Stereotactic Radiosurgery/ Stereotactic Body Radiotherapy for Recurrent Lung Neoplasm: An analysis of outcomes in 100 patients

Abstract

Background.

A significant number of patients develop recurrent lung cancer despite complete resection or treatment with definitive chemoradiation. Stereotactic radiosurgery (SRS)/stereotactic body radiation therapy (SBRT) is emerging as an important modality for the treatment of early-stage lung neoplasm. SRS may also offer an alternative treatment option for patients with recurrent lung disease. We evaluated outcomes after treatment with SRS for recurrent lung neoplasm in a large series of patients.

Methods.

Selected patients with limited recurrent or persistent/progressive disease after one or more prior treatments for lung cancer were offered SRS. Thoracic surgeons evaluated all patients, placed fiducials when needed, and planned treatment in close collaboration with radiation oncologists and medical physicists. In our early experience, a single fraction of 20 Gy was prescribed and was subsequently increased to 45–60 Gy in 3–5 fractions. The primary endpoint evaluated was overall survival.

Results.

We treated 100 patients with recurrent lung cancer (median age 72 years) with SRS. The postprocedure 30-day mortality was 0%; median follow-up was 51 months (range 5 – 123 months). The median overall survival for the entire group was 23 months (95% confidence interval (CI) 19 – 41 months). The probability of 2-year and 5-year overall survival was 49% (CI 40% - 60%) and 31% (CI 23% - 43%), respectively.

Conclusions.

Our experience indicates that SRS is safe, and offers an alternative modality in selected patients with recurrent oligometastatic or persistent lung cancer. Thoracic surgeons should actively participate in SRS and continue to evaluate the efficacy of this treatment strategy.

Introduction

Surgical resection of lung cancer offers the best chance of cure, particularly in the earlier stages of the disease (1). Patients who present at an advanced stage are typically treated with chemoradiation. However, despite complete resection or treatment with definitive chemoradiation, 25–70% of patients develop disease recurrence (2,3). Recurrent disease is the most common cause of cancer-related mortality in lung cancer patients who have undergone an initially successful surgical resection (4). The reported results of reoperation for recurrence have not been encouraging with 2-year survival of 23% (5), and five-year survival of 15% (3–5). With this poor prognosis after recurrence, it is important to develop novel strategies to improve the outcomes of patients with recurrent lung cancer.

Driven by the rapid change in molecular and biologic therapies, our understanding of tumor biology is evolving. The concept of oligometastatic disease was highlighted by Hellman and Weichselbaum who described oligometastases as a state between the stage where the tumor is completely local and the stage where there are widespread metastases. The potential therapeutic implications of this concept are that patients who develop limited recurrent metastases can be treated with curative intent with local therapy and do not need to be treated routinely with palliative intent (6–8). Classic studies investigating lung cancer, including those by Luketich and Ginsberg nearly two decades ago, have documented long-term survival in selected patients, who presented during follow-up with limited recurrent metastases and were treated with complete surgical resection or radiation (9,10). Luketich reported results of surgery or radiation in a very selected group of patients with recurrent disease (including bone, small bowel and skeletal muscle metastases) who had undergone resection of their primary nonsmall cell lung cancer (NSCLC) (9). Remarkably the estimated 10-year survival in this series was 86%. These and other studies have stimulated a renewed interest in treatment with curative intent in selected patients with limited recurrent disease (11).

Stereotactic radiosurgery (SRS), a term originally originally coined by Leksell, is an approach utilizing multiple convergent beams, precise localization with a stereotactic coordinate system, rigid immobilization, and single-fraction treatment. Extracranial SRS is also referred to as stereotactic body radiation therapy (SBRT) or stereotactic ablative radiotherapy (SABR) (12). The application of SRS to tumors in the thorax may offer an alternative in the management of patients with recurrent lung neoplasm (13,14). The objectives of this study were to evaluate the outcomes following SRS for treatment of limited recurrent or persistent/progressive lung cancer following initial treatment and to characterize the factors associated with survival after SRS.

