- Email: [email protected]
Double-blinded, randomized, placebo-controlled study to evaluate the effectiveness of sulphasalazine in preventing acute gastrointestinal complications due to radiotherapy
Radiotherapy and Oncology 57 (2000) 125±129
www.elsevier.com/locate/radonline
Double-blinded, randomized, placebo-controlled study to evaluate the effectiveness of sulphasalazine in preventing acute gastrointestinal complications due to radiotherapy È zenirler b, AysËe Dursun c Diclehan KilicË a,*, ÇIbrahim Egehan a, Seren O a Department of Radiation Oncology, Gazi University Hospital, Ankara, 06510, Turkey Department of Internal Medicine and Gastroenterology, Gazi University Hospital, Ankara, 06510, Turkey c Department of Pathology, Gazi University Hospital, Ankara, 06510, Turkey
b
Received 9 February 2000; received in revised form 30 June 2000; accepted 19 July 2000
Abstract Background and purpose: Acute radiation-induced diarrhea occurs in approximately 80% of the patients receiving pelvic radiotherapy. It is caused by gastrointestinal irritation and in¯ammation. Eicosanoids are thought to be one of the mechanisms of this. Sulphasalazine is an inhibitor of their synthesis in the mucosa. This randomized clinical trial was undertaken to evaluate its effect in preventing acute radiation enteritis (ARE). Materials and methods: Prospectively, 87 patients receiving pelvic radiotherapy were randomized, in a double-blind fashion. Two tablets twice daily of sulphasalazine (500 mg) or placebo were administered orally. Patients were evaluated weekly according to diarrhea grading for the primary study endpoint and according to late effect of normal tissue-subjective objective management analytic (LENT-SOMA) criteria for the secondary endpoint during irradiation. Results: Groups did not differ for age, gender, tumour site or irradiation procedure. Diarrhea occurred in 55 and 86% of the sulphasalazine and placebo groups, respectively (P 0:001). Gastrointestinal toxicity was seen in 80 and 93% of the sulphasalazine and placebo groups according to the maximum LENT-SOMA score (P 0:07). According to the maximum LENT-SOMA score between the two groups, signi®cant differences in favor of sulphasalazine were found for each week. Conclusion: Sulphasalazine (2 g/day) was found to be effective in decreasing the symptoms of ARE. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Pelvic irradiation; Acute radiation enteritis; Sulphasalazine
1. Introduction and scienti®c background Radiation therapy (RT) is frequently administered either primary or adjuvant for a variety of pelvic malignancies, including urological, gynecological and colorectal cancers. Although the bene®ts of this treatment have been well established, a number of patients experience distressing complications as a result of radiation injury to normal tissue [21]. Small bowel tolerance is a highly signi®cant dose-limiting factor because of early adverse effects [4,14]. Acute radiation enteritis (ARE), characterized by abdominal cramping and diarrhea, occurs in approximately 75% of patients and usually begins in the second or third week of RT [4]. These side effects cause discomfort and decreases the therapeutic bene®t by increasing the overall treatment time [18]. * Corresponding author.
Several approaches have been investigated in order to decrease the incidence of both ARE and late sequelae: surgical procedures, such as excluding the small intestine from the irradiated ®eld, partitioning the abdominal and pelvic cavities with endogenous material like omentum or synthetic absorbable mesh or implanted tissue expanders [21]; maneuvers in displacing the small bowel [8]; usage of thiol compounds [20]; prostaglandine (PG) inhibitors [17]; and sucralphate [7]. Radiation induces the synthesis of nuclear regulating proteins. These proteins are able to regulate cytokines and have secondary effects on eicosanoids [3]. Mennie et al. [11] hypothesized that PG release might be another etiological factor in radiation-induced diarrhea. Sulphasalazine has been proven useful in mild or moderately active attacks of ulcerative colitis [9]. It inhibits both the lipoxygenase and cycloxygenase pathway of arachidonic acid metabolism [9]. It has a characteristic and speci®c
0167-8140/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0167-814 0(00)00254-1
126
D. KilicË et al. / Radiotherapy and Oncology 57 (2000) 125±129
af®nity to intestinal tissue. It is a good choice for colitis that effects subepithelial tissues [9,19]. This study was aimed towards assessing the ef®cacy of sulphasalazine enteric tablets, 2 g/day, administered orally throughout the irradiation period to reduce the symptoms of ARE in subjects receiving pelvic irradiation. It was hoped that sulphasalazine would provide a salutary effect without corresponding systemic toxicity. 2. Materials and methods Eighty-seven patients with current histological proof of cancer in the pelvis without metastases beyond the regional lymph nodes, in whom continuous external beam pelvic RT was planned, agreed to take part in the study. They gave their informed consent in accordance with the Declaration of Helsinki, and the study design was approved by the Ethics Committee of our University. Patients were ineligible for the study if they lacked a functioning rectum, if they had stool incontinence, if the stool frequency was $6/day or if the perineum was in the planned irradiation volume. Patients using digoxin, with a history of prior pelvic/abdominal irradiation, in¯ammatory bowel disease, known salicylate hypersensitivity, nephrotic syndrome, hepatic values of twice the normal or Karnofsky performance status of ,70 were also ineligible. All patients had to have no participation in another protocol. No systemic administration of cytotoxic chemotherapy was allowed during RT. Patients received 46±50 Gy in 23±25 fractions of external beam RT to the whole pelvis with megavoltage energy (10 MV). They were treated with opposed anterior±posterior: posterior±anterior (AP:PA) ®elds. For the AP:PA ®elds, the