Predictors and management of chest wall toxicity after lung stereotactic body radiotherapy

Complications of TreatmentPredictors and management of chest wall toxicity after lung stereotactic body radiotherapyTalha Shaikh ⇑,Aruna TurakaDepartment of Radiation Oncology,Fox Chase Cancer Center,Philadelphia,PA,USAa r t i c l e i n f o Article history:Received 17June 2014Received in revised form 19August 2014Accepted 7September 2014Available online xxxxKeywords:Stereotactic body radiotherapy Lung cancerChest wall toxicity Rib fracturea b s t r a c tStereotactic body radiotherapy is the preferred treatment modality for patients with inoperable early stage lung cancer.Chest wall toxicity is a potentially dose limiting side effect and may include fractures or pain secondary to treatment.The pathophysiology of these symptoms is unclear although it is pre-sumed that radiation may alter the bone’s normal tissue environment,affecting maintenance and remod-eling.Chest wall pain is likely neuropathic secondary to injury to the intercostal nerves.Identifying patients with chest wall toxicity can be difficult due to the varying definitions of toxicity as well as het-erogeneous contouring guidelines.Multiple studies have demonstrated a correlation between treatment factors and the incidence of chest wall toxicity.An increase in dose and treatment volume appear to be the most consistent radiation factors associated with toxicity.Patient factors such as body mass index,female gender,tumor location,and age have also been correlated with an increased likelihood of devel-oping side effects.Management of chest wall toxicity is typically conservative using analgesic medica-tions although surgical intervention may be required for displaced fractures.In this review,we examine the treatment,patient,and tumor factors predictive for chest wall toxicity and the implications for the treating physician.Ó2014Published by Elsevier Ltd.IntroductionThe standard management for patients diagnosed with early stage lung cancer remains lobectomy with good local control and overall survival rates [1–3].Unfortunately a significant number of patients are not surgical candidates due to poor performance status,poor lung function,or multiple comorbidities.Historically,non-surgical candidates were offered conventionally fractionated radiotherapy to a total dose of 60–66Gy with multiple studies demonstrating poor overall survival [4–6].Local tumor relapse was often the most common etiology of treatment failure following conventionally fractionated radiotherapy.Since the wide adaptation of lung stereotactic body radiother-apy (SBRT),outcomes have significantly improved with local con-trol rates greater than 75%(Table 1).The literature currently contains a wealth of knowledge pertaining to the excellent efficacy of SBRT regimens as well as strategies for treatment delivery [7,8].The next era in SBRT will focus on minimizing toxicity to maintain these excellent outcomes.The efficacy of SBRT is strongly correlated with the biologic effective dose (BED)with trials demonstrating improved control rates with an increase in BED [9].The growing use of SBRT is not without its risks.The feature that makes the treatment so effective in tumor control is also the etiology of its toxicity.Early SBRT stud-ies demonstrated significant morbidity and mortality associated with treating central lesions.Timmerman et al.found that patients with tumors within 2cm of the proximal bronchial tree had an 11-fold increased risk of toxicity compared to more peripheral lesions [10].It is now well established that patients with central disease should likely undergo a more protracted course of SBRT to minimize toxicity.Currently,multiple studies are being under-taken to evaluate the ideal dosing regimen when treating central lesions.Similarly,early studies also demonstrated the risk of acute skin toxicity which is now less commonly seen [11].As SBRT techniques and treatments have matured,acute toxic-ity has become less common with most patients only experiencing constitutional symptoms.Subacute and late toxicity is often multi-factorial with tumor location and dosimetric factors playing an important role.As the SBRT literature gains longer follow-up,late radiation toxicity has become more evident.One of these toxicities includes chest wall toxicity (CWT)with symptoms not being revealed until six months after the completion of therapy.The/10.1016/j.ctrv.2014.09.0030305-7372/Ó2014Published by Elsevier Ltd.