Acute Radiation Proctitis

A clinical, histopathological and histochemical study

Nils Hovdenak, M.D. Acute Radiation Proctitis

A clinical, histopathological and histochemical study

Nils Hovdenak, M.D.

Institute of Medicine Section of Oncology Haukeland University Hospital, Bergen, Norway 2004 2

1. List of contents

1 List of contents 2

1. Preface and acknowledgements 3

3. General introduction 6

3.1. Principles of radiation 7 3.2. Radiation-induced cell death 8 3.5. Pathology and pathogenesis of radiation proctitis 12 3.7. Radiosensitivity 20 3.8. Radioprotection 22

4. Aims of the study 26

5. Material and Methods 27

6. List of papers 28 Paper 1 28 Paper II 28 Paper III 29 Paper IV 29 7. General discussion 31

7.1. Symptoms 7.2. Histopathology 33 7.3. Gelatinases 35 7.4. Prophylaxis 37 8. Conclusions 39

9. Future prospects 41

9.1. Symptomatic treatment 41 9.2. Prophylaxis 42 9.3. Radiotherapy methods and equipment 10. References 48 In the memory of my parents 3

1. Preface and acknowledgements

The present study has been carried out at The Institute of Medicine, Division of Oncology and

Division of Gastroenterology, University of Bergen, at the Department of Surgery, University

of Arkansas for Medical Sciences, Little Rock, Arkansas, USA, and at the Department of

Pathology, Stanford University School of Medicine and Veterans Healthcare System, Palo

Alto, California, USA.

I came to the Division of Oncology during a sabbatical from my position as gastroenterologist

at The Deaconess Hospital, Bergen in 1994. As a trained gastroenterologist I very soon

became involved in patients who complained of gastrointestinal side-effects of radiotherapy.

A study was planned to elucidate the endoscopic and histopathologic changes occurring in the

rectal mucosa during pelvic radiotherapy for prostate cancer. From the very beginning

professor Olav Dahl was an invaluable source of inspiration and encouragement and I was

deeply thankful when he agreed to be my formal tutor during the work on the present thesis.

His enthusiasm and his vast knowledge in every field of oncology and his always open mind

and helpful attitude have been the greatest inspiration during these years. He made it possible

for me to do the gastrointestinal and endoscopic examinations of patients and encouraged me

to do further research on normal tissue samples, realising the unique source of information in

this material. His continuous interest, help and councelling have been an invaluable

inspiration for me during all the years of work on this thesis.

I was lucky to establish a contact with professor Martin Hauer-Jensen at the University of

Arkansas for Medical Sciences in 1996. At that time, Martin Hauer-Jensen already worked at the cutting edge of radiation enteropathy sciences, internationally well reputed for numerous high quality studies in this field. From the first contact his interest and enthusiasm greatly inspired me and eventually resulted in a 4 months stay in his laboratory in 1997, and subsequently a second period in 2000. Cynthia and Martin's great hospitality made my time in Little Rock unforgettable. I spent two shorter periods as professor Luis F. Fajardo's guest at

Stanford University, Palo Alto, doing the histology work. Fajardo has for years held the position as the world's foremost radiation pathologist and he shared his knowledge, experience, and humour with me as we worked our way through the slides. I am in great debt to him.

I want also to express my thanks to all of my colleagues at the Department of

Medicine, Bergen University Hospital, especially to the head of the gastroenterologic section, professor Arnold Berstad. I have always been met with friendliness, encouragement, and help when needed. The scientific endeavour at the section is inspiring, broad, of high quality and has attracted first-class scientists.

My thanks also go to my co-authors: Ching-Ching Sung, Junru Wang, Thomas Kelly,

Halfdan Sørby, and Åsa Karlsdottir. Without their enthusiasm and work, my thesis would never appear. And we had much good fun too! I thank nurse Anne-Sofie Knag for help during patient handling and endoscopy.

Without the cooperation of the patients this work would be impossible. I am deeply moved when considering the willingness with which they participated in the studies in spite of the serious aspects of their disease. I came on friendly terms with all of them and got invaluable experience of life attitudes. This has also resulted in friendships and continuous contacts later.

I want to express my deep thanks to The Norwegian Cancer Society, The University of

Bergen, and to Kavlis Almennyttige Fond for economic help through grants. 5

First and foremost my thanks go to my wife, Inger. Her patience, and frustrations, have kept me going on. Luckily, she has always insisted on thesis-free holidays, mountain hikes, and visits to our children and grandchildren. She has been, and still is, my cornerstone! 6

3. General introduction

Shortly after the introduction of X-rays in medicine, the first description of intestinal radiation injury appeared in a medical journal (1). A radiation worker had been exposed to unshielded abdominal radiation. A substantial literature on this topic has since then accumulated, fulfilling Walsh's prophesy that "it is obvious that the subject is pregnant with future possibilities" Parallel to the development of radiotherapy (RT) as one of the mainstays in oncology, much effort has been applied to shield both staff and patients from untoward effects of ionising radiation.

If radiation could be applied solely to tumour tissue the dose could freely be escalated to levels that would totally eradicate most malignancies. However, normal tissue will always be exposed during RT. Therefore, injury to non-tumoural organs and tissues is the main limiting factor for curative RT of malignant tumours. Shielding and protection of these structures are consequently of paramount importance in radiation oncology. The purpose of therapeutic radiation is to inflict lethal injury to the clonogenic tumour cells. In comparison to normal tissue cells, tumour cells have reduced capacities to repair radiation induced DNA damage. As a consequence, surrounding normal tissues appear more radio-resistant. No recurrence of malignant tumours would appear after RT if a sufficiently high dose of radiation could be delivered to eradicate all of the clonogenic tumour cells. However, this would carry an unacceptably high risk of severe damage to the normal tissue. The difference in resistance between tumour and normal tissue sets the limitations and possibilities for radiation treatment, expressed as the distance between the dose-response curves in Fig. 1, the "therapeutic ratio".

When highly proliferative normal tissue is included in the radiation field, e.g. intestinal mucosa, this ratio is small, implying high risk for untoward effects due to mucosal injury. An even more complicated situation ensues when RT is combined with chemotherapy and 7

hormone therapy.

100 -l

s 1 £' " Tumor

Normal tissue

0 AA B C Radiation dote

Fig. 2. Dose response curves for tumour and normal tissue. The distance between the curves represent the therapeutic ratio.

3.1. Principles of radiation

Ionizing radiations are of two types: Electromagnetic (x-rays and gamma rays) or particulate (electrons, protons, neutrons, alpha particles and light ions) irradiation.

X-rays are produced when accelerated, high energy electrons are stopped in tungsten or copper. Some of the liberated energy consists of heat, the rest is converted to x-rays, which have a spectrum of energies. The effective energy is often about half to one third of the given maximal energy expressed in MV. Gamma rays come from the decaying nucleus of an unstable radioactive isotope, when excess energy is emitted as monochromatic rays of a single or limited number of energies.

Particulate radiation includes electrons and protons, which can be accelerated into high speed in dedicated machines, and neutrons, which are produced by colliding protons and deuterons, and alpha particles. The use of alpha particles is expanding, administered as conjugates of tumour seeking antibodies. Light ion irradiation has only recently been applied in clinical trials (2,3).

When emitted into tissues, the two types of radiation differ in action. Gamma rays and x-rays are absorbed in tissue by interacting with and dislodging bound orbital electrons, mainly in water molecules, creating lower energetic photons which scatter in different directions and subsequently interact with other electrons on their way through the tissues, creating reactive radicals, which are the basis for their (bio)chemical actions. This type of radiation is called low-linear energy transfer (LET) radiation. On the other hand, protons, neutrons and alpha particles mostly act directly on the nucleus by breaking nuclear bonds some of which the cells are unable to repair.

3.2. Radiation-induced cell death

Ionising radiation, when passing through tissues, causes single-strand and double- strand DNA breaks and elicits bursts of free reactive oxygen species (ROS), which cause damage to cellular DNA, proteins, lipids, and vital complex molecules like enzymes.

DNA damage may lead to cell death in the first cell division after irradiation or within the first few divisions (4). Cell death during mitosis is generally caused by unrepaired or incompletely repaired chromosomal damage and accounts for most of the normal intestinal tissue injury whereas the relative contribution of apoptosis to cell death is uncertain.

Normally, both small intestine and colonic epithelium undergo a low rate of spontaneous apoptosis. Especially in the colon the apoptotic rate is low because of the presence of bcl-2

(5,6) which protects the cell from apoptosis. Radiation-induced apoptosis is dependent on the presence of p53, the expression of which is upregulated and parallels the the increase in apoptotoc rate in intestinal stem cells during irradiation (7). Interestingly, ceramide is a second messenger in this p53-dependent signalling system, as well as in the plasma membrane-derived sphingomyelin pathway, where irradiation activates the enzyme sphingomyelinase to produce ceramide (8). The downstream targets for ceramide action

leading to apoptosis have been described (9).

The importance of apoptosis has been questioned (10). It probably varies during

fractionated RT, as does the radiation-induced compensatory proliferation, depending on

different characteristics of tissue radiation sensitivity, proliferation rate and capacity,

compartment organisation, cell migration capacity. The intestinal cellular turnover is rapid.

Cellular renewal occurs in the crypt epithelium, which is the main target for radiation

cytocidal effects. Therefore, manifestations of radiation injury occur within days when the

mucosal cells are no longer replaced by cells migrating from the crypts.

3.3. Prostate cancer

Prostate cancer is the most common malignancy in males. The incidence has increased

dramatically during the last decennium, presumably due to extensive use of blood-test based

screening for prostate specific antigen (PSA). The escalating incidence is thus mostly due to

the detection of early stage cancer with the potential of curative treatment (11) and a good prognosis. 3.029 new cases of prostate cancer were diagnosed in Norway in 2000, and the number has increased significantly over the previous decade (Fig. 2). Corresponding figures

for the USA are 244.000 diagnosed in 1995, with 40.400 deaths per year. A declining

incidense has though been observed in the USA during recent years (12).

I. Age-adjusted incidence rate 1955-2000 (w.std.) Prostate Fig. 2. Incidence rates of prostate cancer in

Norway (13)

1955 1960 1965 1970 1975 19*0 1985 1990 1995 2000

Year of diagnosis 3.4. Radiation proctitis. Clinical aspects

Due to its relatively fixed position, the rectum is the most frequently involved organ of

the gastrointestinal tract that is subjected to radiation injury during pelvic RT (14), but even the small intestine can be involved in the radiation field. Particularly after laparotomy the

small intestine may become fixed in position by adhesions and therefore subjected to daily irradiation and full dose exposure. The rectum functions as a storage organ and moves into or away from the target volume during the radiation series as a consequence of its varying content of feces or gas (15). Owing to the easy access for biopsy and physiological studies, acute radiation proctitis is a feasible clinical model for the study of general irradiation- induced morphologic, biomolecular, and functional changes without too much discomfort for the patient.

The use of RT for abdominal tumours has been steadily increasing. In our institution, cancer of the prostate accounts for the majority of patients given pelvic RT, while gynecological, testicular, rectal, and urinary bladder cancers also contribute significantly. In spite of meticulous dose-planning, variable amounts of intestine are inevitably exposed to radiation in pelvic RT (15). Symptoms of acute radiation proctitis of varying degree occur in the majority of patients during external pelvic RT. Prevalence figures vary in different studies, depending on RT tecnique, total volume involved, dose, fractionation, and side-effect scoring system. Conformal RT based on computed tomography (CT) simulation and three- dimensional treatment planning was introduced to construct a better tailoring of the radiation to the tumour and to avoid irradiation of normal tissue (16). The use of customised casted blocks or computer-controlled multileaf collimators allows precise shaping of dose distribution. These refinements may permit dose escalation without increased side-effects.

The new technology has been used enthusiastically for several years. Long-term evidence of benefits and limitations is emerging. Problems involved in the evaluation of these studies are

discussed in recent papers (17,18).

However, studies now appear showing promisingly low toxicity in spite of dose

escalation. A randomised study showed no increased toxicity in patients receiving high-dose

conformal RT compared to conventional RT (19), and dose escalation studies have been

carried out in attempts to establish maximum tolerated doses. Even at doses of 79.2 Gy for

prostate cancer, favorably low toxicity was demonstrated (20).

Much interest has focused on brachytherapy for prostate cancer, partly due to the reported

low incidence of short-term gastrointestinal toxicity (21,22). However, higher rates of side-

effects have recently been demonstrated (23,24), and long term follow-up studies are few

Clinically, symptoms of acute radiation injury of the rectum subside after the completion

of the radiation treatment course. Only a minority of patients will eventually develop chronic

disease. The acute toxicity is usually estimated by the RTOG/EORTC acute scoring system

(26), which differentiates symptom severity in grade 0-4 as a global assessment. According to this system toxicity appears low (27-29): No grade 3 or 4 toxicity was found in a study of 266 prostate cancer patients (30). In other studies 5-10 % of patients experienced grade 3 or 4

symptoms (31).

Why is then the study of acute radiation proctitis important? Obviously, the acute bowel

injury carries per se a considerable morbidity in some patients, even to the degree that it can negatively influence the RT if pauses are necessary. But acute toxicity rarely necessitates premature cessation of the treatment. More important, the development of delayed changes

is initiated during the acute phase, the severity of which correlates closely to the development

of delayed structural changes (32-35). Hopefully, an effective prophylaxis based on detailed knowledge of acute mechanisms and the biomolecular phenomena that govern the transition

from acute to chronic injury may forestall or attenuate the development of chronic disease.

3.5. Pathology and pathogenesis of radiation proctitis

Acute radiation-induced histopathologic mucosal changes in the colon are seen in the

epithelium, the glands, the goblet cells, the small vessels, and the stroma. Comprehensive reviews have recently been published (36,37). The surface and glandular epithelium is

shorter, may be partially eroded, with nuclear enlargement, altered chromatin patterns with prominent nucleoli. It is infiltrated with leukocytes, notably eosinophils. The number of goblet cells is reduced and the mucus is depleted. The inflammatory infiltrate in the lamina propria is consists of numerous eosinophils, sometimes accompanied by eosinophilic abscesses. Vascular changes are not prominent in the colon (38), but microthrombi may be seen.

Generally, tissue injury caused by fractionated radiation is in many respects different from other types of injury and from inflammatory bowel disease. It is also probably different from animal experiments where irradiation is applied as one single exposure or as a very limited number of fractions. General principles of trauma response apply, but in a setting characterised by repetitive and frequent injuries over weeks. Cytokines, growth factors, chemokines and a variety of other biologically active molecules are activated in a complex sequence to interact in the processes of inflammation, coagulation, proliferation/regeneration, matrix destruction and remodelling. Each radiation fraction imposes a new stimulus on this system, which has already been activated. The normal tissue exposed to subsequent radiation injury is therefore no longer "normal", but in a continuing process of regeneration, proliferation, remodelling etc. Even during the course of fractionated RT cell division occur, both in tumour and normal tissue. In normal tissue this compensatory proliferation aims at

maintaining the structural and functional integrity of the organ, e.g. the rectal mucosa.

D, 07 Dj D4 03 Figure 3. Idealised situation: repopulation or

NORMAL TISSUE regeneration of cells within the fractional intervals. The therapeutic ratio is improved

through more rapid regeneration of normal

tissue compared to tumour (39).

200 400 600 800 1000 DOSE (rad)

Traditionally, intestinal radiation injury has been categorised as either acute or chronic, depending on the time of clinical presentation. This has been a suitable clinical classification and a good guide for diagnostic work-up and treatment planning. Different pathoanatomic changes characterise the two entities, the chronic being dominated by fibrosis

(40) and vascular pathology (41), the acute by varying degrees of inflammatory reactions

(42,43). The impact of the inflammation on clinical manifestations and the progression to chronic disease has been enigmatic. The inflammation is probably of less significance, as judged by the poor response to anti-inflammatory drugs (42,44,45). A new categorisation has recently been proposed (46), taking broadly into account different levels and types of injury mechanisms. Three interacting categories are defined: cytocidal effects, indirect effects, and functional effects, which individually or in combination can lead to both acute and chronic symptoms and signs. The cytocidal effects refer typically to the "target cell" type of injury, where the response, clinical and morphological, depends on tissue specific radiation sensitivity, proliferation rates etc. Indirect effects relate to secondary changes as results of primary damage to other structures or cells, e.g. vascular injury causing ischemic damage.

Functional effects results from nonlethal effects on intra- and extracellular molecules and gene expression.

When it comes to pathophysiology, pathogenesis, and molecular mechanisms leading to clinical disease, these categorisations have several shortcomings. The classification of

Denham et al (46), however, allows a better understanding of the dynamics that give rise significant structural and functional changes, based on fundamental biomolecular mechanisms.

Normal tissues are organised tissues with interacting, mutually dependent cellular entities and compartments, the interactions of which are orchestrated by multiple and complicated biomolecular functional pathways. Responses to injury to one isolated cellular line as demonstrated in cell cultures can therefore not be extrapolated to explain tissue response in vivo or the clinical picture. Reparative responses are better preserved in organised tissues, in contrast to disorganised, i.e. neoplastic, tissues. Important components in the processes after radiation injury have recently been comprehensively reviewed (47).

Some key processes are taking place in the irradiated mucosa: 1) epithelial injury;

2) inflammation; 3) changes in endothelial cells and coagulation system; 4) regeneration. The processes may be described separately from a didactic point of view. However, an extensive cross-talk is taking place, and morphologic and functional perturbations can therefore not be solely ascribed one mechanism: from ultrastructural studies it has been concluded that vascular changes are antedated by the cellular epithelial damage. This has later been held as the explanation of radiation injury, i.e. reactions caused directly or indirectly by lethal radiation damage to stem cells in the crypts (48-50), rendering a defective epithelial gut barrier function. This is in contrast to later studies (51), which indicate that microvascular events (endothelial injury) initiate the radiation reaction, the damage to intestinal stem cells being a consequence of extensive microvascular injury. Taking a more dynamic view, both mechanisms are important and not mutually exclusive.

3.5A. Epithelial injury. Epithelial stem cells located in the basal and suprabasal zones in the mucosal crypts are rapidly killed during RT, resulting in early cellular depletion in the surface epithelium. The cellular injury and damaged basement membrane zone are associated with defective barrier function, exposing underlying structures to injurious luminal factors that contribute to further inflammation. Defective barrier function may persist even after re- epithelialisation. The importance of luminal factors and barrier function is demonstrated by the reduction of both early and delayed radiation injury by surgical and pharmacologic ablation of pancreatic secretion (52,53).

3.5.2. Inflammation. Radiation to the intestines evokes prominent acute inflammation

(36). The first histologic study of radiation proctitis was published in 1968 (42), describing eosinophilic crypt abscesses and focal tissue eosinophilia as characteristic features, together with derangement of surface and crypt epithelium, diminished mucus production, and shrunken goblet cells. The acute mucosal reaction shows loss of mucus and epithelial lining.

The compromised barrier function renders the mucosa more directly exposed to the

"aggressive" intraluminal milieu. Translocation of microorganisms, antigenes, enzymes, and toxins in combination with radiation-induced ROS initiate and augment the activation of kinase cascades and transcription factors, accellerating the radiation-induced production of several pro-inflammatory cytokines (tumour necrosis factor (TNF)-a, interleukin (IL)-l, IL-8, and interferon (IFN)-y).

3.5.3. Changes in endothelial cells and the coagulation system occur immediately after irradiation as a result of functional radiation effects. ROS are toxic to the endothelial cell. An increased endothelial expression of intercellular adhesion molecules (ICAM), potentiated by

bacterial endotoxin occurs (54). Cascades of pro-inflammatory and pro-coagulant events are

initiated, involving, among others, growth factors, ICAMs, thrombin, plasminogen activator.

Endothelial cell surface thrombomodulin (TM) plays a critical role as a cofactor for thrombin-

dependent formation of the anticoagulant activated protein C. TM is a transmembrane protein

that is abundant on the luminal surface of endothelial cells and acts as an anticoagulant by

inhibiting thrombin to convert fibrinogen to fibrin. TM is down-regulated by irradiation (55),

by II-1, TNF-a, transforming growth factor (TGF)-P, and endotoxin (56-58), released and/or

inactivated by ROS and granulocyte products, and is up-regulated by 11-4 (59), pentoxifylline

(60). Down-regulation is greatly potentiated by the presence of inflammation (61). The net

effect of TM deficiency is a pro-coagulant state with conversion of prothrombin to thrombin,

and subsequently fibrinogen to fibrin with clot formation.

Thrombin is an important regulator of cell proliferation, inflammation and tissue

remodeling. Through mechanisms independent of coagulation, thrombin mediates

lipopolysaccharide-induced tissue injury (62), and regulates endothelial permeability (62,63),

chemotaxis of neutrophils (62,64) and monocytes (62,65), and TGF-(51 production (62,66). It

increases the production of platelet-activating factor (PAF), PAF-mediated neutrophil

adhesion to endothelium, and smooth muscle cell migration and proliferation. Most of these

effects are mediated through the activation of the protease-activated receptor (PAR) family

(62,67). Hence, thrombin is a key factor in both coagulation and inflammation. Specific thrombin inhibitors decrease smooth muscle cell migration and proliferation, and collagen production in vitro (62,68). The gene expression of the matrix metalloproteinase (MMP)-9 is

stimulated (69), and MMP-2 is activated by thrombin (70). Both MMPs are important in

breakdown and remodelling of the irradiated mucosa and in angiogenesis. Platelets are crucial for hemostasis in injury and in addition an important source of biologically active substances of pro-inflammatory and fibrogenic character (71). The importance of endothelial dysfunction and platelet aggregation is illustrated by the amelioration of radiation-induced toxicity by experimental inhibition of ADP-induced platelet aggregation by clopidogrel, indicating anticoagulation as a potential prophylactic measure

(72).

3.5.4. Proliferation/regeneration. After a delay of a few days from the appearance of radiation-induced inflammation in the mucosa, a proliferative, regenerative response is triggered. The triggering mechanisms are poorly understood. An accelerated cell division and subsequent repopulation take place by recruitment of quiscent stem cells into the cell cycle and by shortening of the cycle's duration (73,74). Two additional mechanisms have been suggested: loss of asymmetric stem-cell division, resulting in two daughter stem-cells, and divisions by growth-inhibited cells (75).

3.6. Transition to delayed injury.

The development of chronicity represents the most important clinical aspect of radiation injury. Patients with chronic radiation enteropathy have generally a poor prognosis even after successfull tumour eradication. About 10% die as a consequence of the injury (76-

Delayed radiation-induced injury develops in continuity with, or at varying time after the acute injury and is characterised by increased collagen deposit, fibrosis, depletion of epithelial and stromal cells, neo-angiogenesis, but, in contrast to the acute reaction, little inflammation. There is a correlation between the severity of acute mucosal inflammation as expressed by increased granulocyte transmigration, and subsequent delayed (consequential) injury (35). The presence of different clinical and morphologic manifestations indicates different and multifactorial pathogenetic mechanisms, orchestrated by a combination of functional and indirect effects (46). Epithelial cell depletion due to a combination of vascular damage and clonogenic radiation injury reduces the intestinal barrier function, contributing to ongoing, low-grade subepithelial inflammation. In addition, persisting perturbations of the reparative processes due to non-lethal radiation effects on endothelial cells and fibroblasts result in structural and inflammatory changes. Table 2 summarises the proposed mechanisms in the development of delayed radiation injury.

Repeated intramucosal exudate with unresolved fibrin deposition stimulates collagen production.

Epithelial cell damage leads to bFGF activation and collagen deposition by smooth muscle cells.

Increased expression of TNF-a and PDGF, and downregulation of epithelial NOS activity leads to smooth muscle cell proliferation and collagen deposition.

Downregulation of TM and plasminogen activator activity induces a procoagulant situation through inhibition of intravascular coagulation and decreased fibrinolytic activity, respectively.

Epithelial barrier breakdown and mast cell hyperplasia leads to increased TGF-P production and activation.

Permanent fibroblast phenotypic alterations and accumulation of altered fibroblast population lead to overproduction of matrix.

Table 1. Mechanisms in radiation-induced fibrosis (modified from (47)).

TGF-p plays a pivotal role in the development of delayed structural changes.

Interestingly, TGF-P can lead to completely opposite biological effects, depending on the cell type investigated, e.g. stimulation of proliferation in fibroblasts versus inhibition of proliferation in epithelial cells. This can be explained by the different TGF-P receptors that activate different transcription factor systems, which can be altered by exogenous factors like irradiation (82).

The fibrogenic TGF-P is found in abundance in a-granules in platelets, is released at sites of radiation-induced vascular injury with platelet clotting, and exerts a strong chemotactic action on mast cells (83). The mucosal expression of TGF-P is upregulated immediately after irradiation and in late radiation injury (84,85), associated with mast cell hyperplasia (86) and degranulation (43). Prominent radiation-induced increase of TGF-p expression is particularly found around mature vessels, proliferating capillaries and in fibrosis

(85). TGF-p has also been described as the main inducer of the connective tissue growth factor (CTGF), which is proposed as one of the essential mediators of intestinal radiation fibrosis, contributing to sustained fibrogenic signals in the bowel (87).

TGF-P and mast cells thus play important roles during and after irradiation, involving several biomolecular mechanisms. Mast cells regulate several important biological processes through the production, release and modulation of cytokines, chemokines, and growth factors.

The two types of mast cells, mucosal mast cells (MMCs) and connective tissue mast cells

(CTMCs), differ in function. Broadly speaking, MMCs exert anti-inflammatory actions during the acute phase, while CTMCs stimulate fibrogenic processes in the chronic phase. This is illustrated by experiments in mast cell-deficient rats, where irradiation resulted in strikingly exacerbated mucosal injury but minimal late reactive intestinal wall fibrosis, while TGF-p levels were unchanged (88). Reducing TGF-p activity by administration of a soluble receptor fusion protein resulted in a significant attenuation of radiation fibrosis without effect on mast cell hyperplasia (89). Similar results have been obtained in studies carried out in other organs using neutralising TGF-P antibodies (90), sequestering agents (91), chemical activation blockers and antisense oligonucleotides (92), Cu/Zn superoxide dismutase (93,94), the combination of pentoxifylline and tocopherol (95), and vector-mediated receptor alteration (96). From these studies, TGF-J3 inhibition holds promise as a means of ameliorating chronic radiation injury (97).

Endothelial TM plays a pivotal role in maintaining the coagulation/anticoagulation homeostasis. The conversion of fibrinogen into fibrin and the activation of protease-activated receptor-1 (PAR-1) require thrombin. When bound to TM, thrombin does not exert these functions and instead activates protein C, an anticoagulant and anti-inflammatory factor. This situation is disorganised when TM is deficient, either down-regulated or released by irradiation, creating a pro-coagulant and pro-inflammatory situation. A self-perpetuating process is established, constituting one of the mechanisms responsible for the chronicity of radiation injury (98). It may also be a target for interventions to increase the therapeutic ratio by restoring or preserving endothelial TM and blocking PAR-1.

3.7. Radiosensitivity

Defining predisposing genetic and non-genetic pretreatment characteristics would ease individual tailoring of the treatment to a greater extent than up to now. Significant therapeutic gain could be achieved from the use of predictive assays of radiosensitivity in normal tissue

(99). Table 1 summarises the tests currently used to assess radiosensitivity (100). So far, however, clinically applicable pre-treatment tests have not been widely available. With the advent of micro-array techniques novel analyses may ensue, improving the tailoring of RT in the future. 21

Test Principle Cell(s) tetsted Clonogenic assay Cell death, apoptosis, Fibroblast, keratinocyte differentiation, senescens (biopsy)

Pulsed field gel Measures residual DNA Fibroblast electrophoresis double strand breaks (biopsy)

Single-cell gel electrophoresis Capacity of rejoining in vitro Lymphocytes (The "comet assay") radiation-induced DNA strand (peripheral blood) breaks

Radiation-induced apoptosis Decreased apoptosis after Lymphocytes radiation in AT-patients (peripheral blood)

Fibroblast SF2 Survival fraction after 2Gy Fibroblasts

Cytogenetic tests Table 2. Tests for radiosensitivity

The clinically most important RT complication is chronic structural changes which are dominated by fibrosis. Much interest has therefore focused on the fibroblast/fibrocyte system for predictive tests. In vitro tests of fibroblast survival are theoretically interesting, but are far from applicable in a practical clinical situation. The results from subcutaneous fibroblast tests can not be extrapolated to make changes in the scheduled RT dose for gastrointestinal cancers. The tests are time-consuming and of uncertain value for an individual dose tailoring.

