Stereotactic Ablative Radiotherapy for Centrally Located Early Stage Non Small-Cell Lung Cancer. What We Have Learned

Transcription

1 State of the Art: Concise Review Stereotactic Ablative Radiotherapy for Centrally Located Early Stage Non Small-Cell Lung Cancer What We Have Learned Joe Y. Chang, MD, PhD*, Andrea Bezjak, MD, and Françoise Mornex, on behalf of the IASLC Advanced Radiation Technology Committee Abstract: Image-guided stereotactic ablative radiotherapy (SABR; also called stereotactic body radiotherapy or radiosurgery) has become a standard treatment for medically inoperable peripherally located stage I non small-cell lung cancer (NSCLC) and can achieve local control rates in excess of 90%. However, the role of SABR for centrally located lesions remains controversial because of concerns about the potential for severe toxic effects. When cutting-edge technologies and knowledge-based optimization of SABR planning that considers both target coverage and normal tissue sparing are used, some patients with central lesions can be safely and effectively cured of early stage NSCLC. However, delivery of ablative doses of radiation to critical structures such as bronchial tree, esophagus, major vessels, heart, and the brachial plexus/phrenic nerve could produce severe, potentially lethal toxic effects. Here, we address the current understanding of indications, dose regimens, planning optimization, and normal tissue dose-volume constraints for using SABR to treat central NSCLC. Key Words: SABR, SBRT, Central lesion, Early stage lung cancer, Dose-volume constraints, Toxicity, Survival. (J Thorac Oncol. 2015;10: ) Image-guided stereotactic ablative radiotherapy (SABR; also called stereotactic body radiotherapy) delivers focused, ablative radiation doses to a target while minimizing the dose to surrounding critical normal structures. It has produced local control rates in excess of 90% for patients with early stage non small-cell lung cancer (NSCLC) when the biological *Department of Radiation Oncology, University of Texas MD Anderson Cancer Center, Houston, Texas; Department of Radiation Oncology, Princess Margaret Cancer Center, University of Toronto, Toronto, Canada; and Department of Radiation Oncology, Centre Hospitalier Lyon Sud, Université Claude Bernard, Lyon, France. Disclosure: The authors declare no conflict of interest. Address for correspondence: Joe Y. Chang, MD, PhD, Department of Radiation Oncology, Unit 97, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX jychang@mdanderson.org DOI: /JTO ISSN: /15/ effective dose (BED) to the planning target volume (PTV) is greater than 100 Gy. 1 4 Population-based studies and propensity-matched analyses demonstrate that SABR produces overall survival (OS) and disease-specific survival rates similar to those after lobectomy, and better OS than conventional radiation. 5 7 Currently, SABR is replacing conventional radiotherapy for the treatment of patients with medically inoperable, peripherally located stage I lung cancer 1 7 or multiple-primary early stage NSCLC. 8 The role of SABR in oligometastatic disease 9,10 or recurrent disease in the lung parenchyma 11 is being investigated, and clinical findings have been promising. However, the use of SABR for lesions that are centrally located and thus close to critical normal structures within the thorax is controversial A low rate of freedom from grade 3 5 toxicity (54% at 2 years) in one study in which Gy in three fractions was used 12 and reports of severe toxicity including fatal hemoptysis, pneumonitis, and fistula 14,15 have discouraged many from using this technique. However, other SABR regimens such as Gy in four to five fractions, 13,17 60 Gy in eight fractions, 16,18 and Gy in 10 fractions 19 have been shown to be safe and well tolerated in carefully selected patients. Whether SABR should be considered for central lesions only in the context of clinical protocols continues to be debated because of uncertainty about potential severe adverse effects. Recently, however, findings from several large clinical series reporting long-term clinical outcomes after SABR for central lesions support the use of this technique for curative treatment of centrally located early stage NSCLC, and the authors of these studies have proposed dosevolume constraints for some regimens based on their clinical experience Notably, however, low BEDs (less than 100 Gy) resulted in higher rates of local failure, whereas higher ablative doses to critical structures can cause severe or even fatal toxicity. Thus, maintaining balance between adequate target coverage and minimizing doses exposure to critical structures remains crucial. Unfortunately, published findings are quite inconsistent regarding what constitutes a central lesion, indications for SABR, SABR regimens, techniques, dose prescription methods, calculation algorithms, image guidance, and definition of local failure after SABR. Proper selection of cases, based on tumor location and the ability to meet normal tissue constraints, the careful choice of SABR regimens and Journal of Thoracic Oncology Volume 10, Number 4, April