lateral ®eld borders were 1±2 cm lateral to the widest bony margin of the true pelvic side walls. The superior border of the radiation ®eld was placed no higher than the interspace between the fourth and ®fth lumbar vertebrae. Patients whose treatment planning included the perineum in the planned irradiation volume were excluded. Treatment was delivered to all subjects in the supine position. Treatment with a full bladder to displace the small intestine was encouraged. The daily dose, speci®ed at isocenter, was 2 Gy. A tumour boost was allowed either with brachytherapy or with external beam RT according to the tumour site after the completion of a total pelvic dose of 46±50 Gy. Patients were randomized by a different physician from the one treating before RT. Patients were randomized in a double-blind fashion, to receive two tablets twice daily of either 500 mg sulphasalazine (Salazopyrin-EN w) or a placebo which appeared identical. The placebo tablets were composed of coloring agent, talc, cellulose and puri®ed water. Patients were strati®ed according to gender, type of cancer, prior surgery and total cumulative dose of radiotherapy (including boost ®elds). The toxicity was assessed weekly by the treating physician during the ®rst 5 weeks of RT. Standard toxicity
criteria were used to grade the severity of diarrhea. Diarrhea was graded as: grade 0, no diarrhea or increased stool frequency; grade 1, increase of 2±3 stools/day; grade 2, increase of 4±6 stools/day or nocturnal stool; grade 3, increase of 7±9 stools/day or incontinence; grade 4, increase of $10 stools/day or grossly bloody diarrhea [10]. A late effect of normal tissue-subjective objective management analytic (LENT-SOMA) table for the small intestine/colon was modi®ed and used to grade the severity of acute radiation-induced gastrointestinal complications [13]. Grading was done with the remaining nine parameters after excluding the parameters of ulceration and stricture. The weekly scores of the parameters were summed as it was originally recommended, but were not divided by the number of elements as this was not advocated previously [5]. The toxicity was graded according to the summed LENT-SOMA score. If the patient had a summed score of 0±2, 3±9, 10±18, 19±27 or 28±36, the toxicity was accepted as grade 0, 1, 2, 3 or 4, respectively. In addition to these, if there were two or more 3/4 toxicity scores in any parameters, the patient was accepted as having grade 3/4 toxicity. The primary endpoint of the study was diarrhea. The secondary endpoint was the LENT-SOMA scores of the patients. At the end of pelvic RT, ®ve diarrhea and ®ve LENT-SOMA scores were recorded for each patient. The maximum severity of diarrhea/LENT-SOMA score was the highest score of them all. The treatment with sulphasalazine or placebo continued until pelvic RT was completed, unless the patient began to experience $7 stools/day above the pretreatment baseline, which led to discontinuation of study medication. 2.1. Statistics All gradings were done in a blinded fashion. Data analysis was carried out with SPSS 9.01 for Windows program package (SPSS Inc). The proportion of subjects who experience diarrhea was accepted as 80%, according to ®ndings in the literature and an approximate percentage observed in our previous patients. The sample size required to detect a reduction in the proportion of patients who experience diarrhea from the 80% level to 50% or less if patients were given sulphasalazine was estimated. We accepted a signi®cance level of 0.05 and a power of 0.90. The study design called for the entry of 82 patients (41 in each arm). Patient characteristics (Table 1; tumour type, gender) were compared using the Chi-square test. Quantitative variables (®eld dimensions, pelvic radiation dose) were compared using the non-parametric, Mann±Whitney U-test, and expressed as means ^ standard errors. Chi-square tests were used to ®nd the signi®cance of differences in toxic reactions between patients receiving sulphasalazine or placebo. The statistical differences in the gradings were estimated using a ®xed cutpoint of patients with grade 2 and more reactions. Only two sided results were used. P values of 0.05 and less were considered to be signi®cant.
D. KilicË et al. / Radiotherapy and Oncology 57 (2000) 125±129 Table 1 Patient characteristics a Baseline factors
Gender Male Female Type of cancer Rectum/rectosigmoid Endometrium Cervix uteri Prostate Bladder Pelvic sarcoma Age (years) Median Mean a
Table 3 Severity of diarrhea according to treatment a Sulphasalazine (n 44)
Placebo (n 43)
Number %
Number %
26 18
59 41
23 20
54 46
22 4 5 6 6 1
50 9 11 14 14 2
20 8 3 8 4 ±
47 18 7 19 9 ±
60 59
64 61
No signi®cant difference was observed.
Ninety-®ve percent con®dence intervals (CI) for the proportions were calculated and are mentioned in the text. 3. Results Eighty-seven patients entered the study to receive either sulphasalazine or placebo between August 1997 and April 1999. The main characteristics and irradiation procedures (Table 2) did not differ between the two groups. There were no losses to follow up, no refusals to continue the trial without any side effects, and no complications due to the drug. During irradiation, diarrhea (grades 1±4) occurred in 55% (24/44; 95% CI, 39±69%) of the sulphasalazine and 86% (37/43; 95% CI, 75±96%) of the placebo group, and this difference was found to be statistically signi®cant (P 0:001). Approximately 77% (20/26) of the patients who had no diarrhea during irradiation (grade 0 patients) were in the sulphasalazine group. Diarrhea of grade 2 and more was seen in 27% (12/44; 95% CI, 14±40%) of the sulphasalazine and 49% (21/43; 95% CI, 34±63%) of the Table 2 Irradiation procedure a Sulphasalazine (n 44) Placebo (n 43) Field dimensions Surface at isocenter (cm 2) 251.5 ^ 8.2 Anterior±posterior (cm) 21.2 ^ 0.4 Pelvic radiation dose (Gy) 49.3 ^ 0.2 Boost with external RT % Patients 29 Dose (Gy) 18.4 ^ 1.0 Boost with intracavitary RT % Patients 13 Dose (Gy) 23.3 ^ 0.6 a
127
No signi®cant difference was observed.