Material and Methods

We retrospectively reviewed our experience with SRS for the treatment of recurrent lung neoplasm at the University of Pittsburgh from 2003 to 2011. The study was approved by our Institutional Review Board. Since this was a retrospective study, individual patient consent was waived. Patients were selected for treatment using a multidisciplinary approach. Patients with recurrent lung neoplasm were evaluated by a thoracic surgeon and a radiation oncologist prior to SRS treatment and were staged to evaluate the extent of recurrent disease. Patients with limited recurrence were evaluated for local treatment including evaluation for surgery or ablative therapies. SRS was selected alone or in combination with other modalities for patients with recurrent disease who were high-risk surgical candidates. This series includes patients treated with SRS for intrathoracic/regional recurrent disease and does not include patients with extrathoracic disease.

Treatment protocol

Treatment was delivered using the CyberKnife system (Accuray, Sunnyvale, CA, USA), or Trilogy system (Varian Medical Systems, Palo Alto, CA), as previously described (15, 16). Briefly, up to 4 fiducials (small tumor markers) were placed by thoracic surgeons in and around the tumor for tracking when required. An immobilization device (Alpha cradle, Smithers Medical Products, North Canton, OH, USA) that partially immobilizes the patient to decrease motion and provide a reproducible set up was custom made for each patient. A computed-tomography (CT) scan of the chest and upper abdomen was subsequently performed. The thoracic surgeon, radiation oncologist, and physicist evaluated the CT scan, and a treatment plan was formulated jointly by this radiosurgical team. On the day of the treatment, the patient was repositioned to simulate the original planning setup. The tumor and the planned isocenter of the treatment field were identified using cone-beam CT, orthogonal stereoscopic X-rays, and fluoroscopy. Patients were treated with 20–60 Gy prescribed in 1–5 fractions. Patients were followed at 3 to 6 month intervals (16).

Data collection and statistical analysis

Patient information including demographics, history, tumor characteristics, stage, treatment, and comorbidities (Charlson comorbidity index, CCI) were collected (17). The primary endpoint evaluated was overall survival. Kaplan-Meier plots were constructed using Greenwood confidence limits. The association of clinical parameters, treatment variables, demographic and pathologic factors with the primary endpoint was estimated with proportional hazards regression.

Results

Patient Characteristics

We treated 100 patients with SRS who had received prior treatment for primary lung neoplasm and had subsequently developed recurrent lung neoplasm or had persistent or progressive disease after prior therapy. Their median age was 72 years (range 38–92), and median CCI was 7 (range 2–12). The pathology of malignancy of the primary tumor was confirmed with tissue biopsy in all patients. The originally treated tumor was Stage I in 49 patients (49%), Stage 2 in 9 patients (9%), Stage 3 in 30 (30%) and Stage 4 in 8 patients (8%). Surgery with or without adjuvant therapy was used for the treatment of the original primary tumor in 53 patients. The median size of the recurrent tumor was 2.6 cm (interquartile range (IQR) 1.9–3.8). The patient characteristics are summarized in Table 1.

Table 1.

Patient Characteristics

Patient Characteristic	N (%), unless specified
Age (years)
Median	72
Range	38 – 92
Sex
Male	49 (49)
Female	51 (51)
Tumor Histology
• Squamous	29 (29)
• Adenocarcinoma	37 (37)
• Adenosquamous	2 (2)
• NSCLC(subtype undetermined)	29 (29)
• Other	3 (3)
Stage of Primary Disease a
1	49 (49)
2	9 (9)
3	30 (30)
4	8 (8)
Treatment of Primary Disease
Surgery ± Chemotherapy/RT	53 (53)
Chemo or Chemo/RT	32 (32)
RT only	4 (4)
RFA or other	11 (11)
Interval between Treatments for Primary and First Recurrent Disease b
< 1 year	32 (32)
1– 2 years	27 (27)
2– 3 years	15 (15)
>3 years	25 (25)
Recurrence Treated with SRS
First recurrence	70 (70)
2nd, 3rd or 4th recurrence	30 (30)
Other Concurrent Treatment
Chemotherapy	10 (10)
RFA	8 (8)
None	82 (82)

^aStage was not available for 4 patients

^bTreatment interval was not available for 1 patient

Chemo, chemotherapy; NSCLC, nonsmall cell lung cancer; RFA, radiofrequency ablation; RT, radiotherapy

