Upper gastrointestinal complications associated with gemcitabine ...

J Gastroenterol (2014) 49:1074–1080 DOI 10.1007/s00535-013-0857-3

ORIGINAL ARTICLE—LIVER, PANCREAS, AND BILIARY TRACT

Upper gastrointestinal complications associated with gemcitabine-concurrent proton radiotherapy for inoperable pancreatic cancer Kento Takatori • Kazuki Terashima • Rihito Yoshida • Aya Horai • Shinya Satake • Takayuki Ose • Naoto Kitajima Yoshikazu Kinoshita • Yusuke Demizu • Nobukazu Fuwa

•

Received: 11 December 2012 / Accepted: 2 July 2013 / Published online: 12 July 2013 Ó Springer Japan 2013

Abstract Background Little is known about acute upper gastrointestinal (GI) complications associated with gemcitabine-concurrent proton radiotherapy (GPT) for inoperable pancreatic cancer. We investigated acute GI complications following GPT in patients with inoperable pancreatic cancer using small-bowel endoscopy. Methods This prospective single center observational study was conducted at the Hyogo Ion Beam Medical Center from January 2010 to January 2012. Ninety-one patients who had clinically and medically inoperable pancreatic cancer treated by GPT were analyzed. Endoscopic examinations were performed before and after GPT to clarify the incidence rates of radiation-induced ulcers, GI hemorrhage, and GI perforation associated with GPT. Results Post-treatment endoscopic examinations revealed that 45 (49.4 %) patients had radiation-induced ulcers in the stomach and duodenum. Of those, many ulcerative lesions were found in the lower stomach (51 %) and horizontal part of the duodenum (39 %), regardless of the primary tumor site in the pancreas. Neither GI hemorrhage, nor perforation, was found in post-treatment endoscopy examinations. K. Takatori (&) R. Yoshida A. Horai S. Satake T. Ose N. Kitajima Department of Internal Medicine, Kasai City Hospital, 1-13 Yokoo, Houjou, Kasai, Hyogo 675-2393, Japan e-mail: [email protected] K. Terashima Y. Demizu N. Fuwa Department of Radiology, Hyogo Ion Beam Medical Center, Shingu, Hyogo, Japan Y. Kinoshita Department of Gastroenterology and Hepatology, Shimane University School of Medicine, Izumo, Shimane, Japan

123

Conclusion Approximately half of the patients treated with GPT for inoperable pancreatic cancer exhibited radiation-induced ulcers in the stomach and duodenum. Keywords Proton radiotherapy Pancreatic cancer Radiation ulcer

Introduction Pancreatic cancer is the fourth leading cause of cancer death and there were approximately 43000 new cases reported in 2010 in the United States [1, 2]. Although surgical complete resection is the only curative treatment option for this disease, approximately 80–85 % of affected patients are diagnosed with unresectable cancer and their 5-year survival rate is less than 5 % [3]. Chemotherapy is considered to be the main treatment option for these unresectable cases, while chemo-radiotherapy is expected to improve survival and quality of life, especially for locally advanced pancreatic cancer (LAPC) without metastasis [4, 5]. We previously reported results of a phase I/II trial in which the efficacy and feasibility of gemcitabine-concurrent proton radiotherapy (GPT) for LAPC were promising, with 1-year freedom from local-progression, progressionfree, and overall survival rates of 81.7, 64.3, and 76.8 %, respectively [6]. A favorable dose distribution of so-called Bragg’s peak [7] in proton radiotherapy as compared with photon (X-ray) irradiation might have been related to our good results, while gastrointestinal (GI) toxicity was thought to be a dose-limiting factor due to the close proximity of the pancreas to upper GI tract organs such as the stomach and duodenum. In our previous study, we performed endoscopic examinations only for patients with