⇑Corresponding author at:Department of Radiation Oncology,Fox Chase Cancer Center,333Cottman Avenue,Philadelphia,PA 19111,USA.Tel.:+12157282581;fax:+12152144038.E-mail address:Talha.Shaikh@ (T.Shaikh).reported incidents of CWT after SBRT have varied in the literature (Table 2).This disparity is likely attributable to the heterogeneous dosing regimens,treatment techniques,as well as follow-up inter-vals.In addition,the definition of CWT varies greatly by institution as do the contouring schema.Chest wall toxicity definition and incidenceOne of the difficulties in quantifying CWT is the term itself,which may encompass various features.The definition of CWT can be discursive,encompassing all toxicity related to skin,pain or fractures.In our review,we define CWT as being limited to rib fractures as well as any associated chest wall pain which may be focal or neuropathic with grading according to the Common Termi-nology Criteria for Adverse Events Version 3.0(Table 3)for each respective toxicity.The definition of the chest wall and ribs when defining organs at risk is a relatively new phenomenon (Table 4).In earlier SBRT pro-tocols such as RTOG 0618and RTOG 0813,no definition for con-touring or dosimetric constraints is provided.Newer protocols such as RTOG 1021and RTOG 0915provide a definition for the ribs but do not provide a definition for the chest wall.The contouring atlas for RTOG 1106,a non-SBRT protocol,provides a definition for the chest wall although there is no clear description in the pro-tocol itself.Many institutions have developed their own guidelines to identify the regions of interest.Similar to the descriptors for pneumonitis and esophagitis,identifying chest wall pain and rib fractures can be difficult.In addition,rib fractures are often asymptomatic and thus overlooked when compiling toxicity data.In one study,61%of rib fractures were revealed only through routine imaging [12].Rib fractures and chest wall pain may develop several months after treatment making their etiology difficult to identify.When non-oncologic radiologists are evaluating imaging studies,these minor changes are often overlooked.Radiation-associated bone injury is well discussed in the litera-ture when treating other disease sites.The largest series is from the Joint Center for Radiation Therapy where 1117patients treated with breast conservation therapy were analyzed [13].The authors found a 1.8%incidence of rib fracture after conventionally fraction-ated external beam radiation with a median time to fracture of 12months.The authors found that the fractures were more com-mon in patients treated with low energy photons.Other factors predicting for toxicity were total breast dose and chemotherapyuse.Similarly,Markiewicz et al.examined 1053early stage breast cancer patients who underwent breast conservation therapy [14].With a median follow-up of 6.7years,the authors found a higher incidence of rib fractures in those who received chemotherapy and nodal irradiation.Other breast conservation studies have noted an increase in fractures when patients were treated with lar-ger fraction sizes [15].Due to the era of treatment,all patients in the above studies underwent conventionally fractionated radio-therapy to a total dose of 46–50Gy to the whole breast.Radiother-apy technique relied on bony anatomy and assessment of dosimetry at a single plane.More recent studies have had variable incidence of rib fractures ranging from 0.3%to 3.6%[16,17].The more frequent use of imaging may account for the stable rate of CWT in the modern era.Mechanism of toxicityRadiation to the bone alters the normal tissue environment,affecting maintenance and remodeling effectively rendering the bone brittle and susceptible to injury.Although bone injury has become less common with the use of megavoltage machines and more precise treatment planning techniques,toxicity is still