Treatment with SRS

Fiducials were placed in 76 patients using CT-guidance or bronchoscopy. After fiducial placement, sometimes in conjunction with performance of biopsy and/or CT-guided radiofrequency ablation (RFA), 19 patients (19/77; 25%) developed pneumothorax requiring pigtail catheter placement. No patient developed a prolonged air leak (>5 days). The median interval from the treatment of the primary tumor to treatment of recurrent or progressive disease was 18 months (IQR: 9:35 months). SRS was administered to treat a first recurrence in 70 patients (70%), and to treat a second or subsequent recurrence in 30 patients. The median interval from treatment of the primary tumor to the latest SRS treatment of recurrence was 25 months (IQR: 14–46 months). The location of the treated recurrent disease was bronchopulmonary in 87 patients, regional lymph node in 6 patients, and chest wall in 7 patients. For patients who had a prior lobectomy, the majority of the recurrences were in the other lobes, regional nodes, or the chest wall. In patients who had previously undergone a sublobar resection, the majority of treated recurrences were in the same lobe.

Patients were treated with SRS in 1 to 5 fractions (16 Gy to 60 Gy)—47 patients were treated with a total of 60 Gy in 3 fractions; 31 patients (31%) received 20 Gy or less. The median dose received during SRS was 48 Gy in 4 fractions. The median dose in patients who had received prior radiation was also 48 Gy. The 30-day mortality after treatment with SRS/ SBRT was 0%. The morbidity was minimal. One patient complained of significant fatigue and also had chest wall pain after SRS. Ten patients also received chemotherapy around the time of SRS treatment for recurrence.

Overall Survival

During follow-up (median follow-up 51 months for patients who were alive at last follow-up, range 5–123 months), 68 of 100 patients treated with SRS for recurrence had died. The estimated 1-year overall survival for the entire group was 74% (CI 66%−83%), 2-year overall survival was 49% (CI 40%−60%), 3-year overall survival was 40% (CI 31%−51%), and 5-year overall survival was 31% (CI 23% - 43%) (Figure 1). The median overall survival was 23 months (CI 19– 41 months).

Figure 1.

Kaplan-Meier Plot of overall survival for the entire group. The time shown on the x axis is months from SRS. The shaded area depicts the 95% confidence limits for the probability of overall survival. Number of patients at risk are shown above the x axis at 12 month intervals.

Factors Associated with Overall Survival

We analyzed several factors potentially associated with overall survival and found that overall survival was decreased for male patients, patients treated for a 2nd or later recurrence, patients treated non-surgically for their primary tumor, patients with a larger tumor, and patients treated with <60 Gy (Table 2). Age, initial stage of the primary tumor, and disease-free interval from primary therapy to initial SRS were not associated with overall survival. We examined the joint effect of covariates listed in Table 2, with a proportional hazards regression model. The regression model identified three factors that significantly increased the risk: larger tumor size (p < .0001), whether the SRS was applied to a second or subsequent recurrence (p = .0014), and male gender (p = .0128).

Table 2.

Factors associated with overall survival

Covariate	Reference	HR	95% CI	p
Age	IQR: 66 – 78	0.96	0.74 – 1.26	.7853
Sex	Female vs Male	0.57	0.35 – 0.93	.0242
Stage	Stage 2 vs Stage 1	1.00	0.39 – 2.59	.6148
	Stage 3 vs Stage 1	1.43	0.82 – 2.48
	Stage 4 vs Stage 1	1.31	0.54 – 3.17
Original Treatment	Other vs Surgery	1.75	1.08 – 2.84	.0239
Time to initial recurrence	IQR: 9:36 months	0.80	0.58 – 1.11	.1878
SRS for multiple recurrence	2nd or later vs 1st	1.76	1.06 – 2.89	.0274
SRS Dose	20 Gy or less vs 60 Gy	2.48	1.41 – 4.36	.0036
	22 – 50 vs 60 Gy	2.24	1.19 – 4.21
Tumor Size	IQR: 2.0 – 3.8 cm	1.64	1.27 – 2.13	.0002
CCI	IQR: 6 – 9	1.18	0.81 – 1.71	.3978

CCI, Charlson comorbidity index; CI, confidence interval; HR, hazard ratio; IQR, interquartile range; SRS, stereotactic radiosurgery

Comment

In this study, we report our experience in 100 patients with recurrent or persistent/progressive lung cancer who were treated with SRS/SBRT. After SRS for recurrent cancer, the estimated 1-year, 2- year and 5-year overall survival was 74%, 49%, and 31%, respectively, with a median overall survival of 23 months. We also identified factors associated with survival in patients with recurrent or persistent cancer treated with SRS.