J Gastroenterol (2014) 49:1074–1080

abdominal symptoms or gastrointestinal bleeding complications, of whom 7.5 % (3/40) had Grade 3 gastric ulcers and 2.5 % (1/40) Grade 5 hemorrhagic gastric ulcers several months after completing the therapy [6]. Takagi et al. [8] reported a case of pancreatic cancer with lethally late hemorrhaging after undergoing a pancreatoduodenectomy and heavy ion beam therapy. Although the definite cause for late hemorrhage was not determined in their case, autopsy findings were consistent with radiation enteritis. Since the gastrointestinal complications of proton beam therapy for pancreatic cancer have yet to be fully elucidated, it is necessary to investigate the incidence rate of GI complications such as radiation-induced ulcers in a more precise manner. In the present, in contrast to our previous study, we routinely performed endoscopic examinations for all enrolled patients before and after undergoing GPT for inoperable pancreatic cancer to investigate upper gastrointestinal complications associated with GPT via endoscopic observations.

Methods Treatment protocol Concurrent and adjuvant chemotherapy All patients were scheduled to receive intravenous infusions of gemcitabine (800 mg/m2) for 30 min for the initial 3 weeks during 5 weeks of proton therapy. Gemcitabine was administered when the absolute granulocyte count was [2000/mm3 and the platelet count was [70000/mm3 on the scheduled day. All patients received systemic gemcitabine-based chemotherapy for as long as possible after finishing GPT. Proton radiotherapy Proton radiotherapy is a method of external beam radiotherapy that utilizes ionizing radiation [9]. Proton ion beams show an increase in energy deposition with penetration depth up to a sharp maximum at the end of their range to form the so-called Bragg peak, with nearly no dose deposited in normal tissues beyond that peak. Thus, a favorable dose distribution with a steep dose fall-off occurs at the field borders and more precise dose localization can be achieved with proton beams as compared with conventional X-ray irradiation (photon beams) [7]. The biological effects of proton beams have been evaluated in vitro and in vivo at the Hyogo Ion Beam Medical Center (HIBMC), and their relative biological effectiveness (RBE) value was determined to be 1.1 depending on the depth of the spread-out Bragg peak [10]. We assume that proton

1075

beams have approximately the same RBE value for a particular type of tissue, thus doses expressed in gray equivalent (GyE) are considered to be directly comparable to photon doses. The present patients were treated with proton radiotherapy at a total dose of 67.5 GyE in 25 fractions. The planned area of irradiation was the primary tumor plus apparent lymph nodes and prophylactic irradiation regions containing the draining lymph nodes, paraaortic lymph nodes, and peripheral regions surrounding the celiac artery and superior mesenteric artery. We also took safety margin (5 mm) and respiratory movement (1–5 mm) into consideration when planning irradiation. The dose restrictions for the stomach and duodenum were approximately 50 GyE [11]. Additionally, we planned for the irradiated volume of the stomach and duodenum to be as small as possible. The study protocol is shown in Fig. 1. Endoscopic assessment Endoscopic examinations of the small bowel from the esophagus to horizontal part of the duodenum were performed using a double-balloon endoscope (EN-450P5/20) before and after GPT to investigate acute radiation-related gastroduodenal complications. A post-treatment endoscopic examination was also performed at the completion of therapy. For all endoscopic examinations, we inserted the endoscope to the horizontal part of the duodenum only by pushing without use of a balloon. Radiation associated mucosal lesions were defined as newly developed in the radiation field of the stomach and duodenum, and compared with pre-treatment endoscopic findings. Radiationinduced gastroduodenal complications were defined as one of the following: (1) radiation ulcer, defined as a circumscribed mucosal break C3 mm in diameter with perceptible depth in the stomach and duodenum found in a posttreatment endoscopy examination [12]; (2) radiationinduced mucosal hemorrhage, defined as spontaneous and active bleeding with a background of radiation mucosal damage required for endoscopic hemostatic therapy; (3) perforation of the gastrointestinal tract, defined as a radiation-associated gastrointestinal perforation found in a post-treatment examination with endoscopy or other modalities such as plain X-ray examination or computed tomography. To describe the location of the radiationinduced ulcer, we classified the upper GI tract from the stomach to horizontal part of the duodenum into 6 segments, in which the stomach was divided into 3 regions (upper part: U, middle part: M, lower part: L) and the duodenum into 3 regions (bulbus: B, descending part: D, horizontal part: H). All endoscopic data was collected by 2 experienced endoscopists (K.T., N.K.). To distinguish between pre-existing lesions in the radiation field and new or aggravated lesions induced by proton radiotherapy,

