Journal of Surgical Research 174, 319–325 (2012)doi:10.1016/j.jss.2011.07.044

Stereotactic Body Radiation Therapy for Nonresectable Tumors

of the Pancreas

Kush Goyal, M.D.,* Douglas Einstein, M.D., Ph.D.,‡ Rafael A. Ibarra, M.D.,* Min Yao, M.D., Ph.D.,‡Charles Kunos, M.D., Ph.D.,‡ Rod Ellis, M.D.,‡ James Brindle, Ph.D.,‡ Deepjot Singh, M.D.,†

Jeffrey Hardacre, M.D.,* Yuxia Zhang, M.S.,‡ Jeffrey Fabians, M.S.,‡ Gary Funkhouser, R.T.T.,‡Mitchell Machtay, M.D., Ph.D.,‡ and Juan R. Sanabria, M.D., M.Sc.*,1

*Department of Surgery; †Department of Medicine; and ‡Department of Radiation Oncology, University Hospitals-CaseMedical Center, Case Western Reserve University, Cleveland, Ohio

Originally submitted May 22, 2011; accepted for publication July 29, 2011

Background. Stereotactic body radiation therapy(SBRT) has emerged as a potential treatment optionfor local tumor control of primary malignancies ofthe pancreas. We report on our experience with SBRTin patients with pancreatic adenocarcinoma whowere found not to be candidates for surgical resection.Methods. The prospective database of the first 20

consecutive patients receiving SBRT for unresectablepancreatic adenocarcinomas andaneuroendocrine tu-mor under an IRB approved protocol was reviewed.Prior to SBRT, cylindrical solid gold fiducial markerswere placed within or around the tumor endoscopi-cally (n[ 13), surgically (n[ 4), or percutaneously un-der computerized tomography (CT)-guidance (n[ 3) toallow for tracking of tumor during therapy. Mean radi-ation dose was 25 Gray (Gy) (range 22–30 Gy) deliveredover 1–3 fractions. Chemotherapy was given to 68% ofpatients in various schedules/timing.Results. Patients had amean gross tumor volume of

57.2 cm3 (range 10.1–118 cm3) before SBRT. The meantotal gross tumor volume reduction at 3 and 6 mo afterSBRT were 21% and 38%, respectively (P < 0.05). Me-dian follow-up was 14.57 mo (range 5–23 mo). The over-all rate of freedom from local progression at 6 and 12mo were 88% and 65%. The probability of overall sur-vival at 6 and 12 mo were 89% and 56%. No patienthad a complication related to fiducial markers place-ment regardless of modality. The rate of radiation-

1 To whom correspondence and reprint requests should be ad-dressed at Department of Surgery, Division of Transplant and Hepa-tobiliary Surgery, University Hospitals-Case Medical Center, 11100Euclid Avenue Lakeside 7506, PS 5047, Cleveland, OH 44106.E-mail: juan.sanabria@uhhospitals.org or juan.sanabria@case.edu.

319

induced adverse events was: grade 1–2 (11%) and grade3 (16%). There were no grade 4/5 adverse events seen.Conclusion. Our preliminary results showed SBRT

as a safe and likely effective local treatment modalityfor pancreatic primary malignancy with acceptablerate of adverse events. � 2012 Elsevier Inc. All rights reserved.

Key Words: pancreatic adenocarcinoma; radiosur-gery; CyberKnife; stereotactic body radiation therapy;fiducial placement.

INTRODUCTION

More than 40,000 individuals are diagnosed annuallywith tumors of the pancreas in the United States [1]. Ofthese, less than 20% are amenable to definitive surgicalmanagement due to advanced stage of local disease, dis-tant metastasis, or co-morbid medical conditions [2].Unresectable pancreatic cancer carries a poor progno-sis with limited nonsurgical effective treatment op-tions. At this time, successful systemic therapies forunresectable pancreatic cancer are poorly developed.Gemcitabine and/or 5-FU are routinely offered to pa-tients, but have not had a significant impact on survivalwith a median overall survival of less than 1 y for non-surgically removed primary tumors of the pancreas [3].