2 Chang et al. Journal of Thoracic Oncology Volume 10, Number 4, April 2015 techniques, and the use of image guidance are all crucial for adequate target coverage while avoiding overdoses to critical structures. In this article, we review the existing literature and provide guidelines for the use of SABR for challenging cases. These guidelines can be used not only for early stage central lesions but also for oligometastastic central lesions, another emerging clinical practice currently under investigation. DEFINITION OF CENTRAL AND INDICATIONS FOR SABR The definition of what constitutes a central lesion has varied in published studies, and includes (1) a tumor within 2 cm in all directions of the proximal bronchial tree (carina, right and left main bronchi, and bronchial tree to the second bifurcation), as in RTOG 0236; (2) a tumor within 2 cm in all directions of any mediastinal critical structure, including the bronchial tree, esophagus, heart, brachial plexus, major vessels, spinal cord, phrenic nerve, and recurrent laryngeal nerve (Fig. 1); and (3) a tumor within 2 cm in all directions around the proximal bronchial tree and immediately adjacent to mediastinal or pericardial pleura ( PTV touching the mediastinal pleura ) as in RTOG 0813 (a phase I dose-escalation study of SABR for central lesions). Definition 2 has been used most often in recent studies because of reported toxicity to lung and other critical structures such as esophagus, heart, and nerves etc. after SABR. Therefore, we recommend the definition 2 (Fig. 1) in our routing clinical practice. However, critical normal tissues should be respected when SABR is considered for central lesions. Tumors located at or involving the hilar structures, or tumors invading the bronchial tree or mediastinum, although physically central, are not considered safe targets for central SABR regimens, e.g., those involving delivery of a BED greater than 100 Gy in five or even up to 10 fractions. Treatment involving additional fractions should be considered for such lesions. DOSE REGIMENS, TOXIC EFFECTS, AND NORMAL TISSUE DOSE-VOLUME CONSTRAINTS The proximal bronchial tree was long considered a no fly zone for high-dose radiation, meaning 60 Gy in three fractions (54 Gy in three fractions with heterogeneity correction, BED α/β 10 = 151 Gy, BED α/β 3 = 378 Gy), since early results of a phase II trial published in 2006 indicated that patients with tumors within a 2 cm radius of the trachea and bronchial tree were 11 times more likely to experience grade 3 5 lung toxicity than were those with tumors outside this zone. 12 Indeed, a 4-year update of that study revealed lung toxicity rates of 27.3% for patients with central tumors versus 10.4% for those with peripheral tumors. 24 Others also reported severe adverse effects such as bronchial stenosis, fatal hemoptysis, or fistula after SABR for central tumors when ablative dose were delivered to critical structures. 14,15 To clarify the tolerance of critical structures, the RTOG conducted a phase I prospective study (RTOG 0813) to evaluate the toxicity of SABR in doses escalated from 50 Gy to 60 Gy, in five fractions delivered every other day (except over weekends), with at least 40 hours between treatments; when this review was written, the results of this study were still pending. Since 2006, other SABR regimens have been reported for central lesions. Xia et al. reported using 70 Gy prescribed to the gross tumor volume (GTV) in 10 fractions (BED α/β 10 = 119 Gy, BED α/β 3 = 233 Gy) and achieved a local control rate of 93% at 3 years. 19 No grade 3 or higher toxicity was reported for either central or peripheral lesions. These findings support reconsideration of the use of SABR for centrally located tumors when appropriate dose-volume constraints for normal tissues are used. Investigators at MD Anderson FIGURE 1. Recommended definition of central lesion: a tumor within 2 cm in all directions of any mediastinal critical structure, including the bronchial tree, esophagus, heart, brachial plexus, major vessels, spinal cord, phrenic nerve, and recurrent laryngeal nerve. 578

3 Journal of Thoracic Oncology Volume 10, Number 4, April 2015 Stereotactic Ablative Radiotherapy reported a preliminary version of such constraints for central lesions in 2008, along with outcomes after modern fourdimensional computed tomography-based SABR planning and on-board volumetric image verification for a regimen involving delivery of Gy in four fractions on consecutive days. 13 When 50 Gy was delivered (BED α/β 10 = Gy, BED α/β 3 = 258 Gy), the local control rate at 2 years was 100% and no incidents of fatal toxicity were noted. In contrast, 40 Gy in four fractions was associated with poor local control (57%) and therefore was subsequently abandoned. Moreover, one patient who had received 40 Gy to the brachial plexus experienced severe brachial plexus neuropathy. At that time, preliminary dose-volume constraints were proposed and the protocol was amended to allow compromise of PTV coverage if needed to spare critical structures. An updated report from MD Anderson in 2014 describing the use of SABR with 50 Gy in four fractions for 100 patients with tumors near the bronchial tree or other critical mediastinal structures and brachial plexus showed that median survival time (58 months) and local control rates (96% at 2 years) were comparable to those for peripheral lesions treated with SABR to 50 Gy in four fractions. 