123

1076


Fig. 1 Study protocol

endoscopists compared endoscopic findings obtained before and after proton radiation therapy. All adverse events associated with the GI tract were assessed using the Common Terminology Criteria for Adverse Events v3.0.

protocol was approved by our Institutional Ethics Committee and written informed consent for participation was obtained from all patients. Statistical analysis

Patients We enrolled 126 consecutive patients with either locally unresectable or clinically inoperable pancreatic cancer treated at HIBMC from January 2010 to January 2012. Patients with metastatic disease were included if their distant disease was low-volume and prognosis was favorable with control of the primary tumor. Patients with resectable pancreatic tumors were included if they had several reasons for a diagnosis of clinically inoperable, such as high age, severe comorbidities, and patient will. Of these 126 patients, 35 were excluded from analysis for the following reasons: 9 were observed up to the descending part of the duodenum in post-treatment endoscopic examinations, 7 were post-gastrectomy, 7 were post-pancreatoduodenectomy, 2 were post-gastrojejunostomy, 9 refused a post-treatment endoscopic examination, and 1 had an invasive pancreatic tumor in the stomach. Finally, 91 patients were eligible and analyzed in this study. Prophylactic lansoprazole (30 mg/day) and rebamipide (300 mg/ day) were prescribed to all enrolled patients prior to GPT. Thirty-eight (41.8 %) of the patients had histologically proven adenocarcinoma of the pancreas, while the remainder had a diagnosis of pancreatic cancer based on clinical imaging findings. In addition, 51 (56.0 %) had received prior chemotherapy such as gemcitabine or TS-1Ò (tegafur/gimestat/potassium oxonate), 54 (59.3 %) were positive for anti-helicobacter pylori (HP) or immunoglobulin-G (IgG) antibodies, and 31 (34 %) were taking nonsteroidal anti-inflammatory drugs (NSAIDs). The study

123

Statistical analysis was performed using SPSS 17.0 Software for Windows. Student’s t test was used to compare continuous variables, and v2 and Fisher’s exact tests were used for categorical variables. Binary logistic regression analysis was used to identify risk factors for radiation ulcer. A p value \0.05 was considered to indicate statistical significance.

Results Acute GI complications Pre-treatment endoscopic examinations were conducted in 91 of the enrolled patients, which revealed lesions in the stomach and duodenum of 9 (gastric erosions in 3, duodenal ulcer scars in 3, gastric ulcer scars in 2, early gastric cancer in 1). Post-treatment endoscopic examinations showed that 49.4 % (45/91) of the patients had a total of 51 radiation-induced ulcers in the stomach and duodenum (Figs. 2, 3), which were located in the lower stomach (n = 26, 51 %), descending part of the duodenum (n = 5, 10 %), and horizontal part of the duodenum (n = 20, 39 %). When 9 patients whose horizontal part of the duodenum was not endoscopically examined were included in the analysis, 48 % (48/100) had a total of 54 radiationinduced ulcers in the stomach and duodenum. Regardless of the pancreatic tumor site, many of the radiation-induced lesions were observed in the lower stomach and horizontal


1077 Table 1 Locations of radiation ulcers in relation to pancreatic cancer site in 91 patients Location of mucosal damage

Fig. 2 Radiation-induced ulcer (antrum): a well-circumscribed gastric ulcer in the pyloric end