250.7 ^ 9.4 21.0 ^ 0.3 49.5 ^ 0.2 30 19.3 ^ 0.4 14 21.3 ^ 1.3
Maximum severity of diarrhea Grade
Sulphasalazine (n 44) Placebo (n 43) Number %
Number %
0 1 2 3 4
20 12 9 3 ±
6 16 8 6 7
a
46 27 20 7 ±
14 37 19 14 16
P 0:038.
placebo group (P 0:038). In the sulphasalazine group, grades 1, 2 and 3 diarrhea were seen in 45, 27 and 21% of the patients, respectively. In the placebo arm, 14 and 16% of the patients experienced grades 3 and 4 diarrhea. In the sulphasalazine arm, 7% of the patients experienced grade 3 diarrhea. No grade 4 diarrhea was seen in the sulphasalazine group during irradiation (Table 3). Grades 1±3 toxicity occurred in 80 (35/44; 95% CI, 68±91%) and 90% (40/44; 95% CI, 81±98%) of the sulphasalazine and placebo groups according to the maximum LENT-SOMA score during irradiation (P 0:07; Table 4). No grade 4 toxicity was seen in either the sulphasalazine or placebo group according to LENT-SOMA criteria. No difference was found between the two groups according to the maximum LENT-SOMA score in the ®rst week of RT (P 0:75; Table 5). A signi®cant difference was seen between the two groups in favor of sulphasalazine after the ®rst week. LENT-SOMA scores were found to be lower in the sulphasalazine group than in the placebo group in the second (P 0:003), third (P , 0:001), fourth (P , 0:001) and ®fth (P , 0:001) weeks of RT (Table 5).
4. Discussion In this study; radiation-induced diarrhea was seen 86% of the patients receiving a standardized protocol for pelvic malignancies. Sulphasalazine effectively reduced the rate of this adverse effect of diarrhea to 55% during irradiation (P 0:001). Little is known about the severity of diarrhea occurring Table 4 Severity of LENT-SOMA a Maximum LENT-SOMA score Grade
Sulphasalazine
Placebo
Number %
Number %
0 1 2 3 4
9 29 6 ± ±
3 6 32 2 ±
a
P , 0:001.
20 66 14 ± ±
7 14 74 4 ±
128
D. KilicË et al. / Radiotherapy and Oncology 57 (2000) 125±129
Table 5 The LENT-SOMA score according to treatment by weeks throughout the irradiation Weeks
Drug
LENT-SOMA score
P
Grade
1 2 3 4 5
Sulphasalazine Placebo Sulphasalazine Placebo Sulphasalazine Placebo Sulphasalazine Placebo Sulphasalazine Placebo
0
1
2
3
33 31 25 15 20 5 13 4 15 5
11 11 17 16 20 9 27 8 27 10
± ± 2 10 4 29 4 31 2 27
± 1 ± 2 ± ± ± ± ± 1
0.75 0.003 ,0.001 ,0.001 ,0.001
during pelvic RT. Martenson et al. [10] reported the maximum severity of diarrhea grades in the placebo group as: G1, 36%; G2, 25%; G3, 14%; and G4, 0% using the same diarrhea grading system. These were 30, 27 and 8% for grades 1, 2 and 3 in the Resbeut et al. [18] study, respectively. Although the total dose and fractionation schedules were similar in the two studies, there seems to be a difference in the grades between the two studies and ours. In the Resbeut et al. [18] study, the irradiated volumes were similar in all patients with central pelvic diseases (uterine and prostate cancers). In our study, though there were central, anterior and posterior diseases (bladder and rectum cancers), the irradiated volumes were similar in all subjects with the AP:PA ®eld technique. Some of the patients analyzed by Martenson et al. [10] received concurrent 5-¯uorouracil and diarrhea was graded by a different system. The differences in diarrhea grades were attributed to these factors. ARE is not only diarrhea, but a combination of some discomfort symptoms. This warrants the need to include objective analytic parameters in morbidity scales, such as those already foreseen in the LENT-SOMA table [5]. Although it is speci®cally directed towards late effects, this double-blinded, randomized, prospective study is thought to be an exercise in assessing its feasibility on acute radiation-induced gastrointestinal complications. A dissociation between acute and late gastrointestinal complications due to irradiation was reported in the literature [2]. This system measurement tool was perhaps the best that could be used and was semi-objective. When the weekly maximum LENT-SOMA scores were analyzed, a signi®cant reduction in the severity of symptoms, similar to the reduction in diarrhea grades, could be seen in favor of sulphasalazine in our study. It is important to indicate that there was a therapeutic opportunity. Radiation-induced toxicity is caused by gastrointestinal irritation and in¯ammation [10]. Prodromal, acute and chronic effects of radiation are accompanied by excessive
production of eicosanoids (PGs, thromboxanes (TX) and leukotrienes (LT)). These endogenous mediators of in¯ammatory reactions may be responsible for the vasodilatation, vasoconstriction, increased microvascular permeability, thrombosis and chemotaxis observed after radiation exposure [12]. They can also affect water and electrolyte reabsorption and contractility of the alimentary tract [9]. The excessive production of PGs as a cause of diarrhea during pelvic irradiation was postulated by Mennie et al. [11] after they found that acetylsalicylate reduced the severity of gastrointestinal distress. This role of PGs was not substantiated by plasma PG determinations [1]. Excessive amounts of rectal lumenal PGE2, TXB2 and LTB4 were registered in patients at the completion of pelvic RT and 4±6 weeks later [12]. There is a longstanding debate about the use of 5-ASA (or other PG inhibitors) in radiation-induced gastrointestinal complications. These inhibitors would be more effective if given before the cascade of oxidative enzymes is started by RT insult than after the cascade is already established [9,12]. Three controlled trials assessed the ef®cacy of different oral 5-ASA formulations to prevent ARE [1,10,18]. None of them showed clinical ef®cacy; moreover, the ®rst two trials showed a worsening of diarrhea. Resbeut et al. [18] showed that mesalazine (5-ASA) treatment is ineffective to decrease the rate of diarrhea. They concluded that such worsening may provide some information on the mechanisms of diarrhea. The drug formula could be of major importance. Sulphasalazine has been successfully used in attacks of ulcerative colitis [9]. This compound is split by bacterial azoreductases in the distal terminal ileum and colon into sulphapyridine and 5-ASA [6,9]. 