The impact of these tests on the occurrence of acute or chronic intestinal injury has not been settled. Most studies identify groups of patients at risk of side-effects, but some studies report correlation between fibroblast sensitivity and clinical reactions in the same patient (101-104).

The risk of fibrosis, but not teleangiectasia or erythema, was significantly correlated to fibroblast survival (105). The endpoint of these studies has been the assessment of late structural changes. As expected, there seems to be a lack of correlation between fibroblast sensitivity and acute radiation injury, where inflammation and not fibrosis is dominating.

The influence of patient characteristics and co-morbidity on radiation-induced morbidity has been adressed in several studies. Diabetes mellitus (106) and hypertension (107-109), previous abdominal surgery (110,111), and inflammatory bowel disease (112) contribute to

delayed injury. A recent study showed that smoking is the strongest independent predictor of

overall complications (113), in keeping with a previous study (114), possibly because of

arteriosclerotic changes rendering the tissues less capable of repair. Race may be important as

Hispanics had significantly lower complication rates than whites and blacks (113).

Genetic factors may predispose for radiation side-effects. Certain inherited disorders carry a

pronounced hypersensitivity to conventional doses of RT. A comprehensive review has

recently been published (115). The best known disorders are the ataxia-teleangiectasia (AT)

syndrome (116,117), Nijmegen breakage syndrome (118), Mrel 1 deficiency syndrome (118),

and Fanconi's anaemia (119). In vitro colony survival tests using blood derived

lymphoblastoid cells are pathologic in all these disorders.

Pre-treatment plasma levels of TGF-pl and soluble receptor-bound TGF-pi, probably

genetically determined, were significantly elevated in breast cancer patients who later

developed radiation-induced fibrosis (120). This is a promising approach, but the predictive

value and the organ specificity of the test have not been confirmed.

In the future, genetic screening by new techniques as RNA and DNA expression arrays and

mass spectrometry may come useful for identifying new genes and proteins rendering normal tissue cells extra radiosensitive.

3.8. Radioprotection

Treatment tailoring and mechanical shielding are important measures in protecting normal tissues during RT (108,121,122,122). However, radioprotection refers generally to pharmacological modifications of the tissue responses to irradiation. Only this aspect will be further discussed here.

A variety of principles have been applied to minimise toxicity, thereby allowing higher radiation doses to be delivered to the tumour. Protecting agents must act selectively on normal tissue cells and not induce radioprotection in tumour cells. In addition, both locally

administered and systemic drugs must have an acceptable safety profile. In spite of extensive

research, biomedical and clinical, no available systemic drug has fulfilled all these criteria.

Although acute radiation toxicity does occur and can be of significant clinical severity, the

development of side-effects due to chronic structural changes in normal tissue is the main

dose-limiting factor in RT and the main reason for the search of potent radioprotecting drugs.

The easy access constitutes unique possibilities for local pharmacologic prophylactic and therapeutic modifications of radiation reactions in the rectum.

Radiation is immediately followed by accumulation of reactive oxygen species (ROS) that are highly injurious to cells. Antioxidants are therefore potentially useful radioprotecting agents with minimal side-effects. In most studies vitamins A, C, and E, and carotenoids have been used. Many radiation oncologists do not, however, recommend antioxidants from fear of imposing radioresistance to tumour cells as well. This has been a matter of discussion and disagreement over several years, recently comprehensively reviewed by Prasad et al. (123).

The controversies have arisen on the background of the selected radiation doses and designs of experiments, but there is now accumulating experimental evidence that high doses of antioxidants enhance the effect of irradiation on cancer cells (124), while normal cells are protected against some of the injury (125). Vitamin C succinate induces apoptosis in human prostate cancer cells (126), and in human breast cancer cells (127). Several in vitro and animal model experiments show similar results, and there are no published studies showing that high doses can protect cancer cells against radiation damage. On the other hand, low doses of antioxidants have either no inhibitory effect, or may even stimulate the growth of cancer cells

(128). Many radiation oncology units recommend multivitamin supplies during treatment. If the doses are too small, insufficient effect of RT may be the consequence. Detailed recommendations have been proposed on the basis of an extensive literature review (123). Amifostine (WR-2731), an organic thiophosphate, was originally introduced as a

warfare radiation protectant and has later been used for normal tissue protection during RT. It

is a prodrug, which is cleaved by a membrane alkaline phosphatase yielding a free thiol, the

active metabolite WR-1065 (129). This metabolite is capable of scavenging free radicals,

enhancing endogenous cellular repair mechanisms, and stabilising DNA through positively

charged amino groups (130). The radioprotective action of amifostine is claimed to be

selectively associated with normal tissue, although it protects glioma cells (131). In animal

experiments amifostine exhibit protection of irradiated gastrointestinal mucosal tissue

(132,133). However, on the background of an unfavourable side-effect profile (hypotension)

and the paucity of sufficient clinical data the drug has not achieved widespread application

(134,135).

Glutamine is important for the maintenance, growth and function of the mucosa (136)

and has been recommended as therapy and prophylaxis during RT (137), based on animal

experiments (138-140). Clinical evidence is, however, scant. A randomised study by Richards et al. (abstract FASEB) in 26 patients showed improved histology, but no clinical benefit.

Recently, a larger study showed no significantly beneficial effect of glutamine during pelvic

RT (141).

Sucralfate has been recommended as a radioprotectant after an initial report showed attenuation of acute toxicity during pelvic RT (142). Sucralfate is an aluminium sucrose octasulphate, previously used successfully in the treatment of peptic ulcer (143). It adheres to damaged epithelium, constituting a mechanical shielding against injurious luminal factors.

Sucralfate has also several biological effects, such as binding of basic fibroblast growth factor

(bFGF) at the site of injury (144), stimulation of somatostatin (145) and prostaglandin release

(146), and regulation and stimulation of angiogenesis (147). On a theoretic basis, sucralfate therefore appears to be an ideal drug against acute intestinal radiation injury. However, conflicting clinical results have been reported. Controlled clinical trials have not been able to reproduce the favourable initial effects reported by Henriksson et al. (142). One study has even reported worsening of acute toxicity symptoms (148). On the other hand, sucralfate treatment may be useful in established chronic radiation injury (149). This has, however, to be confirmed in further and larger studies. 4. Aims of the study

a) A sequential description of the clinical course of acute radiation proctitis during pelvic

RT.

b) A sequential description of the rectal mucosal histopathology during pelvic RT as a

possible substrate for clinical toxicity

c) To assess the mucosal protease activity during RT as a possible explanation of the

observed tissue changes.

d) To assess the efficacy of prophylactic sucralfate in acute radiation proctitis a

randomised study was initiated and carried out together with a meta-analysis of

previously available data

e) Most studies on clinical acute toxicity in pelvic RT use either the RTOG/EORTC

score system or focus on diarrhoea/stool frequency. A more differentiated and

sensitive recording was developed and tested to pick up symptoms escaping the

commonly used scores.

f) Study the relation between histopathological findings and the clinical picture. 5. Material and Methods

The present studies started with a randomised, placebo-controlled study of oral sucralfate versus placebo (Paper IV). Similar procedures as described below were applied for further clinical studies.

5.1. Patients. Consecutive patients with organ-confined pelvic cancer scheduled for RT

were recruited at the oncologic outpatient clinic, and included in studies after written

informed consent. Exclusion criteria: Age > 80 years; Crohn's disease, ulcerative colitis,

coeliac disease, gastrointestinal cancer; non-compliant patients.

5.2.Symptom registration was carried out using the RTOG/EORTC acute gastrointestinal

toxicity scoring (26) and our own registration and scoring system (Paper III). Registrations

were done prior to RT, at week 2,4, and 6 during RT, and in some patients also 2 and 8

weeks after the end of RT.

5.3. Endoscopy and biopsy. Proctoscopy was carried out prior to RT, after 2 weeks, in

selected patients after 4 weeks, and after 6 weeks during RT, in selected patients also 2 and

8 weeks after completion of RT. A rigid proctoscope was used. Small cup biopsies were

obtained from the posterior rectal wall 10 cm from the anal. Two biopsies were fixed in

formaldehyde for histology (Paper I), 2 biopsies were snap frozen to -80°C for chemical

analyses (Paper II). Endoscopic findings were graded 0-3 according to Hanauer et al.(150):

0 = normal mucosa; 1 = oedema and/or loss of vascularity, granularity of mucosa; 2 =

friable mucosa or peteciae; 3 = spontaneous hemorrhage or visible ulcers.

5.4. Special analyses were carried out as outlined in the papers. 28

6. List of papers

Paper I 1. Hovdenak N, Fajardo LF, Hauer-Jensen M. Acute radiation proctitis: a sequential

clinicopathologic study during pelvic radiotherapy. Int J Radiat Oncol Biol Phys 2000; 48:

1111-7.

Rectal mucosa biopsies from 33 patients taken before and 2 and 6 weeks into pelvic RT

underwent blinded histologic examination. An ad hoc scoring system was applied,

allowing quantitation of inflammatory changes. Symptoms of acute toxicity and graded

endoscopic changes had been recorded at the same time-points. In contrast to clinical

symptoms, endoscopic scores stabilised and histologic changes regressed from the 2nd to

the 6th week of RT.

Paper II Hovdenak N, Wang J, Sung C-C, Kelly T, Fajardo LF, Hauer-Jensen M. Clinical significance of increased gelatinolytic activity in the rectal mucosa during external beam radiation therapy of prostate cancer. Int J Radiat Oncol Biol Phys 2002; 53:919-27.

Extracellular matrix degradation is evident in both acute and chronic radiation injury.

Gelatinases may be important in this process. We prospectively recorded symptoms,

endoscopic scores, histology, and mucosal macrophages and neutrophils, and examined

changes in matrix metalloproteinase (MMP)-2 and MMP-9 in rectal mucosal samples

before and during RT by gelatin zymography. Pelvic RT caused increased MMP-2 and

MMP-9 activity in the mucosa, correlating with diarrhoea and granulocyte infiltration. 29

Paper III Hovdenak N, Karlsdottir Å, Sørbye H, Dahl O. Profiles and time course of acute radiation toxicity symptoms during conformal radiotherapy for cancer of the prostate. Acta Oncologica

2003 (In press)

Prevalence and severity of specific toxicity symptoms were recorded in 96 patients

undergoing pelvic RT for prostate cancer before, during and after the treatment. Global

assessment of toxicity was done using the RTOG/EORTG acute toxicity scoring system

and an ad hoc total toxicity scoring (TTS) based on interviews and visual analogue scales.

The study demonstrated an increase in prevalence and severity for individual symptoms

and for TTS, which appeared more sensitive than RTOG/EORTG scoring. Multiple

toxicity symptoms contribute to total toxicity. Although not formally validated, TTS is a

feasible and sensitive method that unveils morbidity which is hidden with the use of the

RTOG/EORTG system. Development and timing of symptoms may give clues to

pathogenesis, treatment and prophylaxis.

Paper IV

Hovdenak N, Sørbye H, Dahl O. Sucralfate does not ameliorate acute radiation proctitis

Submitted for publication.

Conflicting results have been reported on the prophylactic use of sucralfate during pelvic

RT. We performed a prospective, randomised study of prophylactic sucralfate versus

placebo in 44 patients with localised pelvic tumours who underwent RT with total dose

66-70Gy in 2 Gy daily fractions. Sucralfate (2g) or placebo tablets were administered 3

times daily during the entire RT course. Endoscopy with recta! biopsy, and symptom

registration were performed before and 2,4 and 6 weeks into the treatment. The study was terminated after inclusion of 44 patients when an interim analysis showed increased diarrhoea in the sucralfate group and no difference in other symptoms, endoscopic appearance or quantitative histology Diarrhoea score was significantly higher in the sucralfate group at week 2 and week 6 (p=0.049 and p=0.033, respectively). A meta- analysis of 5 similar studies failed to show any beneficial effect of sucralfate. Sucralfate can therefore not be recommended in the prophylaxis and may even worsen the symptoms. 31

7. General discussion

Our studies have centered on the observable intestinal phenomena during pelvic radiotherapy. Earlier studies have described qualitative histological changes much in keeping with our findings. Our study is, however, the first to include quantitation that allows statistical calculations, showing an early histopathologic reaction and protease activity, which are attenuated thereafter in spite of ongoing irradiation. These findings are in contrast to the clinical progress, which is generally worsening through the RT course, illustrated by our new and detailed scoring system. The discrepancy between clinical and morphological findings has not been demonstrated previously and calls for new pathogenetic explanations and clinical strategies.

7.1. Symptoms

The clinical picture of acute radiation proctitis has been known through daily experience in the radiation oncology departments for years and is characterised by increasing symptoms, mainly diarrhoea, throughout the radiation course. In recent years, symptom score systems have been applied, mainly for scientific purposes. The RTOG/EORTC acute scoring system

(26) gives no detailed description or quantitation of the different symptoms. It provides an over-all assessment, graded in severity from 0 to 4. The system is used in studies comparing different RT regimens. A few studies (28,151) have presented more detailed symptom descriptions, indicating a low sensitivity of the RTOG/EORTC scoring. Falsely low toxicity estimates may therefore ensue, obscuring the differences between treatments. We looked for a system where the frequency and severity of the most common toxicity symptoms could be assessed and the total symptom burden expressed as a sensitive severity score. A detailed registration could hopefully also offer information on the pathophysiology and the timing and nature of therapeutic and prophylactic measures. Our scoring system has neither been formally validated nor used in studies comparing therapeutic strategies so far. However, it demonstrates a differential and quantitative, sequential description of the clinical situation during and immediately after RT that, together with endoscopic, histological, and immunohistochemical findings, may give clues to hypotheses on pathogenetic and pathophysiologic factors. Most of the symptoms increased during the entire RT course, in contrast to morphologic inflammatory changes which generally increased early during the course and thereafter attenuated. In addition, symptoms clearly originating in organs or tissues outside the radiation field were common and call for other pathophysiologic explanations.

Symptoms like nausea and bloating most likely originate in disturbed gastrointestinal motility. The literature is scarce on motility in patients undergoing RT. Delayed gastric emptying is demonstrated in pelvic RT patients (152), and Yeoh et al. (153) using radionucleic tests demonstrated neuroendocrinologically mediated motility disturbances. We estimated gastric emptying and intragastric volume distribution by three-dimensional ultrasonography in a pilot study in 8 of our patients and found no change during the radiotherapy course (unpublished data). However, these patients did not complain of much nausea. Further studies including more symptomatic patients are planned.

Results from animal experimental studies are difficult to extrapolate to the clinical situation. Single dose or hypofractionated radiation is usually applied in these studies, in contrast to the highly fractionated modern RT. However, results clearly demonstrate profound motility changes (154-158).

Gastrointestinal motility is orchestrated through a very complicated cross-talk between nervous, endocrine, and paracrine mechanisms (159). The enteric nervous system (ENS) is partly autonomic but exerts its functions also in concert with the central nervous system

(CNS) through afferently conveyed signals from intestinal nervous and endocrine structures 33 and through efferent signals from the CNS to the intestinal smooth muscles and intrinsic nervous network. Most of the symptomatic treatment administered for toxicity symptoms exerts its effect through modulation of ENS activity (e.g. loperamide, antiemetics, 5HT- receptor antagonists).

Symptoms of delayed radiation intestinal injury is highly important clinically, characterised by malabsorption (in small bowel involvement), bleeding, and dysmotility. It carries a high morbidity and mortality (79,80). Only a minority develop complications severe enough to require surgical intervention. However, milder symptoms may occur in as much as

60-90% of patients (153,160), substantially reducing the patients' quality of life. These problems are often not recognised or properly dealt with by doctors. Numerous patients do not seak medical help and have to accept severe restrictions on daily life activities.

7.2. Histopathology

Pelvic RT evokes varying degrees of inflammatory changes in the rectal mucosa

(Table 3), detectable endoscopically and histopathologically, associated with variable clinical symptoms.

Epithelium Lamina propria Necrosis Inflammatory infiltration Erosions Crypitis and crypt abscesses Intraepithelial inflammatory infiltration Endothelial changes Haemorrhages Oedema

Table 3. Acute intestinal reactions to irradiation (Modified from (37)). Acute toxicity symptoms have mechanistically been ascribed to this "proctitis" although a difference in timing between objective mucosal changes and symptom progress was indicated several years ago (42). We wanted to define the histologic changes in relation to time

(accumulated radiation dose) and quantitate these characteristics. Endoscopic mucosal biopsies were obtained before and during RT for quantitative histology, and the endoscopic changes were graded.

Early and significant increase in histological inflammatory scores was demonstrated that appeared parallel to the endoscopic changes. From the second week of RT no increase in endoscopic severity was noted. Histopathology improved significantly from the second week towards the end of RT, reflecting a compensatory proliferation and repopulation shortly after the start of RT (161,162), more or less outweighing the radiation-induced cell loss. In a study by Carratu et al. (163) improvement in histology was parallel to functional mucosal changes measured by gut permeability tests, showing early increase followed by normal values at the end of RT. The mechanisms governing this proliferation are unclear, but is related to cell kinetics, and timing and fractionation of irradiation. Radiation-induced activation of preexisting pathways is critically involved in the process (164,165). A major response involves activation of ErbB receptor Tyr kinase in the plasma membrane, the mitogen- activated protein kinase (MAPK) and subsequently anti-apoptotic and pro-proliferative cellular responses (166).

Eosinophilia was prominent histologically, both in the lamina propria, intraepithelially, and in cryptitis and crypt abscesses, in keeping with earlier studies (38,42,167). The significance of this finding is unclear, but the presence of eosinophils indicates functional roles in tissue reaction to trauma. Eosinophils are multifunctional pro-inflammatory leukocytes and exert a host of biological effects. They are associated with several allergic and non-allergic gastrointestional disorders, notably also with inflammatory bowel disease (IBD) (168,169). On degranulation, eosinophils release toxic proteins (major basic protein,

eosinophil cationic protein, eosinophil peroxidase, eosinophil-derived neurotoxin).

Interestingly, tissue eosinophilia is associated with gut dysmotility and neuron damage in

proximity to eosinophils (170). In IBD and in our biopsy study, the eosinophils do not appear

to be degranulating, indicating a regulatory function. Eosinophils release pro-inflammatory

cytokines (IL-2, IL-4, IL-5, IL-10, IL-12, IL-16) (171), TGF-a/p, chemokines (RANTES,

eotaxin), lipid mediators (PAF, leukotriene C4) (172). Still, the significance of these

phenomena is enigmatic. Intestinal accumulation of eosinophils in various diseases and

experimental models is associated with severe clinical symptoms (diarrhoea, weight loss) and

structural gut changes, which are attenuated in eotaxin-deficient mice (170).

The histological and endoscopic changes, with a decrease after an early peak, are in

contrast to the development of clinical toxicity, which increases in intensity and prevalence

throughout the RT course and immediately thereafter. Symptomatic treatment is consequently

administered at a time when objective disease parameters are in the process of healing. On the

other hand, when it comes to the use of prophylactic measures, our findings may have

implications for the timing: intervention should be instituted parallel to the start of RT, or

even before.

7.3. Gelatinases

Atrophy and loss of tissue components are prominent features of radiation injury in the

gut. The mucosal glands become distorted and diminished, making the lamina propria

atrophic. This could be a result of ischemia secondary to vascular insufficiency (51) or a

consequence of direct tissue destructing processes, or both. In acute inflammation cascades of pro-inflammatory cytokines and other agents are activated, triggering several enzyme and kinase systems. The gelatinase activity is therefore of interest in the acute tissue reaction during RT. We showed that the mucosal content of total and active MMPs increased early during RT, with a maximum after 2 weeks. Thereafter a significant elevation was maintained, although no further increase was observed. The changes over time were thus parallel to those observed in histopathology.

The gelatinase activity in delayed radiation-induced structural changes has not been tested. However, the production of MMP is up-regulated by pro-inflammatory cytokines, notably TNF (173,174), by thrombin-thrombomodulin through activated protein C (APC)

(175), and by fibrogenic cytokines and growth factors that are over-expressed in both acute and chronic radiation injury. TGF-(3 is the most prominent of these (176,177). TGF-P may exert its actions partly by activating MMP-2 and MMP-9, resulting in a net reduction in ECM, characteristic for late radiation injury.

Gelatinases belong to the matrix metalloproteinase (MMP) family of over 20 enzymes that cleave various components of the extracellular matrix (178,179). MMPs are characterised by the presence of conserved protein domains: a prodomain, an active domain and a Zn2-binding domain. The two gelatinases matrix metalloproteinase (MMP)-2 and MMP-9 degrade the collagen in the mucosal lamina propria of the basal membrane of the superficial and glandular epithelium and the extracellular matrix (ECM) in the lamina propria (178). MMP-2 is expressed by stromal cells. MMP-9, which is the largest and most complex member of the

MMP family (180), is expressed by haematopoietic and inflammatory cells and epithelial cells and is an important regulator and effector in leukocyte biology (181). Both are also secreted by several malignant tumours, facilitating the local invasion by breaking up the surrounding tissue matrix (182). In inflammatory conditions like radiation proctitis, this breakdown may enhance local infiltration of inflammatory and reparative cells, or, on the other hand, even facilitate the spread of injurious factors gaining access from the rectal lumen after the damage to the epithelial basal membrane. MMPs are a multifunctiomal group of enzymes (183), not confined solely to the

extracellular matrix (184). Increases in ECM degradation, cell migration, inflammation,

platelet aggregation (185), all contribute to structural changes and may be involved in clinical

pathology. Activation (186) and release (187) of TGF-p is probably important in tissue

remodelling after trauma, e.g. irradiation. Anti-inflammatory actions have been reported, for

example, the monocyte chemoattractant protein-3 is cleaved by MMP-2, which then functions

as a chemokine antagonist that dampens inflammation (188). Prostaglandins influence the

activation and expression of MMPs (189,190). The beneficial effect of exogenous

prostaglandin in acute radiation proctitis (191) may therefore indicate an anti-inflammatory

effect of MMPs. However, the net result of RT seems to be an up-regulated pro-inflammatory

MMP activity.

Specific inhibitors of the MMPs have been synthesized recently and have been used in

adjuvant treatment of malignant disease (182). In experimental colitis, treatment with a

synthetic MMP inhibitor prevented inflammation (192,193) and delayed angiogenesis (194).

Specific pharmacologic modulation of MMPs may therefore be a therapeutic or prophylactic

option in RT patients at risk of developing delayed radiation injury.

7.4. Prophylaxis Modern pelvic RT for prostate cancer does not carry a high risk of severe acute toxicity

(Karlsdottir et al., manus submitted). In some other pelvic malignancies radiation fields are

larger, including substantial volume of the colon and the small intestine. The search for prophylactic pharmacologic measures has therefore high priority in an attempt to increase the therapeutic ratio and reduce acute toxicity and late structural intestinal damage. The background of our interventional study was a report from Henriksson et al. on the prophylactic use of oral sucralfate during pelvic RT in 70 patients (142). The study was a clinical, prospective, randomised, and placebo-controlled trial, demonstrating a beneficial effect of sucralfate in terms of less toxicity, expressed as diarrhoea. We wanted to include endoscopic and histological investigations in a similar study. However, our findings did not confirm the clinical results from the Swedish study. We decided to prematurely stop the study when an interim analysis showed a significant increase in diarrhoea in the sucralfate group and no beneficial effect on other toxicity symptoms. This is in keeping with several other studies on sucralfate as outlined in our meta-analysis. Worsening of diarrhoea during sucralfate prophylaxis was also shown in a study by Martenson et al (148). Neither did we find any difference histologically or endoscopically between sucralfate and placebo treated patients. On this background, sucralfate can not be recommended as therapy or prophylaxis during pelvic RT.

However, sucralfate has been used in chronic radiation enteropathy. One study reported significant efficacy in ameliorating symptoms (149). These results should be further explored.

Other studies are of open-label design with few patients (149,195) but showed a striking beneficial effect of rectal and oral sucralfate in keeping with the general impression from several case reports (196-200). A single study from Sweden (201) showed a reduction of chronic diarrhoea in patients who had been on oral sucralfate during RT. However, this was not confirmed in a study from Australia (202).

Effective prophylaxis aims at amelioration or abolition of both acute and late side-effects of RT. Clinically, the late structural changes represents the most serious consequences for the patient, resulting in considerable morbidity and mortality. A pharmacologic approach to the problem requires detailed knowledge of the acute reaction, predisposing factors, and mechanisms leading to chronicity. No prognostic, patient specific, pretreatment features or predisposing factors other than co-morbidity have yet been identified that enhance or inhibit this development. 8. Conclusions

Available data on acute side-effects of fractionated pelvic RT are mostly clinical. Normal intestinal tissue injury has mainly been studied in animal experiments that do not reflect the clinical situation. Although proctoscopy is a minor medical procedure, it is inconvenient for the patients, especially when performed repetitively during RT. In spite of this, all patients asked agreed freely to participate in our studies. This made it possible to present novel endoscopic and histologic data and establish a clinical model feasible for future scientific work.

Based on the data discussed in this thesis, the following conclusions are drawn:

1. Analysis of symptoms of acute toxicity during pelvic RT has been hampered by over-

all scoring systems that give scarce information on details. Using a more detailed

approach with gradings of the individual symptoms, we have been able to reveal

significant complaints that are not readily recognised by systems such as the

RTOG/EORTG acute scoring system. Our method has not been formally validated,

but appears feasible and is easily applicable. Validation studies are on-going.

Prevalence and intensity of individual symptoms as well as the total symptom burden

is shown to increase during the RT course and continue for some time after the

cessation of the treatment. The symptoms of acute radiation proctitis are in contrast to

the histopathology and the MMP activity and may be misleading for planning and

timing of prophylactic and therapeutic measures other than mere symptomatic

approaches. Mechanisms apart from inflammation are likely in the pathogenesis of

acute clinical toxicity in pelvic RT. Many symptoms may reflect intestinal dysmotility. 2. Acute intestinal (i.e. rectal) morphological injury during fractionated pelvic RT is

maximal early during the course and is attenuated thereafter in spite of continuing

radiation and clinical symptoms. The injury is dominated by inflammation and loss of

tissue. A compensatory proliferation results in partial healing of the mucosal lesions

from the second week of RT. The inflammation is characterised by a striking tissue

eosinophilia, the significance of which is not settled. It should be pointed out,

however, that clinically available tissue samples are small and superficial. For safety

reasons, the biopsies do not generally include deeper structures of the gut wall. Injury

in these parts may involve muscular, vascular, neural and neuroendocrine structures as

well as connective tissue which directly or inderectly may influence the mucosa.

3. Matrix metalloproteinases (MMPs) activity is increased early during pelvic RT. This

may be related to the early histologic changes showing loss of tissue components and

infiltration of inflammatory cells in the matrix of the lamina propria. MMPs degrade

the collagen components of the matrix and the basal membrane of the intestinal

surface epithelium, thus facilitating the migration of toxic intraluminal components as

well as inflammatory cells. The inhibition of the MMPs may therefore be a

prophylactic or therapeutic possibility.

4. According to our study, and a review of published trials, sucralfate is ineffective as a

radioprotective drug in pelvic RT. The most likely reason for this failure is that

sucralfate does not interfere with known biomolecular cellular processes governing the

extent and nature of radiation injury. 41

9. Future prospects

There are several reasons for continuing research on acute radiation proctitis. - Acute toxicity carries a significant, although temporary, morbidity. The pathophysiologic mechanisms leading to acute clinical manifestations are still enigmatic, making effective therapy difficult. - An even greater challenge is the fact that the basis for the development of delayed structural changes is established during the RT course, hence modulation of these primary key mechanisms carries immense prophylactic and therapeutic possibilities. -

Finally, research on and development of technical radiation equipment and methods are still important issues and have continuously resulted in improved therapeutic outcome.

9.1. Symptomatic treatment Most of the acute symptoms increase in prevalence and severity during pelvic RT and are not readily explained by inflammation. Most likely, disturbed gastrointestinal motility is the main common pathogenetic denominator for these symptoms (diarrhoea, nausea, bloating).