Fig. 3 Radiation-induced duodenal ulcer: a circumferential ulcer in the horizontal part of the duodenum

part of the duodenum (Table 1). We found no mucosal lesion with spontaneous or active bleeding, and no cases of gastrointestinal perforation at the completion of GPT. None of the present patients complained of abdominal symptoms considered to be related to the presence of a GI ulcer and their condition was scored as grade 1. In addition, there were no differences in regard to clinical symptoms at the completion of GPT and the duration of hospital stay between patients with and without ulcers. Clinical characteristics, including age, gender, performance status, pancreatic tumor site, tumor node metastasis (TMN) classification of T factor, maximum radiation dose, and presence or absence of prior chemotherapy, anti-HP IgG antibody, NSAID administration, biliary stent placement, smoking history were analyzed in 45 (49.4 %) of the 91 patients who developed radiation-induced ulcers (Table 2). Univariate analysis showed no significant differences in regard to these clinical parameters between patients with and without radiation-induced ulcers. Late-phase GI complications The treatment strategy including administration of lansoprazole was not changed regardless of the presence or

Number of ulcers Site of pancreatic tumor Head (40 patients)

Body (44 patients)

Tail (7 patients)

Number of ulcers found in upper part of stomach

0

0

0

Number of ulcers found in middle part of stomach

0

0

0

Number of ulcers found in lower part of stomach

14

11

1

Number of ulcers found in duodenal bulb

0

0

0

Number of ulcers found in descending part of duodenum

3

2

0

Number of ulcers found in horizontal part of duodenum

6

13

1

Total number of ulcers

23 ulcers

26 ulcers

2 ulcers

Total number of patients

40 patients

44 patients

7 patients

absence of acute GI complications. During the 10-month follow-up period, 2 patients (2.2 %) exhibited grade 4 and grade 5 bleeding gastric ulcers at 3 and 10 months, respectively, after completion of GPT. In addition, 1 (1.1 %) with pancreatic head cancer with a metallic biliary stent exhibited a grade 5 duodenal perforation at 5 months after completion of GPT.

Discussion Overall, 49.4 % patients who underwent GPT for medically and clinically inoperable pancreatic cancer had radiation-induced ulcers in the stomach and duodenum revealed in post-treatment endoscopy examinations. In detailed endoscopic observations, a large number of the radiation-induced lesions were seen in the lower stomach (51 %) and horizontal part of the duodenum (39 %), regardless of the primary tumor site in the pancreas. Radiation ulcer is a complication of photon radiotherapy for pancreatic cancer, because the irradiated area is adjacent to the upper GI tract. However there are few reports of incidence. A previous study reported an incidence rate for upper GI ulcers of 37.5 % (9/24) following gemcitabineradiotherapy for inoperable pancreatic cancer [2], while others noted incidence rates for upper GI ulcers with bleeding following gemcitabine-radiotherapy for inoperable pancreatic cancer of between 9.5 % (2/21) [13] and

123

1078 Table 2 Risk factors for radiation-induced ulcers


Variable

Total n = 91, n (%)

With radiation ulcer n = 45, n (%)

Without radiation ulcer n = 46, n (%)

Univariate analysis p value

Age (years, mean ± SD)

64.4 ± 8.34

64.1 ± 8.62

64.6 ± 8.14

0.778a

Male

50 (54.9)

23 (51.1)

27 (58.7)

0.530b

Female

41 (45.1)

22 (48.9)

19 (41.3) 0.953c

ECOG performance status 0

67 (73.6)

34 (75.6)

33 (71.7)

1

20 (22.0)

8 (17.8)

12 (26.1)

4 (4.4)

3 (6.7)

1 (2.2)

2 Tumor location

0.503c

Head

40 (44.0)

21 (46.7)

Body

44 (48.4)

22 (48.9)

19 (41.3) 22 (47.8)

Tail

7 (7.6)

2 (4.4)

5 (10.9)

UICC-TMN 0.143c

T factor T4

70 (76.9)

31 (68.9)

T3

17 (18.7)

10 (22.2)

39 (84.8) 7 (15.2)

T2

2 (2.2)

2 (4.4)

0 (0.0)

T1

2 (2.2)

2 (4.4)

0 (0.0) 0.174b

Prior chemotherapy (?)