20 The incidence and severity of radiation pneumonitis and chest wall pain were also similar; no patient experienced grade 4 or 5 toxicity. The previously proposed dose-volume constraints for each critical structure for a regimen of 50 Gy in four fractions were analyzed and modified based on the findings from the expanded cohort; these dose-volume constraints are shown in Table 1. By way of comparison, the constraints for the brachial plexus, bronchial tree, major vessels, esophagus were lower than in the previous report, and constraints for chest wall and heart were added. Patients in whom these dose-volume constraints could not be met were treated with 70 Gy in 10 fractions, which led to similar local control with tolerable toxicity. 21 This 70 of 10 fractionation reduced late chest wall and brachial plexus toxicity, but it was not clear if it also minimized acute radiation pneumonitis and esophagitis. Moreover, a patient with tumor invading the hilum treated with 70 Gy (hilar D max = 83 Gy) had fatal hemoptysis, which led to the recommendation that tumors that actually invade central structures should not be treated with high-bed schedules. In Europe, Lagerwaard et al. 1 also proposed adaptive dose regimens based on tumor location and reported promising outcomes in That group subsequently reported their results from using 60 Gy in eight fractions (BED α/β 10 = 105 Gy, BED α/β 3 = 210 Gy) for 63 patients with central lesions; at 3 years, the local control rate was 93% and no grade 4 or 5 toxicity had occurred and no dose volume-constraints were provided/proposed. 16 A group at Memorial Sloan Kettering also reported their use of SABR based on intensity-modulated radiation therapy (IMRT) for 108 patients, most of whom received 45 Gy in five fractions (α/β 10 = 85.5 Gy, BED α/β 3 = 180 Gy); the local control rate at 2 years was 79%. 22 However, severe esophageal toxicity, including fistula in a patient with an esophageal D max of 46 Gy, was reported and 6 of 12 patients for whom the median esophageal D max was 30 Gy developed grade greater than or equal to two esophagitis when the PTV overlapped the esophagus. Two patients developed fatal hemoptysis, one with tumor involving the hilum and a maximum dose to the right bronchial tree of 47 Gy in five fractions, and the other with tumor encasing the left superior segmental bronchus with a maximum bronchial tree dose of 48 Gy in five fractions. Another group at the Cleveland Clinic similarly reported a case of esophageal fistula when the esophageal point dose exceeded 51 Gy and the V 48 was >1 cm Investigators at the William Beaumont Hospital compared clinical outcomes for 125 patients after regimens that varied from 48 to 60 Gy in four to five fractions (BED α/β 10 = Gy, BED α/β 3 = Gy) for central versus peripheral lesions using propensity-matched analysis. 23 No significant differences were found in OS or severe toxicity. However, in other reports, a similar dose (40 60 Gy in five fractions) led to fatal hemoptysis when a D max of greater than 50 Gy was delivered to the pulmonary artery and bronchus. 26,27 Local control of lung tumors seems to require a BED of no less than 100 Gy (calculated with the linear quadratic model). 13,22,28 However, doses with BED exceeding 100 Gy represent a double-edged sword in terms of balancing tumor control with normal tissue damage. Therefore, it is crucial to balance coverage of the gross tumor to a BED greater than 100 Gy with dose constraints to avoid damage to nearby critical structures. Based on these findings from several large series, regimens of Gy in four fractions or Gy in five fractions can be considered for centrally located lesions if normal tissue dose-volume constraints can be met; alternatively, 60 Gy in eight fractions or 70 Gy in 10 fractions can also be considered, particularly for patients for whom dose-volume constraints cannot be met using less than or equal to five fractions. More fractionated SABR regimens, such as 70 Gy in 10 fractions, may be preferred to avoid chronic toxicity such as chest wall pain and brachial plexus neuropathy. Even when 8 10 fractions are used, SABR may not be appropriate for tumors invading critical structures such as the esophagus, hilum, major vessels, main bronchus, heart, or brachial plexus. SABR also should not be used for lesions abutting or invading the phrenic, vagus, or recurrent laryngeal nerves. 29 Normal tissue dose-volume constraints will depend on the dose regimens used. Published dose-volume constraints for regimens of 50 Gy in four fractions, 70 Gy in 10 fractions, are shown in Table 1. For Gy in five fractions, RTOG 0813 trail is ongoing and the dose-volume constraints are listed in Table 1 but it still needs clinical data from RTOG 0813 to validate its safety. These guidelines can be used to guide clinical practice for specific dose regimens, and further modifications should be considered when more clinical data particularly prospective studies become available. Notably, severe side effects could continue to appear 1 to 2 years after treatment, and caution must be used when deeming a regimen safe if the median follow-up time is less than 2 years. Although the clinical reports reviewed in the preceding section suggest that most patients with central lesions can be treated safely with SABR as long as prudent planning principles are respected, it should be kept in mind that SABR toxicity could be higher than is currently recognized because of competing risks of death in these patients. Furthermore, 579