Stereotactic radiosurgery (SRS) is a technique thatallows precise delivery of a large ablative radiationdose in 1 to 5 fractions to the target field (neoplasm)while sparing normal surrounding tissue. This tech-nique allows radiation to be delivered with sub-millimeter precision and with a rapid radiation fall-offfrom the target field, thus sparing normal tissuefrom high dose of radiation. Initially developed for

0022-4804/$36.00� 2012 Elsevier Inc. All rights reserved.

JOURNAL OF SURGICAL RESEARCH: VOL. 174, NO. 2, MAY 15, 2012320

intracranial lesions, SRS resulted in local control ratesabove 80%–90% for the treatment of brain metastases[4]. When SRS is used in extracranial tumors, it isalso called stereotactic body radiation therapy(SBRT). Nonetheless, its use in extracranial tumorshad been limited due to the inherent movement of ab-dominal organs that occurs during the respiratory cy-cle. One such device that tracks tumors duringrespiration and automatically adjusts during patientpositioning is the CyberKnife system, which consistsof three key components: (1) an advanced, lightweightlinear accelerator (LINAC), (2) a robotic arm that canpoint the LINAC from a wide variety of angles, and(3) a tumor tracking system. This system tracks a pa-tient’s abdominal tumor during respiration via two si-multaneous mechanisms: (a) internal fluoroscopicmonitoring of fiducial markers placed in or around thetumor; and (b) external monitoring through a camerasystem that model the chest wall motion and adjuststhe linear accelerator movement simultaneously.Thus, SBRT by the Cyberknife system uses real-timetracking of implanted fiducial markers combined withreal-time respiratory motion modeling to achieve sub-millimeter accuracy by continually detecting and cor-recting for tumor motion throughout treatment. Itwas reported that the average treatment delivery preci-sion was 0.3 6 0.1 mm as measured at three differentCyberKnife facilities [5]. We have acquired experiencewith SBRT in the treatment of liver malignancies,and others have shown preliminary encouraging re-sults in the treatment of pancreatic tumors [6–8]. Wereport our initial experience with 19 patients who un-derwent SBRT for unresectable pancreatic adenocarci-noma and one additional patient with a pancreaticneuroendocrine tumor (PNET).

MATERIALS AND METHODS

Patient Population

Aprospectivedatabaseof 20patients treatedwithSBRTatUniversityHospitals-CaseMedicalCenter betweenNovember 2007 andNovember2010 for nonresectable pancreatic tumors was reviewed under an IRBapprovedprotocol.Nineteenpatientsmet the inclusion criteria for enrol-ment including: (1) biopsy proven pancreatic adenocarcinoma, (2) unre-sectable disease, and (3) life expectancy of at least 12 wk. Two of themhad localized pancreatic recurrence after pancreaticoduodenectomyand adjuvant chemo-radiation. Three of themunderwent biliary bypassprior to SBRT and 68% underwent chemotherapy. Another patientreceived SBRT for unresectable PNET; however, this patient was notincluded in the statistical analysis.

Stereotactic Body Radiation Therapy (SBRT)

All patients were evaluated by a multidisciplinary team includinghepatopancreatobiliary (HPB)/oncologic surgeons, medical oncolo-gists, and radiation oncologists. They were discussed in a multidisci-plinary gastrointestinal tumor board. All patients were staged with

contrast enhanced computerized tomography (CT scan), magneticresonance imaging (MRI), and/or positron emission tomographyscan (PET). Subsequent imaging for treatment plan developmentand contouring was obtained as needed.