51 (56.0)

22 (48.9)

29 (63.0)

(-)

40 (44.0)

23 (51.1)

17 (37.0) 0.405b

HP antibody

ECOG Eastern Cooperative Oncology Group, UICC Union for International Cancer Control, TNM tumor node metastasis, HP Helicobacter pylori, NSAID non steroidal anti-inflammatory drug a

Student’s t test

b

Binary logistic regression analysis

c

Chi square and Fisher’s exact tests

A p-value \0.05 was considered to indicate statistical significance

Positive Negative

54 (59.3) 30 (33.0)

27 (60.0) 13 (28.9)

27 (58.7) 17 (37.0)

Unknown

7 (7.7)

5 (11.1)

2 (4.3) 0.237b

NSAIDs administration (?)

31 (34.1)

18 (40.0)

13 (28.3)

(-)

60 (65.9)

27 (60.0)

33 (71.7) 0.869b

Placement of biliary stent (?)

25 (27.5)

13 (28.9)

12 (26.1)

(-)

66 (72.5)

32 (71.1)

34 (73.9) 0.077b

Smoking history (?)

26 (28.6)

(-)

65 (71.4)

36 (80)

29 (63)

49.8 ± 3.75

49.7 ± 4.98

49.9 ± 1.95

Maximum Radiation dose (GyE, mean ± SD)

20.8 % (5/24) [2]. However, the type of radiation (photon) in those was different from that in our study (proton beams) and they only performed an endoscopic examination when indicated, thus the incidence may have been underestimated. Moreover, those ulcerative lesions in the upper GI tract were observed in the late phase (several months after completing chemo-radiotherapy), indicating that the actual incidence rate in the acute phase is unknown. In contrast to those studies, we routinely conducted endoscopic examinations of all enrolled patients both before and after treatment. It is considered that a post-treatment endoscopic examination can detect subclinical mucosal injury with

123

9 (20)

17 (37) 0.831a

possible risk of complications such as GI bleeding and perforation. Our findings revealed that the incidence rate of radiation ulcers was high (49.4 %), though no bleeding lesions were noted at the completion of therapy. As for the locations of the radiation-induced ulcers, the lower stomach (51 %) and horizontal part of the duodenum (39 %) were often seen as affected in post-treatment endoscopic examinations. One reason may simply be that the targeted pancreatic tumor was located in proximity to these regions in most of our cases. However, it is interesting that this tendency held true regardless of the tumor site. Hamilton et al. [14] speculated that a high dose of