5-ASA is the active agent in the sulphasalazine preparate and sulphapyridine acts as a `carrier' [9]. Rauch et al. [17] suggested a bene®cial effect of sulphasalazine on preventing acute radiationinduced diarrhea, but their study was not controlled. Our study could be more useful, because the study of Rauch et al. [17] was small and/or not controlled, and used only a crude grading of diarrhea. We have reported a larger, controlled study with a detailed grading of acute radiation-induced gastrointestinal toxicity. In contrast to quite extensive clinical studies, the mechanisms of the anti-in¯ammatory activity of sulphasalazine have not been entirely elucidated. It has been reported that sulphasalazine inhibits the formation of 5-lipoxygenase products markedly, whereas split products of sulphasalazine, 5-ASA and sulphapyridine are much less potent [15]. On the other hand, the inhibition of cycloxygenase by nonsteroid anti-in¯ammatory drug (NSAID) may promote a LT-mediated in¯ammatory process, presumably by removing the restraining in¯uence of PG [3]. The unexplained diarrhea in patients with ulcerative colitis treated with olsalazine is thought to occur as a consequence of inhibition of water and electrolyte absorption in the small intestine [16]. Parallel to these ®ndings, the oral administration of olsalazine was found to increase the ileal ¯ow rate of liquid and
D. KilicË et al. / Radiotherapy and Oncology 57 (2000) 125±129
might have caused diarrhea in the study of Martenson et al. [10]. In conclusion, our detailed analysis of gastrointestinal toxicity in this randomized clinical trial suggests that sulphasalazine administration reduces the frequency of acute radiation-induced diarrhea and other related symptoms. Further clinical trials to assess the value of measures taken to reduce the late gastrointestinal toxicity by sulphasalazine will be of value for investigators who want to reduce treatment toxicity as an endpoint in RT. References [1] Baughan FF, Canney PA, Buchanan RB, Pickering RM. A randomized trial to assess the ef®cacy of 5-aminosalicylic acid for the prevention of radiation enteritis. Clin Oncol 1993;5:19±24. [2] Burne RG, Kearsley JH, Grove WD, Roberts SJ. The relationship between early and late gastrointestinal complications of radiotherapy for carcinoma of the cervix. Int J Radiat Oncol Biol Phys 1983;9:1445±1450. [3] Cole AT, Slater K, Sokal M. In vivo rectal in¯ammatory mediator changes with radiotherapy to the pelvis. Gut 1993;34:1210±1214. [4] Crook J, Esche BA. Genital tract, uterine cervix. In: Cox JD, editor. Moss' radiation oncology rationale, techniques, results, 7th ed. St. Louis, MO: Mosby±Year, 1994. pp. 655±656. [5] Denekamp J, Bartelink H, Rubin T. Correction for the use of the LENT-SOMA tables. Radiother Oncol 1996;39:191. [6] Gertzen HS, JoÈrnet G, Bukhave K, Lauritsen K, Rask-Madsen J. Effect of azodisol sodium and sulphasalazine on ileostomy output of ¯uid and PGE2 and PGF2a in subjects with a permanent ileostomy. Gut 1986;27:1306±1311. [7] Henrikson R, Franzen L, Littbrand B. Prevention and therapy of radiation-induced bowel discomfort. Scand J Gastroenterol 1992;27(Suppl 3):7±11.
129
[8] Herbert SH, Curran WJ, Solin LJ. Decreasing gastrointestinal morbidity with the use of small bowel contrast during treatment planning for pelvic irradiation. Int J Radiat Oncol Biol Phys 1993;27:1051±1056. [9] Jawetz E. Sulphonamides and trimethoprim. In: Katzung BG, editor. Basic and clinical pharmacology, 6th ed. Norwalk, CA: Appleton & Lange, 1995. pp. 716±722. [10] Martenson JA, Hyland G, Moertel CG, et al. Olsalazine is contraindicated during pelvic radiation therapy: results of a double-blind, randomized trial. Int J Radiat Oncol Biol Phys 1996;35:299±303. [11] Mennie AT, Dalley VM, Dinneen LC, Collier H. Treatment of radiation induced gastrointestinal distress with acetylsalicylate. Lancet 1975;2:342±343. [12] Michalowski AS. On radiation damage to normal tissues and treatment. Acta Oncol 1994;33:139±157. [13] Overgaard J, Barterlink H. LENT-SOMA tables, small intestine/ colon. Radiother Oncol 1995;35:41. [14] Peters LJ. Radiation therapy tolerance limits. Cancer 1996;77:2379± 2385. [15] Pruzanski W, Stefanski E, Vadas P, Ramamurty NS. Inhibition of extracellular release of proin¯ammatory secretory phospholipase A2 by sulfasalazine. Biochem Pharm 1997;53:1901±1907. [16] Raimundo AH, Patil DH, Frost PG, Silk DB. Effects of olsalazine and sulphasalazine on jejunal and ileal water and electrolyte absorption in normal human subjects. Gut 1991;32:270±274. [17] Rauch K, Weilland H. Behandlung der radiogenen kolitis mit salisilazosulfapyridin (azusul®dine). Strahlentherapie 1972;143:660±663. [18] Resbeut M, Marteau P, Cowen C, et al. A randomized, double-blind, placebo-controlled, multicenter study of mesalazine for the prevention of acute radiation enteritis. Radiother Oncol 1997;44:59±63. [19] Ronne IA, Nielsen OH, Christensen A, Langholz E, Binder V, Riis P. Clinical evidence supporting the radical scavenger mechanism of 5aminosalicylic acid. Gastroenterology 1990;98:1462±1469. [20] Rose PG, Haltyer SA, Su CM. The effect of indomethasine on acute radiation-induced gastrointestinal injury: a morphologic study. J Surg Oncol 1992;49:231±238. [21] Saclarides TJ. Radiation injuries of the gastrointestinal tract. Surg Clin N Am 1997;77:261±268.