Symptomatic treatment could therefore logically concentrate on ameliorating these disturbances.

Gastrointestinal motility is closely regulated by neuroendocrine mechanisms that are potential targets for drug therapy, mostly as agonists or antagonists at specific cellular receptors (203). Enteroendocrine cells are situated between epithelial cells in the colon and small intestine (204). Their apical microvilli detect the mechanical and chemical environment in the gut lumen and release appropriate mediators across the basolateral membrane that in a paracrine way influence the mucosal and submucosal afferent nerve fibers, cholecystokinin

(CCK) and serotonin/5-hydroxytryptamine (5-HT) being well documented transmitters. We found significantly increased immunoreacting, serotonin-containing, enterochromaffine cells in the rectal mucosa after 2 weeks of pelvic RT (unpublished data), indicating pathogenetic mechanisms of disturbed motility. The role of serotonin in radiation-induced emesis has been outlined previously (205). After release from the enterochromaffine cells in the gut mucosa, serotonin stimulates the 5-HT type 3 (5-HT3) receptor in near relation to afferent visceral nerves thar activate the chemoreceptor trigger zone in the brain (206). The reported beneficial effect of (5-HT3) receptor inhibitors constitutes indirect evidence for serotonin-mediated dysmotility as a cause of symptoms during abdominopelvic RT (207,208).

However, most studies on radiation-induced dysmotility are experimental animal studies, and very few have adressed the clinical situation. Clinical studies are hampered by technical and methodological difficulties. Absorption tests and breath-tests provide indirect data on transit times (152,209) while peristaltic and propulsive activity data are hard to obtain.

Standardised transabdominal ultrasound recordings may be feasible and has been used in other settings (210,211). It is a non-ivasive, widely available and cheap procedure, and suitable for repetitive examinations. As mentioned previously, we used this technique in patients with prostate cancer during pelvic RT (unpublished data). No significant changes were found in asymptomatic men during the RT. The technique, however, needs further standardisation before further studies are undertaken.

9.2. Prophylaxis A successful prophylaxis may be achieved by expanding the therapeutic ratio (Fig. 4).

This can be done either by selectively increasing the radiosensitivity in the tumour, or selectively protect the normal tissue. Methods of enhancing the radiosensitivity of the tumour cells are, however, beyond the scope of this part of the thesis. 43

Prophylactic and therapeutic biological strategies should be based on a detailed knowledge on biomolecular and genetic mechanisms influencing underlying pathophysiologic mechanisms of the acute radiation injury and the radiosensitivity of tumours and normal tissues. The recent decade has seen a great expansion of such knowledge. Pharmacologic and immunologic regulations of these mechanisms are coming of age and some have been implemented in clinical trials, showing promising results.

We have demonstrated an increase in tissue gelatinase activity during RT. Traditionally,

MMPs have been linked to destruction of basement membranes and extracellular matrix structures. However, recent, enhanced understanding has revealed associations in much more complex biological systems (184), both stimulating and repressing cellular proliferation, apoptosis, chemotaxis, and migration. Inhibition of MMP activity may therefore aggravate a pathological condition, and in other settings be beneficial. Clinical trials have been limited to adjuvant therapy in cancer, with variable results and unclear indications and results (212).

Therapeutic applications of MMP inhibitors may be available for acute (and chronic?) radiation enteropathy in the future, but a greater understanding of the MMPs, their substrates and the biological consequences of their cleavage products is mandatory before such studies can be designed. 44

Interventions should aim at normalisation of unduely exaggerated or prolonged physiological mechanisms evoked during the RT course to prevent the development of serious

acute reactions and, more important, the transition or later appearance of delayed structural

changes. As demonstrated in our studies, acute intestinal reactions take place early, emphasising the importance of early prophylactic interventions.

Our insight into several important biomolecular mechanisms in normal tissue radiation injury has improved considerably over the last years. The following mechanisms are examples of novel knowledge that may form the basis for future therapeutic or prophylactic strategies.

Changes in endothelial function and coagulation mechanisms are essential in the eventual development of delayed injury. The radiation-evoked pro-coagulant and pro-inflammatory state is a potential target for prophylactic intervention (213). Endothelial cell TM is directly down-regulated by radiation (55), resulting in reduced activation of protein C. Activated protein C (APC) serves as an intrinsic anticoagulant and anti-inflammatory agent. Exogenous supplementation of APC is therefore a potentially effective therapeutic and prophylactic approach.

We have demonstrated an increased protease activity in the irradiated mucosa early in the

RT course. The therapeutic implications of this phenomenon have not yet been explored.

Preclinical experiments will be needed to established the clinical significance of our findings.

Specific inhibitors of matrix metalloproteinases are available and should be tested.

The somatostatin analogue, octreotide, was shown to preserve epithelial barrier function and reduce tissue injury in a rat model of radiation enteropathy (53). In a subsequent study

(214) ocreotide was shown to confer dose-dependent protection against both acute and delayed tissue injury, abrogate chronic increase in extracellular matrix-associated TGF-p, collagen I, and collagen III. Octreotide is approved for clinical use in a variety of settings and has few side-effects. The prostaglandin synthesis in the human colon mucosa is inhibited by irradiation (215).

Local administration of the synthetic anti-inflammatory prostaglandin, misoprostol, increases the stem cell survival substatially compared to control values (216). A clinical trial (191) has recently demonstrated a significant reduction in acute and chronic radiation proctitis symptoms in prostate cancer patients receiving local misoprostol. However, larger confirmatory studies are lacking.

In inflammatory bowel disease, anti-inflammatory effects (217) and suppression of in situ immune reactivity are achieved by exogenous poly-unsaturatedfatty acids (PUFA) (218,219).

Of these, eicosapentaenoic acid (EPA) is the substrate for synthesis of anti-inflammatory prostaglandins, inhibiting the synthesis of thromboxanes from arachidonic acid. Fish oil and seal oil are important sources of EPA. Either enriched diet or fish oil supplementation could be explored as a prophylactic means.

Growth factors are important in radiation biology. Epidermal growth factor (EGF) and transformin growth factor-a (TGF-a) are illustrating examples. Upon binding to their receptors they initiate a cascade of signals from the cytoplasm to the nucleus that regulate cell division, proliferation, and differentiation (220). TGF-p is essential in the process of tissue fibrosis (176) and is associated with vascular sclerosis in radiation enteropathy (85). TGF-P is released from platelets at sites of radiation-induced endothelial damage and exerts a strong chemotactic action on mast cells (83). Mast cells are important in chronic fibrogenesis. Recent experimental data concerning the treatment of established radiation fibrotic disorders (88,90-

92,94) show that the TGF-beta pathway may constitute a specific target for antifibrotic agents.

Reduction of TGF-P 1, which is assumed to be the biologically most relevant of the isoforms in the fibrotic response in normal tissue, can be achieved by several methods: neutralising antibodies (90), sequestering agents (91), and antisense oligonucleotides (92). 46

Administration of a soluble recombinant type II TGF-receptor fusion protein in a mouse model of radiation enteropathy resulted in greatly reduced structural radiation injury and

intestinal wall fibrosis (89). This promising approach is awaiting further preclinical and

clinical testing.

9.3. Radiotherapy methods and equipment

Minimising irradiation to normal tissues should still have a high priority in research and development of new technology and RT methods. Radiation dose, fraction size and interval, and shielding are important (108,121,122) and much effort has already been applied to tailor the radiation treatment to the tumour and to spare non-tumourous tissues through computer- guided shielding by multileave collimators and by casted blocks. Intensity modulated radiation therapy (IMRT) has been developed in recent years. This is a technology of optimized computer-controlled treatment planning and computer-controlled modulation of the radiation beam during treatment (221-223). IMRT delivers radiation more precisely to the tumor while limiting the dose to the surrounding normal tissues. This allows dose escalation to 81 Gy without increased acute toxicity in prostate cancer patients (224-227), and significantly reduces irradiated volumes of rectal and bladder walls (225). Even in brachytherapy for prostate cancer, short-term gastrointestinal toxicity occurs (21-24,24), and long term follow-up studies are scant. Stereotactic catheter guided radiation is a new technique has allowed dose escalation without increased delayed toxicity symptoms (228).

New possibilities also include the development of heavy ion RT. Currently, however, such methods are experimental only.

The prevalence and the clinical significance of intestinal radiation injury are often underestimated by the general medical community. Even medical gastroenterologists are 47 unfamiliar with these conditions, and the surgeons see only the top of the iceberg, i.e. patients suffering from the severe complications: strictures, fistulae, perforations, infections, frequently needing major surgical interventions. The majority of patients, however, have more or less resigned, adopting their daily life to varying symptoms and complaints as the price to be paid for the cure of cancer. Their life quality may be severely hampered, and qualified medical help is scarce.

A simple Medline literature search in 6 leading international gastroenterological journals, using the key words: "radiation injury" and "radiation enteropathy", demonstrated the relatively poor interest in the topic among clinical gastroenterologists: The search found only

26 articles, while a general search (any journal) gave the astronomic 40590 hits.

Gastroenterology diagnostic work-up in radiation injury patients often reveals conditions familiar to most gastroenterologists (malabsorption, bacterial overgrowth, motility disturbances, bleeding, malnutrition, deficiencies, etc.), amenable in many instances by dietary changes, drugs, endoscopic treatment, vitamine supplements etc. Increasing the knowledge and clinical alertness among gastroenterologists is therefore an important issue today. 48

10. References

1 Walsh D. Deep traumatism from Roentgen ray exposure. Br Med J1897; 2: 272-273.

2. Debus J, Jackel O, Kraft G, Wannenmacher M. Is there a role for heavy ion beam therapy 1 Recent Results Cancer Res 1998; 150: 170-182.

3. Turesson I, Johansson KA, Mattsson S. The potential of proton and light beams in radiotherapy. Acta Oncol 2003; 42: 107-114.

4. Thompson LH, Suit HD. Proliferation kinetics of x-irradiated mouse L cells studied WITH TIME-lapse photography. II. Int.J.Radiat.Biol.Relat Stud.Phys.Chem.Med. 1969; 15: 347-362.

5. Hendry JH, West CM. Apoptosis and mitotic cell death: their relative contributions to normal-tissue and tumour radiation response. Int.J.Radiat.Biol. 1997; 71: 709-719.

6. Potten CS. The significance of spontaneous and induced apoptosis in the gastrointestinal tract of mice. Cancer Metastasis Rev. 1992; 11: 179-195.

7. Stone HB, Coleman CN, Anscher MS, McBride WH. Effects of radiation on normal tissue: consequences and mechanisms. Lancet Oncol. 2003; 4: 529-536.

8. Haimovitz-Friedman A. Radiation-induced signal transduction and stress response. Radiat.Res. 1998; 150: S102-S108.

9. Verheij M, van Blitterswijk WJ, Bartelink H. Radiation-induced apoptosis—the ceramide-SAPK signaling pathway and clinical aspects. Acta Oncol. 1998; 37: 575- 581.

10. Steel GG. The case against apoptosis. Acta Oncol 2003; 40: 968-975.

11. Otnes B, Harvei S, Fossa SD. Increased incidence of prostatic cancer. The significance of prostate-specific antigen. Tidsskr Nor Laegeforen. 1996; 116: 1795-1799.

12. Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun MJ. Cancer statistics, 2003. CA Cancer J Clin 2003; 53: 5-26. 49

13. Cancer in Norway 1999. Cancer Registry of Norway. 2003. Oslo, Cancer Registry of Norway.

14. Saclarides TJ. Radiation injuries of the gastrointestinal tract. Surg Clin North Am 1997; 77: 261-268.

15. Muren LP, Smaaland R Dahl O. Organ motion, set-up variation and treatment margins in radical radiotherapy of urinary bladder cancer. Radiother Oncol 2003; in press.

16. Dahl O, Kardamakis D, Lind B, Rosenwald JC. Current status of conformal radiotherapy. Acta Oncol 1996; 35: 41-57.

17. Levitt SH, Khan FM. The rush to judgment: Does the evidence support the enthusiasm over three-dimensional conformal radiation therapy and dose escalation in the treatment of prostate cancer? Int J Radiat Oncol Biol Phys 2001; 51: 871 -879.

18. Purdy JA, Michalski JM. Does the evidence support the enthusiasm over 3D conformal radiation therapy and dose escalation in the treatment of prostate cancer? Int J Radiat Oncol Biol Phys 2001; 51: 867-870.

19. Pollack A, Zagars GK, Starkschall G, Childress CH, Kopplin S, Boyer AL, Rosen II. Conventional vs. conformal radiotherapy for prostate cancer: preliminary results of dosimetry and acute toxicity. Int J Radiat Oncol Biol Phys 1996; 34: 555-564.

20. Ryu JK, Winter K, Michalski JM, Purdy JA, Markoe AM, Earle JD, Perez CA, Roach M, III, Sandler HM, Pollack A, Cox JD. Interim report of toxicity from 3D conformal radiation therapy (3D-CRT) for prostate cancer on 3DOG/RTOG 9406, level III (79.2 Gy). Int J Radiat Oncol Biol Phys 2002; 54: 1036-1046.

21. Gelblum D Y, Potters L. Rectal complications associated with transperineal interstitial brachytherapy for prostate cancer. Int J Radiat Oncol Biol Phys 2000; 48: 119-124.

22. Kleinberg L, Wallner K, Roy J, Zelefsky M, Arterbery VE, Fuks Z, Harrison L. Treatment-related symptoms during the first year following transperineal 1251 prostate implantation. Int J Radiat Oncol Biol Phys 1994; 28: 985-990.

23. Han BH, Demel KC, Wallner K, Ellis W, Young L, Russell K. Patient reported complications after prostate brachytherapy. J Urol. 2001; 166: 953-957. 50

24. Davis JW, Kuban DA, Lynch DF, Schellhammer PF. Quality of life after treatment for localized prostate cancer: differences based on treatment modality. J Urol. 2001; 166: 947-952.

25. Norderhaug I, Dahl O, Høisæter PÅ, Heikkila R, Klepp O, Olsen DR, Kristiansen IS, Wæhre H, Bjerklund Johansen TE. Brachytherapy for prostate cancer: a systematic review of clinical and cost effectiveness. Eur Urol 2003; 44: 40-46.

26. Cox JD, Stetz J, Pajak TF. Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int J Radiat Oncol Biol Phys 1995; 31: 1341-1346.

27. Liu L, Glicksman AS, Coachman N, Kuten A. Low acute gastrointestinal and genitourinary toxicities in whole pelvic irradiation of prostate cancer. Int J Radiat Oncol Biol Phys 1997; 38: 65-71.

28. Tait DM, Nahum AE, Rigby L, Chow M, Mayles WP, Dearnaley DP, Horwich A. Conformal radiotherapy of the pelvis: assessment of acute toxicity. Radiother Oncol 1993; 29: 117-126.

29. Zierhut D, Flentje M, Sroka-Perez G, Rudat V, Engenhart-Cabillic R, Wannenmacher M. [The conformal radiotherapy of localized prostatic carcinoma: acute tolerance and early efficacy], Strahlenther.Onkol. 1997; 173: 98-105.

30. Koper PC, Stroom JC, van Putten WL, Korevaar GA, Heijmen BJ, Wijnmaalen A, Jansen PP, Hanssens PE, Griep C, Krol AD, Samson MJ, Levendag PC. Acute morbidity reduction using 3DCRT for prostate carcinoma: a randomized study. Int J Radiat Oncol Biol Phys 1999; 43: 727-734.

31. O'Brien PC, Franklin CI, Dear KB, Hamilton CC, Poulsen M, Joseph DJ, Spry N, Denham JW. A phase III double-blind randomised study of rectal sucralfate suspension in the prevention of acute radiation proctitis. Radiother Oncol 1997; 45: 117-123.

32. Bourne RG, Kearsley JH, Grove WD, Roberts SJ. The relationship between early and late gastrointestinal complications of radiation therapy for carcinoma of the cervix. Int J Radiat Oncol Biol Phys 1983; 9: 1445-1450.

33. Dorr W, Hendry JH. Consequential late effects in normal tissues. Radiother Oncol 2001;61:223-231. 51

34. Weiss E, Hirnle P, Arnold-Bofinger H, Hess CF, Bamberg M. Therapeutic outcome and relation of acute and late side effects in the adjuvant radiotherapy of endometrial carcinoma stage I and II. Radiother Oncol 1999; 53: 37-44.

35. Richter KK, Wang J, Fagerhol MK, Hauer-Jensen M. Radiation-induced granulocyte transmigration predicts development of delayed structural changes in rat intestine. Radiother Oncol 2001; 59: 81-85.

36. Carr KE. Effects of radiation damage on intestinal morphology. Int Rev. Cytol. 2001; 208: 1-119.

37. Fajardo LF, Berthrong M, Anderson RE. Radiation Pathology. New York: Oxford University Press, 2001.

38. Haboubi NY, Schofield PF, Rowland PL. The light and electron microscopic features of early and late phase radiation-induced proctitis. Am J Gastroenterol 1988; 83: 1140-1144.

39. Hendrickson FR, Rodney Withers H. Principles of radiation oncology. In: Holleb Al, Fink DJ, Murphy GP, eds. American Cancer Society Textbook of Clinical Oncology. Atlanta: The American Cancer Society, 1991: 35-46.

40. Coia LR, Myerson RJ, Tepper JE. Late effects of radiation therapy on the gastrointestinal tract. IntJRadiat Oncol Biol Phys 1995; 31: 1213-1236.

41. Fajardo LF, Berthrong M. Vascular lesions following radiation. Pathol Annu. 1988; 23 Pt 1: 297-330.

42. Gelfand MD, Tepper M, Katz LA, Binder HJ, Yesner R, Floch MH. Acute irradiation proctitis in man. Gastroenterology 51 [3], 401 -411. 1968.

43. Sedgwick DM, Ferguson A. Dose-response studies of depletion and repopulation of rat intestinal mucosal mast cells after irradiation. Int J Radiat Biol 1994; 65: 483-495.

44. Freund U, Scholmerich J, Siems H, Kluge F, Schafer HE, Wannenmacher M. [Unwanted side-effects in using mesalazine (5-aminosalicylic acid) during radiotherapy]. Strahlenther. Onkol. 1987; 163: 678-680.

45. Northway MG, Scobey MW, Geisinger KR. Radiation proctitis in the rat. Sequential changes and effects of anti- inflammatory agents. Cancer 1988; 62: 1962-1969. 52

46. Denham JW, Hauer-Jensen M, Peters LJ. Is it time for a new formalism to categorize normal tissue radiation injury? Int J Radiat Oncol Biol Phys 2001; 50: 1105-1106.

47. Denham JW, Hauer-Jensen M. The radiotherapeutic injury~a complex 'wound'. Radiother Oncol 2002; 63: 129-145.

48. Booth C, Potten CS. Gut instincts: thoughts on intestinal epithelial stem cells. J Clin Invest 2000; 105: 1493-1499.

49. Potten CS. A comprehensive study of the radiobiological response of the murine (BDF1) small intestine. Int J Radiat Biol 1990; 58: 925-973.

50. Tamm I, Schriever F, Dorken B. Apoptosis: implications of basic research for clinical oncology. Lancet Oncol 2001,2: 33-42.

51. Paris F, Fuks Z, Kang A, Capodieci P, Juan G, Ehleiter D, Haimovitz-Friedman A, Cordon-Cardo C, Kolesnick R. Endothelial apoptosis as the primary lesion initiating intestinal radiation damage in mice. Science 2001; 293: 293-297.

52. Hauer-Jensen M., Sauer T, Berstad T, Nygaard K. Influence of pancreatic secretion on late radiation enteropathy in the rat. Acta Radiol.Oncol 1985; 24: 555-560.

53. Wang J, Zheng H, Sung CC, Hauer-Jensen M. The synthetic somatostatin analogue, octreotide, ameliorates acute and delayed intestinal radiation injury. Int J Radiat Oncol Biol Phys 1999; 45: 1289-1296.

54. Eissner G, Lindner H, Behrends U, Kolch W, Hieke A, Klauke I, Bornkamm GW, Holler E. Influence of bacterial endotoxin on radiation-induced activation of human endothelial cells in vitro and in vivo: protective role of IL-10. Transplantation 1996; 62: 819-827.

55. Zhou Q, Zhao Y, Li P, Bai X, Ruan C. Thrombomodulin as a marker of radiation- induced endothelial cell injury. Radiat Res 1992; 131: 285-289.

56. Moore KL, Andreoli SP, Esmon NL, Esmon CT, Bang NU. Endotoxin enhances tissue factor and suppresses thrombomodulin expression of human vascular endothelium in vitro. J Clin Invest 1987; 79: 124-130.

57. Nawroth PP, Handley DA, Esmon CT, Stern DM. Interleukin 1 induces endothelial cell procoagulant while suppressing cell-surface anticoagulant activity. Proc Natl Acad Sci USA 1986; 83: 3460-3464. 53

58. Ohji T, Urano H, Yamagishi M, Higashi K, Gotoh S, Karasaki Y. Transforming growth factor betal and beta2 induce down-modulation of thrombomodulin in human umbilical vein endothelial cells. Thromb.Haemost. 1995; 73: 812-818.

59. Kapiotis S, Besemer J, Bevec D, Valent P, Bettelheim P, Lechner K, Speiser W. Interleukin-4 counteracts pyrogen-induced downregulation of thrombomodulin in cultured human vascular endothelial cells. Blood 1991; 78: 410-415.

60. Seigneur M, Dufourcq P, Belloc F, Lenoble M, Renard M, Boisseau M. Influence of pentoxifylline on membrane thrombomodulin levels in endothelial cells submitted to hypoxic conditions. J Cardiovasc Pharmacol 1995; 25: S85-S87.

61. Glaser CB, Morser J, Clarke JH, Blasko E, McLean K, Kuhn I, Chang RJ, Lin JH, Vilander L, Andrews WH,. Oxidation of a specific methionine in thrombomodulin by activated neutrophil products blocks cofactor activity. A potential rapid mechanism for modulation of coagulation. J Clin Invest 1992; 90: 2565-2573.

62. Moulin F, Pearson JM, Schultze AE, Scott MA, Schwartz KA, Davis JM, Ganey PE, Roth RA. Thrombin is a distal mediator of lipopolysaccharide-induced liver injury in the rat. J Surg Res 1996; 65: 149-158.

63. DeMichele MA, Minneår FL. Modulation of vascular endothelial permeability by thrombin. Semin.Thromb.Hemost. 1992; 18: 287-295.

64. Bizios R, Lai L, Fenton JW, Malik AB. Thrombin-induced chemotaxis and aggregation of neutrophils. J Cell Physiol 1986; 128: 485-490.

65. Bar-Shavit R, Kahn A, Fenton JW, Wilner GD. Chemotactic respons of monocytes to thrombin. J Cell Biol 1983; 96: 282-285.

66. Yamabe H, Osawa H, Inuma H, Kaizuka M, Tamura N, Tsunoda S, Baba Y, Shirato K, Onodera K. Thrombin stimulates production of transforming growth factor-beta by cultured human mesangial cells. Nephrol.Dial.Transplant. 1997; 12: 438-442.

67. Dery O, Corvera CU, Steihoff M, Bunnett NW. Proteinase-activated receeptors: novel mechanisms of signaling by serine proteases. Am J Physiol 1998; 274: C1429-C1452.

68. Noda-Heiny H, Sobel BE. Vascular smooth muscle cell migration mediated by thrombin and urokinase receptor. Am J Physiol 1011; 268: C1195-C1201. 54

69. Liu WH, Chen XM, Fu B. Thrombin stimulates MMP-9 mRNA expression through AP-1 pathway in human mesangial cells. Acta Pharmacol Sin. 2000; 21: 641-645.

70. Nguyen M, Arkell J, Jackson CJ. Thrombin rapidly and efficiently activates gelatinase A in human microvascular endothelial cells via a mechanism independent of active MT1 matrix metalloproteinase. Lab Invest 1999; 79: 467-475.

71 Fava RA, Casey TT, Wilcox J, Moses HL, Nanney LB. Synthesis of transforming growth factor-beta by megakaryocytes and its localization to megakaryocyte and platelet alpha-granules. Blood 1990; 76: 1946-1955.

72. Wang J, Albertson CM, Zheng H, Fink LM, Herbert JM, Hauer-Jensen M. Short-term inhibition of ADP-induced platelet aggregation by clopidogrel ameliorates radiation- induced toxicity in rat small intestine. Thromb.Haemost. 2002; 87: 122-128.

73. Ang KK, Landuyt W, Rijnders A, van der SE. Differences in repopulation kinetics in mouse skin during split course multiple fractions per day (MFD) or daily fractionated irradiations. Int J Radiat Oncol Biol Phys 1984; 10: 95-99.

74. Turesson I, Bernefors R, Book M, Flogegard M, Hermansson I, Johansson KA, Lindh A, Sigurdardottir S, Thunberg U, Nyman J. Normal tissue response to low doses of radiotherapy assessed by molecular markers—a study of skin in patients treated for prostate cancer. Acta Oncol 2001; 40: 941-951.

75. Dorr W. Three A's of repopulation during fractionated irradiation of squamous epithelia: Assymetric loss, Acceleration of stem-cell divisions and Abortive dicisions. Int J Radiat Biol 11197; 72: 635-643.

76. Fischer L, Kimose HH, Spjeldnaes N, Wara P. Late progress of radiation-induced proctitis. Acta Chir Scand. 1990; 156: 801-805.

77. Galland RB, Spencer J. Natural history and surgical management of radiation enteritis. Br J Surg 1987; 74: 742-747.

78. Harling H, Balslev I. Long-term prognosis of patients with severe radiation enteritis. Am J Surg 1988; 155: 517-519.

79. Kimose HH, Fischer L, Spjeldnaes N, Wara P. Late radiation injury of the colon and rectum. Surgical management and outcome. Dis Colon Rectum 2003; 32: 684-689. 55

80. Regimbeau J-M, Panis Y, Gouzi J-L, Fagniez P-L. Operative and long term results after surgery for chronic radiation enteritis. Am J Surg 2003; 182: 237-242.

81. Silvain C, Besson I, Ingrand P, Beau P, Matuchansky C, Carretier M, Morichau- Beauchant M. Long-term outcome of severe radiation enteritis treated by total parenteral nutrition. Dig Dis Sci 2003; 37: 1065-1071.

82. Eickelberg O. Endless healing: TGF-beta, SMADs, and fibrosis. FEBS Lett. 2001, 506: 11-14.

83. Gruber BL, Marchese MJ, Kew RR. Transforming growth factor-beta 1 mediates mast cell chemotaxis. J Immunol 1994; 152: 5860-5867.

84. Langberg CW, Hauer-Jensen M, Sung CC, Kane CJ. Expression of fibrogenic cytokines in rat small intestine after fractionated irradiation. Radiother Oncol 1994; 32: 29-36.

85. Richter KK, Langberg CW, Sung CC, Hauer-Jensen M. Association of transforming growth factor beta (TGF-beta) immunoreactivity with specific histopathologic lesions in subacute and chronic experimental radiation enteropathy. Radiother Oncol 1996; 39:243-251.

86. Richter KK, Langberg CW, Sung CC, Hauer-Jensen M. Increased transforming growth factor beta (TGF-beta) immunoreactivity is independently associated with chronic injury in both consequential and primary radiation enteropathy. Int J Radiat Oncol Biol Phys 1997; 39: 187-195.

87. Vozenin-Brotons MC, Milliat F, Sabourin JC, de Gouville AC, Francois A, Lasser P, Morice P, Haie-Meder C, Lusinchi A, Antoun S, Bourhis J, Mathe D, Girinsky T, Aigueperse J. Fibrogenic signals in patients with radiation enteritis are associated with increased connective tissue growth factor expression. Int J Radiat Oncol Biol Phys 2003; 56: 561-572.

88. Zheng H, Wang J, Hauer-Jensen M. Role of mast cells in early and delayed radiation injury in rat intestine. Radiat Res 2000; 153: 533-539.

89. Zheng H, Wang J, Koteliansky VE, Gotwals PJ, Hauer-Jensen M. Recombinant soluble transforming growth factor beta type II receptor ameliorates radiation enteropathy in mice. Gastroenterology 2000; 119: 1286-1296. 56

90. Border W A, Okuda S, Languino LR, Sporn MB, Ruoslahti E. Suppression of experimental glomerulonephritis by antiserum against transforming growth factor beta 1 Nature 1990; 346: 371-374.

91. Border W A, Noble NA, Yamamoto T, Harper JR, Yamaguchi Y, Pierschbacher MD, Ruoslahti E. Natural inhibitor of transforming growth factor-beta protects against scarring in experimental kidney disease. Nature 1992; 360: 361-364.

92. Ferguson MW. Skin wound healing: transforming growth factor beta antagonists decrease scarring and improve quality. J Interferon Res 1994; 14: 303-304.

93. Vozenin-Brotons MC, Sivan V, Gault N, Renard C, Geffrotin C, Delanian S, Lefaix JL, Martin M. Antifibrotic action of Cu/Zn SOD is mediated by TGF-betal repression and phenotypic reversion of myofibroblasts. Free Radic.Biol Med 2001,30: 30-42.

94. Delanian S, Martin M, Bravard A, Luccioni C, Lefaix JL. Cu/Zn superoxide dismutase modulates phenotypic changes in cultured fibroblasts from human skin with chronic radiotherapy damage. Radiother Oncol 2001; 58: 325-331.

95. Delanian S, Balla-Mekias S, Lefaix JL. Striking regression of chronic radiotherapy damage in a clinical trial of combined pentoxifylline and tocopherol. J Clin Oncol 1999; 17: 3283-3290.

96. Qi Z, Atsuchi N, Ooshima A, Takeshita A, Ueno H. Blockade of type beta transforming growth factor signaling prevents liver fibrosis and dysfunction in the rat. Proc.Natl AcadSci US,A 1999; 96: 2345-2349.

97. Martin M, Lefaix J, Delanian S. TGF-betal and radiation fibrosis: a master switch and a specific therapeutic target? Int J Radiat Oncol Biol Phys 2000; 47: 277-290.

98. Richter KK, Fink LM, Hughes BM, Sung CC, Hauer-Jensen M. Is the loss of endothelial thrombomodulin involved in the mechanism of chronicity in late radiation enteropathy? Radiother Oncol 1997; 44: 65-71. 99. Tucker SL, Geara FB, Peters LJ, Brock WA. How much could the radiotherapy dose be altered for individual patients based on a predictive assay of normal-tissue radiosensitivity? Radiother Oncol 1996; 38: 103-113.

100. Alapetite C, Cosset JM, Bourguignon MH, Masse R. Genetic susceptibility to radiations. Which impact on medical practice? Q JNucl.Med 2000; 44: 347-354. 57

101. Brock W A, Tucker SL, Geara FB, Turesson I, Wike J, Nyman J, Peters LJ. Fibroblast radiosensitivity versus acute and late normal skin responses in patients treated for breast cancer. Int J Radiat Oncol Biol Phys 1995; 32: 1371-1379.

102. Burnet NG, Nyman J, Turesson I, Wurm R, Yarnold JR, Peacock JH. Prediction of normal-tissue tolerance to radiotherapy from in-vitro cellular radiation sensitivity. Lancet 1992; 339: 1570-1571.

103. Geara FB, Peters LJ, Ang KK, Wike JL, Brock WA. Prospective comparison of in vitro normal cell radiosensitivity and normal tissue reactions in radiotherapy patients. Int J Radiat Oncol Biol Phys 1993;27: 1173-1179.

104. Johansen J, Bentzen SM, Overgaard J, Overgaard M. Evidence for a positive correlation between in vitro radiosensitivity of normal human skin fibroblasts and the occurrence of subcutaneous fibrosis after radiotherapy. Int J Radiat Biol 1994; 66: 407-412.

105. Johansen J, Bentzen SM, Overgaard J, Overgaard M. Relationship between the in vitro radiosensitivity of skin fibroblasts and the expression of subcutaneous fibrosis, telangiectasia, and skin erythema after radiotherapy. Radiother Oncol 1996; 40: 101- 109.

106. Herold DM, Hanlon AL, Hanks GE. Diabetes mellitus: a predictor for late radiation morbidity. Int.,J.Radiat. Oncol.Biol.Phys. 1999; 43: 475-479.

107. Potish RA. Importance of predisposing factors in the development of enteric damage. Am J Clin Oncol 1982; 5: 189-194.

108. Schultheiss TE, Lee WR, Hunt MA, Hanlon AL, Peter RS, Hanks GE. Late GI and GU complications in the treatment of prostate cancer. Int J Radiat Oncol Biol Phys 1997; 37: 3-11.

109. Skwarchuk MW, Jackson A, Zelefsky MJ, Venkatraman ES, Cowen DM, Levegrun S, Burman CM, Fuks Z, Leibel SA, Ling CC. Late rectal toxicity after conformal radiotherapy of prostate cancer (I): multivariate analysis and dose-response. Int J Radiat Oncol Biol Phys 2000; 47: 103-113.

110. Fine BA, Hempling RE, Piver MS, Baker TR, McAuley M, Driscoll D. Severe radiation morbidity in carcinoma of the cervix: impact of pretherapy surgical staging and previous surgery. Int J Radiat Oncol Biol Phys 1995; 31: 717-723. 58

111. Potish RA, Dusenbery KE. Enteric morbidity of postoperative pelvic external beam and brachytherapy for uterine cancer. Int J Radiat Oncol Biol Phys 1990; 18: 1005- 1010.

112. Willett CG, Ooi CJ, Zietman AL, Menon V, Goldberg S, Sands BE, Podolsky DK. Acute and late toxicity of patients with inflammatory bowel disease undergoing irradiation for abdominal and pelvic neoplasms. Int J Radiat Oncol Biol Phys 2000; 46: 995-998.

113. Eifel PJ, Jhingran A, Bodurka DC, Levenback C, Thames H. Correlation of smoking history and other patient characteristics with major complications of pelvic radiation therapy for cervical cancer. J Clin Oncol 2002; 20: 3651-3657.

114. Kucera H, Enzelsberger H, Eppel W, Weghaupt K. The influence of nicotine abuse and diabetes mellitus on the results of primary irradiation in the treatment of carcinoma of the cervix. Cancer 1987; 60: 1-4.

115. Gatti RA. The inherited basis of human radiosensitivity. Acta Oncol 2001; 40: 702- 711.

116. Hall EJ, Schiff PB, Hanks GE, Brenner DJ, Russo J, Chen J, Sawant SG, Pandita TK. A preliminary report: frequency of A-T heterozygotes among prostate cancer patients with severe late responses to radiation therapy. Cancer JSci Am 1998; 4: 385-389.

117. Hall EJ. Do no harm. Normal tissue effects. Acta Oncologica 40[8], 913-916. 2003.

118. Stewart GS, Maser RS, Stankovic T, Bressan DA, Kaplan MI, Jaspers NG, Raams A, Byrd PJ, Petrini JH, Taylor AM. The DNA double-strand break repair gene hMREl 1 is mutated in individuals with an ataxia-telangiectasia-like disorder. Cell 1999; 99: 577-587.

119. Higurashi M, Conen PE. In vitro chromosomal radiosensitivity in patients and in carriers with abnormal non-Down's syndrome karyotypes.

120. Li C, Wilson PB, Levine E, Barber J, Stewart AL, Kumar S. TGF-betal levels in pre- treatment plasma identify breast cancer patients at risk of developing post- radiotherapy fibrosis. Int J Cancer 1999; 84: 155-159.

121 Wang CJ, Leung SW, Chen HC, Sun LM, Fang FM, Huang EY, Hsiung CY, Changchien CC. The correlation of acute toxicity and late rectal injury in radiotherapy for cervical carcinoma: evidence suggestive of consequential late effect (CQLE). Int J Radiat Oncol Biol Phys 1998; 40: 85-91. 59

122. Denham JW, Hauer-Jensen M, Kron T, Langberg CW. Treatment-time-dependence models of early and delayed radiation injury in rat small intestine. Int J Radiat Oncol Biol Phys 2000; 48: 871-887.

123. Prasad KN, Cole WC, Kumar B, Che PK. Pros and cons of antioxidant use during radiation therapy. Cancer Treat.Rev 2002; 28: 79-91.

124. DeLaney TF, Afridi N, Taghian AG, Sanders DA, Fuleihan NS, Faller DV, Nogueira CP. 13-cis-retinoic acid with alpha-2a-interferon enhances radiation cytotoxicity in head and neck squamous cell carcinoma in vitro. Cancer Res 1996; 56: 2277-2280.

125. Blumenthal RD, Lew W, Reising A, Soyne D, Osorio L, Ying Z, Goldenberg DM. Anti-oxidant vitamins reduce normal tissue toxicity induced by radio- immunotherapy. Int J Cancer 2000; 86: 276-280.

126. Israel K, Yu W, Sanders BG, Kline K. Vitamin E succinate induces apoptosis in human prostate cancer cells: role for Fas in vitamin E succinate-triggered apoptosis. Nutr Cancer 2000; 36: 90-100.

127. Yu W, Israel K, Liao QY, Aldaz CM, Sanders BG, Kline K. Vitamin E succinate (VES) induces Fas sensitivity in human breast cancer cells: role for Mr 43,000 Fas in VES-triggered apoptosis. Cancer Res 1999; 59: 953-961.

128. Prasad KN, Kumar R. Effect of individual and multiple antioxidant vitamins on growth and morphology of human nontumorigenic and tumorigenic parotid acinar cells in culture. Nutr Cancer 1996; 26: 11-19.

129. Capizzi RL. The preclinical basis for broad-spectrum selective cytoprotection of normal tissues from cytotoxic therapies by amifostine. Semin.Oncol 1999; 26: 3-21.

130. Cassatt DR, Fazenbaker CA, Bachy CM, Hanson MS. Preclinical modeling of improved amifostine (Ethyol) use in radiation therapy. Semin.Radiat Oncol 2002; 12: 97-102.

131. Murray D, Rosenberg E, Allalunis-Turner MJ. Protection of human tumor cells of differing radiosensitivity by WR- 1065. Radiat Res 2000; 154: 159-162.

132. I to H, Meistrich ML, Barkley HT, Jr., Thames HD, Jr., Milas L. Protection of acute and late radiation damage of the gastrointestinal tract by WR-2721 Int J Radiat Oncol Biol Phys 1986; 12:211-219. 60

133. Murray D, Altschuler EM, Hunter N, Milas L. Protection by WR-3689 against gamma-ray-induced intestinal damage: comparative effect on clonogenic cell survival, mouse survival, and DNA damage. Radiat Res 1989; 120: 339-351.

134. lindegaard jc, Grau C. Has the outlook improved for amifostine as a clinical radioprotector? Radiother Oncol 2000; 57: 113-118.

135. lindegaard jc. Has the time come for routine use of amifostine in clinical radiotherapy practice? Acta Oncologica 2003; 42: 2-3.

136. Windmueller HG, Spaeth AE. Respiratory fuels and nitrogen metabolism in vivo in small intestine of fed rats. Quantitative importance of glutamine, glutamate, and aspartate. J Biol Chem 1980; 255: 107-112.

137. Klimberg VS, McLellan JL. Glutamine, cancer, and its therapy. Am J Surg 1996; 172: 418-424.

138. Jensen JC, Schaefer R, Nwokedi E, Bevans DW, III, Baker ML, Pappas AA, Westbrook KC, Klimberg VS. Prevention of chronic radiation enteropathy by dietary glutamine. Ann Surg Oncol 1994; 1: 157-163.

139. Klimberg VS, Souba WW, Dolson DJ, Salloum RM, Hautamaki RD, Plumley DA, Mendenhall WM, Bova FJ, Khan SR, Hackett RL, Prophylactic glutamine protects the intestinal mucosa from radiation injury. Cancer 1990; 66: 62-68.

140. Klimberg VS, Salloum RM, Kasper M, Plumley DA, Dolson DJ, Hautamaki RD, Mendenhall WR, Bova FC, Bland KI, Copeland EM, III,. Oral glutamine accelerates healing of the small intestine and improves outcome after whole abdominal radiation. Arch.Surg 1990; 125: 1040-1045.

141. Kozelsky TF, Meyers GE, Sloan JA, Shanahan TG, Dick SJ, Moore RL, Engeler GP, Frank AR McKone TK, Urias RE, Pilepich MV, Novotny PJ, Martenson JA. Phase III double-blind study of glutamine versus placebo for the prevention of acute diarrhea in patients receiving pelvic radiation therapy. J Clin Oncol 2003; 21: 1669-1674.

142. Henriksson R, Franzen L, Littbrand B. Prevention of irradiation-induced bowel discomfort by sucralfate: a double-blind, placebo-controlled study when treating localized pelvic cancer. Am J Med 1991,91: 151S-157S.

143. Jensen SL, Funch JP. Role of sucralfate in peptic disease. Dig. Dis 1992; 10: 153-161. 61

144. Konturek SJ, Brzozowski T, Majka J, Szlachcic A, Bielanski W, Stachura J, Otto W. Fibroblast growth factor in gastroprotection and ulcer healing: interaction with sucralfate. Gut 1993; 34: 881-887.

145. Lucey MR, Park J, DelValle J, Wang LD, Yamada T. Sucrose octasulfate stimulates gastric somatostatin release. Am J Med 1991, 91: 52S-57S.

146. Romano M, Razandi M, Ivey KJ. Effect of sucralfate and its components on taurocholate-induced damage to rat gastric mucosal cells in tissue culture. Dig. Dis Sci 1990; 35: 467-476.

147. Sandor Z, Nagata M, Kusstatscher S, Szabo S. Stimulation of mucosal glutathione and angiogenesis: new mechanisms of gastroprotection and ulcer healing by sucralfate. Scand J Gastroenterol Suppl 1995; 210: 19-21.

148. Martenson JA, Bollinger JW, Sloan JA, Novotny PJ, Urias RE, Michalak JC, Shanahan TG, Mailliard JA, Levitt R. Sucralfate in the prevention of treatment- induced diarrhea in patients receiving pelvic radiation therapy: A North Central Cancer Treatment Group phase III double-blind placebo-controlled trial. J Clin Oncol 2000; 18: 1239-1245.

149. Kochhar R, Sriram PV, Sharma SC, Goel RC, Patel F. Natural history of late radiation proctosigmoiditis treated with topical sucralfate suspension. Dig Dis Sci 1999; 44: 973-978.

150. Hanauer SB, Robinson M, Pruitt R, Lazenby AJ, Persson T, Nilsson LG, Walton- Bowen K, Haskell LP, Levine JG. Budesonide enema for the treatment of active, distal ulcerative colitis and proctitis: a dose-ranging study. U.S. Budesonide enema study group. Gastroenterology 1998; 115: 525-532.

151. Widmark A, Fransson P, Franzen L, Littbrand B, Henriksson R. Daily-diary evaluated side-effects of conformal versus conventional prostatic cancer radiotherapy technique. Acta Oncol 1997; 36: 499-507.

152. Makrauer FL, Oates E, Becker J, Abrams R, O'Connor T, McCallum R, Shumaker J. Does local irradiation affect gastric emptying in humans? Am JMed Sci 1999; 317: 33-37.

153. Yeoh E, Horowitz M, Russo A, Muecke T, Robb T, Maddox A, Chatterton B. Effect of pelvic irradiation on gastrointestinal function: a prospective longitudinal study. Am JMed 1993; 95: 397-406. 62

154. Summers RW, Flatt AJ, Prihoda MJ, Mitros FA. Effect of irradiation on morphology and motility of canine small intestine. Dig.Dis Sei 1987; 32: 1402-1410.

155. Erickson BA, Otterson MF, Moulder JE, Sarna SK. Altered motility causes the early gastrointestinal toxicity of irradiation. Int J Radiat Oncol Biol Phys 1994; 28: 905- 912.

156. Otterson MF, Sarna SK, Leming SC, Moulder JE, Fink JG. Effects of fractionated doses of ionizing radiation on colonic motor activity. Am J Physiol 1992; 263: G518- G526.

157. Bremer K. 5-Hydroxytryptamine (serotonin) subtype 3 antagonists, a major step in prophylaxis and control of cytostatic and radiation-induced emesis. J Cancer Res Clin Oncol 1991; 117: 85-87.

158. Logue JP, Magee B, Hunter RD, Murdoch RD. The antiemetic effect of granisetron in lower hemibody radiotherapy. Clin Oncol (R Coll Radiol.) 1991; 3: 247-249.

159. Costa M, Brookes S J, Hennig GW. Anatomy and physiology of the enteric nervous system. Gut 2000; 47 Suppl4: ivl5-ivl9.

160. Fransson P, Widmark A. Late side effects unchanged 4-8 years after radiotherapy for prostate cancer: a comparison with age-matched controls. Cancer 2003; 85: 678-688.

161 Denekamp J. Changes in the rate of repopulation during multifraction irradiation of mouse skin. Br J Radiol. 1973; 46: 381-387.

162. Morris GM. Review article: effects of radiation on the cell proliferation kinetics of epithelial tissues-therapeutic implications. Br J Radiol. 1996; 69: 795-803.

163. Carratu R Secondulfo M, de Magistris L, Daniele B, Pignata S, D'Agostino L, Frezza P, Elmo M, Silvestro G, Sasso FS. Assessment of small intestinal damage in patients treated with pelvic radiotherapy. Oncol Rep 1998; 5: 635-639.

164. Schmidt-Ullric RK, Mikkelsen RB, Dent P, Todd DG, Valerie K, Kavanagh BD, Contessa JN, Rorrer WK, Chen PB. Radiation-induced proliferation of the human A431 squamous carcinoma cells is dependent on EGFR tyrosine phosphorylation. Oncogene 1997; 15: 1191-1197. 63

165. Schmidt-Ullric RK, Dent P, Grant S, Mikkelsen RB, Valerie K. Signal transduction and cellular radiation responses. Radiat Res 2000; 153: 245-257.

166. Schmidt-Ullric RK, Contessa JN, Dent P, Mikkelsen RB, Valerie K, Reardon DB, Bowers G, Lin PS. Molecular mechanisms of radiation-induced accelerated repopulation. Radiat Oncol Investig 1999; 7: 321-330.

167. Leupin N, Curschmann J, Kranzbuhler H, Maurer CA, Laissue JA, Mazzucchelli L. Acute radiation colitis in patients treated with short-term preoperative radiotherapy for rectal cancer. Am JSurg Pathol 2002; 26: 498-504.

168. Walsh RE, Gaginella TS. The eosinophil in inflammatory bowel disease. Scand J Gastroenterol 1991; 26: 1217-1224.

169. Sarin SK, Malhotra V, Sen GS, Karol A, Gaur SK, Anand BS. Significance of eosinophil and mast cell counts in rectal mucosa in ulcerative colitis. A prospective controlled study. Dig Dis Sci 1987; 32: 363-367.

170. Hogan SP, Mishra A, Brandt EB, Royalty MP, Pope SM, Zimmermann N, Foster PS, Rothenberg ME. A pathological function for eotaxin and eosinophils in eosinophilic gastrointestinal inflammation. Nat.Immunol 2001, 2: 353-360.

171 Kita H. The eosinophil: A cytokine-producing cell? J Allergy Clin Immunol 1996; 97: 889-892.

172. Rothenberg ME. Eosinophilia. N Engl JMed 1998; 338: 1592-1600.

173. Gearing AJ, Beckett P, Christodoulou M, Churchill M, Clements J, Davidson AH, Drummond AH, Galloway WA, Gilbert R Gordon JL,. Processing of tumour necrosis factor-alpha precursor by metalloproteinases. Nature 1994; 370: 555-557.

174. McGeehan GM, Becherer JD, Bast RC, Jr., Boyer CM, Champion B, Connolly KM, Conway JG, Furdon P, Karp S, Kidao S,. Regulation of tumour necrosis factor-alpha processing by a metalloproteinase inhibitor. Nature 1994; 370: 558-561.

175. Pekovich SR, Bock PE, Hoover RL. Thrombin-thrombomodulin activation of protein C facilitates the activation of progelatinase A. FEBS Lett. 2001; 494: 129-132.

176. Canney PA, Dean S. Transforming growth factor beta: a promotor of late connective tissue injury following radiotherapy? Br J Radiol. 1990; 63: 620-623. 64

177. Wang J, Zheng H, Sung CC, Richter KK, Hauer-Jensen M. Cellular sources of transforming growth factor-beta isoforms in early and chronic radiation enteropathy. Am J Pathol 1998; 153: 1531-1540.

178. Durko M, Brodt P. The metalloproteinases and their inhibitors. In: Brodt P, ed. Cell adhesion and invasion in cancer metastasis. New York: Landes Company, 1996: 113- 150.

179. Visse R, Nagase H. Matrix metalloproteinases and tissue inhibitors of metalloproteinases. Circ Res 2003; 92: 827-839.

180. Van den Steen PE, Opdenakker G, Wormald MR, Dwek RA, Rudd PM. Matrix remodelling enzymes, the protease cascade and glycosylation. Biochim Biophys Acta 2001; 1528: 61-73.

181. Opdenakker G, Van den Steen PE, Dubois B, Nelissen I, Van Coillie E, Masure S, Proost P, Van Damme J. Gelatinase B functions as regulator and effector in leukocyte biology. JLeukoc.Biol 2001, 69: 851-859.

182. Chambers AF, Matrisian LM. Changing views of the role of matrix metalloproteinases in metastasis. J Natl Cancer Inst 1997; 89: 1260-1270.

183. Murray GI. Matrix metalloproteinases: a multifunctional group of molecules. J Pathol 2001; 195: 135-137.

184. McCawley LJ, Matrisian LM. Matrix metalloproteinases: they're not just for matrix alone anymore! Curr Opin.Cell Biol 2001; 13: 534-540.

185. Sawicki G, Salas E, Murat J, Miszta-Lane H, Radomski MW. Release of gelatinase A during platelet activation mediates aggregation. Nature 1997; 386: 616-619.

186. Yu Q, Stamenkovic I. Cell surface-localized matrix metalloproteinase-9 proteolytically activates TGF-beta and promotes tumor invasion and angiogenesis. Genes Dev 1012; 14: 163-176.

187. Imai K, Hiramatsu A, Fukushima D, Pierschbacher MD, Okada Y. Degradation of decorin by matrix metalloproteinases: identification of the cleavage sites, kinetic analyses and transforming growth factor-betal release. Biochem J1997; 322: 809- 814. 65

188. McQuibban GA, Gong JH, Tam EM, McCulloch CA, Clark-Lewis I, Overall CM. Inflammation dampened by gelatinase A cleavage of monocyte chemoattractant protein-3. Science 2000; 289: 1202-1206.

189. Vaday GG, Schor H, Rahat MA, Lahat N, Lider O. Transforming growth factor-beta suppresses tumor necrosis factor alpha- induced matrix metalloproteinase-9 expression in monocytes. JLeukoc.Biol 2001; 69: 613-621.

190. Ricke WA, Smith GW, Smith MF. Matrix metalloproteinase expression and activity following prostaglandin F(2 alpha)-induced luteolysis. Biol Reprod. 2002; 66: 685- 691.

191 Khan AM, Birk JW, Anderson JC, Georgsson M, Park TL, Smith CJ, Comer GM. A prospective randomized placebo-controlled double-blinded pilot study of misoprostol rectal suppositories in the prevention of acute and chronic radiation proctitis symptoms in prostate cancer patients. Am J Gastroenterol 2000; 95: 1961-1966.

192. Di Sebastiano P, di Mola FF, Artese L, Rossi C, Mascetta G, Pernthaler H, Innocenti P. Beneficial effects of Batimastat (BB-94), a matrix metalloproteinase inhibitor, in rat experimental colitis. Digestion 2001; 63: 234-239.

193. Sykes AP, Bhogal R, Brampton C, Chander C, Whelan C, Parsons ME, Bird J. The effect of an inhibitor of matrix metalloproteinases on colonic inflammation in a trinitrobenzenesulphonic acid rat model of inflammatory bowel disease. Aliment Pharmacol Ther 1999; 13: 1535-1542.

194. Lockhart AC, Braun RD, Yu D, Ross JR, Dewhirst MW, Humphrey JS, Thompson S, Williams KM, Klitzman B, Yuan F, Grichnik JM, Proia AD, Conway DA, Hurwitz HI. Reduction of wound angiogenesis in patients treated with BMS-275291, a broad spectrum matrix metalloproteinase inhibitor. Clin Cancer Res 2003; 9: 586-593.

195. Gul YA, Prasannan S, Jabar FM, Shaker AR, Moissinac K. Pharmacotherapy for chronic hemorrhagic radiation proctitis. World J Surg 2002; 26: 1499-1502.

196. Kochhar R, Sharma SC0BB, Mehta SK. Rectal sucralfate in radiation proctitis. Lancet 1988; 400.

197. Ladas SD, Raptis SA. Sucralfate enemas in the treatment of chronic postradiation proctitis. Am J Gastroenterol 1989; 84: 1587-1589.

198. Melko GP, Turco TF, Phelan TF, Sauers NM. Treatment of radiation-induced prodtitis with sucralfate enemas. Ann Pharmacother 1999; 33: 1274-1276. 66

199. Sasai T, Hiraishi H, Suzuki Y, Masuyama H, Ishida M, Terano A. Treatment of chronic post-radiation proctitis with oral administration of sucralfate. Am J Gastroenterol 1998; 93: 1593-1595.

200. Stockdale AD, Biswas A. Long-term control of radiation proctitis following treatment with sucralfate enemas. Br J Surg 1996; 84: 379.

201. Henriksson R, Franzen L, Littbrand B. Prevention and therapy of radiation-induced bowel discomfort. Scand J Gastroenterol Suppl 1992; 191: 7-11.

202. O'Brien PC, Franklin CI, Poulsen MG, Joseph DJ, Spry NS, Denham JW. Acute symptoms, not rectally administered sucralfate, predict for late radiation proctitis: longer term follow-up of a phase III trial-Trans- Tasman Radiation Oncology Group. Int.J Radiat Oncol Biol Phys 2002; 54: 442-449.

203. Tack J. Receptors of the enteric nervous system: potential targets for drug therapy. Gut 2000; 47: iv20-iv22.

204. Grundy D. The intestinal mucosa as a target and trigger for enteric reflexes. Gut 2000; 47: iv44-iv45.

205. Scarantino CW, Ornitz RD, Hoffman LG, Anderson RF. On the mechanism of radiation-induced emesis: the role of serotonin. Int J Radiat Oncol Biol Phys 1994; 30: 825-830.

206. Feyer PC, Stewart AL, Titlbach OJ. Aetiology and prevention of emesis induced by radiotherapy. Support Care Cancer 1998; 6: 253-260.

207. Abbott B, Ippoliti C, Bruton J, Neumann J, Whaley R, Champlin R. Antiemetic efficacy of granisetron plus dexamethasone in bone marrow transplant patients receiving chemotherapy and total body irradiation. Bone Marrow Transplant 1999; 23: 265-269.

208. Scarantino CW, Ornitz RD, Hoffman LG, Anderson RF. Radiation-induced emesis: effects of ondansetron. Semin. Oncol 1992; 19: 38-43.

209. Fernandez-Banares F, Villa S, Esteve M, Roca M, Cabre E, Abad-Lacruz A, Martin- Comin J, Gassull MA. Acute effects of abdominopelvic irradiation on the orocecal transit time: its relation to clinical symptoms, and bile salt and lactose malabsorption. Am J Gastroenterol 1991; 86: 1771-1777. 67

210. Dietrich CF, Brunner V, Seifert H, Schreiber-Dietrich D, Caspary WF, Lembcke B. Intestinale B-Bild Sonographic bei Patienten mit einheimischer Sprue. Ultraschall in Med 1999; 20: 242-247.

211 Rettenbacher T, Hollerweger A, Macheiner P, Huber S, Gritzmann N. Adult celiac disease: US signs. Radiology 1999; 211: 389-394.

212. Coussens LM, Fingleton B, Matrisian LM. Matrix metalloproteinase inhibitors and cancer: trials and tribulations. Science 2002; 295: 2387-2392.

213. Hauer-Jensen M, Wang J, Denham JW. Mechanisms and modification of the radiation response of gastrointestinal organs. In: Nieder C, Milas L, Ang KK, eds. Modification of radiation response. Berlin, Heidelberg, New York: Springer-Verlag, 2003: 49-72.

214. Wang J, Zheng H, Hauer-Jensen M. Influence of Short-Term Octreotide Administration on Chronic Tissue Injury, Transforming Growth Factor beta (TGF- beta) Overexpression, and Collagen Accumulation in Irradiated Rat Intestine. J Pharmacol Exp Ther 2001,297: 35-42.

215. Gal D, Strickland DM, Lifshitz S, Buchsbaum HJ, Mitchell MD. Effect of radiation on prostaglandin production by human bowel in vitro. Int J Radiat Oncol Biol Phys 1984; 10: 653-657.

216. Hanson WR, DeLaurentiis K. Comparison of in vitro murine intestinal radiation protection by E-prostaglandins. Prostaglandins 1987; 33: 93-104.

217. Arslan G, Brunborg LA, Frøyland L, Brun JG, Valen M, Berstad A. Effects of duodenal seal oil administration in patients with inflammatory bowel disease. Lipids 2002; 37: 935-940.

218. Almallah YZ, El-Tahir A, Heys SD, Richardson S, Eremin O. Distal procto-colitis and n-3 polyunsaturated fatty acids: the mechanism(s) of natural cytotoxicity inhibition. Eur J Clin Invest 2000; 30: 58-65.

219. Almallah YZ, Ewen SW, El-Tahir A, Mowat NA, Brunt PW, Sinclair-Heys SD, Eremin O. Distal procto-colitis and n-3 polyunsaturated fatty acids (n-3 PUFAs): the mucosal effect in situ. J Clin Immunol 2000; 20: 68-76.

220. Ullrich A, Schlessinger J. Signal transduction by receptors with tyrosine kinase activity. Cell 1990; 61: 203-212. 68

221. Potters L, Steinberg M, Wallner P, Hevezi J. How one defines intensity-modulated radiation therapy. Int J Radiat Oncol Biol Phys 2003; 56: 609-610.

222. Teh BS, Woo SY, Butler EB. Intensity modulated radiation therapy (IMRT): a new promising technology in radiation oncology. Oncologist. 1999; 4: 433-442.

223. Webb S. Advances in three-dimensional conformal radiation therapy physics with intensity modulation. Lancet Oncol 2000; 1: 30-36.

224. Mohan DS, Kupelian PA, Willoughby TR. Short-course intensity-modulated radiotherapy for localized prostate cancer with daily transabdominal ultrasound localization of the prostate gland. Int J Radiat Oncol Biol Phys 2000; 46: 575-580.

225. Zelefsky MJ, Fuks Z, Happersett L, Lee HJ, Ling CC, Burman CM, Hunt M, Wolfe T, Venkatraman ES, Jackson A, Skwarchuk M, Leibel SA. Clinical experience with intensity modulated radiation therapy (IMRT) in prostate cancer. Radiother Oncol 2000; 55: 241-249.

226. Zelefsky MJ, Fuks Z, Hunt M, Lee HJ, Lombardi D, Ling CC, Reuter VE, Venkatraman ES, Leibel SA. High dose radiation delivered by intensity modulated conformal radiotherapy improves the outcome of localized prostate cancer. J Urol. 2001; 166: 876-881.

227. Zelefsky MJ, Fuks Z, Hunt M, Yamada Y, Marion C, Ling CC, Amols H, Venkatraman ES, Leibel SA. High-dose intensity modulated radiation therapy for prostate cancer: early toxicity and biochemical outcome in 772 patients. Int J Radiat Oncol Biol Phys 2002; 53: 1111-1116.

228. Fransson P, Bergstrom P, Lofroth PO, Franzen L, Henriksson R, Widmark A. Daily- diary evaluated side effects of dose-escalation radiotherapy of prostate cancer using the stereotactic Beam-Cath technique. Acta Oncol 2003; 42: 326-333. Paper I

tat J. Radiation Oncology Biol Phys.. Vol. 48. No. 4. pp. 1111-1117, 2000 Copyright © 2000 Elsevier Science Inc. Printed in the USA- All rights reserved jTIOl 0360-3016/00/S-see front matter ELSEVIER Pn S0360-3016f00)00744-6

CLINICAL INVESTIGATION Normal Tissue

ACUTE RADIATION PROCTITIS: A SEQUENTIAL CLINICOPATHOLOGIC STUDY DURING PELVIC RADIOTHERAPY

NILS HOVDENAK, M.D.,* LUIS F. FAJARDO, M.D.,+ AND MARTIN HAUER-JENSEN, M.D., PH.D.* *Department of Oncology, University of Bergen, Bergen, Norway; department of Pathology, Stanford University School of Medicine and Veterans Healthcare System, Palo Alto, CA; department of Surgery, University of Arkansas for Medical Sciences and Central Arkansas Veterans Healthcare System, Little Rock, AR Purpose: Rectal toxicity is often dose limiting during pelvic radiation therapy. This prospective study examined the sequential development and associations of clinical, endoscopic, and histopathologic rectal toxicity during ongoing radiation therapy. Methods and Materials: Thirty-three patients with nongastrointestinal pelvic carcinomas underwent proctoscopy with biopsy before radiation therapy, after 2 weeks treatment, and toward the end of the treatment course (6 weeks). Symptoms of acute toxicity were recorded, and endoscopic changes were graded. Histologic changes in the surface epithelium, glandular layer, and lamina propria were assessed using an ad hoc scoring system. Macrophage accumulation was evaluated in anti-CD68 stained sections. Results: Pretreatment endoscopy and biopsies were unremarkable. Clinical symptoms progressed toward the end of the treatment course. In contrast, endoscopic pathology was maximal at 2 weeks. Biopsies obtained during treatment exhibited atrophy of the surface epithelium, acute cryptitis, crypt abscesses, crypt distortion and atrophy, and stromal inflammation. Histologic changes, particularly those in the surface epithelium, were consistently more pronounced at 2 weeks than they were at 6 weeks. Conclusion: In contrast to clinical symptoms, endoscopic changes stabilize and histologic changes regress from the 2nd to the 6th week of treatment. These results may have implications for the design and timing of prophylactic and therapeutic interventions to reduce radiation proctitis. © 2000 Elsevier Science Inc.

Radiation iqjury, Radiation proctitis, Histopathology, Rectum, Human, Macrophage.

INTRODUCTION The underlying mechanisms of delayed radiation proctitis and optimal timing of preventive and therapeutic inter- The small and large intestine are dose-limiting organs dur- ventions are not well defined. ing pelvic and abdominal radiation therapy. The small Some histologic features of acute colorectal radiation bowel may be displaced partially or completely outside the injury have been described (5-7). However, because of the radiation field by patient positioning (1), special treatment paucity of sequential studies in humans, little is known tables (2), or surgical abdominopelvic partitioning with about the time course of development of individual changes omentum (3) or absorbable mesh (4). In contrast, the rectum during treatment and the association between histopathol- is fixed in position and unavoidably exposed during irradi- ogy and endoscopic and clinical signs of toxicity. The ation of pelvic tumors. present prospective study used objective, quantitative meth- The incidence of both early and delayed intestinal ods to characterize the progression of histologic changes in radiation toxicity depends on radiation dose, treatment the rectal mucosa during pelvic radiotherapy, and to assess time, and fractionation. Total doses of less than 40 Gy how these changes reflect radiation toxicity at the endo- seldom cause clinically significant symptoms. However, scopic and clinical level. nonalimentary pelvic tumors are usually treated to doses in the range of 50-70 Gy, which entail significant acute morbidity, including diarrhea, abdominal cramps, and hematochesia. In some patients, these toxicities may re- METHODS AND MATERIALS quire interruption of therapy or other modifications that Patients and radiation therapy forestall optimal completion of the original treatment The study comprised 31 men and 2 women who under- plan. Some patients develop symptoms of chronic injury, went radiation therapy without concomitant chemotherapy usually after a latency period of several months to years. for nonalimentary pelvic carcinomas (prostate, 27; bladder,

Reprint requests to: Dr. Martin Hauer-Jensen, Arkansas Cancer Accepted for publication: 6 June 2000. Research Center, 4301 West Markham, Little Rock, AR 72205. E-mail: [email protected] 1112 L J. Radiation Oncology • Biology • Physics Volume 48, Number 4, 2000

4; endometrium, 2) at the Department of Oncology, Uni- Table 1. Grading of histologic changcs was based on versity Hospital, Bergen, from 1994 to 1995. Patients with microscopic examination of all available sections of each patient a past or present history of significant gastrointestinal dis- sample ease were excluded. All patients underwent external beam Structure Characteristic radiation therapy with individual treatment planning using 6- to 18-MV photons to a total tumor dose of 54 - 66 Gy Surface epithelium Loss of cellular height/flattening of cells over 6-7 weeks, administered as once-daily 2.0-Gy frac- Cellular inflammatory infiltrate Glands (crypts) Lumcnal migration of epithelial nuclei tions (1.75 Gy per fraction in 1 patient) with weekend Loss of goblet cells breaks. A 4-field box technique with appropriate custom- Mitotic activity* ized blocking was applied. The present series was part of a Cryptitis (migration of segmented randomized, placebo-controlled trial using sucralfate for the neutrophils through the crypt wall) possible amelioration of acute radiation-induced bowel Eosinophilic crypt abscesses* symptoms. All patients gave informed written consent to the Loss of glands Atrophy of glands* study, which was approved by the University of Bergen Distortion of glands* Institutional Review Board (IRB) for human studies. Lamina propria Inflammation stroma Edema Congestion of blood vessels

Endoscopy, clinical symptoms, and quantitative histology The individual abnormalities weTc assessed as norma) (score = Rigid proctoscopy was performed immediately before 0) or abnormal, ranked according to severity and arranged in RT, at the end of the 2nd week of treatment, and at the end quartiles: 0 = normal; 1 = mildly abnormal; 2 = moderately of treatment (6 weeks). At each time point, the rectal mu- abnormal; 3 = markedly abnormal; 4 = severely abnormal. * Mitotic activity was graded as normal, increased, or de- cosa was examined macroscopically, changes were re- creased, as compared with controls of approximately similar age. corded, and 2- to 3-cup forceps biopsies were obtained from * Parameters were recorded as present or absent. the posterior rectal wall, 10 cm from the anal verge. The endoscopic appearance was assessed according to Hanauer el al. (8). This scoring system uses a 4-point scale, in which 0 = normal mucosa: 1 = edema and/or loss of visible according to severity by use of a predefined scale. Nine vascularity, granularity of the mucosa; 2 = friable mucosa of the 13 changes were graded from 0 to 4 (0 = normal, (visible, induced bleeding on examination) or peteciae; and 1 = mildly abnormal, 2 = moderately abnormal, 3 = 3 = spontaneous hemorrhage or visible ulcers. Assessment markedly abnormal, and 4 = severely abnormal); three and scoring of endoscopic pathology was always done be- parameters were classified as present or absent; and mi- fore biopsy. The presence or absence of signs/symptoms of totic activity was scored as normal, decreased, or in- acute toxicity (abdominal pain, loose stools, bloody stools, creased. Actual counts of the listed abnormalities were tenesmus, and nausea) were recorded at the times of endos- not appropriate or not considered necessary. This ad hoc copy using a standardized questionnaire designed for and grading system was based on the radiation pathologist's administered in accordance with the approved clinical pro- previous observations and experience with radiation tocol. proctitis, as well as on the range of lesions in the spec- The rectal biopsies were fixed in 10% neutral buffered imens included in the present study. Additional informa- formalin, processed and paraffin embedded, and sectioned tion and details of the scoring system are shown in Table at 4 p,m. Sets of sections were stained with hematoxylin and 1. eosin for histologic assessment and with periodic acid Schiff (PAS) to assess goblet cells and mucin content. Macrophage accumulation was assessed immunohistochcmically using a Statistics monoclonal mouse anti-human CD68 primary antibody Statistical analysis was performed with StatXact 4 (Cytel (DAKO Corp., Carpinteria, California), a biotinylated sec- Software, Cambridge, MA), a software package for exact ondary antibody, and peroxidasc/DAB staining. nonparametric inference. Overall and individual compari- The sections were coded in random order. A patholo- sons were performed using unordered or singly ordered gist with special expertise in radiation injury (L.F.F.) contingency tables with x2 statistics as appropriate. Asso- performed all microscopic examinations and quantita- ciations were assessed using Spearman's rank-correlation tions of histopathologic changes. The pathologist did not coefficient or the Jonckheere-Terprstra test, depending on have access to information regarding patient identity, the number of levels of the variables. There was no signif- clinical symptoms, endoscopic results, or whether a par- icant difference for any parameter between placebo-treated ticular biopsy was obtained before or during radiation patients (n = 16) and sucralfate-treated patients (n = 17). therapy. Each slide was assessed systematically for Therefore, the data from these 2 groups of patients were changes in 1) the surface epithelium; 2) the glandular pooled. However, all analyses were performed stratified by compartment of the mucosa; and 3) the lamina propria drug (sucralfate or placebo); p values less than 0.05 were stroma (Table 1). A total of 13 parameters were graded considered statistically significant. Acute radiation proctitis • N. HOVDENAK el al. 1113

Tabic 2. Number of patients reporting clinical symptoms of acute toxicity before radiation therapy (lime = 0) and after the 2 and 6 weeks of pelvic radiation therapy (time = 2 and time = 6, respectively)

Abdominal Loose Bloody Time pain stool stool Tenesmus Nausea

0 2 2 1 0 0 2 6 15 2 2 3 6 9 16 8 13 8

RESULTS Endoscopic changes and clinical symptoms In keeping with other studies and common clinical expe- rience, the prevalence of symptoms of acute toxicity in- creased throughout the radiation therapy course. The total Fig. I. Spider chart showing the average score for 9 alterations number of symptoms increased from baseline to 2 weeks scored from 0 to 4 according to severity (sec Table 1). The changes (p < 0.001) and from 2 weeks to 6 weeks (p < 0.001). The arc grouped according to compartment (surface epithelium, glands, proportions of patients who reported having one or more of lamina propria) and shown separately for biopsies obtained before the symptoms included on the questionnaire were 61% and irradiation (black), 2 weeks after initiation of radiation therapy 86% after 2 weeks and 6 weeks of radiation therapy, re- (light gray), and 6 weeks after initiation of radiation therapy (dark gray). All changes, except edema, exhibit highly significant spectively (compared to 11% before radiation therapy). The changes. Sec text for individual p values. frequencies of the individual symptoms at each examination time are summarized in Table 2. There was no statistically significant difference between placebo treated and sucral- tion of leukocytes through the crypt wall), crypt abscesses, fate treated patients for any of the 5 clinical symptoms at inflammatory cell infiltration in the surface epithelium and any time point (p values ranging from 0.21 to 1.0). lamina propria, and a striking accumulation of eosinophilic All baseline endoscopic exams were unremarkable (en- granulocytes. The rectal glands (crypts) exhibited nuclear doscopic score 0). The endoscopic scores increased signif- migration toward the luminal pole of the cells, atrophy, and icantly from the pretreatment examination to the examina- distortion with loss of goblet cells and/or of complete tions performed al 2 and 6 weeks during radiation therapy glands. Edema in the lamina propria stroma was not a (p < 0.001). In contrast to the clinical symptoms, however, consistent finding. median endoscopic score was maximal al 2 weeks (median. There was highly significant regression of morphologic 2 [range, 0-3], vs. median, 1 (range, 0-3], at 6 weeks), but abnormalities toward the end of the radiation therapy the difference between the 2-week score and 6-week score course, i.e., from 2 weeks to 6 weeks (p = 0.0002). When was not statistically significant (p = 0.82). Again, there was histologic parameters were analyzed individually, the im- no significant difference between placebo-treated and su- provement was statistically significant for surface cell cralfate-treated patients at either time point (p = 0.33 and height reduction (flattening) (p = 0.035) and inflammation p - 0.74 at 2 and 6 weeks, respectively). of the surface epithelium (p = 0.01). Endoscopic score was associated with the severity of Immunohistochemical staining of the CD68 antigen loose stools at 2 weeks (p < 0.001), but (because of low showed increased density of macrophages at 2 weeks com- prevalence) not with the other symptoms. There was no pared to pretreatment. but no further change at 6 weeks. The association between endoscopic pathology and individual stained macrophages were mainly localized beneath the clinical symptoms al 6 weeks. There were moderate corre- epithelium in close proximity to and intermingled with other lations between endoscopic scores and overall number of inflammatory cells in the lamina propria. clinical symptoms at both time points (rho2 = 0.22, p = The association between overall histologic score and en- 0.01 and rho2 = 0.19, p = 0.03, respectively). doscopic score was significant at 2 weeks (rho2 = 0.20, p = 0.02), but not at 6 weeks (p = 0.24). Overall histologic Quantitative histopathology score was associated with loose stools at 2 weeks (p = 0.01) All pre-irradiation biopsies showed unremarkable rectal and with abdominal pain at 6 weeks (p = 0.01), but not with mucosa (scores ranging from 0 to 0.25). In contrast, all the other individual symptoms. The association between biopsies obtained during radiation therapy exhibited signif- overall histologic scores and number of clinical symptoms icant abnormalities in all mucosal compartments (i.e., sur- was borderline significant at 2 weeks (rho2 = 0.14, p = face epithelium, glandular elements, and lamina propria). 0.05) and nonsignificant at 6 weeks {p = 0.46). The findings are summarized in Fig. 1. Inflammatory Surface epithelium. Changes in the surface epithelium changes were prominent and consisted of cryptitis (migra- consisted of prominent flattening of the columnar cells and 1114 L J. Radiation Oncology • Biology • Physics Volume 48, Number 4, 2000

Fig. 3. Cryptius; 14 days (and 18 Gy) after initiation of therapy. The deep portions of these two adjacent glands are infiltrated by segmented neutrophils and occasional lymphocytes; the stroma also contains a number of neutrophils. Fig. 2. Normal rectal epithelium from biopsy obtained before radiation therapy. Notice regularity of glands and mucin vacuoles. edema. Prominent aggregations of macrophages containing invasion of the epithelium by neutrophilic and eosinophilic strongly PAS-positive (diastase-resistant) cytoplasmic ma- granulocytes. These changes appeared to be of early onset terial were present in the lamina propria m some sections, and were statistically highly significant for both cell height suggestive of mucin phagocytosis. reduction (p < 0.0001) and inflammation (p = 0.0007) at 2 weeks. Both changes decreased significantly from 2 to 6 DISCUSSION weeks (p = 0.035 and p = 0.01, respectively). Rectal glands. The most prominent findings in this com- There is a paucity of clinical, prospective, and sequential partment were: infiltration of the deep glandular wall by studies with multiple endoscopic exams and rectal biopsies segmented neutrophils (cryptitis; neutrophil cryptitis), col- at fixed intervals. Hence, little is known about the time lections of eosinophils in the deep glandular lumen (crypt course of development of individual mucosal abnormalities abscesses; eosinophilic crypt abscesses), loss of goblet cells, during radiation therapy and the relationship of mucosal and apical nuclear migration (migration of nuclei of epithe- histopathology to other parameters of radiation toxicity. The lial cells toward the glandular lumen). All of these changes were statistically highly significant compared to baseline (p < 0.0001). There was also prominent glandular atrophy (p = 0.0025), glandular distortion (p = 0.001), and loss of glands (p < 0.0001). All changes in the glandular compart- ment were maximal at 2 weeks and decreased from Week 2 to Week 6. Thus, at 6 weeks, atrophy and distortion of glands were no longer statistically different from before irradiation. Characteristic histologic findings are shown in Figs. 2-6. In contrast to crypt abscesses, which consisted mainly of eosinophils, inflammatory infiltrates in the crypts were mostly comprised of neutrophils. Atypical mitotic figures and atypical nuclei in epithelial cells were striking findings in biopsies obtained during irradiation. However, the number of mitotic figures did not change significantly during treatment. Stroma of the lamina propria. Biopsies obtained during ongoing radiation therapy exhibited prominent and statisti- cally significant inflammation of the lamina propria (p < Fig. 4. Two of three glands showing cell atrophy resulting in cuboidal or flat epithelium. A collection of eosinophils and neu- 0.0001). Although there was significant congestion of lam- trophils (crypt abscess) is seen in the lumen of the central gland. ina propria vessels at Week 2 (p = 0.002), there was no Mucin is conspicuously absent. Sample obtained on Day 18, after consistent or statistically significant evidence of stromal 26 Gy. Acute radiation proctitis • N. HOVDENAK el al. 1115

delayed radiation proctitis (9), the efficacy of this drug for preventing acute gastrointestinal radiation toxicity is more controversial. Hence, while some studies promote sucralfate as an effective modulator of acute intestinal radiation tox- icity (10), others have found no beneficial effects (11). The present study did not show significant drug effects on any of the clinical, endoscopic, or histologic parameters and thus does not support the use of sucralfate in the acute setting. Examination of the sequential biopsies showed that pro- nounced mucosal injury develops relatively early during the radiation course and, surprisingly, improves significantly toward the end of treatment. Hence, all histologic parame- ters exhibited maximal values at 2 weeks and regressed from 2 to 6 weeks despite continued daily radiation treat- ments and worsening clinical symptoms. Because biopsies Fig. 5. Gland distortion and moderate cryptitis in a 14-day speci- were obtained only at two time points during irradiation, men (18 Gy). The glands are widely separated by stroma that however, it is not known whether 2 weeks is the time of contains modest numbers of plasma cells, macrophages, and gran- maximal response, at what point the various lesions begin to ulocytes. improve, or when/if they resolve completely. Cup forceps biopsies are relatively superficial and do not present study used systematic procurement of biopsies at provide material for assessment of changes in the deeper predetermined times before and during ongoing pelvic ra- layers of the rectal wall. However, it is likely that a defec- diation therapy, combined with "blinded," quantitative as- tive epithelial barrier allows intestinal contents, e.g., micro- sessment of individual parameters. This allowed objective organisms, foreign antigens, and proteolytic enzymes, ac- characterization of the sequential development of specific cess to stromal tissue compartments. This elicits radiation-induced changes in the rectal mucosa, as well as inflammatory and fibrogenic responses that may further assessment of the relationships among histopathologic, damage the mucosa and possibly set the stage for chronic macroscopic (endoscopic), and clinical parameters of acute changes in the deep layers of the bowel wall. radiation toxicity. In accordance with previous experimental studies in rat The anterior rectal wall is in close proximity to and partly intestine (12), we found a significant decrease in the number in continuity with the therapeutic target organs. To avoid of mucin-containing goblet cells. A decrease in mucin pro- potential artifacts in the rectal mucosa from direct apposi- duction may ftuther reduce the protective barrier function of tion, biopsies were always obtained from the posterior wall the mucosa. Mucin capture by macrophages in the lamina of the rectum and at a standardized distance from the anal propria was consistent with goblet cell destruction with verge. release of mucin within the mucosa and subsequent phago- In contrast to the well-documented benefit of sucralfate in cytosis. Although cryptitis most likely precedes crypt abscess formation, these two alterations were graded separately because cryptitis was more prevalent than crypt abscesses, and very often the two changes were not concomitant. The •r . .viv zy-"*» large number of eosinophils in the inflammatory reactions was another striking feature. The significance of this phe- nomenon is not clear. Intestinal eosinophilia is common in various pathologic conditions, including allergy, drug reac- tions, and some parasitic infections. It is conceivable that decreased epithelial barrier function allows access of anti- gens/allergens to subepithelial tissues, eliciting a reactive eosinophilic inflammatory response. This notion is consis- tent with preclinical studies that show increased inflamma- 'ijf ttS'FttS&Aite**''Sn» Isf/'V -C " tory cell transmigration (13) and connective tissue mast cell accumulation in the deeper layer of the bowel wall after intestinal irradiation (14). Macrophages play a central role in mucosal inflamma- •<#4mm 2ir tion, but are not readily recognized in hematoxylin/eosin Fig. 6. Considerable gland loss and atrophy is already present 13 preparations. Therefore, immunohistochemical staining of days (16 Gy) after initiation of therapy. Atrophy of the surface the CD68 antigen was performed to obtain information epithelium is best seen on the left. about changes in the monocyte/macrophage cell lineage in 1116 L J. Radiation Oncology • Biology • Physics Volume 48, Number 4, 2000

the present study. Macrophages not only constitute a prom- mal (5), the severity of early (predominantly mucosal) in- inent cellular component of the inflammatory process, but jury, as described in our study, predicts and may be mech- also modulate inflammation through expression of a large anistically involved in the development of delayed toxicity number of potent mediators, including cytokines and che- as consequential late effects (17-21). mokines. Hence, in addition to reflecting the intensity of The improvement of histologic injury during the latter inflammation, macrophages in the rectal mucosa during part of the treatment course may also be due to compensa- radiation therapy are an important source of continued pro- tory mechanisms in the epithelium. Progression of radiation inflammatory stimuli. For example, expression of macro- injury is determined by the interplay among pro- and anti- phage inflammatory protein (MTP) is stimulated by luminal inflammatory processes, as well as by cell loss and repopu- products, e.g., butyrate (15), which may gain access to the lation. The time to onset of compensatory proliferation mucosal tissue through damaged surface epithelium. The during radiation therapy is a controversial issue. It is likely remarkable regression of changes in the epithelium from 2 to be no more than 2 weeks, however, and cell turnover may to 6 weeks during irradiation may be associated with a be twice normal at the end of a 6-week treatment course decrease in this luminal influence, consistent with the over- (22). Consequently, increased proliferation may outweigh all decrease in inflammatory response in the glands and cell loss during ongoing radiation. This may also explain stroma of the lamina propria. why the histologic changes in our study did not progress The significance of atypical nuclei and atypical mitoses from the 2nd to the 6th week of treatment, but rather remains to be elucidated. Some studies have reported an exhibited significant and consistent improvement. These increased incidence of adenocarcinoma development after issues could be resolved by assessing cytokinetic parame- therapeutic irradiation (16). It is not clear to what extent the ters systematically in well-oricntcd mucosal biopsies. atypia observed in our study reflects increased potential for subsequent malignant transformation. However, these atyp- CONCLUSIONS ical cells probably have short lives (days to weeks), whereas radiation-induced carcinomas appear late (years). Pelvic radiation therapy is associated with prominent The mechanisms underlying the histologic improvement morphologic changes in the superficial epithelium, glands, from Week 2 to Week 6, despite ongoing radiation treat- and stroma of the lamina propria in the rectal mucosa. ment and worsening symptoms, are unclear. Damage to Histologic and endoscopic changes develop during the first deeper structures beyond the reach of our biopsies, e.g., pan of the treatment course and stabilize or regress toward submucosa, enteric nerve plexus, blood vessels, or muscu- the end of treatment, despite ongoing irradiation and pro- laris propria, may possibly be involved in the persistence of gressive clinical symptoms. Further studies are needed to symptoms and subsequent development of chronic changes address the role of cell proliferation, pro- and anti-inflam- despite mucosal normalization. Delayed radiation injury in matory mediators (e.g.,. cytokines, growth factors, and ei- the colon may develop 6 months-30 years after exposure, cosanoids), as well as the significance of specific tissue usually 1-5 years. Symptomatic patients usually present reactions at the molecular level, for example, endothelial with bleeding or disturbed transit due to fibrosis. Although dysfunction, proteases, and acute and chronic oxidative these delayed lesions arc predominantly vascular and stro- stress.

REFERENCES 1. Caspers RJ. Hop WC. Irradiation of true pelvis for bladder and ation-induced proctitis. Am J Gastroenterol 1988;83:1140- prostatic carcinoma in supine, prone or Trendelenburg posi- 1144. tion. Int J Radial Oncol Biol Phys 1983;9:589-593. 8. Hanauer SB. Robinson M, Pruitl R, et al. Budesonide enema 2. Das II, Lanciano RM. Movsas B, et al. Efficacy of a belly for the treatment of active, distal ulcerative colitis and proc- board device with CT-siraulation in reducing small bowel titis: A dose-ranging study. U.S. Budesonide Enema Study volume within pelvic irradiation fields. Int ] Radiat Oncol Biol Group. Gastroenterology 1998;l 15:525-532. Phys 1997;39:67-76. 9. Kochhar R, Patel F, Dhar A. el al. Radiation-induced proc- 3. Lechner P, Cesnik H. Abdominopelvic omentopexy: prepara- tosigmoiditis. Prospective, randomized, double-blind con- tory procedure for radiotherapy in rectal canccr. Dis Colon trolled trial of oral sulfasalazine plus rectal steroids versus Rectum 1992;35:1157-1160. rectal sucralfate. Dig Dis Sci 1991;36:103-107. 4. Deulsch AA, Stem HS. Technique of insertion of pelvic 10. Henriksson R, Franzen L, Litlbrand B. Effects of sucralfate on Vicryl* mesh sling to avoid postradiation enteritis. Dis Colon acute and late bowel discomfort following radiotherapy of Rectum 1989;32:628-630. pelvic cancer. J Clin Oncol 1992;10:969-975. 5. Berthrong M, Fajardo LP. Radiation injury in surgical pathol- 11. O'Brien PC, Franklin CI, Dear KBG, et al. A phase m ogy. Part II. Alimentary tract. Am J Surg Pathol 1981;5:153- double-blind randomised study of rectal sucralfate suspension 178. in the prevention of acute radiation proctitis. Radiother Oncol 6. Gelfand MD, Tepper M, Katz LA, el al. Acute irradiation 1997:45:117-123. proctitis in man. Development of eosinophilic crypt abscesses. 12. Becciolini A, Fabbrica D, Crcmonini D, el al. Quantitative Gastroenterology 1968;54:401-411. changes in the goblet cells of the rat small intestine after 7. Haboubi NY, Schofield PF, Rowland PL. The light and irradiation. Acta Radiol Oncol 1985;24:291-299. electron microscopic features of early and late phase radi- 13. Richter KK. Fagerhol MK, Cair JC, el al. Association of Acute radiation proctitis • N. HOVDENAK el al. 1117

granulocyte transmigration with structural and cellular param- 18. Schultheiss TE, Lee WR, Hunt MA, et al. Late GI, and GU eters of injury in experimental radiation enteropathy. Radial complications in the treatment of prostate cancer. Int J Radiat Oncol Invest 1997;5:275-282. Oncol Biol Phys 1997;37:3-11. 14. Richter KK, Langberg CW, Sung C-C, et al. Increased trans- 19. Wang C-J, Leung SW, Chen H-C, el al. The correlation of forming growth factor j3 (TGF-0) immunoreactivity is inde- acute toxicity and late rectal injury in radiotherapy for pendently associated with chronic injury in both consequential cervical carcinoma: Evidence suggestive of consequential and primary radiation enteropathy. Int J Radiat Oncol Biol late effect (CQLE). Int J Radiat Oncol Biol Phys 1998;40: Phys 1997;39:187-195. 85-91. 15. Ohno Y, Lee J, Fusynyan RD, et al. Macrophage inflammatory 20. Weiss E, Himle P, Amold-Bofinger H, et al. Therapeutic protein-2: chromosomal regulation in rat small intestinal epithe- outcome and relation of acute and late side effects in the lial cells. Proc Natl Acad Sci USA 1997;94:10279-10284. adjuvant radiotherapy of endometrial carcinoma stage I and n. 16. Fischer L, Kimose HH, Spjeldnaes N, et al. Late progress of Radiother Oncol 1999;53:37-44. radiation-induced proctitis. Acta Chir Scand 1990; 156:801- 21. Kline JC, Buchler DA, Boone ML, et al. The relationship of 805. reactions to complications in radiation therapy of cancer of the 17. Bourne RB, Kearsley JH, Grove WD, et al. The relationship cervix. Radiology 1972;105:413-416. between early and late gastrointestinal complication of radia- 22. Morris GM. Effects of radiation on the cell proliferation tion therapy for carcinoma of the cervix. Int J Radial Oncol kinetics of epithelial tissues—Therapeutic implications. Br J Biol Phys 1983;9:1445-1450. Radiol 1996;69:795-803.

Paper II

CLINICAL SIGNIFICANCE OF INCREASED GELATINOLYTIC ACTIVITY IN THE RECTAL MUCOSA DURING EXTERNAL BEAM RADIATION THERAPY OF PROSTATE CANCER

NILS HOVDENAK, M.D.,* JUNRU WANG, M.D., PH.D.,* CHING-CHING SUNG, M.S.,* THOMAS KELLY, PH.D.,* LUIS F. FAJARDO, M.D.,§ AND MARTIN HAUER-JENSEN, M.D., PH.D.** •Department of Oncology, Bergen University Hospital, Bergen, Norway; Departments of tSurgery and 'Pathology, University of Arkansas for Medical Sciences and Central Arkansas Veterans Healthcare System, Little Rock, AR; 'Department of Pathology, Stanford University School of Medicine and Veterans Healthcare System. Palo Alto, CA

Reprinted from INTERNATIONAL JOURNAL OF RADIATION, ONCOLOGY, BIOLOGY, AND PHYSICS VOLUME 53, NUMBER 4, JULY 15,2002

^mswjrt Int. J Radiation Oncology Biol Phys.. Vol. 53. No 4. pp. 919-927, 2002 JdBh8E| Copyright € 2002 Elsevier Science Inc. JS^Drøffi^ Pnnied in the USA. At] rights reserved * ^V jjl 0360-301 M)2/S-see front matter

ELSEVIER Pn S0360-3016<02)02808-0

CLINICAL INVESTIGATION Prostate

CLINICAL SIGNIFICANCE OF INCREASED GELATINOLYTIC ACTIVITY IN THE RECTAL MUCOSA DURING EXTERNAL BEAM RADIATION THERAPY OF PROSTATE CANCER

NILS HOVDENAK, M.D..* JUNRU WANG. M.D., PH.D.,* CHING-CHING SUNG, M.S.,* THOMAS KELLY. PH.D.,* LUIS F. FAJARDO, M.D.,§ AND MARTIN HAUER-JENSEN, M.D.. PH.D.** •Department of Oncology. Bergen University Hospital, Bergen, Norway; Departmenis of 'Surgery and 'Pathology. University of Arkansas for Medical Sciences and Central Arkansas Veterans Healthcare System, Little Rock, AR; "Department of Pathology, Stanford University School of Medicine and Veterans Healthcare Sysiem, Palo Alto, CA Purpose: Rectal toxicity (proctitis) is a dose-limiting factor in pelvic radiation therapy. Mucosal atrophy, i.e.. net extracellular matrix degradation, is a prominent feature of radiation proctitis, but the underlying mechanisms are not known. We prospectively examined changes in matrix metalloprotcina.se (MMP)-2 and MMP-9 (gelati- nase A and B) in the rectal mucosa during radiation therapy of prostate cancer, as well as the relationships of these changes with symptomatic, structural, and cellular evidence of radiation proctitis. Methods and Materials: Seventeen patients scheduled for external beam radiation therapy for prostate cancer were prospectively enrolled. Symptoms of gastrointestinal toxicity were recorded, and endoscopy with biopsy of the rectal mucosa was performed before radiation therapy, as well as 2 and 6 weeks into the treatment course. Radiation proctitis was assessed by endoscopic scoring, quantitative histology, and quantitative immunohisto- chemistry. MMP-2 and MMP-9 were localized immunohistochemically, and activities were determined by gelatin zymography. Results: Symptoms, endoscopic scores, histologic injur)', and mucosal macrophages and neutrophils increased fronT5aseline to 2 weeks. Symptoms increased further from 2 weeks to 6 weeks, whereas endoscopic and cellular evidence of proctitis did not. Compared to pretreatment values, there was increased total gelatinolytic activity of MMP-2 and MMP-9 at 2 weeks (p = 0.02 and p = 0.004, respectively) and 6 weeks (p = 0.006 and p = 0.001, respectively). Active MMP-2 was increased at both time points (p = 0.0001 and p = 0.002). Increased MMP-9 and MMP-2 at 6 weeks was associated with radiation-induced diarrhea ip = 0.007 and p — 0.02, respectively) and with mucosal neutrophil infiltration (rho = 0.62). Conclusions: Pelvic radiation therapy causes increased MMP-2 and MMP-9 activity in the rectal mucosa. These changes correlate with radiation-induced diarrhea and granulocyte infiltration and may contribute to abnormal connective tissue remodeling in radiation proctitis. © 2002 Elsevier Science Inc. Prostate cancer. Radiation therapy. Matrix metalloproteinases, Rectum, Inflammation.

INTRODUCTION The ECM is a functionally active scaffolding meshwork that, in addition to maintaining normal tissue structure, is Pelvic radiation therapy is almost invariably accompanied by some degree of acute proctitis and, occasionally, fol- pivotal in the integration of cellular interactions. Its main lowed by a progressive inflammatory-fibrotic condition that components are the basement membrane and the interstitial often appears several months to years later. Atrophy of the stroma, the latter being comprised of a large variety of rectal mucosa, i.e., net loss of mucosal extracellular matrix molecules, including collagens, laminin, gelatins, elastin, (ECM), is a characteristic feature of early radiation proctitis. proteoglycans, and fibronectin. The matrix metalloprotein- Furthermore, while there is increased connective tissue dep- ases (MMP), along with their activators and inhibitors, play osition in the deeper layers of the bowel wall during the a central role in regulating the equilibrium between ECM chronic phase of radiation proctitis (intestinal wall fibrosis), synthesis and breakdown. the mucosal atrophy generally persists. Although these Major efforts in cancer research have focused on the role changes have obvious functional implications, the mecha- of MMPs in tumor progression and metastasis (for example, nisms underlying the radiation-induced ECM degradation in reviewed in Ref. 1) and on MMPs as potential targets for the intestinal mucosa arc unknown. antitumor therapy (2). However, MMPs are also involved in

Reprint requests to: Martin Hauer-Jensen. M.D., Ph.D.. Arkan- Supported by the Kavli Foundation and Central Arkansas Ra- sas Cancer Research Center, 4301 West Markham. Slot 725, Little diation Therapy Institute. Rock. AR 72205. Tel: (501) 686-7912; Fax: (501) 421-0022; Received Nov 19, 2001. and in revised form Feb 12. 2002. E-mail: [email protected] Accepted for publication Feb 27, 2002.

919 920 L J. Radiation Oncology • Biology • Physics Volume 48, Number 4, 2000 many benign disorders, especially those associated with (1) before the commencement of radiation therapy; (2) 2 inflammation and abnormal ECM remodeling. Hence, in- weeks into the radiation therapy course (This time point is creased MMP expression occurs in arthritis, peptic ulcer henceforth referred to as "2 weeks"); and (3) 6 weeks into disease (3), celiac disease (4). and inflammatory bowel the radiation therapy course (i.e., toward the end of the disease (5, 6), and MMP inhibitors show promise as mod- course and henceforth referred to as "6 weeks"). ulators of gastrointestinal inflammation (7). Whereas MMPs The objective of this work was to investigate the associ- seem to be overexpressed in intestinal inflammation, tissue ation of MMP activity with specific symptoms and signs of inhibitor of metalloproteinase-1 (TIMP-1) is not (8), sug- toxicity. Therefore, relevant gastrointestinal symptoms/ gesting a net increase in proteolytic activity in inflamed signs were recorded individually in preference to using a mucosa (9). global toxicity scoring system, such as the EORTC/RTOG Ionizing radiation upregulates MMP expression in many scale. The presence or absence of six individual symptoms in vitro systems (10). although some investigators report the and signs of acute toxicity (pain, diaiThea [loose stool], opposite effect (11-13). Preclinical and clinical studies that tenesmus [fecal urgency with crampy rectal pain], bloody address the effects of radiation on MMPs in vivo are few stools, bloating, and nausea) was recorded at each time and even less conclusive. Seifert et al. found that radiation point using a standardized questionnaire, administered by a enhanced anastomotic MMP activity in the rat colon (14). In gastroenterologist (16). contrast, a small clinical study in patients undergoing neo- Proctoscopy was performed, and endoscopic changes adjuvant radiation therapy of rectal cancer did not show were graded according to a four-point scale where 0 = significant alterations of MMP-2 (gelatinase A) or MMP-9 normal mucosa: 1 = edema and/or loss of visible vascular- (gelatinase B) activities in rectal mucosa that was not di- ity, granularity of the mucosa; 2 = friable mucosa (visible rectly involved with cancer (15). induced bleeding on examination) or peteciae; and 3 = Changes in MMP activity during radiation therapy have spontaneous hemorrhage or visible ulcers (16. 17). At each potential implications not only for the mechanisms of nor- time point, 2-3 cup forceps biopsies were obtained from the mal tissue toxicity, but also for interactions between normal posterior rectal wall, placed immediately on dry ice, and tissues and tumors. We performed a prospective clinical stored at — 70°C for subsequent batch analysis. Additional trial in which a cohort of patients scheduled for external biopsies were obtained for immunohistochemical analysis beam treatment of prostate cancer consented to repeated and quantitative histology, formalin-fixed, and processed rectal endoscopies with biopsies before and during radiation routinely The average dose to the posterior rectal wall was therapy. The sequential design allowed assessment of 77% ± 4.7% of the prescribed dose, estimated at a point 10 changes in MMP-2 and MMP-9 in the rectal mucosa and cm proximal to the anal verge and opposite the midline correlation with clinicopathologic parameters of radiation point on the anterior wall, using the computerized treatment proctitis. The results demonstrated that pelvic radiation plans. therapy causes consistent and highly significant increases in the gelatinolytic (MMP-2 and MMP-9) activity in rectal Mucosal histology and inflammation mucosa and that these changes are associated with radia- Cellular aspects of mucosal inflammation were assessed tion-induccd diarrhea and neutrophil infiltration in the tis- immunohistochemically by staining for macrophages sue. (CD68), granulocytes (myeloperoxidase), mast cells (tryptase), and interleukin-la (IL-la). Formalin-fixed, par- affin-embedded biopsies were sectioned at 3-4 /xm. Spe- METHODS AND MATERIALS cific antibodies against CD68 (M0876, 1:50 dilution. Dako, Patients Carpinteria, CA), myeloperoxidase (A0398. 1:200 dilution, Seventeen patients (median age: 68 years, range: 58-76 Dako). tryptase (M7052, 1:100 dilution, Dako), and inter- years) scheduled to undergo radiation therapy without con- leukin-la (02-1150, 1:25 dilution. Cistron Biotechnology. comitant chemotherapy for organ-confined prostate carci- Pine Brook, NJ) were applied and visualized using conven- noma were enrolled prospectively. Patients with current or tional avidin-biotin staining methods, diaminobenzidine past gastrointestinal disease were not considered eligible. chromogen, and hematoxylin counterstaining. The trial protocol was approved by the Regional Medical The number of cells staining positive for myeloperoxi- Ethics Committee (Institutional Review Board for Clinical dase, CD68, tryptase. or IL-la in each specimen were Research), and written informed consent was obtained from counted manually using light microscopy and a square each patient before enrollment. eyepiece grid that covered 62,500 finr at the magnification All patients underwent external beam radiation therapy used (40X objective). For each antibody, the total number with individual treatment planning that used a four-field box of stained cells was determined in 10 grids per section. The technique with customized blocking and 6-18-MV photons grids were applied so as to encompass the lamina propria, to a total tumor dose 54 - 66 Gy over 6-7 weeks (once-daily crypts, and surface epithelium, while avoiding the cut edges 2.0-Gy fractions with weekend breaks). of the biopsy specimen and the submucosa when present. The patients were interviewed, examined endoscopically, A pathologist with special expertise in radiation injury and biopsied by the same investigator at three time points: systematically assessed and quantified changes in the sur- Metalloprotcinases in radiation proctitis • N. HOVDENAK et al. 921

face epithelium, glandular compartment of the mucosa, and zymogen forms and active forms of MMP-2 (72 kD, 62 kD) the lamina propria stroma, as documented and described in and MMP-9 (92 kD. 83 kD). as verified by MMP standards detail elsewhere (16). A total of nine parameters pertaining applied to each gel. Images were captured in a gel docu- to changes in the surface epithelium (reduced cellular mentation system (Gel Print 20001, Scientific Consultants height, inflammatory' infiltrate), rectal glands (lumenal mi- Inc., Baton Rouge, LA), and the eiectrophoretic bands were gration of epithelial nuclei, loss of goblet cells, neutrophil quantitatively analyzed by densitometry using SigmaGel cryptitis, loss of glands), and lamina propria stroma (inflam- (Jandel Scientific Software, San Rafael. CA). mation, edema, blood vessel congestion) were scored from To assess the localization of the MMPs in the rectal 0 to 4. Three additional parameters (eosinophilic crypt ab- mucosa, representative formalin-fixed biopsies were stained scesses. glandular atrophy, and distortion of rectal glands) using specific antibodies, conventional avidin-biotin stain- were classified as present or absent. ing methods, diaminobenzidine chromogen, and hematoxy- The investigators who performed the quantitative immu- lin counterstaining. Appropriate negative control slides with nohistochemistry (N.H.) and the quantitative histologic omission of primary antibody and substitution of the pri- analysis (L.F.F.) did not have access to information regard- mary immune antibody with nonimmune IgG (DAKO. ing patient identity, symptoms, endoscopy findings, or bi- Carpinteria, CA) were included. MMP-2 staining was per- opsy sequence (i.e.. whether a particular biopsy was ob- formed using a mouse monoclonal antibody against tained before or during radiation therapy). 4-aminophenylmercuric-activated native 72 kD human MMP-2 (MS-806-P0, 1:100 dilution, Laboratory Vision, MMP-2 and MMP-9 zymography Fremont, CA) that recognizes both the 72 kD zymogen and and immunohistochemistry the 66 kD active forms. Specificity of this antibody was Biopsies for quantitative MMP gelatin zymography were established by ELISA and western blotting (personal com- processed as a single batch. The frozen biopsies were pul- munication, Dr. Jack Windsor. Indiana University). MMP-9 verized mechanically and suspended in 1 mL extraction staining was performed with a mouse monoclonal antibody

buffer (Tris-HCl 0.5 M, NaCI 0.2 M, CaCI2 0.01 M, Triton- against amino acids 626 - 644 of human MMP-9 (56-2A4, 100 0.1%, pH = 7.6). Cells were disrupted using a mini- 1:100 dilution. Oncogene. San Diego, CA ) that specifically beadbeater at high speed for 2 min. The material was recognizes the zymogen (92 kD) and active (83 kD) forms centrifuged (15,000 g. 30 min at 4°C), and protein concen- of MMP-9 (18). tration of the supernatant was measured with a commercial protein assay (Bio-Rad Laboratories, California). Samples Statistics containing 20 /xg protein were mixed with 2X Sample The data were analyzed with StatXact 4 for Windows Buffer (Novex, San Diego, CA) and applied on preformed (Cytel Software, Cambridge. MA), a software package for gelatin-containing polyacrylamidc gel plates of uniform exact nonparamciric inference. Differences in symptom 1.0-mm thickness (Novex). scores, endoscopy scores, inflammation parameters, and The identity of the collagenases was confirmed by im- MMP activity levels among the three time points were munoblotting using MMP-2 and MMP-9-specific antibod- examined using the Friedman test and post hoc Wilcoxon ies. Further confirmation was obtained by preincubating in signed rank test for repealed samples for pairwise compar- 0.1 p.M 4-aminophenylmcrcuric acetate to activate MMP. isons. Associations among the different end points were ethylene diamine tetra-acetate to demonstrate inhibition of assessed with Spearman's rank correlation coefficient. Dif- MMPs, and the protease inhibitors phenyl methane sulfonyl ferences in MMP activities between groups of patients fluoride and N-ethyl maleimide to show that the proteolytic classified according to dichotomous variables (e.g., pres- activity observed in our samples was not because of serine ence or absence of specific symptoms) were assessed with proteases or cysteine proteases, respectively. the Mann-Whitney V test. Two-sided tests were used All plates used in the present study were from the same throughout, and p values less than 0.05 were considered batch. All three samples from each patient (baseline. 2 statistically significant. weeks. 6 weeks) were run together on the same plate, together with active MMP-2 and MMP-9 standards. Elec- RESULTS trophoresis was performed in a special chamber (Novex) for 90 min: voltage 125 V and current 30-40 mA (start) to Symptoms 8-12 mA (end). Renaturing, developing, staining, and The individual symptoms and signs of radiation proctitis washing were carried out under gentle agitation. The gel at the different time points are summarized in Table 1. The plates were renatured with Triton X-100 (Novex), 25% v/v percentage of patients having one or more symptoms/signs in distilled water for 30 min, placed in developing buffer was 65% at 2 weeks and 71% at 6 weeks (compared to 24% (Novex) for 30 min. and then further developed in fresh before radiation therapy). Only three of the 17 patients did developing buffer for 18 h al 37°C. The plates were then not report symptoms or signs of acute toxicity during radi- stained with Colloidal Blue (Novex) for 3 h and washed in ation therapy. The total number of symptoms/signs in- distilled water for 7 h. The gel plates exhibited areas of creased from baseline to 2 and 6 weeks (p = 0.03 and p < gelatinolytic activity in patient samples corresponding to 0.001, respectively). The progression of symptoms/signs 922 L J. Radiation Oncology • Biology • Physics Volume 48, Number 4, 2000

Table 1. Number of patients reporting individual symptoms 50% for presence of eosinophilic crypt abscesses; 6%, before radiation therapy and 2 weeks and 6 weeks into the 38%, and 21% for presence of glandular atrophy; and 6%, radiation therapy course (n = 17) 44%, and 14% for presence of glandular distortion. Symptom Pretreatment 2 weeks 6 weeks The immunohistochemical analysis data are summarized in Table 3. There were significant increases in the numbers Pain 1 1 4 of myeloperoxidase-positive cells (granulocytes) and Diarrhea 2 9 11 CD68-positive cells (macrophages) at 2 weeks compared to Tenesmus 0 1 4 Blood in stools 0 1 3 baseline (p = 0.005 and p — 0.02, respectively), as well as Bloating 3 8 8 at 6 weeks (p = 0.01 and p = 0.04, respectively). However, Nausea 1 1 5 there was no significant difference between the 2- and 6-week time points. The numbers of tryptase-positive cells (mast cells) and IL-la-positive cells were not altered sig- from 2 weeks to 6 weeks was also statistically significant nificantly in biopsies obtained during radiation therapy in (p = 0.01). comparison to biopsies obtained before treatment.

Endoscopy MMP-2 and MMP-9 Endoscopic scores increased from baseline to 2 weeks Compared to the pretreatment baseline, gelatin zymogra- (p < 0.001) and from baseline to 6 weeks (p < 0.001). phy showed increased MMP-9 activity in rectal mucosa There was no increase in endoscopic severity of proctitis during radiation therapy in all 17 patients and increased from 2 to 6 weeks; in fact, there was a trend that approached MMP-2 activity in 16 of 17 patients. At 2 weeks, median statistical significance (p - 0.08) toward decreasing scores (95% confidence interval) MMP-2 and MMP-9 activity (data not shown). levels had increased to 232% (155%-284%,p = 0.02) and 390% (131%-1281%, p = 0.004) relative to the preirradia- Mucosal histology and inflammation tion baseline levels (Fig. 1). The median (95% confidence Histologic examination of the biopsies was consistent interval) percent activated MMP-2 increased from 12% with our previously reported larger cohort study of pa- (7%-23%) at baseline to 30% (18%-33%) at 2 weeks (p = tients undergoing pelvic radiation therapy (16); i.e., there 0.0001) (Fig. 2). The total levels of both MMPs and the were pronounced histologic changes in the epithelium, percent active MMP-2 remained significantly increased at 6 glandular layer, and lamina propria at both 2 weeks and weeks (p = 0.006, p = 0.001, and p = 0.002, respectively). 6 weeks (statistically significant for all changes com- There was no significant difference in MMP activity levels pared to pretreatment values), and most histologic between the biopsies at 2 weeks and 6 weeks. changes were more pronounced at 2 weeks than at 6 Two weeks into the radiation therapy course, there was weeks, despite the higher cumulative dose received at the a borderline significant association between diarrhea and latter time point. The scores for the nine histologic pa- increased MMP-2 activity (p = 0.05). However, the rameters that were graded 0-4 are summarized in Table relative increase in MMP activity at this time point did 2. The percentages of patients exhibiting changes scored not correlate with the other symptoms, the severity of as absent or present at the three observation times (pre- endoscopic proctitis, or the inflammation markers. In treatment, 2 weeks, and 6 weeks) were 0%, 38%, and contrast, both MMP-9 and active MMP-2 correlated with

Table 2. Median and interquartile range of scores for nine histologic parameters of radiation proctitis (0 = normal. 1 = mildly abnormal, 2 = moderately abnormal, 3 = markedly abnormal, 4 = severely abnormal) Pretreatment 2 weeks 6 weeks Surface epithelium Loss of cellular height 0(0-0) 1 (0-3.75) 0(0-1.5) Inflammatory infiltrate 0(0-0) 0.5 (0-2) 0(0-0.25) Rectal glands Lumenal migration of epithelial nuclei 0(0-1) 2(1-4) 2(2-3) Loss of goblet cells 0 (0-0.5) 2(1-4) 2(2-3) Cryptitis 0(0-0) 2(0-2) 2(1-2.25) Loss of glands 0(0-0) 1(0-3) 1 (0-2.25) Lamina propria Inflammation 0(0-0) 1.5 (0.25-3) 1(1-2) Edema 0(0-0) 0 (0-0.75) 0(0-0.25) Blood vessel congestion 0(0-0) 0(0-2) 0(0-2) Note: Biopsies were obtained from rectal mucosa before radiation therapy (baseline) and at 2 and 6 weeks during ongoing radiation therapy. Metalloprotcinases in radiation proctitis • N. HOVDENAK et al. 923

Table 3. Number of cells positive for myeloperoxidase (granulocytes), CD68 (macrophages), tryptase (mast cells), and IL-la in biopsies obtained from rectal mucosa of patients before radiation therapy and at 2 and 6 weeks during the course of ongoing radiation therapy

Antigen Pretreatment 2 weeks 6 weeks

Myeloperoxidase 49 (9-74) 162(99-216)* 152 (97-171)* CD68 12(6-33) 35 (13-54)* 39(11-79)* Tryptase 52(43-60) 70 (47-100) 60 (52-99) IL-la 47 (25-91) 65(23-132) 86(21-133) Note: The values are median and interquartile range of the number positive cells per 625,000 pim2. 'Significantly different from pretreatment baseline value. the number of mucosal granulocytes at 6 weeks (p = 0.62 cleave and alter the activity of many extracellular and and p = 0.04 for both associations), and patients who membrane-bound peptides, e.g., growth factors and re- reported diarrhea at 6 weeks exhibited significantly ceptors. The two gelatinases, MMP-2 (gelatinase A) and greater increases in mucosal MMP-2 activity (p = 0.02) MMP-9 (gelatinase B), degrade collagens type IV, V, and MMP-9 activity (p = 0.007). VII, and X (19) and thus, taken together, are capable of Representative immunohistochemical staining patterns breaking down a substantial part of the ECM. MMP-2 is for MMP-2 and MMP-9 at baseline and 2 weeks into the expressed constitutively by stromal cells and a number of radiation therapy course are shown in Fig. 3. There was other cell types. MMP-9 is produced mainly by hemato- strong MMP-2 staining in the pericryptal myofibroblast poietic and inflammatory cells, but, consistent with the sheath, inflammatory cells, and smooth muscle cells. The present study, is also expressed in colonic epithelium (20, rectal epithelium exhibited weak immunoreactivity. MMP-2 21). Mast cells express both MMP-2 and MMP-9 (22). immunoreactivity levels were significantly increased in The production of MMP-9 in tissue is linked mainly to specimens obtained during radiation therapy, particularly in the presence of inflammation, whereas MMP-2 seems to subepithelial inflammatory cells and pericryptal myofibro- play more organ- and injury-specific roles. blasts. In contrast, MMP-9 immunoreactivity was similar in The increase in MMP activity in the rectal mucosa during biopsies obtained before and during radiation therapy, de- radiation therapy in the present clinical study is in agree- spite the significantly increased activity levels demonstrated ment with Seifert et al.. who found prolonged postoperative by zymography. gelatinolytic activity in colon anastomoses after intraoper- ative irradiation (14). Kumar et al. (15), on the other hand, who examined MMP-2 and MMP-9 in mucosal biopsies DISCUSSION from seven patients before and 48 h after neoadjuvant MMPs are calcium-dependent enzymes containing zinc that degrade ECM components. In addition, they also

MMP-9

• •

• • • •

ii -+1 ail ... + r 2 weeks 6 weeks 2 weeks 6 weeks (p=0.02) (p=0.006) (p=0.004) (p=0.001)

Fig. 1, Total MMP-2 and MMP-9 gelatinolytic activity (zymogen Baseline 2 weeks 6 weeks plus activated enzyme, relative to preirradiation baseline values) in rectal mucosal biopsies obtained before and during pelvic radiation Fig. 2. Percent activated MMP-2 relative to total (zymogen plus therapy. Small circles indicate individual values obtained by den- activated enzyme) MMP-2 in rectal mucosal biopsies at baseline sitometric analysis of gelatin zymograms, and boxes show median and during ongoing pelvic radiation therapy (means and standard and interquartile range. The horizontal dotted line indicates preir- errors of values obtained by densitometric analysis of gelatin radiation baseline (100%). zymograms). 924 L J. Radiation Oncology • Biology • Physics Volume 48, Number 4, 2000

Fig. 3. Immunohistochemically stained biopsies of rectal mucosa demonstrating the localization of MMP-2 and MMP-9 before pelvic radiation therapy and 2 weeks into the course of pelvic radiation therapy. (A and B) MMP-2 before and during radiation therapy, respectively. Immunoreactivity in pericryptal myofibroblasts, inflammatory cells, and smooth muscle cells. MMP-2 immunoreactivity levels increased significantly during radiation therapy, particularly in subepi- thelial inflammatory cells and pericryplal myofibroblasts. (C and D) MMP-9 before and during radiation therapy, respectively. Immunoreactivity in fibroblasts, inflammatory cells, endothelial cells, and epithelial cells. MMP-9 immunoreactivity was similar in biopsies obtained before and during radiation therapy, despite increased activity as demonstrated by zymography. (B and D) Note prominent glandular atrophy and stromal changes in biopsies obtained during radiation therapy. Original magnification of all images 400X.

radiation therapy for rectal cancer (25 Gy in 5 fractions), did "normal" mucosa adjacent to the tumor exhibited some not find significant changes in MMP activity in the normal reactive changes; and (5) lower statistical power due to mucosa. Possible reasons for the difference between our fewer patients. In addition, Kumar et al. did not report study and theirs include the following: (7) the shorter du- whether their patients experienced symptoms of rectal tox- ration of radiation therapy in their study; (2) the possibility icity and did not describe histologic or inflammatory that, even before commencement of radiation therapy, the changes in their biopsies (15). Metalloprotcinases in radiation proctitis • N. HOVDENAK et al. 925

Radiation proctitis is a result of clonogenic epithelial cell Of the symptoms recorded in the present study, diar- death, breakdown of the mucosal barrier, and mucosal in- rhea and bloody stools are attributable to direct damage flammation. Although in some cell types radiation may of the rectal mucosa, whereas tenesmus, nausea, and upregulate MMPs directly, the increase in mucosal MMPs bloating are more likely the result of local or systemic during radiation therapy that was observed in our study is neuromuscular or humoral effects. Furthermore, it is rea- more likely secondary to inflammation and stromal cell sonable to assume that diarrhea depends particularly activation in the lamina propria. Hence, the production of strongly on the severity of local radiation mucositis, thus both MMP-2 and MMP-9 is induced by a large number of explaining the association of this symptom with greater inflammatory and fibrogenic cytokines and growth factors increases in MMP-2 and MMP-9 activity levels, as well (23). many of which are consistently and prominently over- as the correlation between the MMP activity levels and expressed during the intestinal radiation response. For ex- mucosal granulocyte infiltration. ample, transforming growth factor /31, which is upregulated Despite the significanl and correlative changes in the in both acute and delayed intestinal radiation toxicity (24, activities of MMP-2 and MMP-9 found in the present study, 25), increases the expression of MMP-2 and MMP-9, both it is not clear to what extent these MMPs are directly transcriptionally and post-transcriptionally (23). This mech- involved in the mechanisms of symptomatic, structural, or anism may be partly responsible for the coexistence of functional pathology in radiation proctitis. However, in mucosa] atrophy and bowel wall fibrosis during the delayed addition to the prominent roles of MMP-2 and MMP-9 in phase of intestinal radiation injury, i.e.. net ECM degrada- ECM degradation, cell migration, and inflammation, a num- tion in the compartment that exhibits inflammation, but net ber of additional functions support their putative involve- ECM deposition in compartments where inflammation is ment in intestinal radiation toxicity. For example. MMP-2 less pronounced or absent. causes platelet aggregation through a pathway distinct from All MMPs are synthesized, and most MMPs are secreted those involving the release of endoperoxide/thromboxane as zymogens (proMMPs). Their activity in the extracellular A, and adenosine diphosphate (ADP) (40). Platelet aggre- environment is controlled by activators (26) and TIMPs gation is associated with release of growth factors and other (27). The regulation of MMP activity in vivo is complex and inflammatory mediators, and inhibition of aggregation ame- depends on a number of additional factors, including the liorates intestinal radiation toxicity (41). It is possible that localization of the MMPs. activators, and inhibitors in the increased levels of MMP-2 contribute to enhanced local microenvironmeni. For example, although TIMPs inhibit platelet aggregation in the irradiated bowel. the activity of MMP-2 in the extracellular space, a major Although MMPs are generally considered proinflamma- mechanism of MMP-2 activation at the cell membrane is the tory, they also have anti-inflammatory properties. There- formation of a ternary complex of membrane-type I MMP fore, despite the aforementioned effects of MMPs. the pos- (MT1-MMP), TTMP-2. and proMMP-2, resulting in sibility that increased MMP levels may actually serve to MMP-2 activation by an adjacent MT1-MMP. Thrombin, reduce some aspects of radiation mucositis cannot be ex- activated protein C. and urokinase, on the other hand, po- cluded. For example, MMP-2 cleaves monocyte chemoat- tently activate MMP-2 without requiring MT1-MMP (28- tractant protein-3 (42). Although cleaved monocyte che- 31). Cathepsin G, MMP-2, stromelysin (MMP-3), trypsin, moattractant protein-3 still binds to CC-chemokine and mast cell chymase activate MMP-9 (32-34), whereas receptors-1, -2. and -3, it no longer promotes chemotaxis, plasmin, thrombin, and MMP-1 do not (18). but instead acts as a general chemokine antagonist to Investigation of the relative significance of MMP activa- dampen inflammation. The role of certain prostaglandins in tion and inhibition mechanisms was beyond the scope of the the regulation of MMP-9 expression and release and the present study. However, our results clearly demonstrate that impressive beneficial effect of misoprostol in radiation pelvic radiation therapy causes a net increase (absolute and proctitis may also lend support to the notion that increased relative) in the activation of MMP-2. Radiation therapy MMP activity represents a physiologic response to the nor- renders the endothelium strongly procoagulant as a result of mal tissue radiation insult (43). deficiency of thrombomodulin (35) and induction of tissue factor (36, 37). Hence, mucosal MMP-2 may be activated CONCLUSIONS during pelvic in-adiation by thrombin generated locally at sites of procoagulant endothelium (29). Furthermore, reac- Radiation proctitis is associated with increased mucosal tive oxygen species (ROS) are generated in large amounts MMP-2 and MMP-9 activities that correlate with the pres- by radiation exposure and subsequently by activated inflam- ence of radiation-induced diarrhea and granulocyte infiltra- matory cells. In addition to their nonspecific effects on tion. Further studies should be performed to address the various tissue components. ROS also exert a number of significance of these findings relative to the development of specific actions, including efficient activation of MMP-2 symptomatic, structural, and functional aspects of the intes- (38). Interestingly, peroxynitrite. formed by the combina- tinal radiation response. Pharmacologic modulation of tion of ROS with nitric oxide, causes nitration of MMP-2 MMPs may have a role in the clinic as a method to reduce and endows it with collagenolytic activity (39), thus ex- the severity of proctitis in patients undergoing pelvic radi- panding the spectrum of MMP-2 substrates. ation therapy. 926 L J. Radiation Oncology • Biology • Physics Volume 48, Number 4, 2000

REFERENCES 1. Chambers AF, Matrisian LM. Changing views of the role of human fibrosarcoma cells. Purification and activation of the matrix metalloproteinases in metastasis. J Natl Cancer Inst precursor and enzvmatic properties. J Biol Chem 1992:267: 1997;89:1260-1270. 21712-21719. 2. Parsons SL, Watson SA. Steele RJ. Phase I/1I trial of batimas- 19. Durko M, Brodl P. The metalloproteinases and their inhibi- tat. a matrix mctalloproleinasc inhibitor, in patients with ma- tors. In: Brodt P. editor. Cell adhesion and invasion in cancer lignant ascites. Eur J Surg Oncol 1997;23:526-531. metastasis. New York: Landes Company: 1996. p. 113-150. 3. Otani Y, Sakurai Y. Kameyama K, et al. Matrix metallopro- 20. Gan X, Wong B. Wright SD, et al. Production of matrix teinase gene expression in chronic gastric ulcer: A potential metalloproteinase-9 in CaCo-2 cells in response to inflamma- role of eosinophils in perforation. J Clin Gastroenterol 1997: tory stimuli. J Interferon Cytokine Res 2001;21:93-98. 25(S1):101-104. 21. Tariton JF. Whiting CV, Tunmore D, et al. The role of 4. Daum S. Bauer U, Foss HD, et al. Increased expression of up-regulated serine proteases and matrix meialloproteinases in mRNA for matrix metalloproteinases-1 and -3 and tissue the pathogenesis of a murine model of colitis. Am J Pathol inhibitor of metalloproteinases-1 in intestinal biopsy speci- 2000;157:1927-1935. mens from patients with coeliac disease. Gut 1999;44:17-25. 22. Fang KC, Wolters PJ, Steinhoff M. et al. Mast cell expression 5. Stallmach A. Chan CC. Ecker KW. et al. Comparable expres- of gelatinase A and B is regulated by kit ligand and TGF-p. sion of matrix metalloproteinases I and 2 in pouchitis and J Immunol 1999:162:5528-5535. ulcerative colitis. Gut 2000;47:415-422. 23. Mauviel A. Cytokine regulation of metalloproteinase gene 6. Bailey CJ. Hembry RM. Alexander A. el al. Distribution of expression. J Cell Biochem 1993;53:288-295. matrix metalloproteinases stromelysin, gelatinases A and B. 24. Canney PA. Dean S. Transforming growth factor beta: A and collagenase in Crohn's disease and normal intestine. promotor of late connective tissue injury following radiother- J Clin Pathol 1994;47:113-116. apy? Br J Radiol 1990;63:620-623.' 7. Di Sebastiano P, Di Mola FF. Artese L. et al. Beneficial 25. Wang J. Zheng H. Sung C-C, et al. Cellular sources of effects of batimastat (BB-94), a matrix metalloproteinase in- transforming growth factor /3 (TGF-p) isoforms in early and hibitor, in rat experimental colitis. Digestion 2001;63:234- chronic radiation enteropathy. Am J Pathol 1998:153:1531- 239. 1540. 8. Heuschkel RB. MacDonald TT. Monteleone G, et al. Imbal- 26. Nagase H. Activation mechanisms of matrix metalloprotein- ance of stromelysin-1 and TIMP-1 in the mucosal lesions of ases. Biol Chem 1997;378:151-160. children with inflammatory bowel disease. Gut 2000;47:57- 27. Gomez DE. Alonso DF. Yoshiji H, el al. Tissue inhibitors of 62. metalloproteinases: Structure, regulation and biological func- 9. von Lampe B. Barthel B, Coupland SE. et al. Differential tions. Eur J Cell Biol 1997:74:111-122. expression of matrix metalloproteinases and their tissue inhib- 28. Zucker S, Conner C. DiMassimo BI. et al. Thrombin induces itors in colon mucosa of patients with inflammatory bowel the activation of progelatinase A in vascular endothelial cells. disease. Gut 2000;47:63-73. Physiologic regulation of angiogenesis. J Biol Chem 1995; 10. Nirmala C. Jasti SL, Sawya R. et al. Effects of radiation on the 270:23730-23738. levels of MMP-2. MMP-9 and TIMP-1 during morphogenic 29. Nguyen M. Arkell J, Jackson CJ. Thrombin rapidly and effi- glial-endothelial cell interactions. Int J Cancer 2000:88:766- ciently activates gelatinase A in human microvascular endo- 771. thelial cells via a mechanism independent of active MT1 11. Brooks PC. Broek D, Roth J, et ai. Differential effects of matrix metalloproteinase. Lab Invest 1999:79:467-475. ionizing radiation on angiogenesis and matrix metalloprotein- 30. Nguyen M, Arkell J. Jackson CJ. Activated protein C directly ases (MMPs) activity in vitro and in vivo (Abstr.). Int J Radiat activates human endothelial gelatinase A. J Biol Chem 2000; Oncol Biol Phys 2001:51(S1):156. 275:9095-9098. 12. Zhao W, O'Malley Y, Wei S. et al. irradiation of rat tubule 31. Mazzicri R, Masiero L, Zanetta L, et al. Control of type IV epithelial cells alters the expression of gene products associ- collagenase activity by components of the urokinase-plasmin ated with the synthesis and degradation of extracellular ma- system. A regulatory mechanism with cell-bound reactants. trix. Int J Radiat Biol 2000;76:391-402. EMBO J 1997;16:2319-2332. 13. Wang JL. Sun Y. Wu S. Gamma-irradiation induces matrix 32. Fridman R. Toth M, Pena D, et al. Activation of progelatinase metalloproteinase II expression in a p53-dependent manner. B (MMP-9) bv gelatinase A (MMP-2). Cancer Res 1995:55: Mol Carcinog 2000;27:252-258. 2548-2555. 14. Seifert WF. Wobbcs T, Hoogenhout J, et al. Intra-operative 33. Sang QX, Birkedal-Hansen H. Van Wart HE. Proteolytic and irradiation prolongs the presence of matrix metalloproteinase non-proteolytic activation of human neutrophil progelatinase activity in large bowel anastomoses of the rat. Radial Res B. Biochim Biophys Acta 1995:1251:99-108. 1997:147:354-361. 34. Fang KC. Raymond WW, Lazarus SC, et al. Dog mastocy- 15. Kumar A, Collins HM, Scholefield JH, et al. Increased toma cells secrete a 92-kDa gelatinase activated extracellu- lype-IV collagenase (MMP-2 and MMP-9) activity following larly by mast cell chymase. J Clin Invest 1996:97:1589-15%. preoperative radiotherapy in rectal cancer. Br J Cancer 2000; 35. Richter KK. Fink LM. Hughes BM. et al. Differential effect of 82:960-965. radiation on endothelial cell function in rectal cancer and 16. Hovdenak N, Fajardo LF, Hauer-Jensen M. Acute radiation normal rectum. Am J Surg 1998:176:642-647. proctitis: A sequential clinicopathologic study during pelvic 36. Verheij M, Dewit LGH, van Mourik JA. The effect of ionizing radiotherapy. Int J Radiat Oncol Biol Phys 2000;48:1111- radiation on endothelial tissue factor activity and its cellular 1117. localization. Thromb Haemost 1995:73:894-895. 17 Hanauer SB, Robinson M, Pruitt R, et al. Budesonide enema 37. Richter KK, Fink LM. Hughes BM, et al. Is the loss of for the treatment of active, distal ulcerative colitis and proc- endothelial thrombomodulin involved in the mechanism of titis: A dose-ranging study. U.S. Budesonide enema study chroniciiy in late radiation enteropathy? Radiother Oncol group. Gastroenterology 1998;115:525-532. 1997:44:65-71. 18. Okada Y, Gonoji Y. Naka K, et al. Matrix metalloproteinase 38. Belkhiri A. Richards C, Whalcy M. et al. Increased expression 9 (92-kDa gelatinase/type IV collagenase) from HT 1080 of activated matrix metalloproteinase-2 by human endothelial Metalloprotcinases in radiation proctitis • N. HOVDENAK et al. 927

cells after sublethal H,0? exposure. Lab Invest 1997:77:533- of ADP-induced platelet aggregation by clopidogrel amelio- 539. rates radiation-induced toxicity in rat small intestine. Thromb 39. Rajagopalan S, Meng XP. Ramasamy S, et al. Reactive oxy- Haemost 2002;87:122-128. gen species produced by macrophage-derived foam cells reg- 42. McQuibban GA, Gong JH. Tam EM, et al. Inflammation ulate the activity of vascular matrix metalloproteinases in dampened by gelatinase A cleavage of monocyte chemoat- vitro: Implications for atherosclerotic plaque stability. J Clin tractant protein-3. Science 2000;289:1202-1206. Invest 1996:98:2579. 43. Khan AM. Birk JW, Anderson JC, et al. A prospective ran- 40. Sawicki G, Salas E, Murat J, et al. Release of gelatinase A domized placebo-controlled double-blinded pilot study of mi- during platelet activation mediates aggregation. Nature 1997: soprostol rectal suppositories in the prevention of acule and 386:616-619. chronic radiation proctitis syndrome in prostate cancer pa- 41. Wang J, Albertson CM, Zheng H, et al. Short-term inhibition tients. Am J Gastroenterol 2000;95:1961-1966.

Paper III

* Taylor&Francis ORIGINAL ARTICLE • healthsciences

Profiles and Time Course of Acute Radiation Toxicity Symptoms during Conformal Radiotherapy for Cancer of the Prostate

Nils Hovdenak, Åsa Karlsdottir, Halfdan Sørbye and Olav Dahl

From the Division of Gastroenterology (N. Hovdenak) and the Section of Oncology (A. Karlsdottir, H. Sørbye, 0. Dahl), Institute of Medicine, Haukeland University Hospital, University of Bergen, Norway

Correspondence to: Nils Hovdenak, Institute of Medicine, Haukeland University Hospital, N0-5021 Bergen, Norway. Tel: +47 55972130. Fax: +47 55972950. E-mail: [email protected]

Acta Oncologica Vol. 42, No. 7, pp. 741-748, 2003

Symptoms of gastrointestinal toxicity are dose-limiting for pelvic radiotherapy (RT). Existing toxicity registrations (RTOG/EORTC) are helpful in defining maximal tolerated doses, but tend to underestimate the total toxicity burden by excluding several minor complaints. We have applied a more detailed and quantitative recording of symptoms and related these scores to RT-induced endoscopic and histopathologic changes. Prevalence and severity of specific toxicity symptoms were recorded before, during (weeks 2 and 6) and 2 and 8 weeks after RT in 96 patients undergoing external beam RT for localized prostate cancer. RTOG/EORTC acute toxicity and ad hoc total toxicity scores (TTS) were recorded. TTS scores were calculated by adding scores based on visual analog scale (VAS) grading of individual symptoms. Fifty of the patients also underwent sequential proctoscopy with mucosa] biopsy. Individual symptoms increased, but differed in prevalence and intensity during and after RT. TTS increased during the entire treatment course in spite of normalizing histopathologic and endoscopic changes from week 2 onwards. Twenty-seven patients had no RTOG/EORTC toxicity, four had grade 3 and none had grade 4 toxicity. All patients with grade 0 had increased TTS. Thus, TTS appeared more sensitive than RTOG/EORTC scoring. The study demonstrates that multiple toxicity symptoms contribute to total toxicity in response to pelvic RT. TTS is a feasible and sensitive method for detecting and quantifying acute toxicity and unveils morbidity which remains hidden with the RTOG/EORTC score system. The development and timing of symptoms may give clues to pathogenesis, treatment, and prophylaxis.

Received 6 November 2002 Accepted 28 May 2003

Symptoms of acute radiation toxicity occur in the majority recently reviewed (9). Apart from acute mucosal inflamma- of patients during radiotherapy (RT) for malignant pelvic tion. dysmotility seems to play an important role. Current disease. The symptoms vary in intensity and character, and methods of toxicity recording have low sensitivity. The aim may be sufficiently severe to forestall optimal completion of of the present study was to assess the prevalence, severity the planned treatment. Radiation toxicity depends on total and time related progression of abdominal symptoms irradiation dose, fraction size, fraction interval, treatment during pelvic RT by accurate recording of individual time, and treatment volume. The severity of acute radiation symptoms and calculation of a score for total toxicity toxicity symptoms is one of the main predictors for the burden. The results are compared to the existing RTOG development of chronic rectal injury (1, 2) and for tumor acute toxicity score and related to RT-induced endoscopic response (3, 4). Much effort is therefore applied in and histopathological changes, and discussed in relation to individual tailoring of the treatment, and several measures previously known factors that may be of pathogenic have been developed to minimize extra-tumoral irradiation importance. (positioning of the patient, special treatment boards ('belly board'), surgical abdominopelvic partitioning with omen- MATERIAL AND METHODS tum or displacement of the gut by intra-abdominal absorb- able meshes). Pharmacologic prophylaxis and treatment Radiotherapy have so far not been very helpful (5, 6). Patients with organ confined prostate cancer (Tlc-T3, N0- Symptoms tend to increase in frequency during RT and X, M0) scheduled to undergo curative pelvic RT were decrease after cessation of radiation (7, 8). The mechanisms included consecutively. Patients with significant gastroin- by which symptoms are elicited are poorly understood, as testinal disease (inflammatory bowel disease, gastrointest-

© Taylor & Francis 2003. ISSN 0284-186X Acta Oncologica DOI: 10.1080/02841860310011302 742 JVC Hovdenak et al Acta Oncologica 42 (2003)

mal cancer, coeliac disease) were not included. All patients bleeding on examination) or peteciae; and 3 = spontaneous underwent conformal external beam RT administered as hemorrhage or visible ulcers. single daily doses of 2 Gy to a total dose of 66-70 Gy over 6|-7 weeks, typically as four fields, given five days a week Histology with weekend breaks. The treatment was planned individu- During proctoscopy, rectal mucosal cup biopsies were ally and was given as 6-18 MV photons. T1 and T2 tumors obtained from the posterior rectal wall 10 cm above the were treated with a margin of 10 mm against the rectum, T3 anal verge. The results from quantitative histology have tumors were initially given whole pelvic field up to 50 Gy, been reported previously (12). followed by radiation to the prostate only, with 5 mm margin against the rectum. The rectum was shielded by Ethics customized blocks or multileave collimators. Calculated doses to the anterior and posterior rectal wall at the level of The protocol was approved by the Regional Committee for the mucosal biopsy were 100% and 40-60% of prescribed Medical Research Ethics. Written, informed consent was dose, respectively, and in no case more than one third of the obtained from all patients before inclusion in the study. rectal circumference was included in the field. Statistics Symptoms Symptom severity, endoscopic scores, and TTS at different timepoints were analyzed with paired t-tests using Graph- Symptoms were recorded prior to treatment (referred to as Pad Prism (GraphPad Software Inc., USA). Symptom 'week 0'), at 2, 4, and 6 weeks into treatment (referred to as prevalences were analyzed using McNemar's test. Statistical 'week 2', 'week 4', and 'week 6', respectively), and 2 weeks significance was defined as p-values less than 0.05. and 8 weeks after end of treatment (referred to as 'week 9' and 'week 15', respectively). The prevalences of abdominal pain, loose stools, macroscopic blood in the stools, tenes- RESULTS mus, bloating, and fecal incontinence were recorded in all A total of 96 patients were included (mean age: 65.6+6.9 patients. The last 47 patients examined also graded the years, range: 49-78 years). Generally, symptoms appeared severity of symptoms (pain, loose stools, tenesmus, bloat- early and then gradually increased in prevalence (Fig. 1) ing, and nausea), using a visual analogue scale (VAS). A and severity (Fig. 2) during the rest of the treatment. No symptom registration form was filled in at each visit based patient had to reduce, postpone, or stop RT because of on the doctor's detailed interview. The VAS scales were radiation toxicity. marked by the patient supervised by the doctor. An ad hoc global grading based on these symptoms was constructed by adding the VAS scores of each symptom at each time- Loose stools point, expressing a 'total symptom score' (TSS). The Loose stools were an early and the most common symptom. RTOG/EORTC acute toxicity scale (10), shown in Table The prevalence increased significantly up to week 2 and was 1, was applied in all patients. still present in the majority even 2 months after cessation of treatment. The subjective discomfort occurred early and Endoscopy was considerable as expressed on the VAS scale. Rigid proctoscopy was performed at week 0, week 2, and week 6. Endoscopic appearance was assessed using a four- Tenesmus point scoring system (11, 12) where 0 = normal mucosa; Tenesmus increased significantly in prevalence during the 1 = edema and/or loss of normal vascular pattern, granu- entire treatment period, occurred in one fifth of the patients, larity of the mucosa; 2 = friable mucosa (visible, induced and was still prevalent at week 15 (p < 0.05). However, most

Table 1 RTOGIEORTC acute radiation morbidity scoring criteria for pelvic radiation (10)

1 3

No Increased frequency or Diarrhoea requiring parasym- Diarrhoea requiring parenteral Acute or subacute obstruction, change change in quality of bowel patholytic drugs/mucous discharge support/severe mucous or fistula or perforation; G1 habits not requiring medi- not necessitating sanitary pads/ blood discharge necessitating bleeding requiring transfusion; cation/rectal discomfort not rectal or abdominal pain requiring sanitary pads/abdominal dis- abdominal pain or tenesmus requiring analgesics analgesics tension (flat radiograph de- requiring tube decompression monstrates distended bowel or bowel diversion loops) Acta Oncologica 42 (2003) Acute radiation proctitis 743

Fig. 1 Prevalence of toxicity symptoms (mean) before, during (weeks 2 and 6), and after (weeks 9 and 15) radiotherapy. The p-values refer to pretreatment. *** = p <0.001, " =p <0.01, * =p <0.05

of the patients scored low on tenesmus intensity on the VAS VAS scales was generally low, but increased significantly scale. during the later part of the treatment, reaching its maximum 2 weeks after end of RT (week 9), but was not Nausea significantly increased at week 15 compared to pre-treat- ment. The prevalence of nausea increased significantly early and tended to increase further during treatment (p = 0.09). persisting unchanged at week 15. However, this appeared Bloating to be a troublesome symptom only for a few patients, the Bloating is a common symptom in the general population majority scoring low on the VAS scale. and occurred in over 30% of patients before treatment. However, compared to week 0 a significantly increased Abdominal pain prevalence was observed during the entire treatment and The prevalence of abdominal pain increased significantly post-treatment period, although the subjective discomfort from week 0 to week 2 and week 4, but did not change as expressed by the VAS recordings was not increased at any significantly thereafter. Pain intensity as recorded on the time. 744 JVC Hovdenak et al Acta Oncologica 42 (2003)

Bloaing

%M

Fig. 2. Symptom severity expressed as VAS-scores (mean) before, during (weeks 2 and 6), and after (weeks 9 and 15) radiotherapy *** - p < 0.001, " = p<0.01. * = p<0.05.

Blood in the stools below this during the RT course. One patient scored Blood in the stools was seldom recorded. However, a consistently 0 throughout the RT. significantly increased prevalence occurred early during Grade 3 RTOG/EORTC acute gastrointestinal toxicity the treatment period and was maintained at week 15. occurred in four patients: in three at week 6 and in one at week 9. None experienced grade 4 toxicity (Table 2). 19 patients had maximum RTOG/EORTC grade 2 symptoms, Fecal incontinence and 30 patients had maximum grade 1 symptoms, while 43 Fecal incontinence (Fig. 3) was recorded in a few patients at patients had no or only minor symptoms, not recorded by week 2, without any further increase thereafter. No data the RTOG/EORTC toxicity scale. From week 2 no sig- from week 9 and week 15 were recorded. nificant change was observed in RTOG/EORTC grade.

The TSS score (Fig. 4) increased significantly during the radiation course. The scores had not fully returned to Endoscopy pretreatment values at week 15 (p = 0.07). If the 25% All patients examined had normal pretreatment endoscopic percentile was arbitrarily used as a cut off for clinical findings (Fig. 5). Scores of radiation injury were maximal significance, only five patients (11%) scored consistently after two weeks. Two weeks after cessation of radiation Acta Oncologica 42 (2003) Acute radiation proctitis 745

Table 2 Fecal incontinence RTOG grading during and after radiotherapy

RTOG Week 2 Patients Week 6 Patients Week 9 Patients grade (%) (%) (%)

0 63 (66) 41 (43) 55 (57) 1 26 (27) 35 (36) 27 (28) 2 7(7) 17(18) 13(13) 3 0 (0) 3(3) 1 (1) 4 0(0) 0(0) 0(0)

weeks

Fig. 3. Fecal incontinence before and during radiotherapy- ** = p<0.01, * = p<0.05. No follow-up data from week 15 were available.

