
4]; however, the role of postoperative radiotherapy (PORT) remains controversial. Although studies have found that PORT can reduce locoregional recurrence (LR), no survival benefit was observed. Moreover, meta-analyses of randomized trials and retrospective studies have reported that PORT had a detrimental effect on OS in patients with N0 and N1 NSCLC, despite being beneficial in patients with N2 disease[
5‐
8]. As a consequence, PORT is often recommended for patients with resected pathological N2 disease, but not for those with N0 or N1 disease. However, these studies may have been compromised by limitations such as flaws in the study design and outdated radiation techniques.
The reported crude risk of LR for resected N1 disease is high[
9‐
11], which suggests that a subset of N1 patients with high risk of LR may benefit from further local treatment after surgery. The purpose of the present study was to assess the clinical and pathological risk factors that influence LR in patients with resected N1 NSCLC in order to identify and select those patients at high risk of LR for further studies on PORT.
Methods
Study population
This retrospective medical record review study was approved by our institutional review board. The inclusion criteria were as follows: (1) pathologically confirmed T1–3N1M0 NSCLC, (2) consecutive treatment between January 2007 and December 2008 in our hospital, (3) anatomical pulmonary resection with lobecotomy, sleeve lobectomy, bilobecotomy or pneumonectomy and (4) complete records of clinical and pathological data. The exclusion criteria were as follows: (1) positive surgical margins, (2) prior neoadjuvant therapy involving chemotherapy, radiotherapy or PORT, (3) synchronous primary lung tumors or a history of lung cancer and (4) perioperative death. Patients were staged according to the 7
th edition of the American Joint Committee on Cancer (AJCC) tumor-node-metastasis (TNM) classification system[
12].
Definition of recurrence
The primary end point of this study was freedom from LR (FFLR). Local recurrence was defined as the recurrence of a tumor at the bronchial stump or anastomosis, and regional recurrence was defined as recurrence in the mediastinum, hilum or supraclavicular fossa. Other sites of recurrence, including contralateral lung and metastatic lymph nodes in the neck or axilla were defined as distant metastasis (DM). Concurrent recurrence was defined as LR and DM being detected within 30 days of each other; otherwise, it was considered sequential recurrence.
The diagnosis of disease recurrence and recurrence sites was based on a review of the medical records, including operative notes, follow-up imaging reports and cytological and pathological reports. All local recurrences were confirmed by sputum cytology or bronchoscopic biopsy. Nodal recurrences were confirmed by biopsy using ultrasound-guided fine-needle aspiration (FNA) or endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA). New or enlarging lymph nodes (≥1 cm short axis) were confirmed by computed tomography (CT) or through hypermetabolism on positron-emission tomography (PET)-CT, which was found to be consistent with disease progression during subsequent clinical follow-ups. Sites of recurrent lymph nodes were identified using the lymph node map proposed by the International Association for the Study of Lung Cancer (IASLC)[
13].
Follow-up
The median follow-up time was 53.8 months (range, 1.4–81.8 months). Physical examination, chest x-rays and/or CT scans and abdominal ultrasonography were performed every 3 months for the first 2 years following adjuvant chemotherapy, then every 6 months for 5 years and every year thereafter. PET-CT was performed if recurrent disease could not be identified or suspected by routine CT. Other tests were carried out at the discretion of the treating physician.
Of the 199 patients that were eligible for this study, 158 (79.4%) had received adjuvant platinum-based chemotherapy after surgery.
OS was measured from the date of surgery to the date of death from any cause; disease-free survival (DFS) was measured from the date of surgery to the date of LR, DM or death from any cause; FFLR was measured from the date of surgery to the date of identification of LR; freedom from DM (FFDM) was measured from the date of surgery to the date of identification of DM.
Statistical analysis
The Kaplan-Meier method was used to estimate OS, DFS, FFLR and FFDM. Differences between groups were calculated using the log-rank test. Receiver operating characteristic (ROC) curves were constructed, and the Youden Index was used to determine the optimal cut-off points for part of continuous variables, suh as number of dissected lymph nodes and lymph node ratio. Univariate and multivariate Cox regression analyses were performed to assess risk factors associated with LR. A P-value <0.05 was considered statistically significant.
Discussion
The LR rate in patients with resected N1 NSCLC is reported to be 9%–46% (Table
2). In comparison, our data showed a 5-year LR rate of 22.7%. LR risk may now be higher as adjuvant chemotherapies can reduce or delay DM and thereby increase survival times. In addition, better LR control has been associated with improved OS in locally advanced NSCLC patients[
14,
15], indicating that a decreased LR rate after surgery might also improve OS, which suggests that PORT could be beneficial in patients with resected N1 NSCLC who are at high risk for LR.
Table 2
Risk factors for locoregional recurrence (LR) and distant metastasis (DM) in patients with resected N1 non–small cell lung cancer (NSCLC)
| 335 | 8 | 23 (crude) | 35 | LVI; tumor size |
| 198 | 25 | 55 | 40 | Visceral pleural invasion; VATS approach; no. of involved N1 nodes |
| 60 | 40 | 32 | 46 | Chemotherapy |
| 107 | 0 | 47 | 38 | Positive preoperative bronchoscopy findings |
| 1402 | 9 | 16 | 9 | Surgical procedure; LVI; visceral pleural invasion; tumor size |
| 180 | NM | 23 (crude) | 22 (crude) | Visceral pleural invasion |
This study | 199 | 79.4 | 42.6 | 22.7 | Incomplete MLN dissection; positive LN at station 10 |
To date, studies on PORT have found no survival benefits for patients with resected N1 NSCLC, and therefore, PORT is not recommended for these patients. Moreover, compared with chemotherapy alone, PORT following adjuvant chemotherapy was found to be associated with detrimental survival in N1 disease patients in the Adjuvant Navelbine International Trialist Association (ANITA) randomized study[
7]. The ANITA trial was a randomized study, but the patients that were recommended for PORT were selected by the researchers, so were not randomized[
7]; consequently, patient selection bias may have been introduced into the results.