Patients with recurrent lung cancer after surgical resection comprise a difficult group of patients to treat. Sugimura reported the outcomes of 390 patients who developed recurrent cancer after complete surgical resection (11). The median post-recurrence survival was 8.1 months, while the 1- and 2-year overall survival was 37% and 17%, respectively. Only 19.5% of patients were treated with surgery (SRS included)for the first disease recurrence. Patients with recurrent disease (any site) treated with surgery (including SRS) had a median survival of 22.9 months; patients treated with chemotherapy and radiation had a median survival of 14.9 months. The median survival was 3.2 months when the recurrence was not treated. In patients with pulmonary recurrence alone, patients whose treatment incorporated surgery had a median survival of 32.8 months vs. 13.4 months for non-surgical treatment and 8.4 months for no treatment.

In our current study, the results of treatment of recurrent lung cancer with SRS/SBRT appear to be improved as compared with the results of external beam radiation therapy (EBRT). Okamato and colleagues reported a median survival of 8 months in 34 patients who were originally treated with radiation, and retreated with EBRT for local recurrence of lung cancer (18). Tada reported overall 1-year and 2-year survival of 26% and 11%, respectively, and a median survival of 7.1 months in 19 patients treated with repeat radiation for locally recurrent lung cancer (19). Tada also reported on the use of radiation treatment for locoregional recurrence in 31 patients after complete resection (20), with 1- year, 2-year, and 4-year overall-survival of 61%, 30%, and 15%, respectively. Standard treatment for recurrent disease involves EBRT, which requires the patient to come for treatment several days a week, for a few weeks, depending on the dose schema and site of treatment. SRS has the advantage of delivering a high biologically effective dose (BED) in 3–5 fractions, typically delivered over a 2-week period; treatment can be completed in a shorter time.

In addition, SRS/SBRT appears to be cost-effective. Lanni and colleagues performed a cost analysis comparing EBRT with SRS in 86 patients with Stage I NSCLC, and found that SRS was significantly less expensive compared with EBRT (P<0.01), and was associated with improved outcomes (21). Similarly, Mitera performed a cost-effectiveness analysis comparing conventional fractioned radiotherapy with SRS in patients with stage I NSCLC, and found the cost effectiveness of SRS was favorable (22).

Several studies have assessed the outcomes of surgery and/ or chemoradiation for recurrent lung cancer. Hishida reported the results of a very selected cohort of 28 patients with recurrent lung cancer after complete resection (23). All the patients underwent surgical resection for the recurrent cancer. The sites of recurrence were lung, brain, adrenal gland, chest wall, stomach, skin, pelvic lymph node, and malar bone. The median follow-up after treatment of recurrence was 33 months. The estimated 1-year, 2-year, and 5-year survival after resection were 89%, 59% and 32%, respectively. The median survival was 25 months. Hung and colleagues reported the outcomes after local recurrence (with or without distant metastases) in Stage I NSCLC patients who had undergone complete resection (4), presenting the results of 74 patients with local recurrence only in detail. The median follow-up after recurrence was 28 months, and 1-year and 2-year survival was 48% and 18.7%, respectively. In multivariate analysis, treatment of recurrent disease was associated with better survival, and surgical treatment was associated with better survival as compared with treatment with chemoradiation, similar to the findings of other studies (11,23). Interestingly, post-recurrence survival was not significantly different between patients with local recurrence only and patients with local and distant recurrences. These results are somewhat similar to the Hishida study, where a solitary extrapulmonary recurrence or distant recurrence was not a negative prognostic factor when treated with complete surgical resection. Comparison of many published series with our current series is difficult due to the heterogeneous nature of both the patient population and the treatments administered. Our series is different from some studies in that we present our experience with SRS treatment of patients who developed limited locoregional/ intrathoracic recurrence and did not include patients with isolated or extensive extrathoracic distant metastatic disease. Our series assessed a complex group of patients with various stages of disease and prior therapies including chemoradiation for advanced disease.

While it is evident that surgical resection is beneficial in highly selected patients with recurrent lung cancer, pulmonary resection may be associated with increased risks particularly in compromised patients who underwent prior lung resection or definitive chemoradiation. In one study evaluating outcomes of surgical resection for local recurrence of NSCLC, the operative mortality was 20% (4). In another study, a CCI of 3–4 was predictive of increased risk of major postoperative complications after lung resection (24). The patients in our study had significant comorbidities with a median CCI of 7, and despite comorbidities, there was no postprocedure mortality after SRS. SRS is applicable for selected patients when surgical resection is not feasible, due to comorbidities, limited pulmonary function from prior resection, or other factors that place them at high risk. Another ablative option which is potentially applicable for recurrent lung cancer is image-guided ablation (25,26) and further work is required to evaluate its role in management.

Dose and toxicity

Dose escalation may allow better control of disease, and Onishi reported a lower local recurrence rate when a BED >100 Gy was used (25). In our current study, the estimated median BED was >100 Gy. It is evident that an increase in dose improves local control; however administering 60–66 Gy in three fractions may also be associated with increased toxicity, particularly when treating central lesions. Timmerman reported excessive toxicity in central tumors treated with SRS (60–66 Gy in three fractions) in a phase II study (26). In order to decrease this risk, we utilized a modified, lower-dose schema (typically 48 Gy in 4 fractions) for central lesions. This differentiates our series from other series reporting results of SRS for the treatment of peripheral Stage I NSCLC. Recurrent disease represents aggressive tumor biology, and the patients in our series had recurrent or progressive disease despite prior treatment including radical resection and definitive chemoradiation. The aggressive tumor biology likely impacted overall survival of the patients. In many of these patients with aggressive tumor biology, the full dose of radiation could not be administered to limit toxicity due to a history of chemoradiation. Notably, with a modified dose schema, we were able to safely treat patients with SRS, despite history of prior radiation.

Strengths and Limitations

The strengths of the study are that, to our knowledge, this is one of the largest studies to date reporting the use of SRS/SBRT/SABR for the treatment of recurrent lung cancer. In addition, the follow-up period (median 51 months) is one of the longest reported in the literature. One interesting aspect of this study is that thoracic surgeons were actively involved in patient selection, placement of fiducials, obtaining tissue diagnoses, delineation of targets for SRS, and treatment planning, in close collaboration with the radiation oncologists and physicists who comprise our radiosurgical team. The thoracic surgeons’ expertise with regard to the thoracic anatomy, physiologic assessment of the patient, and the ability to make an evaluation of the surgical risk is valuable in the selection of patients for SRS.

The current study does have limitations including those common to retrospective studies, such as selection bias. The patients treated in this study also comprise a very heterogeneous group, a selected group with limited regional or intrathoracic recurrence, and the dose schema that was used in this study varied. Further, this study was not a randomized study, and does not include a control group. Finally although the follow-up duration is longer than that of most published studies (3,11,23), longer follow-up is needed for a more complete evaluation of survival endpoints.

Future Directions

There are several factors which merit further investigation including optimal patient selection, appropriate dose and fractionation, balancing the efficacy of the treatment with the toxicity, and evaluating the role of a multimodality approach with the addition of targeted therapies in conjunction with SRS/SBRT and/or surgery (13,14). In addition, further prospective studies are needed, and are currently in progress in our institution (NCT01345552) and others, to investigate SRS/SBRT for recurrent disease (13). Further work is also required to investigate the role of RFA and other thermal ablative therapies in the management of patients with recurrent lung cancer(27,28).

Summary

In summary, we report the use of SRS in 100 patients with limited recurrent or persistent lung neoplasm. The management of recurrent lung cancer can be complex, and in selected circumstances surgery, radiation and chemotherapy may play a role in its management. Our experience indicates that SRS is safe, even in the setting of prior radiation, and offers an alternative modality in selected patients with recurrent oligometastatic lung cancer. SRS/SBRT/SABR may be a useful modality for treatment of selected patients with recurrent lung cancer. Thoracic surgeons should actively participate in SRS and continue to evaluate the efficacy of this treatment strategy for this selected group of patients along with oncologists.