Cylindrical solid gold fiducial markers 3 to 5 mm in size (Best Med-ical International, Springfield, VA) were placed endoscopically(n¼ 12), surgically (n¼ 4), or under CT guidance (n¼ 3); 3 to 6 fiducialwere placed within or around the tumor tissue and at a minimum dis-tance of 2 cm between adjacent markers. One week was provided be-tween markers placement and imaging studies for SBRT treatmentplanning simulation to allow for fiducial settling. Patients werethen brought to the SBRT suite. They were immobilized using ana cradle, and fittedwith a synchrony vest during simulation and treat-ment. Patients underwent imaging in the SBRT immobilized position.These scans were imported into the Multiplan treatment planningsystem ver. 2.05 (Accuray Inc., Sunnyvale, CA) and digitally fused.Tumor definition, normal tissue constraints, and final treatmentplan were approved by the attending radiation oncologist, the attend-ing surgeon, and the medical physicist. Regional lymph nodes werenot included in the treatment plan; 100 to 300 6 MV X-ray beamswere used for each plan.

Prior to each treatment fraction, patients were premedicated withdexamethasone (Roxane, OH) and Ondansetron (Ethex, MO), andthey were continuously monitored during treatment. SBRT was per-formed under real-time kilovoltage camera fiducial tracking andreal-time respiratory motion modeling using a separate SynchronyRespiratory Tracking System (Accuray Inc., Sunnyvale, CA). Averagetreatment time per fraction was 2 h. Patients were aimed to be as-sessed every 3 mo after completion of treatment by physical examand imaging. CT, MRI, and/or PET scans were performed at eachfollow-up. Themaximum tumor diameter and the gross tumor volume(GTV) were measured using the Multiplan treatment planning sys-tem ver. 2.05 (Accuray Inc., Sunnyvale, CA), and the images were im-ported into ADAC pinnacle radiation therapy planning system with3D volume algorithms.

Local response to SBRTwas graded by RECIST 1.1 (Response Eval-uation Criteria in Solid Tumors ver. 1.1) to describe change in treatedtumor lesion [9, 10]. This grading system has four tumor responsegrades. Complete Response: disappearance of all target lesions; Par-tial Response: at least a 30% decrease in the sum of the longest diam-eter (LD) of target lesions, taking as reference the baseline sum LD;Progressive disease: at least a 20% increase in the sum of the LDof tar-get lesions, taking as reference the smallest sum LD recorded sincethe treatment started or the appearance of one or more new lesions;Stable disease: neither sufficient shrinkage to qualify for PR nor suf-ficient increase to qualify for PD, taking as reference the smallest sumLD since the treatment started [10]. To further evaluate a partial tu-mor response to SBRT we developed a grading system taking in con-sideration tumor volume [7]. Partial response grade I: at least a 10%decrease in tumor volume but less than 30% from original tumor vol-ume; Partial response grade II: a decrease in volume �30% but <50%from original tumor volume; Partial response grade III: a decrease intumor volume �50%. In some instances, the tumor volume was simi-lar to the original but the enhancement andPET activity vanished;weconsidered these particular cases as a grade III partial response.

Local or distant recurrent disease was graded as well by the follow-ing scale, Grade 1: local recurrence (tumor progression within or atthe periphery of the radiation field) with two subgroups, Grade 1a:1 local recurrence and Grade 1b: >1 local recurrences; Grade 2: dis-tant intra-abdominal recurrence [new tumor>3 cm away from the ra-diation field or in another organ); Grade 3: distant extra-abdominalrecurrence; and Grade 4: a combination of local and distant recur-rences [7].

Adverse Events

Adverse events after SBRT were graded on 1–5 scale according toNational Cancer Institute Common Terminology Criteria for Adverse

GOYAL ET AL.: STEREOTACTIC BODY RADIATION THERAPY 321

Events (NCI CTCAE v3.0). Causes were attributed to either surgery,placement of fiducial markers, chemotherapy, radiation therapy, orrelated to medical comorbidities.

Statistical Analysis

A database of clinical, imaging, and radiation variables was createdand maintained in a prospective manner. Subject and tumor charac-teristics were expressed in mean, standard deviation, and percent-ages. Paired t-tests were used to compare tumor volume before andafter treatment. Kaplan-Meier product-limit curves were generatedto calculate overall survival (OS) and freedom from local progression(FFLP). Local progression is defined as any progression of the diseaseadjacent to the treated area. OSwas calculated from the date of SBRTto the day of last follow-up or death. Log-rank test were used to com-pare estimates of survival between subgroups of patients. Patientswhowere still alive were censored onMarch 2011. Statistical routineswere performed using JMP Statistical Discovery Software ver. 9.0(SAS Institute, Cary, NC). P values <0.05 were considered statisti-cally significant.