4 Chang et al. Journal of Thoracic Oncology Volume 10, Number 4, April 2015 TABLE 1. Normal Tissue Dose-Volume Constraints for Central Lesions Treated with SABR MD Anderson Experience RTOG 0813 Regimen 50 Gy in Four Fractions 70 Gy in 10 Fractions Gy in Five Fractions Dose Constraints [References] Normal Tissue Volume Max Dose Volume Max Dose Volume Max Dose Endpoint to be Avoided Lung Total MLD 6 Gy 20 MLD 9 Gy 20 V 12.5 < 1500 cm 3 Lung function/pneumonitis V 5 30% 20 V 40 7% 21 V 13.5 <1000 cm 3 V 10 17% 20 V 20 12% 20 V 30 7% 20 Ipsilateral imld 10 Gy 20 NA NA iv 10 35% 20 iv 20 25% 20 iv 30 15% 20 Trachea V 35 1 cm 320 V 40 1 cm 321 D max < 60 Gy 21 V 18 <4 cm 3 D max < 105% of PTV Pneumonitis/stenosis/fistula Bronchial tree V Gy 20 V 50 < 1 cm 321 D max < 60 Gy 21 V 18 < 4 cm 3 D max < 105% of PTV Pneumonitis/hemoptysis Hilar major vessels V Gy 20 V 50 < 1 cm 321 D max < 75 Gy 21 Pneumonitis/hemoptysis Other chest great vessels V Gy 20 V 50 < 1 cm 321 D max < 75 Gy 21 V 47 < 10 cm 3 D max < 105% of PTV Pneumonitis/hemoptysis Esophagus V Gy 20 V 40 1 cm 321 D max 50 Gy 21 V 27.5 < 5 cm 3 D max < 105% of PTV Esophagitis/stenosis/fistula Heart/pericardium V 40 1 cm 320 D max 45 Gy 20 V 45 < 1 cm 321 Dmax 60 Gy 21 V 32 <15 cm 3 D max <105% of PTV Cardiac disorder/pericarditis V 20 5 cm 349 Brachial plexus V cm 320 D max 35 Gy 20 V 50 < 0.2 cm 321 D max < 55 Gy 21 V 30 < 3 cm 3 D max < 32 Gy Brachial neuropathy Spinal cord V Gy 20 V 35 1 cm 321 D max < 40 Gy 21 V 22.5 < 0.25 cm 3 D max < 30 Gy Myelitis V 13.5 < 0.5 cm 3 Chest wall/skin V cm 3 (cw) 50 V cm 321 D max 82 Gy 21 V 30 < 10 cm 3 D max < 32 Gy Chest wall pain/skin toxicity V cm 3 (skin) 50 V cm 321 V cm 321 MLD, mean lung dose; V x, volume of tissue exposed to x Gy or more; D max, maximum dose; PTV, planning target volume; NA, not available; SABR, stereotactic ablative radiotherapy. 580

5 Journal of Thoracic Oncology Volume 10, Number 4, April 2015 Stereotactic Ablative Radiotherapy attributing clinical symptoms to SABR toxicity as opposed to the natural progress of comorbid conditions can be extremely difficult. Finally, particular caution should be used with regard to additional insults to already damaged normal tissues, such as biopsy, to prevent severe or even fatal complications. DOSE PRESCRIPTION AND TREATMENT- PLANNING OPTIMIZATION Local tumor control after SABR seems to depend on the dose delivered to the PTV, and a BED greater than 100 Gy seems to be required to achieve optimal local control. However, even when the prescribed dose is the same, the actual dose delivered to the PTV can vary drastically depending on where (to which isodose line or point) the dose was prescribed. For example, a dose of 50 Gy prescribed to the 60% isodose line could deliver a maximum dose of 96 Gy to the PTV, with a total mean PTV dose of 83 Gy; in contrast, a dose of 50 Gy prescribed to the 100% isodose line could deliver only 32 Gy to some of the PTV, with a total mean PTV dose of 50 Gy, which is more than 30 Gy less than when dose is prescribed to the 60% isodose line (Fig. 2). The total mean PTV dose is even lower if the dose is prescribed to the isocenter. In addition, dose calculation algorithms used by treatment-planning systems, such as pencil beam versus Monte Carlo calculations, can also cause dose variations of up to 15% or more. Therefore, it is very important to make sure that the PTV receives adequate dose coverage and that an appropriate dose calculation algorithms is used. In addition, the total PTV mean dose should be considered in the comparison of different prescription approaches. Considering that optimal local control has been achieved by the historical approach of creating a much higher dose in the GTV using multiple beams aimed at the isocenter, we should avoid using homogenous dose distribution within the target and thus preventing a higher GTV dose when IMRT is used. SABR plans should be optimized to have high-dose regions in the GTV and sharp fall-off outside the PTV. Threedimensional conformal radiation therapy or IMRT (including volumetric-modulated arc therapy [VMAT]) SABR plans are typically used and optimized using 6 to 12 coplanar or noncoplanar 6-MV photon beams (three-dimensional conformal radiation therapy or IMRT) or 1- to 3-arc VMAT. The SABR dose should be prescribed to the PTV between the 60% and 90% (typically 80 85%) isodose lines, using appropriate calculation algorithms with heterogeneity correction. IMRT and VMAT plans should be designed with optimization goals to achieve these results. Specifically, the GTV should be optimized to a dose between 10% and 30% higher than the prescription isodose line, and a 50% higher dose to the isocenter is acceptable (Fig. 2). The fall-off outside the PTV should be optimized such that a dose of less than 50% of the prescription isodose is beyond 2 cm of the PTV. Both of these requirements should be evaluated on a patient-by-patient basis. The falloff recommendation in particular should be sacrificed when needed to achieve critical structure sparing (i.e., allowing the dose to fall-off less sharply in one direction so that it can falloff more quickly closer to a critical structure). The fall-off requirement should also be balanced against sparing of the contralateral lung. The total mean PTV dose should be considered to evaluate the actual dose received by PTV; the greater