www.elsevier.com/locate/radonline
Double-blinded, randomized, placebo-controlled study to evaluate the effectiveness of sulphasalazine in preventing acute gastrointestinal complications due to radiotherapy È zenirler b, AysËe Dursun c Diclehan KilicË a,*, ÇIbrahim Egehan a, Seren O a Department of Radiation Oncology, Gazi University Hospital, Ankara, 06510, Turkey Department of Internal Medicine and Gastroenterology, Gazi University Hospital, Ankara, 06510, Turkey c Department of Pathology, Gazi University Hospital, Ankara, 06510, Turkey
b
Received 9 February 2000; received in revised form 30 June 2000; accepted 19 July 2000
Abstract Background and purpose: Acute radiation-induced diarrhea occurs in approximately 80% of the patients receiving pelvic radiotherapy. It is caused by gastrointestinal irritation and in¯ammation. Eicosanoids are thought to be one of the mechanisms of this. Sulphasalazine is an inhibitor of their synthesis in the mucosa. This randomized clinical trial was undertaken to evaluate its effect in preventing acute radiation enteritis (ARE). Materials and methods: Prospectively, 87 patients receiving pelvic radiotherapy were randomized, in a double-blind fashion. Two tablets twice daily of sulphasalazine (500 mg) or placebo were administered orally. Patients were evaluated weekly according to diarrhea grading for the primary study endpoint and according to late effect of normal tissue-subjective objective management analytic (LENT-SOMA) criteria for the secondary endpoint during irradiation. Results: Groups did not differ for age, gender, tumour site or irradiation procedure. Diarrhea occurred in 55 and 86% of the sulphasalazine and placebo groups, respectively (P 0:001). Gastrointestinal toxicity was seen in 80 and 93% of the sulphasalazine and placebo groups according to the maximum LENT-SOMA score (P 0:07). According to the maximum LENT-SOMA score between the two groups, signi®cant differences in favor of sulphasalazine were found for each week. Conclusion: Sulphasalazine (2 g/day) was found to be effective in decreasing the symptoms of ARE. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Pelvic irradiation; Acute radiation enteritis; Sulphasalazine
1. Introduction and scienti®c background Radiation therapy (RT) is frequently administered either primary or adjuvant for a variety of pelvic malignancies, including urological, gynecological and colorectal cancers. Although the bene®ts of this treatment have been well established, a number of patients experience distressing complications as a result of radiation injury to normal tissue [21]. Small bowel tolerance is a highly signi®cant dose-limiting factor because of early adverse effects [4,14]. Acute radiation enteritis (ARE), characterized by abdominal cramping and diarrhea, occurs in approximately 75% of patients and usually begins in the second or third week of RT [4]. These side effects cause discomfort and decreases the therapeutic bene®t by increasing the overall treatment time [18]. * Corresponding author.
Several approaches have been investigated in order to decrease the incidence of both ARE and late sequelae: surgical procedures, such as excluding the small intestine from the irradiated ®eld, partitioning the abdominal and pelvic cavities with endogenous material like omentum or synthetic absorbable mesh or implanted tissue expanders [21]; maneuvers in displacing the small bowel [8]; usage of thiol compounds [20]; prostaglandine (PG) inhibitors [17]; and sucralphate [7]. Radiation induces the synthesis of nuclear regulating proteins. These proteins are able to regulate cytokines and have secondary effects on eicosanoids [3]. Mennie et al. [11] hypothesized that PG release might be another etiological factor in radiation-induced diarrhea. Sulphasalazine has been proven useful in mild or moderately active attacks of ulcerative colitis [9]. It inhibits both the lipoxygenase and cycloxygenase pathway of arachidonic acid metabolism [9]. It has a characteristic and speci®c
0167-8140/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0167-814 0(00)00254-1
126
D. KilicË et al. / Radiotherapy and Oncology 57 (2000) 125±129
af®nity to intestinal tissue. It is a good choice for colitis that effects subepithelial tissues [9,19]. This study was aimed towards assessing the ef®cacy of sulphasalazine enteric tablets, 2 g/day, administered orally throughout the irradiation period to reduce the symptoms of ARE in subjects receiving pelvic irradiation. It was hoped that sulphasalazine would provide a salutary effect without corresponding systemic toxicity. 2. Materials and methods Eighty-seven patients with current histological proof of cancer in the pelvis without metastases beyond the regional lymph nodes, in whom continuous external beam pelvic RT was planned, agreed to take part in the study. They gave their informed consent in accordance with the Declaration of Helsinki, and the study design was approved by the Ethics Committee of our University. Patients were ineligible for the study if they lacked a functioning rectum, if they had stool incontinence, if the stool frequency was $6/day or if the perineum was in the planned irradiation volume. Patients using digoxin, with a history of prior pelvic/abdominal irradiation, in¯ammatory bowel disease, known salicylate hypersensitivity, nephrotic syndrome, hepatic values of twice the normal or Karnofsky performance status of ,70 were also ineligible. All patients had to have no participation in another protocol. No systemic administration of cytotoxic chemotherapy was allowed during RT. Patients received 46±50 Gy in 23±25 fractions of external beam RT to the whole pelvis with megavoltage energy (10 MV). They were treated with opposed anterior±posterior: posterior±anterior (AP:PA) ®elds. For the AP:PA ®elds, the lateral ®eld borders were 1±2 cm lateral to the widest bony margin of the true pelvic side walls. The superior border of the