RESULTS

Patient characteristics and demographics are sum-marized in Table 1. Our cohort of patients with pancre-atic adenocarcinoma consisted of eight males and 11females (n ¼ 19) with a mean age of 74.5 y (range: 54–91 y). The median follow-up among survival patientswas 9 mo. (range: 5.8–23.1 mo). Mean tumor volumewas 57.2 cm3 (range: 10.1–118 cm3). Fourteen patientswere treated with 20–25 Gray (Gy) in one fraction andfive patients were treated with 24–30 Gy in three frac-tions. The median prescription dose was 25 Gy (20–30Gy) at a median prescription isodense line of 70%.

We have initial follow-up imaging after SBRT ata median of 3 mo (range: 1–5 mo) in 84% of patientsand a second follow-up imaging at a median of 9 mo(range: 4.7–22.9) in 53% of patients. We were unableto determine initial tumor response in one patientdue to medical related death before follow-up imagingat 2 mo. Two patients did not have follow-up imaging

TABLE 1

Demographics and Baseline Tumor Characteristics ofPatients with Nonresectable PancreaticAdenocarcinoma Treated with SBRT

Subjects 19

Mean age (range) 74 (54–91)Gender (M:F) 8:11Initial tumor characteristics

Number of tumors 1Mean max diameter (range) 4.7 (2.3–9.1)Mean volume (range) 57.2 (10.1–118)

Prior therapy (no. of patients)Surgery 6 (32%)Chemotherapy 13 (68%)Radiation therapy 1 (5%)

Age (y); max diameter (cm); tumor volume (3 cm).

in our center. Local control was achieved in 81% ofthe patients. Freedom from local progression (FFLP)rates at 6 mo and 1 y were 88% and 65%, respectively(Fig. 1). The median time to local progression was11.43 mo (range: 3.9–15.47). The median overall sur-vival (OS) for the group was 14.37 mo. The estimated6 mo and 1 y survival were 89% and 56% (Fig. 2). Inthe subgroup of patients with metastatic disease atthe time of SBRT the estimated 6 mo and 1 y survivalwas 75% and 25%, however, this was not statisticallydifferent from the non-metastatic group with a Log-rank test value of P ¼ 0.29 (Fig. 3).

The mean tumor volume reduction among the 16 pa-tients with initial follow-up imaging studies at 3 mowas 21% (P ¼ 0.005 by single sided paired t-test). Theoverall mean tumor volume reduction at last follow-up imaging was 38% (P ¼ 0.001 by single sided pairedt-test, Fig. 4). Using RECIST criteria, two patients(13%) had complete response, five patients (31%) hadpartial response, and six patients (38%) had stable dis-ease at the time of last follow-up. Only three patients(19%) had progressive disease and three patients didnot have follow-up imaging. Using our developed par-tial tumor response grading system, 38% of patientshad a grade 3 response, 19% of patients had a grade 2response, and 13% had a grade 1 response at the timeof their last follow-up. Six patients (32%) developed dis-tantmetastasis to the liver at amedian time of 12.45moafter SBRT (range: 3.9–15.47 mo, Table 2).

No complications were observed during placement offiducial markers. Two patients (11%) presented grade1–2 adverse events including fatigue, nausea, and oneasymptomatic pyloric ulcer that responded to medicaltreatment. Three patients (16%) presented grade 3 gas-trointestinal ulcers, which responded to medical

FIG. 1. Overall freedom from local progression from the time ofSBRT. Median time to local progression was 11.43 mo (range: 3.9–15.47 mo). Freedom from local progression (FFLP) rates at 6 mo and1 y were 88% and 65%. (Color version of figure is available online.)

FIG. 2. Overall survival from the time of SBRT. Median overallsurvival for the group was 14.37 mo. The estimated 6 mo and 1 y sur-vival was 89% and 56%. (Color version of figure is available online.)