than or equal to 20% prescribed dose for the total PTV mean dose seems appropriate, although more data are needed to validate this. Dose-volume constraints for nearby critical structures based on the published literature should be considered, and compromise to PTV coverage may be needed to maintain the normal-structure constraints. However, the internal GTV (igtv) plus a margin of 5 mm (PiGTV) must receive at least 95% of the prescribed dose, at least 95% of the PiGTV must be covered by at least the prescribed dose, and 100% of the igtv must be covered by at least the prescribed dose while respecting the dose-volume constraints for normal tissues. If FIGURE 2. Treatment plans and dose-volume histograms illustrate differences in doses to the mean PTV when doses are prescribed (Rx) to the 60%, 80%, and 100% isodose lines. PTV, planning target volume. 581

6 Chang et al. Journal of Thoracic Oncology Volume 10, Number 4, April 2015 these requirements cannot be met, then regimens that involve additional fractions should be considered. Use of SABR beam angle/weighting optimization with tight aperture margins is crucial to create a sharp dose gradient that provides adequate target coverage while avoiding overdosing critical structures (Fig. 3). To further reduce longterm toxicity, we must continue to seek strategies for minimizing late toxicity by keeping critical structure dose as low as possible while simultaneously not compromising PTV coverage. SABR beam angles should be optimized based on balancing target coverage and sparing certain critical structures to improve the therapeutic ratio and efficacy (Fig. 3). Daily CT-on-rail or a cone-beam CT scans or implemented fiducial tracking should be used during each radiotherapy fraction to verify and adjust coverage of the target volume and to spare critical structures as needed. PROTON-BASED STEREOTACTIC RADIOTHERAPY The application of protons for delivering SABR to early stage NSCLC has emerged as a tool that may reduce the risk of normal tissue toxicity in patients with complicated presentations. 30 The advantage of proton therapy lies in its ability to minimize dose to normal tissues distal to the tumor. Thus, proton therapy may offer clinical benefits for patients with limited pulmonary reserve, tumors in close geometric proximity to critical normal structures, or those who have received prior thoracic radiation. In such cases, reducing radiation damage to normal tissues is an absolute priority. This may translate into clinical benefit and via reducing the risk of clinical toxicity in some patients. Loma Linda University has the longest-term experience with proton SABR. In one study by Bush et al. 31 involving treatment escalation from 51 Gy to 70 Gy in 10 fractions, local control and survival were both improved with the higher doses. No patient experienced radiation pneumonitis, suggesting that the dose could be increased further, possibly improving outcomes. However, uncertainties of motion and proton range remain. Currently, larger PTV margin might be needed for proton-based SABR. Obviously, proton cannot spare critical structures if these structures are within PTV. A phase II randomized clinical trial comparing proton versus photon-based SABR for centrally located or recurrent lung parenchymal early stage NSCLC is currently ongoing ( clinicaltrials.gov/show/nct ). As the availability of proton therapy expands and the experience with this form of therapy matures, improvements will continue to be made in its implementation. Emerging techniques such as intensitymodulated proton therapy and on-board volumetric image guidance within proton facilities will continue to reduce uncertainties and refine dose conformality. 32 EVALUATING LOCAL AND REGIONAL LYMPH NODE CONTROL The use of CT or positron emission tomography (PET)/ CT for follow-up after SABR remains controversial. Indeed, SABR induces significant consolidation of lung parenchyma and has been associated with residual activity on PET, which may contribute to differences in reported local control rates. 33 This difficulty with interpreting follow-up images could be worse for central lesions, because the proximity to the bronchial tree may increase the incidence of atelectasis after SABR. Some evidence exists to suggest that certain morphologic changes on CT after SABR can predict local recurrence, although further validation is needed. 34,35 Also, high standardized uptake values (SUVs greater than five) on PET more than 6 months after SABR should raise suspicion for local recurrence, and close follow-up is indicated. 33 If the SUV remains high on sequential images, biopsy should be considered to confirm local recurrence. The rate of lymph node recurrence after SABR is between 5% and 10%, 36 which is comparable to recurrence FIGURE 3. Beam-angle-optimized SABR plans for delivering 50 Gy in four fractions while keeping target coverage and avoiding overdoses to nearby critical structures for tumors in four locations: near the brachial plexus (A); near the major vessels (B); near the bronchial tree, esophagus and spinal cord (C); and near the esophagus and heart (D). Color wash dose distributions are shown with corresponding scales. The gray area inside the high-dose region indicates where the dose was higher than the scale maximum (6238 cgy). BP, brachial plexus; PV, pulmonary vessel; BT, bronchial tree; E, esophagus; SC, spinal cord; A, aorta; H, heart; GTV, gross tumor volume; SABR, stereotactic ablative radiotherapy. 582