radiation ®eld was placed no higher than the interspace between the fourth and ®fth lumbar vertebrae. Patients whose treatment planning included the perineum in the planned irradiation volume were excluded. Treatment was delivered to all subjects in the supine position. Treatment with a full bladder to displace the small intestine was encouraged. The daily dose, speci®ed at isocenter, was 2 Gy. A tumour boost was allowed either with brachytherapy or with external beam RT according to the tumour site after the completion of a total pelvic dose of 46±50 Gy. Patients were randomized by a different physician from the one treating before RT. Patients were randomized in a double-blind fashion, to receive two tablets twice daily of either 500 mg sulphasalazine (Salazopyrin-EN w) or a placebo which appeared identical. The placebo tablets were composed of coloring agent, talc, cellulose and puri®ed water. Patients were strati®ed according to gender, type of cancer, prior surgery and total cumulative dose of radiotherapy (including boost ®elds). The toxicity was assessed weekly by the treating physician during the ®rst 5 weeks of RT. Standard toxicity
criteria were used to grade the severity of diarrhea. Diarrhea was graded as: grade 0, no diarrhea or increased stool frequency; grade 1, increase of 2±3 stools/day; grade 2, increase of 4±6 stools/day or nocturnal stool; grade 3, increase of 7±9 stools/day or incontinence; grade 4, increase of $10 stools/day or grossly bloody diarrhea [10]. A late effect of normal tissue-subjective objective management analytic (LENT-SOMA) table for the small intestine/colon was modi®ed and used to grade the severity of acute radiation-induced gastrointestinal complications [13]. Grading was done with the remaining nine parameters after excluding the parameters of ulceration and stricture. The weekly scores of the parameters were summed as it was originally recommended, but were not divided by the number of elements as this was not advocated previously [5]. The toxicity was graded according to the summed LENT-SOMA score. If the patient had a summed score of 0±2, 3±9, 10±18, 19±27 or 28±36, the toxicity was accepted as grade 0, 1, 2, 3 or 4, respectively. In addition to these, if there were two or more 3/4 toxicity scores in any parameters, the patient was accepted as having grade 3/4 toxicity. The primary endpoint of the study was diarrhea. The secondary endpoint was the LENT-SOMA scores of the patients. At the end of pelvic RT, ®ve diarrhea and ®ve LENT-SOMA scores were recorded for each patient. The maximum severity of diarrhea/LENT-SOMA score was the highest score of them all. The treatment with sulphasalazine or placebo continued until pelvic RT was completed, unless the patient began to experience $7 stools/day above the pretreatment baseline, which led to discontinuation of study medication. 2.1. Statistics All gradings were done in a blinded fashion. Data analysis was carried out with SPSS 9.01 for Windows program package (SPSS Inc). The proportion of subjects who experience diarrhea was accepted as 80%, according to ®ndings in the literature and an approximate percentage observed in our previous patients. The sample size required to detect a reduction in the proportion of patients who experience diarrhea from the 80% level to 50% or less if patients were given sulphasalazine was estimated. We accepted a signi®cance level of 0.05 and a power of 0.90. The study design called for the entry of 82 patients (41 in each arm). Patient characteristics (Table 1; tumour type, gender) were compared using the Chi-square test. Quantitative variables (®eld dimensions, pelvic radiation dose) were compared using the non-parametric, Mann±Whitney U-test, and expressed as means ^ standard errors. Chi-square tests were used to ®nd the signi®cance of differences in toxic reactions between patients receiving sulphasalazine or placebo. The statistical differences in the gradings were estimated using a ®xed cutpoint of patients with grade 2 and more reactions. Only two sided results were used. P values of 0.05 and less were considered to be signi®cant.
D. KilicË et al. / Radiotherapy and Oncology 57 (2000) 125±129 Table 1 Patient characteristics a Baseline factors
Gender Male Female Type of cancer Rectum/rectosigmoid Endometrium Cervix uteri Prostate Bladder Pelvic sarcoma Age (years) Median Mean a
Table 3 Severity of diarrhea according to treatment a Sulphasalazine (n 44)
Placebo (n 43)
Number %
Number %
26 18
59 41
23 20
54 46
22 4 5 6 6 1
50 9 11 14 14 2
20 8 3 8 4 ±
47 18 7 19 9 ±
60 59
64 61
No signi®cant difference was observed.
Ninety-®ve percent con®dence intervals (CI) for the proportions were calculated and are mentioned in the text. 3. Results Eighty-seven patients entered the study to receive either sulphasalazine or placebo between August 1997 and April 1999. The main characteristics and irradiation procedures (Table 2) did not differ between the two groups. There were no losses to follow up, no refusals to continue the trial without any side effects, and no complications due to the drug. During irradiation, diarrhea (grades 1±4) occurred in 55% (24/44; 95% CI, 39±69%) of the sulphasalazine and 86% (37/43; 95% CI, 75±96%) of the placebo group, and this difference was found to be statistically signi®cant (P 0:001). Approximately 77% (20/26) of the patients who had no diarrhea during irradiation (grade 0 patients) were in the sulphasalazine group. Diarrhea of grade 2 and more was seen in 27% (12/44; 95% CI, 14±40%) of the sulphasalazine and 49% (21/43; 95% CI, 34±63%) of the Table 2 Irradiation procedure a Sulphasalazine (n 44) Placebo (n 43) Field dimensions Surface at isocenter (cm 2) 251.5 ^ 8.2 Anterior±posterior (cm) 21.2 ^ 0.4 Pelvic radiation dose (Gy) 49.3 ^ 0.2 Boost with external RT % Patients 29 Dose (Gy) 18.4 ^ 1.0 Boost with intracavitary RT % Patients 13 Dose (Gy) 23.3 ^ 0.6 a
127
No signi®cant difference was observed.