A meta-analysis has also reported similar detrimental effects of PORT in N1 NSCLC patients[
8]; however, due to improved technologies, PORT may not be detrimental in all patients with N1 disease. The trials included in the meta-analysis had been carried out using older radiation techniques, such as those that delivered doses using wedges with opposed lateral portals with an excessive volume; used different fractionation schedules ranging from 1.8–3.0 Gy; used different total doses ranging from 30–60 Gy; or used cobalt-60, which has a worse biological efficacy and therapeutic ratio than those achieved with current linear accelerators[
16‐
19]. Treatment-related morbidities and deaths were more prevalent with these outdated radiation techniques than with more modern techniques, which may account for the detrimental survival rates. This has been supported by two recent randomized trials on NSCLC, which investigated PORT with modern radiation techniques. The first showed that PORT was associated with lower LR and higher 5-year OS in patients with postoperative pathological N0 NSCLC[
20]; the other found that PORT with modern techniques significantly reduced LR, although there no significant improvement in OS and DFS was observed in patients with postoperative pathological stage T1–3 N0–2 M0 NSCLC[
21]. The toxicity of PORT was acceptable in both these trials. These findings suggest that PORT based on modern technologies may promote therapeutic efficacy and could be beneficial for patients with resected pathological N1 NSCLC who are at high risk for LR.
To our knowledge, our present study is the first to identify and report that IMD and positive lymph nodes at station 10 may be risk factors for LR in patients with resected N1 NSCLC, and as such, these patients may warrant further attention. Patients with positive lymph nodes at station 10 are at a higher risk for micrometastases in MLNs through the lymphatic drainage pathway of NSCLC. Studies have reported that the number of resected lymph nodes and the extent of dissected lymph node stations were associated with survival in patients with resectable NSCLC[
22‐
24]. Several studies have recommended that at least six lymph nodes from three MLN stations should be excised[
23,
25,
26]; however, others recommend resection of at least six lymph nodes from mediastinal stations and station 10 for accurate nodal staging and prediction of prognosis in patients with resected NI NSCLC[
27,
28]. Taken together, these results suggest that the extent of dissected or examined MLN stations may be significantly associated with LR and verify the importance of performing sufficient MLN dissections or examinations.
In this study, we found that patients with resected NI NSCLC and <3 dissected or examined MLN stations had lower FFLR rates than those with ≥3 dissected and examinated MLN stations. Furthermore, our findings showed that the 5-year FFLR rate in patients with IMD (75.9%) was worse than that observed in patients with positive lymph nodes at station 10 (93.3%), indicating that IMD was the greater risk factor for LR (Figure
6). Patients with both risk factors had the highest 5-year LR rate (48%), which was comparable to reported values in resected N2 NSCLC[
29]; these patients should therefore be recommended PORT. In contrast, patients with IMD alone should only be considered for PORT after careful consultation between the surgeon and radiation oncologist.
Tumor size has been shown to be a prognostic factor for survival in patients with NSCLC[
30‐
32]. Saynak et al. and Lopez Guerra et al. reported that tumor size was independently associated with LR in resected N1 NSCLC patients[
33,
34]. Other reports include three studies which found that lymphovascular invasion was a risk factor for LR in N1 NSCLC patients[
33‐
35]. However, in the study by Lopez Guerra et al.[
33], the number of patients with N1 NSCLC was small (16%) compared to the total sample population, and the 5-year LR was much lower (9%) than those reported in previous studies. Several studies have reported visceral pleural invasion as a risk factor for LR[
33,
36,
37], and Varlotto et al. observed an increased risk of LR following adjuvant chemotherapy[
38]. Surgical procedures, such as lobectomy or pneumonectomy and positive preoperative bronchoscopy have also been associated with a risk of LR in N1 NSCLC patients[
30,
35,
39]. However, these reports are inconsistent, and their findings will need further confirmation before being used in patient selection for PORT in N1 NSCLC.
Our retrospective study had the following limitations: It was a single-institution study, and therefore may have had patient selection bias. Some of the patients underwent insufficient MLN dissection or examination, and may have been staged with N2 disease if they had undergone sufficient MLN dissection, which could have introduced staging bias. Patients with concurrent LR and DM may not accurately reflect the pattern of failure observed in this study, and LR that developed before or after DM may not have been identified promptly due to occult symptoms or incomplete radiologic examination. Despite these limitations, our findings support further investigations on the role of PORT in patients with resected N1 NSCLC who have the risk factors for LR identified in this study.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
CF drafted the manuscript. SG, ZH, LJ, JML participated in data collection, and helped to analyze the data. JH, SG participated in the coordination of the study. LW conceived of the study, participated in its design. All authors made substantial contributions to acquisition and statistical analysis of data, and read and approved the final manuscript.