7 Journal of Thoracic Oncology Volume 10, Number 4, April 2015 Stereotactic Ablative Radiotherapy rates after surgical resection. Generally, central lesions are considered to have a higher probability of regional lymph node recurrence compared with peripheral lesions. 37 The modern staging workup, including PET/CT, endobronchial ultrasonography, and mediastinoscopy has helped to detect disease in lymph nodes, and hence stage the disease, more accurately. Moreover, a recent case-matched analysis indicated that an incidental dose of greater than 20 Gy to hilar lymph nodes, which is likely in the treatment of central lesions, may help to prevent lymph node recurrence after SABR. 37 However, at this moment, there is no data to support treating lymph nodes prophylactically in central lesions. IMMUNOTHERAPY AND SABR: I-SABR Although SABR can produce local control rates in excess of 90%, distant metastasis remains a dominant pattern of failure, occurring in 10% to 20% of cases. 36 Like regional lymph node recurrence, the risk of distant metastasis from central lesions could be higher than from peripheral lesions. In addition, due to the concern of the toxicity of proximal critical structures, the target coverage of some central lesions might have to be compromised. Further improvements in the therapeutic ratio of SABR for central lesions require novel approaches to overcome the compromised target coverage and/or distant metastases. Although radiation therapy is generally used to treat localized targets within PTV, it is also known to have systemic effects through its ability to recruit biological effectors outside the treatment field. 38 High doses of radiotherapy, such as those used in SABR, have been shown to induce objective tumor regression outside the irradiated field; this phenomenon is called the abscopal effect. 39,40 SABR results in local tumor sterilization with the release of tumor cell fragments that contain molecules that may be immunogenic, and thus radiation can be considered to function as an in situ vaccine. Emerging evidence suggests that lung tumors can evoke specific antitumoral immune responses in cancer and that manipulating the immune system may be a valid therapeutic approach. The immune response to these tumor antigens is regulated by cells and molecules with the ability to inhibit immune responses via immunologic response checkpoint pathways. Antibodies against participants in the checkpoint signaling pathways, such as PD1 and CTLA-4, and tumor cell vaccines seem to be attractive approaches for inducing antitumor effects and improving disease-free survival in NSCLC Immunotherapy with anti-ctla-4 antibodies such as ipilumumab can release the immune-response inhibition and work synergistically with a SABR-induced immune response. Indeed, the combination of SABR and ipilumumab has led to complete clinical response, including response of multiple distant metastases in melanoma. 46 The combination of SABR and IL-2 immunotherapy has also achieved surprisingly high clinical response rates (greater than 50% complete response on PET) in melanoma and renal cell carcinoma. 47 SABRinduced immune response was considered the main mechanism for these abscopal effects. Furthermore, preliminary clinical findings suggest that adding immunotherapy to SABR may be effective against advanced refractory lung cancer. 48 By triggering an immune response to wipe out residual disease, combining immunotherapy with SABR has the potential to significantly improve clinical outcome by reducing regional and distant metastasis, and may allow us to deliver less than 100 Gy to the some of the target while maintaining greater than 95% local control that may be needed for some of the central lesions that compromised target coverage has to be considered due to toxicity. SUMMARY SABR is a double-edged sword that can kill cancer cells but can also damage surrounding critical structures; this is a particular concern for lesions in the central thorax. Welldesigned SABR requires a sharp dose gradient from ablative dose to tolerable dose and balance target coverage and critical normal tissue dose-volume constraints. Appropriate case selection and SABR dose regimens based on target location are crucial to reduce toxicity. Critical structures such as the esophagus, bronchial tree, spinal cord, major vessel, heart, brachial plexus, and trachea should not receive an ablative dose. Involved hilar lymph nodes and mediastinal lymph nodes should not be treated with SABR because of their proximity to these critical structures. The dose-volume constraints proposed here are meant as a guide to help clinicians to decide where the limits should be for achieving this balance between targeting the tumor and protecting nearby critical normal tissue structures. Prospective clinical trials to validate these data are needed and should be strongly encouraged. The dose to critical structures should be kept as low as possible. Because distant metastasis remains a dominant pattern of failure in patients after SABR, the identification of a molecular marker to predict distant metastasis would help clinicians to identify which patients would need adjuvant systemic treatment. Novel forms of immunotherapy may also help to