250.7 ^ 9.4 21.0 ^ 0.3 49.5 ^ 0.2 30 19.3 ^ 0.4 14 21.3 ^ 1.3
Maximum severity of diarrhea Grade
Sulphasalazine (n 44) Placebo (n 43) Number %
Number %
0 1 2 3 4
20 12 9 3 ±
6 16 8 6 7
a
46 27 20 7 ±
14 37 19 14 16
P 0:038.
placebo group (P 0:038). In the sulphasalazine group, grades 1, 2 and 3 diarrhea were seen in 45, 27 and 21% of the patients, respectively. In the placebo arm, 14 and 16% of the patients experienced grades 3 and 4 diarrhea. In the sulphasalazine arm, 7% of the patients experienced grade 3 diarrhea. No grade 4 diarrhea was seen in the sulphasalazine group during irradiation (Table 3). Grades 1±3 toxicity occurred in 80 (35/44; 95% CI, 68±91%) and 90% (40/44; 95% CI, 81±98%) of the sulphasalazine and placebo groups according to the maximum LENT-SOMA score during irradiation (P 0:07; Table 4). No grade 4 toxicity was seen in either the sulphasalazine or placebo group according to LENT-SOMA criteria. No difference was found between the two groups according to the maximum LENT-SOMA score in the ®rst week of RT (P 0:75; Table 5). A signi®cant difference was seen between the two groups in favor of sulphasalazine after the ®rst week. LENT-SOMA scores were found to be lower in the sulphasalazine group than in the placebo group in the second (P 0:003), third (P , 0:001), fourth (P , 0:001) and ®fth (P , 0:001) weeks of RT (Table 5).
4. Discussion In this study; radiation-induced diarrhea was seen 86% of the patients receiving a standardized protocol for pelvic malignancies. Sulphasalazine effectively reduced the rate of this adverse effect of diarrhea to 55% during irradiation (P 0:001). Little is known about the severity of diarrhea occurring Table 4 Severity of LENT-SOMA a Maximum LENT-SOMA score Grade
Sulphasalazine
Placebo
Number %
Number %
0 1 2 3 4
9 29 6 ± ±
3 6 32 2 ±
a
P , 0:001.
20 66 14 ± ±
7 14 74 4 ±
128
D. KilicË et al. / Radiotherapy and Oncology 57 (2000) 125±129
Table 5 The LENT-SOMA score according to treatment by weeks throughout the irradiation Weeks
Drug
LENT-SOMA score
P
Grade
1 2 3 4 5
Sulphasalazine Placebo Sulphasalazine Placebo Sulphasalazine Placebo Sulphasalazine Placebo Sulphasalazine Placebo
0
1
2
3
33 31 25 15 20 5 13 4 15 5
11 11 17 16 20 9 27 8 27 10
± ± 2 10 4 29 4 31 2 27
± 1 ± 2 ± ± ± ± ± 1
0.75 0.003 ,0.001 ,0.001 ,0.001
during pelvic RT. Martenson et al. [10] reported the maximum severity of diarrhea grades in the placebo group as: G1, 36%; G2, 25%; G3, 14%; and G4, 0% using the same diarrhea grading system. These were 30, 27 and 8% for grades 1, 2 and 3 in the Resbeut et al. [18] study, respectively. Although the total dose and fractionation schedules were similar in the two studies, there seems to be a difference in the grades between the two studies and ours. In the Resbeut et al. [18] study, the irradiated volumes were similar in all patients with central pelvic diseases (uterine and prostate cancers). In our study, though there were central, anterior and posterior diseases (bladder and rectum cancers), the irradiated volumes were similar in all subjects with the AP:PA ®eld technique. Some of the patients analyzed by Martenson et al. [10] received concurrent 5-¯uorouracil and diarrhea was graded by a different system. The differences in diarrhea grades were attributed to these factors. ARE is not only diarrhea, but a combination of some discomfort symptoms. This warrants the need to include objective analytic parameters in morbidity scales, such as those already foreseen in the LENT-SOMA table [5]. Although it is speci®cally directed towards late effects, this double-blinded, randomized, prospective study is thought to be an exercise in assessing its feasibility on acute radiation-induced gastrointestinal complications. A dissociation between acute and late gastrointestinal complications due to irradiation was reported in the literature [2]. This system measurement tool was perhaps the best that could be used and was semi-objective. When the weekly maximum LENT-SOMA scores were analyzed, a signi®cant reduction in the severity of symptoms, similar to the reduction in diarrhea grades, could be seen in favor of sulphasalazine in our study. It is important to indicate that there was a therapeutic opportunity. Radiation-induced toxicity is caused by gastrointestinal irritation and in¯ammation [10]. Prodromal, acute and chronic effects of radiation are accompanied by excessive
production of eicosanoids (PGs, thromboxanes (TX) and leukotrienes (LT)). These endogenous mediators of in¯ammatory reactions may be responsible for the vasodilatation, vasoconstriction, increased microvascular permeability, thrombosis and chemotaxis observed after radiation exposure [12]. They can also affect water and electrolyte reabsorption and contractility of the alimentary tract [9]. The excessive production of PGs as a cause of diarrhea during pelvic irradiation was postulated by Mennie et al. [11] after they found that acetylsalicylate reduced the severity of gastrointestinal distress. This role of PGs was not substantiated by plasma PG determinations [1]. Excessive amounts of rectal lumenal PGE2, TXB2 and LTB4 were registered in patients at the completion of pelvic RT and 4±6 weeks later [12]. There is a longstanding debate about the use of 5-ASA (or other PG inhibitors) in radiation-induced gastrointestinal complications. These inhibitors would be more effective if given before the cascade of oxidative enzymes is started by RT insult than after the cascade is already established [9,12]. Three controlled trials assessed the ef®cacy of different oral 5-ASA formulations to prevent ARE [1,10,18]. None of them showed clinical ef®cacy; moreover, the ®rst two trials showed a worsening of diarrhea. Resbeut et al. [18] showed that mesalazine (5-ASA) treatment is ineffective to decrease the rate of diarrhea. They concluded that such worsening may provide some information on the mechanisms of diarrhea. The drug formula could be of major importance. Sulphasalazine has been successfully used in attacks of ulcerative colitis [9]. This compound is split by bacterial azoreductases in the distal terminal ileum and colon into sulphapyridine and 5-ASA [6,9]. 