fre-quently seen.Radiation related bone toxicity may also result in pelvic insufficiency fractures,mandibular osteonecrosis,as well as long bone fractures.The exact pathophysiology of chest wall pain is not understood.It is assumed to be caused by injury to the intercostal nerves result-ing in neuropathic pain rather than an inflammatory reaction,which is more likely to cause acute toxicity.The role of large single and total doses has been implicated in nerve dysfunction in other disease sites [18,19].The most commonly cited radiation-related nerve injury is brachial plexopathy which occurs in approximately 1%of breast cancer patients receiving conventionally fractionated radiation therapy [20].Patients typically present with symptoms of motor or sensory dysfunction,which can happen several months to years after treatment.When treating lung cancer,these symp-toms are rather transient with most patients having symptomatic control using over the counter analgesics.In lung cancer patients,chest wall pain has long been an established toxicity.For example in patients undergoing surgical resection,chronic chest wall pain may be seen post thoracotomy [21,22].One of the reasons for the difficulty in defining and minimizing CWT is the simplistic linear quadratic model often used when iden-tifying toxicity.The model likely is not able to accurately describeTable 1Toxicity and local control rates for patients with Stage I non-small cell lung cancer treated with SBRT.Studyn Dose/fxMedian F/U LC (%)CWT Other toxicity Lagerwaard [40]20660Gy/3–8fx 129749Koto [41]3145–60Gy/3–8fx 327501Onishi [42]25718–75Gy/1–22fx 3886417Zimmerman [43]6837.5Gy/3–5fx 189462Baumann [44]13845Gy/3fx 359224Nyman [45]4545Gy/3fx 438032Timmerman [10]7060–66Gy/3fx 179508Ricardi [46]4345Gy/3fx 289213Xia [47]4350Gy/5fx 279601Nagata [48]4248Gy/4fx 309802Fakiris [49]7060–66Gy/3fx 5094012Uematsu [50]5050–60Gy/5–10fx 369420Wulf [51]2026–37.5Gy/1–3fx 119200Grills [52]50520–64Gy/1–15fx19941550CWT:Chest wall toxicity.Fx:Fraction.F/U:Follow-up.LC:Local control.2T.Shaikh,A.Turaka /Cancer Treatment Reviews xxx (2014)xxx–xxxthe actual BED for late osseous toxicity.Although the volumetric constraints may help predict toxicity,the effective dosimetric values for patients being treated with three or five fractions are quite different.Brashears et al.noted fractures in ribs even when maintaining tolerance values below thresholds extrapolated from classic radio-logic models after accelerated partial breast irradiation [23].The authors found that traditional models for estimating the BED may be inaccurate when examining patients being treated with accelerated partial breast irradiation high dose rate brachytherapy.They concluded that the simplistic linear quadratic model may fail to describe the actual BEDs for high dose rate brachyherapy.Similarly the alpha–beta ratio of three for late osseous toxicity is likely too low and unable to accurately predict toxicity.The com-plexity is evident when noting that patients with point doses as low as 33Gy have been documented to have rib fractures [24].Table 4Contouring definitions for chest wall and ribs in SBRT.Study Target Definition Andolino [12]CW 3cm expansion from ipsilateral lung or liver Ribs Individual ribs contouredCreach [26]CW 3cm axial expansion on the ipsilateral lung with exclusion of the mediastinal structures and ipsilateral lung.If contour was outside of patient,it was brought into the external skin edge.Effort was made to include the thoracic nerve roots Dunlap [25]CW 3cm expansion of lung minus lung volume,mediastinal soft tissue,vertebral body Kim [32]CW Ipsilateral hemibody that excluded the lungs and mediastinum Nambu [53]Ribs Contoured fractured ribsStephans [27]CW Arc of all ipsilateral soft tissue outside of lung from edge of sternum circumferentially to the edge of the vertebral body including the spinal nerve root exitTaremi [30]Ribs Contoured ribs from the costovertebral to the costosternal/costal cartilageWelsh [37]CW Outer edge of patient’s skin/chest wall automatically contoured minus total lung contourBongers [34]CW Expansion of the lungs 2cm in lateral,posterior,and anterior directions except in the direction