radiation may increase the risk for gastric and duodenal ulcerations in patients undergoing abdominal radiation therapy for testicular cancer in a study of the relationship between radiation dose and GI toxicity. However, the maximum dose irradiated to the stomach and duodenum did not significantly differ between our patients with and without a GI ulcer. We confirmed the expected irradiation dose to the target lesion by pre-treatment CT and dosevolume histogram findings. On the other hand, the GI tract contains peristaltic organs such as the stomach and duodenum, thus, it was difficult to predict the exact dose irradiated to the GI tract. It is important to note that an unexpected high dose may be given and, consequently, severe GI toxicity might occur. Moreover, we can not rule out the possibility that the planned irradiations were not done precisely, because there was no relationship between the occurrence of radiation ulcers and maximum irradiated dose. With this point in mind, we currently perform GPT after placement of a fiducial marker adjacent to the pancreatic tumor using an angiographic procedure to more accurately achieve the planned irradiation. We also analyzed several clinical characteristics in 45 (49.4 %) of our patients who had GI ulcers in the stomach and duodenum. Prior chemotherapy (p = 0.174) did not have affect on the incidence of radiation-induced ulcers. Similarly, Didolkar et al. [15] reported in their study of image-guided stereotactic radiosurgery for locally advanced pancreatic adenocarcinoma that the incidence of GI toxicity (grade III or IV) did not correlate with prior chemotherapy, while neither Helicobacter pylori infection (p = 0.405) nor NSAID administration (p = 0.237) was related to the incidence of radiation-induced ulcers. NSAID administration and atrophic gastritis induced by Helicobacter pylori infection are risk factors for ulcerative disease in the upper GI tract [16, 17]. However, all eligible patients in our study had already used as prophylactics a proton pump inhibitor (lansoprazole 30 mg per day) and mucosal protectant (rebamipide 300 mg per day) from the beginning of proton radiotherapy. Therefore, it was difficult to evaluate whether those factors exerted an influence on development of radiation-associated ulcers. It is important to determine the influence of these factors in relation to GI toxicity under the condition of no prophylactic drugs in a future study. However, given the high incidence of radiation-induced ulcers in our and previous studies of chemo-radiotherapy of pancreatic cancer [2, 13, 18], in which potentially lethal GI damage can occur, it was reasonable to give muco-protecting drugs to all enrolled patients in our study. This study has some limitations. First, we excluded 9 (7.1 %) patients because of their refusal to undergo posttreatment examinations, thus the incidence rate of radiationinduced ulcers may have been under-estimated. Second, we

1079

enrolled 53 (58 %) patients who did not have histologically proven adenocarcinoma of the pancreas. However, the reported sensitivity of CT for revealing pancreatic carcinoma is high, ranging from 89 to 97 % [19]. Therefore, our data for the incidence rate of radiation-induced ulcers in patients with pancreatic cancer is considered to be reasonable. Third, our results were mainly obtained during the acute phase of proton-radiation associated upper GI complications. Since the HIBMC is a specialized institute for treatment with heavy particle radiotherapy, follow-up examinations and chemotherapy treatments were performed at other hospitals after the radiation therapy was completed. Thus, the post-treatment endoscopic examinations conducted at our institute were done within a few days after completing treatment. It has been reported that radiationassociated ulceration may occur some time after completion of treatment, with a peak effect 1–2 months after therapy [20]. Also, our previous results suggested that 7.5 % (3/40) of patients had grade 3 gastric ulcers and 2.5 % (1/40) grade 5 hemorrhagic gastric ulcers [6]. In the present study, even though the follow-up observational period was not adequate, 2 patients (2.2 %) exhibited grade 4 and grade 5 bleeding gastric ulcers, respectively, while 1 (1.1 %) developed a grade 5 duodenal perforation. In this regard, we recommend that all treated patients receive follow-up endoscopic examinations regardless of the presence of mucosal damage, while we are also in the process of collecting and analyzing longer term follow-up data. In addition, it is not clear whether acute and late GI toxicity induced by proton radiation is milder than that induced by photon radiation for pancreatic cancer, since no randomized control trials have been performed in this field. Therefore, the safety and efficacy of proton radiotherapy for pancreatic cancer should be clarified in the future. In conclusion, approximately half of our patients who underwent GPT for medically and clinically inoperable pancreatic cancer had GI ulcers at the completion of treatment. Further improvements of GPT are warranted, while the development of late-phase GI complications should be carefully monitored in treated patients. Conflict of interest of interest.

The authors declare that they have no conflict

References 1. Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin. 2010;60:277–300. 2. de Lange SM, van Groeningen CJ, Meijer OW, Cuesta MA, Langendijk JA, van Riel JM, et al. Gemcitabine-radiotherapy in patients with locally advanced pancreatic cancer. Eur J Cancer. 2002;38:1212–7. 3. Li D, Xie K, Wolff R, Abbruzzese JL. Pancreatic cancer. Lancet. 2004;363:1049–57.