5-ASA is the active agent in the sulphasalazine preparate and sulphapyridine acts as a `carrier' [9]. Rauch et al. [17] suggested a bene®cial effect of sulphasalazine on preventing acute radiationinduced diarrhea, but their study was not controlled. Our study could be more useful, because the study of Rauch et al. [17] was small and/or not controlled, and used only a crude grading of diarrhea. We have reported a larger, controlled study with a detailed grading of acute radiation-induced gastrointestinal toxicity. In contrast to quite extensive clinical studies, the mechanisms of the anti-in¯ammatory activity of sulphasalazine have not been entirely elucidated. It has been reported that sulphasalazine inhibits the formation of 5-lipoxygenase products markedly, whereas split products of sulphasalazine, 5-ASA and sulphapyridine are much less potent [15]. On the other hand, the inhibition of cycloxygenase by nonsteroid anti-in¯ammatory drug (NSAID) may promote a LT-mediated in¯ammatory process, presumably by removing the restraining in¯uence of PG [3]. The unexplained diarrhea in patients with ulcerative colitis treated with olsalazine is thought to occur as a consequence of inhibition of water and electrolyte absorption in the small intestine [16]. Parallel to these ®ndings, the oral administration of olsalazine was found to increase the ileal ¯ow rate of liquid and
D. KilicË et al. / Radiotherapy and Oncology 57 (2000) 125±129
might have caused diarrhea in the study of Martenson et al. [10]. In conclusion, our detailed analysis of gastrointestinal toxicity in this randomized clinical trial suggests that sulphasalazine administration reduces the frequency of acute radiation-induced diarrhea and other related symptoms. Further clinical trials to assess the value of measures taken to reduce the late gastrointestinal toxicity by sulphasalazine will be of value for investigators who want to reduce treatment toxicity as an endpoint in RT. References [1] Baughan FF, Canney PA, Buchanan RB, Pickering RM. A randomized trial to assess the ef®cacy of 5-aminosalicylic acid for the prevention of radiation enteritis. Clin Oncol 1993;5:19±24. [2] Burne RG, Kearsley JH, Grove WD, Roberts SJ. The relationship between early and late gastrointestinal complications of radiotherapy for carcinoma of the cervix. Int J Radiat Oncol Biol Phys 1983;9:1445±1450. [3] Cole AT, Slater K, Sokal M. In vivo rectal in¯ammatory mediator changes with radiotherapy to the pelvis. Gut 1993;34:1210±1214. [4] Crook J, Esche BA. Genital tract, uterine cervix. In: Cox JD, editor. Moss' radiation oncology rationale, techniques, results, 7th ed. St. Louis, MO: Mosby±Year, 1994. pp. 655±656. [5] Denekamp J, Bartelink H, Rubin T. Correction for the use of the LENT-SOMA tables. Radiother Oncol 1996;39:191. [6] Gertzen HS, JoÈrnet G, Bukhave K, Lauritsen K, Rask-Madsen J. Effect of azodisol sodium and sulphasalazine on ileostomy output of ¯uid and PGE2 and PGF2a in subjects with a permanent ileostomy. Gut 1986;27:1306±1311. [7] Henrikson R, Franzen L, Littbrand B. Prevention and therapy of radiation-induced bowel discomfort. Scand J Gastroenterol 1992;27(Suppl 3):7±11.
129
[8] Herbert SH, Curran WJ, Solin LJ. Decreasing gastrointestinal morbidity with the use of small bowel contrast during treatment planning for pelvic irradiation. Int J Radiat Oncol Biol Phys 1993;27:1051±1056. [9] Jawetz E. Sulphonamides and trimethoprim. In: Katzung BG, editor. Basic and clinical pharmacology, 6th ed. Norwalk, CA: Appleton & Lange, 1995. pp. 716±722. [10] Martenson JA, Hyland G, Moertel CG, et al. Olsalazine is contraindicated during pelvic radiation therapy: results of a double-blind, randomized trial. Int J Radiat Oncol Biol Phys 1996;35:299±303. [11] Mennie AT, Dalley VM, Dinneen LC, Collier H. Treatment of radiation induced gastrointestinal distress with acetylsalicylate. Lancet 1975;2:342±343. [12] Michalowski AS. On radiation damage to normal tissues and treatment. Acta Oncol 1994;33:139±157. [13] Overgaard J, Barterlink H. LENT-SOMA tables, small intestine/ colon. Radiother Oncol 1995;35:41. [14] Peters LJ. Radiation therapy tolerance limits. Cancer 1996;77:2379± 2385. [15] Pruzanski W, Stefanski E, Vadas P, Ramamurty NS. Inhibition of extracellular release of proin¯ammatory secretory phospholipase A2 by sulfasalazine. Biochem Pharm 1997;53:1901±1907. [16] Raimundo AH, Patil DH, Frost PG, Silk DB. Effects of olsalazine and sulphasalazine on jejunal and ileal water and electrolyte absorption in normal human subjects. Gut 1991;32:270±274. [17] Rauch K, Weilland H. Behandlung der radiogenen kolitis mit salisilazosulfapyridin (azusul®dine). Strahlentherapie 1972;143:660±663. [18] Resbeut M, Marteau P, Cowen C, et al. A randomized, double-blind, placebo-controlled, multicenter study of mesalazine for the prevention of acute radiation enteritis. Radiother Oncol 1997;44:59±63. [19] Ronne IA, Nielsen OH, Christensen A, Langholz E, Binder V, Riis P. Clinical evidence supporting the radical scavenger mechanism of 5aminosalicylic acid. Gastroenterology 1990;98:1462±1469. [20] Rose PG, Haltyer SA, Su CM. The effect of indomethasine on acute radiation-induced gastrointestinal injury: a morphologic study. J Surg Oncol 1992;49:231±238. [21] Saclarides TJ. Radiation injuries of the gastrointestinal tract. Surg Clin N Am 1997;77:261±268.