prevent or eliminate tumor recurrence, metastasis, and secondary lung cancer. Although the location of the tumor to be treated with SABR is important for achieving an optimal therapeutic ratio, a more important factor is how much dose is delivered to the target versus nearby critical normal tissues. A central location in and of itself is not a no fly zone for SABR; the key is to know and maintain the dose-volume constraints for each critical structure and to optimize SABR to avoid dangerous side effects. Ablative doses are fatal to tumors; they may well also be fatal to critical normal tissues. Therefore, SABR should be designed carefully to kill the tumor but respect the critical normal tissues. If this goal cannot be met, a different treatment strategy is needed. ACKNOWLEDGMENTS This study was supported in part by Cancer Center Support (Core) Grant CA to The University of Texas MD Anderson Cancer Center from the National Cancer Institute, National Institutes of Health. The authors gratefully acknowledge the support and assistance of the members of the Advanced Radiation Technology Committee, of Steven Register and Jordan Sutton for figure preparation, and of Christine Wogan for manuscript editing. 583

8 Chang et al. Journal of Thoracic Oncology Volume 10, Number 4, April 2015 REFERENCES 1. Lagerwaard FJ, Haasbeek CJ, Smit EF, Slotman BJ, Senan S. Outcomes of risk-adapted fractionated stereotactic radiotherapy for stage I non-smallcell lung cancer. Int J Radiat Oncol Biol Phys 2008;70: Timmerman R, Paulus R, Galvin J, et al. Stereotactic body radiation therapy for inoperable early stage lung cancer. JAMA 2010;303: Chang JY, Liu H, Balter P, et al. Clinical outcome and predictors of survival and pneumonitis after stereotactic ablative radiotherapy for stage I non-small cell lung cancer. Radiat Oncol 2012;7: Taremi M, Hope A, Dahele M, et al. Stereotactic body radiotherapy for medically inoperable lung cancer: Prospective, single-center study of 108 consecutive patients. Int J Radiat Oncol Biol Phys 2012;82: Shirvani SM, Jiang J, Chang JY, et al. Lobectomy, sublobar resection, and stereotactic radiation for stage 1 non-small cell lung cancers in the elderly. JAMA Surg 2014;149: Verstegen NE, Oosterhuis JW, Palma DA, et al. Stage I-II non-small-cell lung cancer treated using either stereotactic ablative radiotherapy (SABR) or lobectomy by video-assisted thoracoscopic surgery (VATS): Outcomes of a propensity score-matched analysis. Ann Oncol 2013;24: Palma D, Visser O, Lagerwaard FJ, Belderbos J, Slotman BJ, Senan S. Impact of introducing stereotactic lung radiotherapy for elderly patients with stage I non-small-cell lung cancer: A population-based time-trend analysis. J Clin Oncol 2010;28: Chang JY, Liu YH, Zhu Z, et al. Stereotactic ablative radiotherapy: A potentially curable approach to early stage multiple primary lung cancer. Cancer 2013;119: Tree AC, Khoo VS, Eeles RA, et al. Stereotactic body radiotherapy for oligometastases. Lancet Oncol 2013;14:e28 e Palma DA, Salama JK, Lo SS, et al. The oligometastatic state-separating truth from wishful thinking. Nat Rev Clin Oncol 2014;11: Kelly P, Balter PA, Rebueno N, et al. Stereotactic body radiation therapy for patients with lung cancer previously treated with thoracic radiation. Int J Radiat Oncol Biol Phys 2010;78: Timmerman R, McGarry R, Yiannoutsos C, et al. Excessive toxicity when treating central tumors in a phase II study of stereotactic body radiation therapy for medically inoperable early-stage lung cancer. J Clin Oncol 2006;24: Chang JY, Balter PA, Dong L, et al. Stereotactic body radiation therapy in centrally and superiorly located stage I or isolated recurrent non-smallcell lung cancer. Int J Radiat Oncol Biol Phys 2008;72: Milano MT, Chen Y, Katz AW, Philip A, Schell MC, Okunieff P. Central thoracic lesions treated with hypofractionated stereotactic body radiotherapy. Radiother Oncol 2009;91: Song SY, Choi W, Shin SS, et al. Fractionated stereotactic body radiation therapy for medically inoperable stage I lung cancer adjacent to central large bronchus. Lung Cancer 2009;66: Haasbeek CJ, Lagerwaard FJ, Slotman BJ, Senan S. Outcomes of stereotactic ablative radiotherapy for centrally located early-stage lung cancer. J Thorac Oncol 2011;6: Rowe BP, Boffa DJ, Wilson LD, Kim AW, Detterbeck FC, Decker RH. Stereotactic body radiotherapy for central lung tumors. J Thorac Oncol 2012;7: Nuyttens JJ, van der Voort van Zyp NC, Praag J, et al. Outcome of fourdimensional stereotactic radiotherapy for centrally located lung tumors. Radiother Oncol 2012;102: Xia T, Li H, Sun Q, et al. Promising clinical outcome of stereotactic body radiation therapy for patients with inoperable stage I/II non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2006;66: Chang JY, Li QQ, Xu QY, et al. Stereotactic ablative radiation therapy for centrally located early stage or isolated parenchymal recurrences of nonsmall cell lung cancer: How to fly in a no fly zone. Int J Radiat Oncol Biol Phys 2014;88: Li QQ, Swanick CW, Allen PK, et al. Stereotactic ablative radiotherapy (SABR) using 70 Gy in 10 fractions for non-small cell lung cancer: Exploration of clinical indications. Radiother Oncol 2014;112: Modh A, Rimner, A, Williams E, et al. Local control and toxicity in a large cohort of central lung tumors treated with stereotactic body radiotherapy. Int J Radiat Oncol Biol Phys 2014;1: Mangona, VS, Aneese AM, Marina O, et al. Toxicity after central versus peripheral lung stereotactic body radiotherapy (SBRT): A propensity score matched-pair analysis. Int J Radiat Oncol Biol Phys October 18, 2014 [Epub ahead of print]. 