123

1080 4. Burris HA 3rd, Moore MJ, Andersen J, Green MR, Rothenberg ML, Modiano MR, et al. Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: a randomized trial. J Clin Oncol. 1997;15:2403–13. 5. Cardenes HR, Moore AM, Johnson CS, Yu M, Helft P, Chiorean EG, et al. A phase II study of gemcitabine in combination with radiation therapy in patients with localized, unresectable, pancreatic cancer: a Hoosier Oncology Group Study. Am J Clin Oncol. 2011;34:460–5. 6. Terashima K, Demizu Y, Hashimoto N, Jin D, Mima M, Fujii O, et al. A phase I/II study of gemcitabine-concurrent proton radiotherapy for locally advanced pancreatic cancer without distant metastasis. Radiother Oncol. 2012;103:25–31. 7. Schulz-Ertner D, Tsujii H. Particle radiation therapy using proton and heavier ion beams. J Clin Oncol. 2007;25:953–64. 8. Takagi K, Takada T, Amano H, Yoshida M, Miura F, Isaka T, et al. Late hemorrhage after pancreatoduodenectomy and heavy ion beam therapy. J Hepatobiliary Pancreat Surg. 2007;14:331–5. 9. Wilson RR. Radiological use of fast protons. Radiology. 1946;47: 487–91. 10. Kagawa K, Murakami M, Hishikawa Y, Abe M, Akagi T, Yanou T, et al. Preclinical biological assessment of proton and carbon ion beams at Hyogo Ion Beam Medical Center. Int J Radiat Oncol Biol Phys. 2002;54:928–38. 11. Kavanagh BD, Pan CC, Dawson LA, Das SK, Li XA, Ten Haken RK, et al. Radiation dose-volume effects in the stomach and small bowel. Int J Radiat Oncol Biol Phys. 2010;76:S101–7. 12. Yeomans ND, Naesdal J. Systematic review: ulcer definition in NSAID ulcer prevention trials. Aliment Pharmacol Ther. 2008; 27:465–72.

123

J Gastroenterol (2014) 49:1074–1080 13. Shibuya K, Oya N, Fujii T, Doi R, Nakamura A, Matsuo Y, et al. Phase II study of radiation therapy combined with weekly lowdose gemcitabine for locally advanced, unresectable pancreatic cancer. Am J Clin Oncol. 2011;34:115–9. 14. Hamilton CR, Horwich A, Bliss JM, Peckham MJ. Gastrointestinal morbidity of adjuvant radiotherapy in stage I malignant teratoma of the testis. Radiother Oncol. 1987;10:85–90. 15. Didolkar MS, Coleman CW, Brenner MJ, Chu KU, Olexa N, Stanwyck E, et al. Image-guided stereotactic radiosurgery for locally advanced pancreatic adenocarcinoma results of first 85 patients. J Gastrointest Surg. 2010;14:1547–59. 16. Adamu MA, Schottker B, Weck MN, Brenner H. Helicobacter pylori infection is strongly associated with gastric and duodenal ulcers in a large prospective study. Clin Gastroenterol Hepatol. 2012;10:487–93. 17. Sostres C, Gargallo CJ, Arroyo MT, Lanas A. Adverse effects of non-steroidal anti-inflammatory drugs (NSAIDs, aspirin and coxibs) on upper gastrointestinal tract. Best Pract Res Clin Gastroenterol. 2010;24:121–32. 18. Murphy JD, Adusumilli S, Griffith KA, Ray ME, Zalupski MM, Lawrence TS, et al. Full-dose gemcitabine and concurrent radiotherapy for unresectable pancreatic cancer. Int J Radiat Oncol Biol Phys. 2007;68:801–8. 19. Wong JC, Lu DS. Staging of pancreatic adenocarcinoma by imaging studies. Clin Gastroenterol Hepatol. 2008;6:1301–8. 20. Coia LR, Myerson RJ, Tepper JE. Late effects of radiation therapy on the gastrointestinal tract. Int J Radiat Oncol Biol Phys. 1995;31:1213–36.