of the mediastinum with inclusion of intercostal muscles but excluding other muscles and skinWoody [29]CW Contoured by expanding ipsilateral lung at least 3cm in anterior,posterior,medial,lateral directions to cover relevant ribs and soft tissues and then adjusted to exclude spine,mediastinum,contralateral chest wallMutter [28]CW 3cm expansion of lung minus lung volume,mediastinal soft tissue,vertebral body Pettersson [31]Ribs Individual ribs receiving total dose of 21GyRTOG 1021/0915RibsRibs within 5cm of the PTV should be contoured by outlining the bone and marrow.Typically,several portions of adjacent ribs will be contoured as one structure.Adjacent ribs,however,should not be contoured in a contiguous fashion (i.e.,do not include the inter-costal space as part of the ribs)RTOG 0618/0813N/AN/AGy:Gray.CW:Chest wall.Table 2Incidence of chest wall toxicity in dedicated studies.StudynMedian F/UCW pain (%)Rib fracture (%)Median onset (mo)Dose/FxGrade (pain)Grade (fracture)III III IV I II III Andolino [12]34719mo 105918–72Gy/6–24fx 272421–––Creach [26]14622.5mo 15812.650–54Gy/3–5fx 11920–––Dunlap [25]6011.1mo 3287.121–60Gy/3–5fx 11170–––Kim [32]11822mo 27381736–60Gy/3–4fx ––––3684Nambu [53]17727mo 212321.248–70Gy/4–10fx 11700–––Stephans [27]4518.8mo 22–8.860Gy/3fx 4600–––Taremi [30]4624.9mo 46372154–60Gy/3fx 9930101914Welsh [37]26510.3mo 223650Gy/4fx 222300–––Bongers [34]50033mo 112860Gy/3–8fx47100008Woody [29]10225mo 20–8.148–60Gy/3–10fx 61310–––Mutter [28]12616mo 434940–60Gy/3–5fx 1916190–––Pettersson [31]3329mo–398.845Gy/3fx–––––––CW:Chest wall.F/U:Follow-up.Fx:Fraction.Table 3Common terminology criteria for adverse events v3.0.Adverse event Grade 12345Pain Mild pain not interfering with function Moderate pain;pain or analgesics interfering with function,but not interfering with ADL Severe pain;pain or analgesics severely interfering with ADL Disabling –FractureAsymptomatic,radiographic findings only (e.g.,asymptomatic rib fracture on plain X-ray,pelvic insufficiency fracture on MRI,etc.)Asymptomatic,radiographic findings only (e.g.,asymptomatic rib fracture on plain X-ray,pelvic insufficiency fracture on MRI,etc.)Symptomatic and displaced or open wound with bone exposure;operative intervention indicatedDisabling;amputation indicatedDeathT.Shaikh,A.Turaka /Cancer Treatment Reviews xxx (2014)xxx–xxx3Treatment factors predicting toxicityMultiple studies have demonstrated a strong correlation between various treatment factors and the incidence of CWT (Table 5).Since treatment factors are often able to be manipulated,these are the more commonly targeted factor to reduce treatment side effects.As with other organs at risk,dose constraints theoret-ically limit the risk of toxicity,although constraints for CWT remain difficult to discern.The most widely established constraint for CWT remains volu-metric dosing.One of the earliest studies examining SBRT CWT by Dunlap et al.found that the volume receiving greater than 30Gy (V30)strongly correlated with significant toxicity [25].The authors found that there was a 30%risk of CWT when 35cm 3of the chest wall volume received 30Gy.Similarly,Creach et al.found that V30and V40were predictive of CWT [26].Using thresholds of 0.7%and 0.19%for the V30and V40,respectively,correlated with a 15%risk of CWT.Stephans et al.also found multiple volumetric parameters to be effective in predicting toxicity [27].On univariate analysis,V30–V70were predictive of toxicity while on multivari-ate analysis V30–V60remained significant.Other studies have also demonstrated V30as being a predictor for CWT [28,29].Examining the literature it becomes evident that although volume-based anal-ysis not an ideal predictor of CWT,it is likely the best one available at this moment.The exact point dose at which a patient may develop a rib frac-ture is difficult to elucidate with various results in the literature.Taremi et al.found that the dose to 0.5cc (D 0.5)of the rib was asignificant predictor of fractures with a 50%risk of fracture associ-ated with a D 0.5of 60Gy [30].Meanwhile,Pettersson et al.found the D 2was associated with a 50%of risk of fracture using a cut-off of 27.3Gy [31].Furthermore,Andolino et al.found that using an EC 50model,D 5and D 15receiving 40Gy predicted 10%and 30%toxicity respectively [12].Furthermore,Kim et al.found D 8receiving 58Gy predicts for a 50%risk of fracture [32].Examining the literature it becomes evident that various methods of treat-ment and analysis make cross correlation between studies difficult.Although it is clear that the point dose is a strong predictor of tox-icity,the ideal constraint has yet to be defined.Similar to the breast cancer literature,total dose and fraction-ation may also be important predictors of toxicity.Andolino et al.found that the dose per fraction and total dose correlated with CWT [12].In their analysis,patients receiving maximum dose of 50Gy to the chest wall and rib were more likely to develop pain or fracture.Stephans et al.also found a trend toward statistical sig-nificance when examining the maximum point dose [27].There was no evidence of CWT in patients receiving a minimum absolute chest wall point dose of less than 67.5Gy.This group also found that patients treated with 60Gy and three fractions had a greater incidence of toxicity versus those treated with 50Gy in five fractions.As the linear quadratic formula may be too simplistic in predict-ing CWT,other methodologies have been discussed.Woody et ed the concept of a modified equivalent uniform dose (mEUD)originally proposed by Park et al.[29,33].The mEUD may be a more sensitive predictor due its ability to account for inhomogeneity as well as different fractionations.In their cohort of 102patients,they found that the mEUD was a better predictor of chest wall pain ver-sus the V30.Due to the sparse data,the role of other treatment factors is often extrapolated from the literature.Hoppe et al.retrospectively evaluated 50patients with Stage I NSCLC who underwent SBRT [11].They found that 14%of patients developed >Grade 2acute skin toxicity.Prognostic factors predictive of skin toxicity included distance from the tumor to the skin,maximum back skin dose,and treatment with three beams.The impact of number of beams likely holds true for CWT as well.Presently,it is well established that using a larger number of beams will help with both overall confor-mality as well as dose distribution.As long-term data for SBRT accrues,treatment planning tech-niques are becoming more refined to minimize toxicity.Currently,the RTOG recommends a max point dose of 30Gy or 40Gy to the rib,in one or four fractions respectively.There is no consensus on volumetric constraints for ongoing RTOG SBRT protocols.Although achieving these constraints is important,physicians should not compromise coverage in order to achieve these goals.As the guidelines are being established,institutions will often fol-low their established in-house constraints.Patient and tumor related factors predicting toxicityAs the role of treatment factors is better understood in the development of CWT,patient variables still remain difficult to dis-cern.Theoretically,factors such as diabetes,collagen vascular dis-ease,and osteoporosis,should make a patient susceptible to the development of fibrosis and associated toxicity.Due to the limited data available at this time,this correlation is difficult to appreciate.The more predictable relationship is the proximity of the disease to the chest wall.As malignancies develop closer to the chest wall,the difficulty in achieving dose fall off becomes more evident.The-oretically,tumor distance from the chest wall should be the most significant predictor of toxicity.Multiple trials have demonstrated the impact of tumor distance with CWT [30,32,34].AlthoughTable 5Factors predictive for chest wall toxicity in dedicated studies.Study n Treatment factorsPatient &tumor factors Andolino347Dose per fraction –Total doseMaximum dose to chest wall Maximum dose to rib GTVCreach 146V35–45(UVA)V30-V40(MVA)V20-V60(UVA)Mean dose (UVA)Body mass indexV40(MVA)Connective tissue disease Dunlap 60V30–Kim 118D 8Gender (Female)Location (Lateral)Nambu 177BEDTumor-chest wall distanceStephans 45PTVGreatest tumor dimension V30-V70(UVA)Taremi46AgeD 0.5DistanceMaximum point dose Gender (Female)Welsh 265V30BMIBongers 500PTVTumor-chest wall distance Diameter Age Woody 102V30BMIMaximum point dose mEUD Mutter 126V30–Pettersson 33D 2–BED:Biologic effective dose.Dxx:Dose to xx volume.GTV:Gross tumor volume.mEUD:Modified equivalent uniform dose.MVA:Multivariable analysis.PTV:Planning target volume.UVA:Univariable analysis.Vxx:%Volume of the chest wall receiving at least xx Gy.4T.Shaikh,A.Turaka /Cancer Treatment Reviews xxx (2014)xxx–xxxachieving dose constraints is difficult,proximity to the chest wall is not a contraindication to SBRT.Of 21patients treated by Stephans et al.with lesions adherent to the chest wall,only seven ultimately developed toxicity [27].At our institution,we will use a risk-adap-tive methodology to determine the dose fractionation schedule to deliver a BED of greater than 100Gy.Although SBRT is typically defined as less than five fractions within the United States,a com-mon European Society for Medical Oncology supported fraction-ation includes 7.5Gy in 8fractions [35].Historically,obesity has also been an important predictor for toxicity when treating other disease sites [36].This factor appears to hold true when examining patients who develop CWT.Welsh et al.examined 265patients with tumors within 2.5cm of the chest wall.Fourteen patients developed acute pain with 45patients developing chronic pain [37].They also found that body mass index (BMI)was strongly correlated with the development of chest wall pain.Patients with a BMI greater than 29had a stron-ger risk of chronic pain with diabetes mellitus a significant predic-tor in these patients.Similarly,Woody et al.also found that BMI were predictors of chest wall pain [29].It remains unclear whether BMI itself is the underlying etiology of CWT or whether it predicts for chest wall thickness and subsequent effects.Other patient factors remain difficult to transform into daily practice.Taremi et al.found that older age was a predictor of tox-icity on both multivariate and univariate analysis [30].On the other hand,Bongers et al.found that younger age predicted for more severe toxicity [34].The role of gender has also been demon-strated on multiple trials with the female gender being a predictor of toxicity [30,32].The role of tobacco use has been linked to radi-ation related toxicity [38].Stephans et al.found a trend towards CWT in patients who continued to smoke during treatment [27].Management and conclusionsManagement for patients with CWT is relatively conservative with treatment involving oral anti-inflammatory medications.For pain uncontrolled by over the counter medications,occasionally gabapentin or narcotics may be required.Theoretically,if a patient experiences a rib fracture that is grade 4or greater they may require surgical intervention.To our knowledge there is no reported grade 4+rib fracture in the literature although there is an episode of chest wall pain [12].As long-term toxicity data for SBRT accrues,the traditional models for estimating CWT will likely also need to be reassessed.In addition to addressing the radiobiological properties of SBRT,other dosimetric and patient variables will also be important.In the interim,methodologies of minimizing chest wall dose remains important.For lesions proximal to the chest wall,reducing the dose per fraction remains the easiest method of maintaining dose constraints.Another means of reducing chest wall dose may be angle optimization in order to improve does fall off.Ultimately set up reproducibility is vital,as even small changes in tumor motion will profoundly impact the amount of dose received by the chest wall.Taremi et al.examined 46inoperable patients treated with SBRT with a median follow-up of 25months [30].Per their analy-sis,D 0.5,gender,and age were all correlated of toxicity.These fac-tors were subsequently used to construct a nomogram to predict radiation induced bone injury following SBRT.In the example pre-sented in their manuscript,a 75-year-old female treated with 54Gy in 3fractions and a D 0.5of 60Gy has a 65%risk of a fracture.Although not a perfect,this is a unique and effective tool for pre-dicting CWT.Given that patients with peripheral tumors have an elevated risk of CWT it is important to devise a means of reducing chest walldose when treating these types of lesions.Ding et al.retrospec-tively evaluated ten patients treated with SBRT to a total dose of 60Gy in 5fractions [39].When evaluating these treatment plans using Volume Modulated Arc Therapy (VMAT),the authors found a significant improvement in the V30to the chest wall and ribs although there was an increase in the lung V20of one patient.Another advantage of VMAT is the rapid delivery time thus mini-mizing the risk of target and patient motion.At our institution for non-operable SBRT candidates with lesions proximal to the chest wall,we attempt to meet dosimetric constraints set forth by the RTOG.We define the chest wall as a 2cm expansion of the ipsilateral lung minus the lung itself.The mediastinal structures and vertebral bodies are manually extracted from our contour.For the ribs we contour the ribs proximal to the tumor itself,within 1cm superior and inferior to the GTV.Although patient factors appear to have an important role in guid-ing treatment,at this time we do not discriminate when offering SBRT.In cases when we are unable to meet dosimetric constraints,we will modify the SBRT regimen using a risk-adapted fraction-ation approach.Patients unable to meet dosimetric constraints are offered a more protracted courses of radiation.A recent study by Coroller et al.examined this approach and demonstrated an 8.3%incidence of CWT with good local control rates (CITE).Limitations of the literature include a heterogeneous group of treatment methodologies,varying definitions when defining the rib or chest wall,small sample sizes,and limited follow-up.It is evident that patients with peripheral lesions are at a considerable risk for CWT.Although the ideal predictors for toxicity are not yet clear,efforts should be made to meet the current recommended guidelines while maintaining tumor coverage.As the SBRT litera-ture matures,predicting and preventing CWT will become better established.Conflicts of interestThere are no conflicts of interest to declare for any author.References[1]McKenna Jr RJ,Houck W,Fuller CB.Video-assisted thoracic surgery lobectomy:experience with 1,100cases.Ann Thorac Surg 2006;81:421–5.discussion 5–6.[2]Roviaro G,Varoli F,Vergani C,Nucca O,Maciocco M,Grignani F.Long-termsurvival after videothoracoscopic lobectomy for stage I lung cancer.Chest 2004;126:725–32.[3]Shiraishi T,Shirakusa T,Hiratsuka M,Yamamoto S,Iwasaki A.Video-assistedthoracoscopic surgery lobectomy for c-T1N0M0primary lung cancer:its impact on locoregional control.Ann Thorac Surg 2006;82:1021–6.[4]Jeremic B,Classen J,Bamberg M.Radiotherapy alone in technically operable,medically inoperable,early-stage (I/II)non-small-cell lung cancer.Int J Radiat Oncol Biol Phys 2002;54:119–30.[5]Qiao X,Tullgren O,Lax I,Sirzen F,Lewensohn R.The role of radiotherapy intreatment of stage I non-small cell lung cancer.Lung Cancer (Amsterdam,Netherlands)2003;41:1–11.[6]Wisnivesky JP,Bonomi M,Henschke C,Iannuzzi M,McGinn T.Radiationtherapy for the treatment of unresected stage I-II non-small cell lung cancer.Chest 2005;128:1461–7.[7]Senthi S,Lagerwaard FJ,Haasbeek CJ,Slotman BJ,Senan S.Patterns of diseaserecurrence after stereotactic ablative radiotherapy for early stage non-small-cell lung cancer:a retrospective ncet Oncol 2012;13:802–9.[8]Guckenberger M,Allgauer M,Appold S,Dieckmann K,Ernst I,Ganswindt U,et al.Safety and efficacy of stereotactic body radiotherapy for stage 1non-small-cell lung cancer in routine clinical practice:a patterns-of-care and outcome analysis.J Thorac Oncol 2013;8:1050–8.[9]Onishi H,Shirato H,Nagata Y,Hiraoka M,Fujino M,Gomi K,et al.Stereotacticbody radiotherapy (SBRT)for operable stage I non-small-cell lung cancer:can SBRT be comparable to surgery?Int J Radiat Oncol Biol Phys 2011;81:1352–8.[10]Timmerman R,McGarry R,Yiannoutsos C,Papiez L,Tudor K,DeLuca J,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;24:4833–9.[11]Hoppe BS,Laser B,Kowalski AV,Fontenla SC,Pena-Greenberg E,Yorke ED,et al.Acute skin toxicity following stereotactic body radiation therapy for stage I non-small-cell lung cancer:who’s at risk?Int J Radiat Oncol Biol Phys 2008;72:1283–6.T.Shaikh,A.Turaka /Cancer Treatment Reviews xxx (2014)xxx–xxx5。