24. Fakiris AJ, McGarry RC, Yiannoutsos CT, et al. Stereotactic body radiation therapy for early-stage non-small-cell lung carcinoma: four-year results of a prospective phase II study. Int J Radiat Oncol Biol Phys 2009;75: Stephans KL, Djemil T, Diaconu C, et al. Esophageal dose tolerance to hypofractionated stereotactic body radiation therapy: Risk factors for late toxicity. Int J Radiat Oncol Biol Phys 2014;90: Nishimura S, Takeda A, Sanuki N, et al. Toxicities of organs at risk in the mediastinal and hilar regions following stereotactic body radiotherapy for centrally located lung tumors. J Thorac Oncol 2014;9: Corradetti MN, Haas AR, Rengan R. Central-airway necrosis after stereotactic body-radiation therapy. N Engl J Med 2012;366: Onishi H, Araki T, Shirato H, et al. Stereotactic hypofractionated highdose irradiation for stage I nonsmall cell lung carcinoma: Clinical outcomes in 245 subjects in a Japanese multiinstitutional study. Cancer 2004;101: Shultz DB, Trakul N, Maxim PG, Diehn M, Loo BW Jr. Vagal and recurrent laryngeal neuropathy following stereotactic ablative radiation therapy in the chest. Pract Radiat Oncol 2014;4: Register SP, Zhang X, Mohan R, Chang JY. Proton stereotactic body radiation therapy for clinically challenging cases of centrally and superiorly located stage I non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2011;80: Bush DA, Cheek G, Zaheer S, et al. High-dose hypofractionated proton beam radiation therapy is safe and effective for central and peripheral early-stage non-small cell lung cancer: Results of a 12-year experience at Loma Linda University Medical Center. Int J Radiat Oncol Biol Phys 2013;86: Chang JY, Li H, Zhu XR, et al. Clinical implementation of intensity-modulated proton therapy for thoracic malignancies. Int J Radiat Oncol Biol Phys 2014;90: Zhang X, Liu H, Balter P, et al. Positron emission tomography for assessing local failure after stereotactic body radiotherapy for non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2012;83: Mattonen SA, Huang K, Ward AD, Senan S, Palma DA. New techniques for assessing response after hypofractionated radiotherapy for lung cancer. J Thorac Dis 2014;6: Mattonen SA, Palma DA, Haasbeek CJ, Senan S, Ward AD. Early prediction of tumor recurrence based on CT texture changes after stereotactic ablative radiotherapy (SABR) for lung cancer. Med Phys 2014;41: Senthi S, Lagerwaard FJ, Haasbeek CJ, Slotman BJ, Senan S. Patterns of disease recurrence after stereotactic ablative radiotherapy for early stage non-small-cell lung cancer: A retrospective analysis. Lancet Oncol 2012;13: Lao L, Hope AJ, Maganti M, et al. Incidental prophylactic nodal irradiation and patterns of nodal relapse in inoperable early stage NSCLC patients treated with SBRT: A case-matched analysis. Int J Radiat Oncol Biol Phys 2014;90: Formenti SC, Demaria S. Systemic effects of local radiotherapy. Lancet Oncol 2009;10: Zeng J, Harris TJ, Lim M, Drake CG, Tran PT. Immune modulation and stereotactic radiation: Improving local and abscopal responses. Biomed Res Int 2013;2013: Demaria S, Formenti SC. Radiation as an immunological adjuvant: Current evidence on dose and fractionation. Front Oncol 2012;2: Lynch TJ, Bondarenko I, Luft A, et al. Ipilimumab in combination with paclitaxel and carboplatin as first-line treatment in stage IIIB/IV nonsmall-cell lung cancer: Results from a randomized, double-blind, multicenter phase II study. J Clin Oncol 2012;30: Topalian SL, Hodi FS, Brahmer JR, et al. Safety, activity, and immune correlates of anti-pd-1 antibody in cancer. N Engl J Med 2012;366: Brahmer JR, Tykodi SS, Chow LQ, et al. Safety and activity of anti- PD-L1 antibody in patients with advanced cancer. N Engl J Med 2012;366: Butts C, Socinski MA, Mitchell PL, et al.; START trial team. Tecemotide (L-BLP25) versus placebo after chemoradiotherapy for stage III nonsmall-cell lung cancer (START): A randomised, double-blind, phase 3 trial. Lancet Oncol 2014;15:

9 Journal of Thoracic Oncology Volume 10, Number 4, April 2015 Stereotactic Ablative Radiotherapy 45. Butts C, Murray N, Maksymiuk A, et al. Phase II study of belagenpumatucel-l, a transforming growth factor 2 antisense gene-modified allogeneic tumor cell vaccine in non-small-cell lung cancer. J Clin Oncol 2006;24: Postow MA, Callahan MK, Barker CA, et al. Immunologic correlates of the abscopal effect in a patient with melanoma. N Engl J Med 2012;366: Seung SK, Curti BD, Crittenden M, et al. Phase 1 study of stereotactic body radiotherapy and interleukin-2 tumor and immunological responses. Sci Transl Med 2012;4:137ra Wang YY, Wang YS, Liu T, et al. Efficacy study of CyberKnife stereotactic radiosurgery combined with CIK cell immunotherapy for advanced refractory lung cancer. Exp Ther Med 2013;5: Evans JD, Gomez DR, Chang JY, et al. Cardiac 18 F-fluorodeoxyglucose uptake on positron emission tomography after thoracic stereotactic body radiation therapy. Radiother Oncol 2013;109: Welsh J, Thomas J, Shah D, et al. Obesity increases the risk of chest wall pain from thoracic stereotactic body radiation therapy. Int J Radiat Oncol Biol Phys 2011;81: