Lung reirradiation with stereotactic body radiotherapy (SBRT)

Abstract

Introduction

There is limited data on the use of SBRT in reirradiation of lung tumors. We reviewed outcomes following SBRT after previous thoracic radiotherapy at the University of Minnesota Medical Center.

Methods

From August 2006 through October 2012, fourteen lung tumors in thirteen patients with either biopsy confirmed or presumed non-small cell lung cancer in patients who were medically unable to undergo biopsy, were retreated with SBRT. Eligible patient charts were reviewed to evaluate survival, recurrence patterns and toxicity following reirradiation.

Results

The median age of patients at the time of SBRT was 67.9 years. The median duration of follow-up was 11.4 months. Ten patients received prior conventional thoracic irradiation (median dose 6120 cGy). Two patients received prior SBRT with curative intent. The median time to reirradiation with SBRT was 19.7 months. Following reirradiation with SBRT, four patients (33%) are alive and disease free. Eight patients (67%) experienced progressive disease. There were five distant and two regional recurrences. There was one isolated local recurrence. Local control was 92% with a median survival of 24 months (95% CI: 8-38 months). 1- and 2-year overall survival were 80% (95% CI: 41%-95%) and 36% (95% CI: 6%-68%) respectively. There was one grade 2 and one grade 3 toxicity. No grade 4 or 5 toxicities were seen.

Conclusions

SBRT is a reasonable salvage therapy for lung tumor recurrence or second primary lung malignancy in patients previously treated with thoracic radiotherapy, offering good local control and resulting in acceptable toxicity. Further evaluation of this treatment option is warranted.

1. INTRODUCTION

Local failure after the treatment of early staged, inoperable lung cancer with 3D conformal radiotherapy approaches 40% [1-3]. In addition, development of a metachronous lung cancer has been estimated at 1-10% per year [4]. Local relapses or second primary lung malignancies in patients who have previously received thoracic radiation therapy presents a difficult management situation for which there are not many treatment options. Surgical salvage and or systemic therapy are dependent on patients’ pulmonary function and performance status, and often not recommended. Salvage chemotherapy alone, for those who are eligible, offers a low probability of disease control [5,6]. Reirradiation with conventional radiotherapy has been used for palliation of symptoms with some success but seldom results in long term control [7-13].

Stereotactic body radiation therapy (SBRT) offers excellent local control in patients with early stage non-small cell lung cancer (NSCLC) who are deemed medically inoperable or refuse surgery, with rates consistently above 85% [14- 16]. In patients with 1-3 pulmonary metastases, SBRT has demonstrated excellent local control at 2 years, approaching 100% [17]. Due to the increased therapeutic ratio of SBRT, its role is being investigated in the treatment of locoregional relapses or second primary lung malignancy in patients who have previously received thoracic radiation therapy. There are still limited data available on the use of SBRT in reirradiation of lung tumors [18-21]. We reviewed the outcomes of patients treated with SBRT after previous thoracic radiotherapy at the University of Minnesota Medical Center.

2. MATERIALS AND METHODS

2.1 Patients

All patients treated at the University of Minnesota Medical center with thoracic SBRT were reviewed. From August 2006 through October 2012, fourteen tumors in thirteen patients with either biopsy confirmed NSCLC or patients with presumed NSCLC who were medically unable to undergo biopsy, were retreated with SBRT. After receiving a single fraction, one patient did not return to complete the prescribed course of SBRT or follow-up, and was therefore excluded from analysis. All patients were medically inoperable, as determined by thoracic surgery. Each were staged clinical T1-2aN0M0. Patients were exculuded for primary tumor size greater than 5cm, Zubrod performance status 4, as well as any positive nodal or distant metastasis. Patients were not exclded by any initial diagnosis or prior radiation delivered. Patients previously received thoracic radiation for either Stage I-III NSCLC (n=10), Limited Stage SCLC (n=1) or Stage IV Wilms tumor (n=1). All patients received positron emission tomography (PET)/computed tomography (CT) scans prior to SBRT to exclude regional and/or metastatic disease. Four of the fourteen (30.8%) tumors were centrally located as defined by Timmerman et al. [22].

2.2 Treatment

SBRT immobilization consisted of fixation in a customized stereotactic body frame with abdominal compression. Planning image acquisition was performed with use of four-dimensional (4D) CT. An internal gross tumor volume (ITV) was created by combining gross tumor volume (GTV) on normal respiration, maximal intensity projection, and average projection of the planning 4D-CT. A planning tumor volume (PTV) was created by expanding the ITV by 5-8 mm. Block edge was placed at the PTV expansion. Normal tissue constraints used on all patients followed guidelines as outlined in Table 1 and included consideration of the patients’ previous radiation treatments. Spinal cord max point dose listed as a range in Table 1, for example, was determined after evaluating prior cord dose. Considering the SBRT PTV was in previously irradiated, and therefore already damaged lung, whole lung dose constraints were not restricted. Planning was performed on Pinnacle™ treatment planning system with collapsed-cone (CC) convolution calculation algorithm. Treatment was delivered with 10 non-coplanar 6 megavoltage photon beams. Patients were treated with five fractions of 9-10 Gy, for a total prescription dose of 45-50 Gy. Dose was prescribed to the 80-95% isodose line covering the planning tumor volume (PTV) with tissue heterogeneity corrections in all but one patient. Although the prescription isodose was at the discretion of the treating physician, it was based on the optimum coverage of the ITV and PTV. Fractions were separated by 1-3 days. The maximum number of fractions per week was three.

Table 1.

Normal tissue constraints for SBRT planning

Organ	Volume (cc)	Total dose (dose per fraction) (Gy)
Spinal cord	Max Point	20 (4)  30 (6)
Brachial Plexus	Max Point	32 (6.4)
Trachea and bronchus	Max Point	30 (6)
Esophagus	<5	27.5 (5.5)
Heart	Max Point	40 (8)
Skin	Max Point	32 (6.4)

2.3 Follow-up and toxicity assessment

Patients were evaluated with initial chest CT scan six weeks after the completion of SBRT. Subsequently patients underwent CT scanning alternating with PET/CT scanning every three months for 2 years and every six months during years 3-5 following treatment. The endpoints evaluated were local control, disease recurrence patterns, overall survival and toxicity. Local recurrences were defined as greater than 20% increase in treated tumor size on CT or an increase in the maximum standard uptake value on PET. Regional recurrences were defined as development of new lesions on PET/CT or positive biopsy outside of the treated PTV, in the ipsilateral lung or mediastinal lymph nodes. Distant recurrences were defined as progressive abnormalities on PET/CT or positive biopsy in the contralateral lung or anywhere outside the thorax. Follow-up time was calculated from the time of last SBRT fraction until the last patient contact by a medical provider for oncologic surveillance. The time of disease recurrence or death was defined at the time of first identified abnormality on imaging or date of death, respectively. NCI-Common Terminology Criteria for Adverse Events (CTCAE) v4.0 was used to evaluate toxicity.

3. RESULTS

3.1 Patients

Patient characteristics are detailed in Table 2. There were four women and eight men. The median age of patients at the time of reirradiation with SBRT was 67.9 years (range 45.9-86.7 years). The median duration of follow-up for patients after reirradiation was 11.4 months (range 1.56-38.3 months). Ten patients received prior chemotherapy and conventional thoracic radiation with a median dose of 6120 cGy (range 1200-7000 cGy). Two patients received previous SBRT for early stage NSCLC with curative intent. Only patients who received prior SBRT with recurrence adjacent to the previous SBRT field (within the same lobe of the lung) and were reirradiated with SBRT were included in this report. Patients who received prior SBRT to another lobe or contralateral lung were not included in this report. The median time to reirradiation with SBRT was 19.7 months (range 4.7-84.7 months).The median size of tumor recurrences treated with SBRT was 4.6 cm³ (range 1.0-28.4 cm³). The median size of the treatment PTV was 37.9 cm³ (range 19.5-119.6 cm³). Eight patients received 9 Gy per fraction for 5 fractions (BED for tumor α/β=10, was 85.5.) In seven patients dose was prescribed to the 85% isodose line. In one patient, due to a large volume, dose was prescribed to the tumor isocenter. Five patients received 10 Gy per fraction for 5 fractions prescribed to the 80-95% isodose line (BED for tumor α/β=10, was 100.). Adjuvant chemotherapy was reserved for patients who had disease progression following SBRT.

Table 2.

Eligible Patient Characteristics (n=12)

Patient Characteristics	Median (Range)
Age (years)	67.9 (45.9-86.7)
Prior thoracic RT dose (Gy)	61.2 (12-70)
Interval between RT (mo)	19.7 (4.7-84.7)
Retreatment GTV (cm3)	4.6 (1.0-28.4)
Retreatment CTV/ITV (cm3)	5.6 (1.4-43.3)
Retreatment PTV (cm3)	37.9 (19.5-119.6)
	N (%)
Male	8 (66.7%)
Female	4 (33.3%)
Initial Diagnosis
NSCLC	10 (83.3%)
SCLC	1 (8.3%)
Wilms Tumor	1 (8.3%)
Prior thoracic RT
CF	9 (75.0%)
HF	1 (8.3%)
SBRT	2 (16.7%)
Location Recurrence
LUL	5 (38.5%)
LLL	3 (23.1%)
RUL	2 (15.4%)
RML	3 (23.1%)
RLL	0
Retreatment SBRT Rx
9 Gy x 5	8 (61.5%)
10 Gy x 5	5 (38.5%)

Abbreviations: RT = Radiation therapy; NSCLC = Non-small cell lung cancer; SCLC = Small cell lung cancer; CF = conventional fractionation; HF = Hyperfractionation; SBRT = Stereotactic Body Radiation Therapy

3.2 Treatment Failure and Survival

Following reirradiation with SBRT, four of the twelve (33%) are alive and disease free at a median of 14 months from the completion of SBRT. The details of patients alive and free of disease can be seen in Table 3. Two of the twelve patients are alive with progressive disease; both are controlled locally and have both regional and distant progression. Eight patients (67%) had progressive disease. Three patients had isolated distant recurrences, and two failed regionally and distantly. There were two isolated regional recurrences, and one patient developed an isolated local recurrence. Overall local control of the retreated tumors was 92%. The sole local recurrence occurred following 4500 cGy in 5 fractions. There have been six deaths, all of which were in patients with progressive disease, at a median of 17.6 months from completion of SBRT. Patient outcomes after re-irradiation with SBRT are outlined in Table 4. The estimated median survival time is 24 months (95% CI: 8-38 months). 1- and 2-year overall survival are 80% (95% CI: 41%-95%) and 36% (95% CI: 6%-68%) respectively. Kaplan-Meier curve for survival from time of retreatment with SBRT (survival) and a curve showing the time from retreatment to documented local failure (local control) are plotted in Figure 1.

Table 3.

Patients Alive and Free of Disease following retreatment with SBRT

Patient	Initial Diagnosis	Initial Radiation Dose	Interval (years)	SBRT Fractionation
1	LS-SCLC	4500 cGy BID	5.93 6.96	5000 cGy/5 fx 5000 cGy/5 fx
2	IA NSCLC	7000 cGy	1.64	5000 cGy/5 fx
3	IA NSCLC	4800 cGy/ 4 fx (SBRT)	1.77	5000 cGy/5 fx
4	IA NSCLC	6840 cGy	0.84	4500 cGy/5 fx

Abbreviations: LS-SCLC=Limited Stage-Small Cell Lung Cancer; NSCLC = Non-small cell lung cancer; BID= twice daily treatments; fx= number of fractions

Table 4.

Patient outcomes following SBRT retreatment

SBRT outcome	N (%)
Alive	6 (50.0%)
NED	4 (33.3%)
Distant + Regional Recurrence	2 (16.7%)
Deceased	6 (50.0%)
Local Recurrence	1 (8.3%)
Regional Recurrence	2 (16.7%)
Distant recurrence	3 (25.0%)

Abbreviations: NED = No evidence of disease

Figure 1.

Local Control and Overall Survival after reirradiation with SBRT.

3.3 Toxicity

All analyzed patients completed SBRT as scheduled without measurable acute toxicity. There was one grade 2 toxicity, described as lobar atelectasis occurring six months following SBRT. This occurred in a patient initially diagnosed with Stage IA NSCLC. She previously had been treated with 6480 cGy in 38 fractions in 2009 at an outside hospital. The left and right lung V20 were 30% and 2%, respectively. There was no combined lung Dose Volume Histogram (DVH) for her initial radiation therapy. There was one patient with grade 3 pneumonitis, described as progressive dyspnea and oxygen dependence four months following SBRT. This patient was initially diagnosed with stage IIIA (T4N1M0) NSCLC. She recived 6120 cGy in 2001 at an outside hospital. There were no available records of this initial radiation therapy. Oxygen dependence may have been related to progressive COPD or radiation pneumonitis. There were no grade 4 or 5 toxicities.

4. DISCUSSION

Treatment of local relapses or second primary lung malignancies in patients who have previously received thoracic radiation therapy is a unique situation for which there are few treatment options. Due to its potential to eradicate disease locally while delivering limited dose to nearby normal tissue, the role of SBRT is being evaluated in these situations.

Consistent with previous reports on retreatment of lung cancer with SBRT [18-21], our results suggest encouraging local control and limited toxicity. A summary of results with SBRT reirradiation are shown in Table 4. In four previously published series, local control following reirradiation with SBRT ranges from 52 to 92%. Grade 3 toxicity ranges from 0 to 33%, and appears acceptable. Only one series documented grade 4 to 5 toxicity, following reirradiation with SBRT, and all of these were central tumors previously receiving SBRT [19].

Our series most closely mirrors that of the largest series by Kelly et al. Kelly and colleagues reported 36 patients treated with SBRT after prior conventionally fractionated thoracic radiation [18]. Regiments included 50 Gy in 4 fractions and 40 Gy in 4 fractions. After a median follow up of 15 months, local control was 92%. There were 12 patients (33%) who experienced grade 3 toxicity, but no grade 4 or 5 toxicity occurred. The earlier patients in our series were treated with 45 Gy in 5 fractions, and the latter patients have been treated with 50 Gy in 5 fractions, which is our currently preferred fractionation for this population. Similarly, local control in our series was 92%, and we believe this correlates with a higher biological dose.

Seung et al. described 8 patients who had local recurrence after definitive radiation for lung cancer [20]. SBRT was delivered in four different fractionation schemes, at the discretion of the treating physician and included 12 Gy x 4, 10 Gy x 5, 8 Gy x 5, or 20 Gy x 3. At a median follow up of 18 months local control was 86%. There was no observed grade 3 or higher toxicity. Peulen et al. reported 32 lesions in 29 patients treated with SBRT after prior lung SBRT [19]. Several fractionation schemes were prescribed; the most common regimens were 15 Gy x 2-3 and 10 Gy x 4 with dose prescribed to the 67% isodose. After a median follow up of 12 months, local control was 52%. There were eight patients (28%) who experienced grade 3-4 and three patients (10%) who experienced grade 5 toxicity. All patients with grade 4-5 toxicity had central tumors. Local failure was higher in other series, in addition to the surprisingly large number of therapy related deaths. Toxicity maybe attributable to both the location of tumors treated as well as the maximum biological dose delivered. Trakul et al. reported 17 lesions in 15 patients located in previously irradiated regions (either prior conventionally fractionated or stereotactic radiotherapy) treated with SBRT [21]. The most common prescribed regimens were 20-25 Gy x 1 and 10 Gy x 3-4. With a median follow up of 15 months, local control was 65.5%. There was no grade 3-5 toxicity in the reirradiated patients.

Table 5.

Reirradiation of lung cancer with Stereotactic Body Radiation Therapy (SBRT)/ Stereotactic Ablative Radiotherapy (SABR). A summary of published reports.

Institution	n	Number tumors	Most common regimens (Gy/Fx)	Median F/u (mo)	LC (%)	Median OS (mo)	1y OS (%)	2y OS (%)	3 y OS (%)	Grade 3 toxicity (%)	Grade 4-5 toxicity (%)
MDA Kelly et al.	36	36	40/4, 50/4	15	92			59		33	0
Oregon Seung et al.	8	8	48/4, 50/5 40/5, 60/3	18	86					0	0
European Peulen et al.	29	32	30/2, 45/3 40/4	12	52	19.3	59	43	23	24	14
Stanford Trakul et al.	15	17	20-25/1, 30/3 40/4	15	65.5		80			0	0
U of MN	12	13	45/5, 50/5	11.4	92	24	80	36		8	0

In select patients with recurrent lung cancer, re irradiation with SBRT is both feasible and effective, with high local control rates and low grade 3-5 toxicity. Preventing a local recurrence may translate into improved survival. Among published studies, including ours, 1- and 2- year overall survival estimates range from 59-80% and 36-59% respectively. However, the majority will eventually fail regionally and/or distantly, and ultimately die from their disease. The greatest challenge will be to differentiate those who are most likely to see local control translate into better survival.

The current study has limitations including selection bias, inherent to retrospective studies. The sample size is small and heterogeneous, including patients with progressive recurrent lung cancer as well as second primary early stage lung cancer. The doses, treatment modality and timing of their previous thoracic irradiation are variable. Patients also had significant comorbidities that rendered them medically inoperable. Finally, follow up is relatively short, with a median of 11.4 months, and patients remain at risk for both the development of late toxicity and disease recurrence. Nonetheless, our findings are encouraging.

CONCLUSIONS

SBRT may be a reasonable treatment option for lung tumor recurrence or second primary lung malignancy in patients previously treated with thoracic radiotherapy, offering good local control and resulting in acceptable toxicity. However, controlling regional and systemic progression of disease remains a challenge. Considering the published retrospective reports on SBRT following previous thoracic radiotherapy, which support its potential for success with reasonable toxicity, prospective studies for salvage or second primary NSCLC, should be developed.