JOURNAL OF SURGICAL RESEARCH: VOL. 174, NO. 2, MAY 15, 2012322

therapy. One patient had a sterile fluid collectiondrained revealing necrotic pancreatic tissue. Two pa-tients with baseline pain had pain management proce-dures performed after SBRT including celiac plexusblock and RFA ablation of posterior spinal rami. Onepatient died within 3 mo of SBRT prior to obtainingpost-treatment imaging. Each death was reviewedand considered not related to SBRT complications.

Additionally, a 55-y old morbidly obese male receivedSBRT for a pancreatic nonfunctional neuroendocrinetumor of 2.5 cm diameter (15.5 cm3 volume). It wastreated to a dose of 30 Gy in three fractions to the80% isodose curve. This patient had a complete tumorresponse at 31 mo after SBRT. Patient is still aliveand free of disease 3 y after treatment and 4 y afterdiagnosis.

non-metastaticmetastatic

FIG. 3. Overall survival for metastatic and non-metastaticgroups. In the initially metastatic group the median survival timewas 8.43 mo versus 14.77 mo in the non-metastatic group. Howeverthis was not statistically significant. (Log-Rank test P ¼ 0.29).

DISCUSSION

Analysis of 19 patients treated with SBRT for unre-sectable adenocarcinomas of the pancreas are pre-sented. Sixty-nine percent of patients responded toSBRT as judged by >10% decrease in gross tumor vol-ume (GTV) in 11 patients and arrest of tumor growthin two patients. No patient had local tumor recurrenceon imaging 3 mo after SBRT. At 9 mo follow-up two pa-tients recurred locally and one patient recurred locallyand distally to the liver. Six patients (35%) developeddistant metastasis to the liver at a median time of12.45 mo after SBRT (range: 3.9–15.47 mo).

Two patients (11%) were admitted 6 and 8 mo afterSBRT with bleeding gastric and duodenal ulcers,which responded to medical therapy. On review oftheir delivered plan, radiation doses to affected muco-sal areas were below normal tolerance dose, makingtheir etiology unlikely from SBRT. Because of the re-ported risk of GI ulceration, fractionated SBRT withsmaller doses per fraction may reduce the risk of lu-minal gastrointestinal toxicity while maintaining lo-cal control. We modified our protocol to provideSBRT in three fractions along with chemotherapysimilar to the regimen reported by the Stanfordgroup [11]. The published literature reports a rateof GI ulceration in approximately 10% in patientstreated with SBRT for pancreatic adenocarcinoma[2, 3, 6, 11–16].

Two patients treated in our series presented withrecurrent pancreatic adenocarcinoma after pancreati-coduodenectomy and chemoradiation. One of them de-veloped distant recurrence to the liver 2 mo afterSBRT. Treatment of unresectable tumor recurrencehas been established as a method for local control inthese lesions. This suggests local control can beachieved with SBRT in locally recurrent pancreaticcancer that is refractory to multimodal treatment.Nevertheless, overall survival ultimately depends onthe development of systemic disease, which furthersupports the use for a multimodal approach withSBRT and chemotherapy. We report a local controlrate of 65% at 12 mo. This is comparable to the currentliterature in which the FFLP at 12 mo is between48.5% to 84% (Table 3) [6, 8], indicating that only35% of our patients could potentially die of uncon-trolled local disease rather than systemic failure [6].In addition to the increase in overall survival obtainedwith SBRT in the treatment of pancreatic adenocarci-noma, the importance of achieving local control shouldnot be undermined. Preventing or delaying local re-currence with SBRT not only decreases tumor burden,but may also offer palliative benefit. Untreated localdisease can lead to significant pain, gastric outlet ob-struction, biliary obstruction, and other morbidities

FIG. 4. Tumor response in patientswithnonresectable pancreatictumors treated with Stereotactic Body Radiation Therapy (SBRT).Thirty-two percent of patients (6/19) with adenocarcinoma of the pan-creas had a partial tumor response grade 3, 16% of patients (3/19) hada grade 2 response, and 11% had a partial tumor response grade 1 atthe time of their last follow-up. Tumor volume decreased from 57.2 þ35.7 cc to 33.3þ 34.9 cc (P¼ 0.001 by one-tailed paired t-test). One pa-tient died frommedical issues before imaging 3 mo post-therapy. Twopatients did not have follow-up imaging in our center. (Color versionof figure is available online.)

GOYAL ET AL.: STEREOTACTIC BODY RADIATION THERAPY 323

that decrease quality of life [2]. Thus, SBRT shouldalso be considered as a palliative option for unresect-able pancreatic adenocarcinoma.

Gemcitabine or 5-FU is commonly prescribed to pa-tients with pancreatic adenocarcinoma and cholangio-carcinoma. At our institution all candidates for SBRT

TABLE 2

Tumor Response and Recurrence of Malignancy inPatients Treated with SBRT for Nonresectable

Pancreatic Adenocarcinoma

Subjects 19*

Last follow-up imaging (mo 6 SD) 7.3 6 5.5Tumor response (RECIST)

n/a 3 (16%)PR 5 (26%)SD 6 (32%)CR 2 (11%)PD 3 (16%)

Partial tumor response graden/a 3 (16%)Grade 1 2 (11%)Grade 2 3 (16%)Grade 3 6 (32%)

Tumor recurrenceGrade 0 (no recurrence) 11 (58%)Grade 1a 3 (16%)Grade 1b 0Grade 2 1 (5%)Grade 3 1 (5%)Grade 4 3 (16%)

*Follow-up imaging was available only in 16 patients.

were evaluated by medical oncologists. Sixty-eightpercent of our patients underwent chemotherapy withgemcitabine. We acknowledge that many of these pa-tients with locally advanced cancer have other medicalco-morbidities or poor performance status thatmay preclude chemotherapy. It is our practice that allpatients with unresectable pancreatic malignanciesare offered multimodal treatment, including chemo-therapy when medically feasible, in addition to the lo-cally ablative treatment of SBRT. Thus, using SBRTearlier in the natural history of cancer progression (be-fore second-line chemotherapy) may lead to improvedoutcomes [17].

Pancreatic neuroendocrine tumors (PNETs) are rela-tively rare cancers. They are usually treatedwith surgi-cal resection as the only potential cure. Contessa et al.in 2009 was the first in reporting a series of 36 patientstreated with external beam radiotherapy for palliationof unresectable PNETs. They reported a 90% rate ofclinical response rate achieving symptomatic palliationand a median overall survival of 2 y [18]. We have suc-cessfully treated one patient with a nonfunctionalPNET obtaining a complete response at 31 mo andachieving 3 y survival and local control after SBRT. Itis difficult to draw a conclusion from a single case; how-ever, this type of tumor appears to be radiosensitive andsuitable for SBRT treatment.

SBRT for unresectable pancreatic tumors has excel-lent local control rates and low adverse events fromradiotherapy or fiducial marker placement [2, 3, 6, 8,11–15]. In previous publications fiducial markerswere placed safely by an endoscopic approach nearthe porta hepatis, gastroesophageal junction, and pan-creas in 11/13 (84.6%) patients. Fiducial markers couldnot be placed endoscopically in two patients because thelinear scope was unable to pass to the pancreatic headdue to gastric outlet obstruction or retrocrural locationof tumor [19]. Currently, only gold fiducials are avail-able for use with 19 gauge FNA needle that can beman-ually back-loaded into the tip of the needle.Compatibility with a 22 gauge needle would facilitateeasier deployment of fiducial markers within a tumor.The ability to deploy multiple markers in one passwould shorten procedure duration and decrease riskof bleeding or infectious complications [20]. In ourstudy, multiple approaches were used to place fiducialmarkers. Twelve patients had endoscopic placementfor pancreatic tumors without complication. Three pa-tients had CT-guided placement to the pancreatic tu-mor. One pancreatic tumor had a hypervascularappearance and duodenal narrowing on endoscopy,making it difficult to pass the endoscope to second por-tion of duodenum. Four patients had fiducial markersplaced surgically without any complications. In our ex-perience, surgical placement of fiducial markers is

TABLE 3

Local Control and Median Overall Survival Rates inPatients after SBRT for Adenocarcinoma of the

Pancreas

SeriesNo. of

patientsPrior

XRT (%)FFLP(%)

MedianOS (m)

Stanford phase I (3) 15 2 (13) 77 11Stanford boost phase II (8) 16 16 (100) 94 8.3Stanford gemcitabine (9) 16 0 (0) 81 11.4Stanford composite (10) 77 9 (12) 84 6.4Danish phase II (7) 22 0 (0) 57 5.7Deaconess (11) 24 3 (13) 79 NRGeorgetown (12) 28 28 (100) 86 5.3Pittsburgh (16) 71 15 (21) 48.5 10.3Case 19 0 (0) 65 14.4Total 288 73 (25) 74.6 9.1

FFLP ¼ freedom from local progression; OS ¼ overall survival.

JOURNAL OF SURGICAL RESEARCH: VOL. 174, NO. 2, MAY 15, 2012324

enhanced by use of the intra-operative ultrasound todetect the tumor(s).

RECIST is a commonly used method to report tumorresponse to treatment based on measurement of maxi-mal diameter in the abdominal tumor(s) [9, 10]. How-ever, this grading system is not the optimal way toassess the entire tumor’s response to therapy.Measure-ment of largest diameter is a snapshot into tumor effectthat can occur after chemotherapy or loco-regionaltherapy versus effect on tumor volume. Gross tumorvolume (GTV) is precisely measured in three dimen-sions during treatment planning for SBRT. The sameprocess can be used after loco-regional therapy, includ-ing SBRT to assess tumor volume. We report thechange in tumor volume after SBRT based on measur-ing GTV using the Multiplan system. Amongst the cur-rent literature, there are vague and inconsistentdefinitions of tumor response and recurrence. Withouta precise and uniform classification system, it is diffi-cult to compare results among studies and centers.We encourage the development of a grading systemfor abdominal tumor response and recurrence usingchange in tumor volume after SBRT. We encourageits use in similar cohorts in an attempt to further eval-uate different treatment modalities among centers.

It has been suggested that radiographic responseanalysis should be set no earlier than 4–6 mo afterSBRT to assess tumor response in clinical trials [21].Future long-term data from randomized clinical trialsare needed to determine the role of SBRT in the treat-ment of tumors of the pancreas. It appears that SBRTcan play a primary role in the local control of these ma-lignancies. The sample size and follow-up of this studyare smaller to other published SBRT studies but thisstudy adds to the existing literature pertaining toSBRT for pancreatic malignancies. Within the limita-

tions of a relatively small sample size we have demon-strated that SBRT is a safe form of therapy forunresectable tumors of the pancreas with a 65% localresponse rate with 14 mo mean follow-up. Future stud-ies should focus on the development of strategies todefine the role of SBRT in the treatment of pancreatictumors.

CONCLUSIONS

SBRT is a safe and effective local treatment option forunresectable pancreatic adenocarcinoma. In the multi-disciplinary management of these tumors, SBRT addsto our armamentarium of local treatment modalities,especially for non-operable patients. Fiducial markerscan be placed endoscopically, under CT-guidance, orsurgically, for planning and treatment of pancreatic tu-mors withminimal adverse events. Further prospectivestudies are ongoing to determine long-term responseand survival after SBRT for pancreatic malignancies.

ACKNOWLEDGMENTS

The authors are grateful to the staff and residents in the Depart-ment of Surgery, Medicine, Radiology, and Radiation Oncology, Uni-versity Hospital-Case Medical Center, Cleveland, Ohio. RAI wassupported by a Ruth L. Kirschstein National Research Service AwardNIH/NIDDK (T32-DK007319).

REFERENCES

1. American Cancer Society. Cancer Facts and Figures. Atlanta:American Cancer Society, 2008. 1-1-0008. Ref Type: Report.

2. Chang BW, Saif MW. Stereotactic body radiation therapy(SBRT) in pancreatic cancer: Is it ready for prime time? JOP2008;9:67.

3. Koong AC, Le QT, Ho A, et al. Phase I study of stereotactic radio-surgery in patients with locally advanced pancreatic cancer. IntJ Radiat Oncol Biol Phys 2004;58:1017.

4. Young RF. The role of the gamma knife in the treatment of ma-lignant primary and metastatic brain tumors. CA Cancer J Clin1998;48:177.

5. Yu C, Main W, Taylor D, et al. An anthropomorphic phantomstudy of the accuracy of Cyberknife spinal radiosurgery. Neuro-surgery 2004;55:1138.

6. Chang DT, Schellenberg D, Shen J, et al. Stereotactic radiother-apy for unresectable adenocarcinoma of the pancreas. Cancer2009;115:665.

7. Goyal K, Einstein D, Yao M, et al. Cyberknife stereotactic bodyradiation therapy for nonresectable tumors of the liver: Prelim-inary results. HPB Surg, 2010:2010.

8. Rwigema JC, Parikh SD, Heron DE, et al. Stereotactic body ra-diotherapy in the treatment of advanced adenocarcinoma of thepancreas. Am J Clin Oncol 2011;34:63.

9. Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines toevaluate the response to treatment in solid tumors. EuropeanOrganization for Research and Treatment of Cancer, NationalCancer Institute of the United States, National Cancer Instituteof Canada. J Natl Cancer Inst 2000;92:205.

10. Eisenhauer EA, Therasse P, Bogaerts J, et al. New responseevaluation criteria in solid tumours: Revised RECIST guideline(version 1.1). Eur J Cancer 2009;45:228.

GOYAL ET AL.: STEREOTACTIC BODY RADIATION THERAPY 325

11. Schellenberg D, Goodman KA, Lee F, et al. Gemcitabine chemo-therapy and single-fraction stereotactic body radiotherapy for lo-cally advanced pancreatic cancer. Int J Radiat Oncol Biol Phys2008;72:678.

12. Hoyer M, Roed H, Sengelov L, et al. Phase-II study on stereotac-tic radiotherapy of locally advanced pancreatic carcinoma. Ra-diother Oncol 2005;76:48.

13. Koong AC, Christofferson E, Le QT, et al. Phase II study to as-sess the efficacy of conventionally fractionated radiotherapy fol-lowed by a stereotactic radiosurgery boost in patients withlocally advanced pancreatic cancer. Int J Radiat Oncol BiolPhys 2005;63:320.

14. Lominska CE, Nasr NM, Silver NL, et al. Salvage stereotac-tic radiosurgery for locally recurrent previously irradiatedpancreatic cancer. Int J Radiat Oncol Biol Phys 2008;72:S276.

15. Mahadevan A, Shanmugam L, Kaplan I, et al. Fractionated ra-diosurgery for pancreas cancer. Int J Radiat Oncol Biol Phys2007;69:S307.

16. Chang DT, Schellenberg D, Shen J, et al. Stereotactic radiother-apy for unresectable adenocarcinoma of the pancreas. Cancer2009;115:665.

17. Lee MT, Kim JJ, Dinniwell R, et al. Phase I study of individual-ized stereotactic body radiotherapy of liver metastases. J ClinOncol 2009;27:1585.

18. Contessa JN, Griffith KA, Wolff E, et al. Radiotherapy for pan-creatic neuroendocrine tumors. Int J Radiat Oncol Biol Phys2009;75:1196.

19. Pishvaian AC, Collins B, Gagnon G, et al. EUS-guided fidu-cial placement for CyberKnife radiotherapy of mediastinaland abdominal malignancies. Gastrointest Endosc 2006;64:412.

20. Van DJ, Varadarajulu S, Jin Z. EUS 2008 Working Group docu-ment: Evaluation of EUS-guided implantation therapy (withvideo). Gastrointest Endosc 2009;69:S49.

21. KavanaghBD, Scheftera TE,Wersall PJ. Liver, renal, and retro-peritoneal tumors: Stereotactic radiotherapy. Front Radiat TherOncol 2007;40:415.