Total symptom score (TSS)

9 15 weeks

Fig. 5. Endoscopy scores before and during radiotherapy.

early during ongoing pelvic RT and are not attenuated until several weeks after the end of treatment. The individual symptoms as well as the total toxicity burden caused by these symptoms are significant and have an impact on daily well-being. Loose stools are common and cause considerable and continuing discomfort, while bloating, nausea, and abdom- inal pain do not cause much concern. Rectal bleeding does occur in a small proportion, usually without causing Fig. 4. The total symptom scores (TTS) before, during (weeks 2 and 6), and after (weeks 9 and 15) radiotherapy. The p-values refer anemia, and apparently unrelated to inflammation. Fecal to pretreatment. *** = p < 0.001. incontinence is a very troublesome symptom in a significant number of patients during and after RT. The frequency (week 9) a significant reduction compared to week 6 was increases after the end of treatment (13), and is one of the observed (p <0.01). most incapacitating symptoms in chronic radiation procti- tis. Biopsies The RTOG/EORTC acute toxicity scoring system (10) The histopathologic changes have been published pre- provides a global assessment and does not differentiate viously (12), showing an early, statistically significant between the prevalent symptoms. When expressed by this increase in inflammation, with attenuation after week 2 system, toxicity appears low, as shown in previous studies despite ongoing RT. At week 6 inflammatory parameters (14-16). Koper et al. (17) found no grade 3 or 4 acute were not significantly different from pretreatment. toxicity among 266 prostate cancer patients. Only 5-10% of patients will experience symptoms of RTOG/EORTC grade 3 or 4 (18, 19). In our study only four patients (4%) had DISCUSSION grade 3, and none grade 4. The four grade 3 patients had The present study gives a detailed, sequential, and overall high TSS scores. In contrast to the single asymptomatic picture of the acute side effects of pelvic RT for prostate patient with no increase in TSS, 43% scored RTOG/ cancer. Most symptoms of acute radiation toxicity increase EORTC = 0. 746 JVC Hovdenak et al Acta Oncologica 42 (2003)

The use of the ad hoc grading system as a severity index 3) Dysmotility. There are few reports on motility studies assumes that the symptoms considered were of equal in patients with acute radiation enteropathy. Radio- clinical importance. This assumption and the lack of formal nucleic absorption tests (25) lend support to motility validation limit the applicability of the system. However, it disturbances, probably through local neuroendocrine demonstrates a steady increase in the total toxicity burden mechanisms. Changes in gastric emptying rate are throughout the treatment course. Two months after cessa- independent of radiation field (pelvic, thoracic, ab- tion of RT. the TSS had abated and was not significantly dominal) (26), suggesting neurogenic mechanisms. different from pretreatment values, which is in agreement Most studies on motility have concentrated on the with other studies (8, 18, 20). However, not all patients were small intestine. Results from such studies may, how- symptom-free at week 15, a trend was observed (p = 0.07) ever, to a certain extent be applicable to colonic towards persistence of symptoms, of which loose stools and phenomena. Experimental animal studies do not abdominal pain were dominant. In contrast to the RTOG/ necessarily reflect the situation in humans undergoing EORTC score system, our method appears more feasible for highly fractionated RT. Usually, in animal studies the clinical setting by showing a differentiated picture and radiation is given in one dose to the abdomen, by quantifying the total toxicity burden even in patients resulting in profound motility changes (27). An who do not present increased RTOG/EORTC scores. The increased incidence of giant colonic migrating con- results underline the need for more sensitive registration of tractions coinciding with episodes of diarrhoea is seen acute pelvic radiation toxicity. High sensitivity score during fractionated radiation (28, 29). Release of 5-HT systems are important tools when comparing different induces vomiting and diarrhoea. The antiemetic and treatment regimes. Recently, after the completion of the antidiarrhoeic effects of specific 5-HT3 receptor current study, an elaborate, prostate cancer specific, ques- antagonists during RT indicates an increased 5-HT tionnaire-based programme (21), encompassing urinary, release (30, 31). gastrointestinal, and sexual functions has been proposed. 4) Luminal factors. Most probably, hostile luminal fac- Validation of the programme demonstrates high reliability tors (bacteria, viruses, enzymes, antigens) influence the and responsiveness. This programme has recently been used morphologic and functional changes in the gut wall in a prospective evaluation after conformal RT for prostate during radiation. No study on colonic intraluminal cancer in 287 patients (22) and in 267 patients treated by the microbiological milieu during radiation is available. urethral catheter BeamCath® technique (20). Radiation-induced decrease in fecal bifidus excretion The development of acute clinical radiation toxicity is was reversed by oral application of L. acidophilus (32), multifactorial and should be sought in the pathogenesis of and patients receiving oral L. acidophilus prophylaxis the intestinal injury. Important mechanisms considered are: had reduced incidence of diarrhoea during pelvic RT (1) inflammation; (2) malabsorption/hypersecretion; (3) (33). dysmotility; and (4) changes in luminal factors. These mechanisms are interdependent through an extensive A new categorization of intestinal radiation injury has 'cross-talk' involving neural, endocrine, and paracrine recently been proposed (34), in which three basic effects are signalling. described that alone or in combination underlie the pathophysiology and the clinical picture. The cytocidal 1) Inflammation. Endoscopically, inflammation is rapidly effects occur within few days after radiation exposure in attenuated after cessation of RT (Fig. 5). Histopatho- tissues with rapid cell renewal, such as epithelial surfaces. logically, the inflammation in the superficial mucosal Indirect effects occur as results of damage to other tissue layers decreases from week 2 in contrast to the time- structures or as 'by-stander' phenomena, while functional dependent development of symptoms. However, per- effects reflect changes in molecular and gene expression in sisting inflammation was demonstrated in the lamina damaged tissue. Radiation injury occurs as a result of propria and the glands (12), possibly contributing to interaction of several mechanisms. The acute inflammation the clinical picture. Following a single radiation dose, reflects both cytocidal, functional, and indirect effects, thickening of the muscularis mucosae has been de- namely rapid development of injury, activation of stress monstrated (23), but no systematic data exist on protein genes, and upregulation of cytokines, growth inflammatory changes in or around the smooth muscle factors, and chemokines. layers or in the ganglia of the enteric nervous system The progression from acute to chronic radiation injury during fractionated RT. occurs in 5-20% (5) and is of great clinical significance. 2) Malabsorptionlhypersecretion. Radiation-induced im- Detailed studies of the acute reaction is of basic importance pairment of water and electrolyte absorption has been in this issue as modifications by pharmacotherapy or other demonstrated in rat experiments (24). Injury of the measures during the acute response may influence the risk small bowel may result in malabsorption of bile acids, of developing late injury. Two or more simultaneously vitamin B12, fat, and glucose. occurring biomolecular changes may have opposite effects Acta Oncologica 42 (2003) Acute radiation proctitis 747

oil gut motility, the significance of changes in morphology 11. Hanauer SB, Robinson M, Pruitt R, et al. Budesonide enema and/or function in different cells and compartments is for the treatment of active, distal ulcerative colitis and proctitis: unknown, and the 'cross-talk' between cells and tissues is a dose-ranging study. U.S. Budesonide enema study group. Gastroenterology 1998; 115: 525-32. only partly understood. Therapeutic and prophylactic 12. Hovdenak N, Fajardo LF, Hauer-Jensen M, Acute radiation intervention strategies depend on characterization of me- proctitis: a sequential clinicopathologic study during pelvic chanisms of injury at a cellular and molecular level. radiotherapy. Int J Radiat Oncol Biol Phys 2000; 48: 1111-7. In summary, we have demonstrated a differential devel- 13. Yeoh EK., Russo A, Botten R, et al. Acute effects of therapeutic opment of acute pelvic radiation toxicity symptoms during irradiation for prostatic carcinoma on anorectal function. Gui 1998; 43: 123-7. and immediately after the radiation course, and an increas- 14. Liu L, Glicksman AS. Coachman N, Kuten A, Low acute ing total symptom burden in most patients, not readily gastrointestinal and genitourinary toxicities in whole pelvic picked up by the conventional RTOG/EORTC registration irradiation of prostate cancer. Int J Radiat Oncol Biol Phys system. A simple but detailed and clinically feasible 1997; 38: 65-71. registration method has been used. Most of the symptoms 15. Tait DM, Nahum AE, Rigby L, et al. Conformal radiotherapy may reflect motility disturbances, some of which may be of the pelvis: assessment of acute toxicity. Radiother Oncol secondary to radiation induced inflammation, but for the 1993; 29: 117-26. 16. Zierhut D. Flentje M, Sroka-Perez G, Rudat V, Engenhart- majority not directly related to the inflammatory changes. Cabillic R, Wannenmacher M. The conformal radiotherapy of A detailed description of the individual symptoms illus- localized prostatic carcinoma: acute tolerance and early trates different pathogenic mechanisms as targets for efficacy. Strahlenther Onkol 1997; 173: 98-105. development of specific pharmacotherapy and prophylaxis 17. Koper PC, Stroom JC, van Putten WL, et al. Acute morbidity and may give clues for future research. reduction using 3DCRT for prostate carcinoma: a randomized study. Int J Radiat Oncol Biol Phys 1999; 43: 727-34. 18. O'Brien PC, Franklin CI, Dear KB, et al. A phase III double- blind randomised study of rectal sucralfate suspension in the ACKNOWLEDGEMENTS prevention of acute radiation proctitis. Radiother Oncol 1997, The study was in part supported by The Norwegian Cancer Society. 45: 117-23. 19. Kumar A. Collins HM, Scholefield JH, Watson SA. Increased type-IV collagenase (MMP-2 and MMP-9) activity following REFERENCES preoperative radiotherapy in rectal cancer. Br J Cancer 2000: 82: 960-5. 1. Denham JW, O'Brien PC, Dunstan RH, et al. Is there more 20. Fransson P. Lofroth PO, Franzen L, Henriksson R. Bergstrom than one late radiation proctitis syndrome? Radiother Oncol P, Widmark A. Acute side effects after dose-escalation treat- 1999; 51: 43-53. ment of prostate cancer using the new urethral catheter 2. Schultheiss TE, Lee WR. Hunt MA, Hanlon AL, Peter RS, BeamCath techniquem. Acta Oncol 2001; 40: 756-65. Hanks GE. Late GI and GU complications in the treatment of 21. Fransson P. Tavelin B, Widmark A. Reliability and responsive- prostate cancer. Int J Radiat Oncol Biol Phys 1997; 37: 3-11. ness of a prostate cancer questionnaire for radiotherapy- 3. Dahl O, Horn A, Mella O. Do acute side-effects during induced side effects. Support Care Cancer 2001, 9: 187-98. radiotherapy predict tumour response in rectal carcinoma? 22. Fransson P, Bergstrom P, Lofroth PO, Widmark A. Prospective Acta Oncol 1994; 33: 409-13. evaluation of urinary and intestinal side effects after BeamCath 4. Wang CJ, Leung SW, Chen HC, et ai. The correlation of acute stereotactic dose-escalated radiotherapy of prostate cancer. toxicity and late rectal injury in radiotherapy for cervical Radiother Oncol 2002; 63: 239-48. carcinoma: evidence suggestive of consequential late effect 23. Summers RW, Hayek B. Changes in colonic motility following (CQLE). Int J Radiat Oncol Biol Phys 1998; 40: 85-91. abdominal irradiation in dogs. Am J Physiol 1993; 264: 5. Babb RR. Radiation proctitis: a review. Am J Gastroenterol G1024-30. 1996; 91: 1309-11. 24. Francois A, Dublineau I, Lebran F, Ksas B, Griffiths NM. 6. Hong JJ, Park W, Ehrenpreis ED. Review article: current Modified absorptive and secretory processes in the rat distal therapeutic options for radiation proctopathy. Aliment Phar- colon after neutron irradiation: m vivo and in vitro studies. macol Ther 2001; 15: 1253-62. Radiat Res 1999: 151: 468-78. 7 Tait DM, Nahum AE, Meyer LC, et al. Acute toxicity in pelvic 25. Yeoh EK, Lui D, Lee NY. The mechanism of diarrhoea radiotherapy; a randomised trial of conformal versus conven- tional treatment. Radiother Oncol 1997; 42: 121 -36. resulting from pelvic and abdominal radiotherapy; a prospec- 8. Widmark A, Fransson P, Franzen L, Littbrand B, Henriksson tive study using selenium-75 labelled conjugated bile acid and R. Daily-diary evaluated side-effects of conformal versus cobalt-58 labelled cyanocobalamin. Br J Radiol 1984; 57: conventional prostatic cancer radiotherapy technique. Acta 1131-6. Oncol 1997: 36: 499-507. 26. Makrauer FL, Oates E, Becker J, et al. Does local irradiation 9. Classen J, Belka C, Paulsen F, Budach W, Hoffmann W, affect gastric emptying in humans? Am J Med Sci 1999; 317: Bamberg M. Radiation-induced gastrointestinal toxicity. 33-7. Pathophysiology, approaches to treatment and prophylaxis. 27. Summers RW, Flatt AJ, Prihoda MJ, Mitros FA. Effect of Strahlenther Onkol 1998; 174(Suppl 3): 82-4. irradiation on morphology and motility of canine small 10. Cox JD, Stetz J, Pajak TF. Toxicity criteria of the Radiation intestine. Dig Dis Sci 1987; 32: 1402-10. Therapy Oncology Group (RTOG) and the European Organi- 28. Erickson BA, Otterson MF. Moulder JE, Sarna SK. Altered zation for Research and Treatment of Cancer (EORTC). Int J motility causes the early gastrointestinal toxicity of irradiation. Radiat Oncol Biol Phys 1995; 31: 1341-6. Int J Radiat Oncol Biol Phys 1994; 28: 905-12. 748 JVC Hovdenak et al Acta Oncologica 42 (2003)

29. Otterson MF, Sarna SK, Leming SC, Moulder JE, Fink JG. irradiation abnormal bacterial flora and intestinal irradiation Effects of fractionated doses of ionizing radiation on colonic reaction through Bacterium bifidus substitution therapy. motor activity. Am J Physiol 1992; 263: G518-26. Strahientherapie 1973; 145: 588-99. 30. Bremer K. 5-Hydroxytryptamine (serotonin) subtype 3 antago- 33. Salminen E, Elomaa I, Minkkinen J, Vapaatalo H, Salminen S. nists, a major step in prophylaxis and control of cytostatic and Preservation of intestinal integrity during radiotherapy using radiation-induced emesis. J Cancer Res Clin Oncol 1991; 117: live Lactobacillus acidophilus cultures. Clin Radiol 1988; 39: 85-7. 435-7. 31. Logue JP, Magee B, Hunter RD, Murdoch RD. The antiemetic 34. Denham JW, Hauer-Jensen M, Peters LX Is it time for a new effect of gramsetron in lower hemibody radiotherapy. Clin formalism to categorize normal tissue radiation injury? Int J Oncol (R Coll Radiol) 1991; 3: 247-9. Radiat Oncol Biol Phys 2001; 50: 1105-6. 32. Mettler L, Romeyke A, Brieler G. Effects of para- and post- Paper IV

Sucralfate does not ameliorate acute radiation

proctitis

Nils Hovdenak, Halfdan Sørbye, Olav Dahl.

Division of Gastroenterology and Section of Oncology, Institute of Medicine,

Haukeland University Hospital, University of Bergen, Norway

Correspondance: Nils Hovdenak, Institute of Medicine, Bergen University Hospital, N-5021

Bergen, Norway.

Running title: Sucralfate in radiation proctitis. Abstract

During pelvic radiotherapy (RT) many patients develop radiation induced gastrointestinal symptoms which may interfere with the treatment. Prophylaxis during RT should ideally prevent both the acute reaction and the development of delayed injury.

Sucralfate, an aluminium sucrose octasulphate, has been used for both acute and delayed radiation side-effects. However, conflicting results have been published. We report here a prospective, randomised, placebo-controlled study of prophylactic sucralfate during pelvic

RT. In addition, a meta-analysis of available data from the literature has been carried out.

Fifty-one patients with localised pelvic tumours scheduled for curative conformal pelvic RT (total dose 64-70 Gy over 6 1/2-7 weeks in 2 Gy daily fractions) were included.

Peroral sucralfate 2 g 3 times daily or identically appearing placebo tablets were administered during the RT course. Symptom registration, endoscopy, and biopsies were performed immediately before RT, 2 weeks and 6 weeks into the treatment course, and 2 weeks after the completion of RT. Mucosal cup forceps biopsies were obtained through a rigid proctoscope.

Graded endoscopic appearance and quantitative histology were registered. An interim analysis of 44 evaluable patients showed significantly increased diarrhoea in the sucralfate group and the trial was stopped. No difference was seen in other symptoms, endoscopic appearance, or histology.

A meta-analysis comprising 5 published studies showed no statistically significant beneficial effect of sucralfate on acute symptoms.

We conclude that sucralfate can not be recommended in the prophylaxis of acute radiation proctopathy and may even worsen the symptoms. Introduction

Acute radiation proctitis occurs during external pelvic radiotherapy (RT) for malignant diseases such as prostate cancer and gynecologic malignancies. The majority of patients experience symptoms of varying intensity, sometimes interfering with the treatment and reducing the quality of life during and for some time after the radiation course (1,2).

Radiation proctitis is the consequence of radiation exposure of the intestinal mucosa, which is especially vulnerable owing to its rapid cellular renewal. In 5-20% of patients delayed structural changes appear months or years after the RT, causing considerable morbidity and even mortality (3).

Prophylaxis during RT has a dual goal: attenuation of the acute radiation proctitis, and amelioration or prevention of the development of chronic radiation-induced structural changes. Much effort has been applied to shielding normal tissue, especially the gut during radiation therapy. This can be accomplished by meticulous dose-planning, new radiation techniques, such as conformal radiation (4), intensity modulated radiation therapy (5-7), and in some cases also by pretreatment surgical displacement of intestine (8). Pharmacologic treatment or prophylaxis has, however, not been very successful.

Sucralfate, an aluminium sucrose octasulphate, has previously been used successfully

in the treatment of peptic ulcer (9). It adheres to damaged epithelium, constituting a

mechanical shielding against aggressive luminal factors. Several biological effects have been

demonstrated, such as binding of basic fibroblast growth factor (bFGF) at the site of injury

(10) and stimulation of somatostatin release (11). On a theoretic basis, sucralfate therefore

appears to be an ideal drug against acute intestinal radiation injury. However, conflicting

clinical results have been reported on the effect of sucralfate on acute radiation intestinal

injury. We present our own data from a randomised clinical trial on the use of sucralfate during pelvic RT, together with compiled data from previous studies in a meta-analysis, where we focus on diarrhoea as the main side effect of pelvic RT.

Material and methods

1) The study was prospective, placebo-controlled, randomised, double blind, and compared peroral placebo and sucralfate as prophylaxis during RT for pelvic cancer.

Symptom registration and proctoscopy with rectal mucosal biopsy were carried out at each

Patients: Consecutive patients with localised pelvic tumour scheduled for curative pelvic

RT were included. Patients with significant current or previous gastrointestinal disease

(ulcerative colitis, Crohn's disease, coeliac disease) were excluded.

Radiation treatment: External beam RT was administered as single daily doses of 2 Gy to a total dose of 64-70 Gy over 6,5 - 7 weeks given 5 days per week. Individual CT-based planning was used. Customised blocks or multileave collimators were used for rectal shielding. The patients were treated in the supine position. Prostate tumours penetrating the capsule (T3, T4), tumours of the urinary bladder and uterine cervical cancers were treated with pelvic fields to 50 Gy followed by a boost, while T1 and T2 prostate tumours had limited treatment volumes with maximal margins of 2cm (Table 1).

Drugs: The patients received 2 tablets 3 times daily with either sucralfate (1 gram per tablet) or identically appearing placebo tablets, starting on the first day of treatment and continuing throughout the RT course. The drug (Antepsin®) and the placebo tablets were generously supplied by Orion Pharma A/S, Oslo, Norway.

Symptom registration was carried out before RT and after 2, 4, and 6 weeks (referred to as

"week 0", "week 2", "week 4", and "week 6", respectively) into the treatment course and 2 weeks after completion of the RT ("week 9"). The_doctnr filled in n questionnaire, at these visits and the patient reported in a diary through one week before each visit. Recorded symptoms were: abdominal pain, diarrhoea, tenesmus, and bloating. The severity of each symptom was graded 1-4 by the patient (1= absent; 2= mild; 3= moderate; 4= severe).

Endoscopy was carried out at week 0, week 2, and week 6, using a rigid proctoscope.

Endoscopic mucosal injury was scored according to previous reports (12,13). Mucosal biopsies were obtained by cup forceps from the posterior rectal wall, 10 cm proximal to the anal verge. The specimens were formalin fixed and paraffin embedded for quantitative histopathology.

Statistics: Symptom data were analysed by ANOVA-analysis and two-sided t-test where appropriate. A p-value<0.05 was considered statistically significant. A sample size of 114 patients (57 per group) were planned included to show a 30% difference (power 90%) between the groups.

Ethics: The study was approved by the Regional Committee for Medical Research Ethics, complying with health authorities' regulations. Written, informed consent was obtained from all patients before inclusion in the study.

2) A literature search was carried out through Medline/PubMed covering the years 1990 through 2002, using the key-words "sucralfate" and "radiation". Seven randomised studies were retrieved. Our own results were combined with these in a global meta-analysis using the published data in the Comprehensive Metaanalysis data package (Biostat, Inc., Englewood,

New Jersey, USA). Diarrhoea is usually reported as the main side effect. We therefore focused on this complaint in the meta-analysis. Results

The randomised study

Fifty-one patients were eligible and included (sucralfate: 24; placebo: 27). Of these, 44 patients adhered to the study protocol and completed the RT (sucralfate: 20; placebo: 24). In the sucralfate group, one patient stopped the medication because of constipation, while 3 patients did not adhere to the study protocol. In the placebo group, RT was stopped because of disease progression in one patient, while 2 patients dropped out because of drug non- compliance. No drug-related side-effects other than constipation were observed. Patient characteristics are shown in Table 1. There was no difference between the groups with respect to age, pretreatment bowel symptoms and endoscopic appearance.

Due to negative reports in the literature, an unscheduled interim analysis was performed after the inclusion of 51 patients. During RT, an insignificant increase was observed in abdominal pain, tenesmus, and bloating in both groups, but no difference was seen between the groups (Figure 1).

Diarrhoea (Figure 2) increased significantly during the RT course in both the sucralfate and the placebo group (p= 0.001, and p= 0.005, respectively). At both week 2 and week 6, however, mean diarrhoea score was significantly higher (p=0.049 and p=0.033, respectively) among the sucralfate-treated patients than in the placebo group. The combined frequencies of grade 3 + 4 at week 2 were: 25% and 12.5%, and at week 6: 55% and 16.7% for sucralfate and placebo, respectively.

Although the number of patients was low, data were also dicotomised according to estimated radiation volumes: pelvic fields or limited fields (Table 1). In the group treated with limited fields (N=15), diarrhoea score increased significantly during the RT course in the sucralfate group (p= 0,046), but not in the placebo group. In patients with pelvic fields

(N=29), both the sucralfate group and the placebo group had increasing diarrhoea scores during RT (p= 0.003 and p= 0,003, respectively), but no difference was observed between the groups, although a trend towards more diarrhoea was found at week 4 and week 6 (p=0.08).

Based on data from the literature and the interim results, the trial was prematurely terminated as further inclusions were considered unlikely to change the results into benefit for the sucralfate group.

Data from endoscopy and histology have been published earlier (12), showing no statistically significant differences between the sucralfate and the placebo group.

Meta-analysis and previous studies

Five studies were identified, including a total of 699 patients (14-18). All studies were of a randomised design giving data on the frequency of gastrointestinal symptoms (i.e. diarrhoea) during radiation. All estimates were extracted from full articles. Statistical results were reported as p-values in per protocol analysis in all articles. In order to make the different studies comparable and allow uniform calculations we have categorised the extracted estimates as the frequency of Grade 2-3 diarrhoea. The results from the meta-analysis are shown in Figure 3. The combination of all studies (including our own data) did not disclose any significant benefit in favour of prophylactic effect of the drug. Two additional randomised, double-blind, placebo-controlled studies reported data that were unsuitable for inclusion in the meta-analysis: Stellamans et al. (19) found no statistically significant difference in diarrhoea between sucralfate and placebo in 80 patients. In a study of 120 patients by Vails et al. (20) no difference was seen in the number of stools per week, but placebo patients had significantly less formed stools and required more loperamide tablets. 8

Discussion

Diarrhoea is the main symptom during pelvic radiotherapy (2), affecting per se the wellbeing of patients. Diarrhoea is also contributing to fecal incontinence, which is the most perplexing complaint during and after RT. It is therefore reasonable to consider diarrhoea as a main parameter in evaluating the efficacy of medical prophylaxis.

Conflicting data exist on the prophylactic use of sucralfate in acute radiation proctitis.

Neither the presented meta-analysis, nor the present clinical study support such use for the endpoint diarrhoea. Our results show that sucralfate in acute radiation proctitis is ineffective and may even increase symptoms. Compared to placebo, a statistically significant increase in diarrhoea was observed during sucralfate prophylaxis. This is in keeping with observations made by Martenson et al., who found increased fecal incontinence and use of protective pads in sucralfate-treated patients (17). Our study was terminated before inclusion of the planned number of patients was achieved. Type 2 errors may therefore occur in our statistical calculations. However, the results are in accordance with previously published studies.

Sucralfate, an aluminium salt of the sulfated disaccharide, sucrose octasulfate, has previously been used in the treatment and prophylaxis of peptic gastric and duodenal ulcers

(9,21). It adheres to damaged mucosa, forming a protecting shield against aggressive luminal factors (22), enhances prostaglandin production (23), binds growth factors (10), stimulates

somatostatin release (11), and has angiogenic effects (24,25).

Theoretically, the documented biological actions of sucralfate should make it an ideal drug for radiation proctitis. However, our results do not support the prophylactic use of the

drug during pelvic RT. The reasons for this failure are unclear. It is minimally absorbed from

the intestine and the full oral dose is therefore supposed to reach the rectum. One reason for

failure to reach the target organ could be either intestinal absorption or deactivation. However, it is ineffective even applied locally as enemas in patients (18) and experimentally in rats (26).

Most of the documentation on sucralfate stems from the treatment of peptic ulcers, i.e. in an acid milieu. If an acid pH is a prerequisite, the intestinal milieu may be inadequate in inducing sucralfate activity. Alternatively, the mucosal radiation injury may not be severe enough to bind and activate sucralfate (27).

Sucralfate has also been used in the treatment of chronic, established radiation proctitis. Scientific evidence for this treatment is scarce. A Swedish study demonstrated less chronic diarrrhoea in patients who had taken sucralfate orally during RT (28). This observation was, however, not confirmed in a later study using rectal application (29).

Kochhar et al. (30) compared oral sulfasalazine plus rectal steroids with rectal sucralfate in a prospective, randomised, double-blind study. Sucralfate resulted in significantly better clinical response and had fewer side-effects. An open-label designed study with few patients (31) demonstrated a striking beneficial effect of rectal sucralfate, which is in keeping with the general impression from several case reports (32-36). A dose-dependent decrease in clinical signs and endoscopic and histologic changes has been demonstrated experimentally in rats

(24). Although available studies have methodological shortcomings (37), all demonstrate some beneficial effects and almost no side-effects.

The idea of pharmacologic protection of the rectal mucosa during pelvic RT is still very tempting. The aim is both to ameliorate the acute toxicity symptoms and, certainly more importantly, to minimise chronic changes. Extensive research has been carried out on the use of the aminothiol, amifostine, as recently reviewed (38-41). In an open label study, rectal administration of amifostine during pelvic RT resulted in reduced late toxicity (42). However, no widespread use of amifostine is seen because of an unfavourable side-effect profile and the paucity of well documented clinical benefit for proctitis (40). Exogenous prostaglandin analogues is an interesting option according to a recently published randomised study with misoprostol suppositories during RT, showing significant reduction in acute and chronic radiation proctitis symptoms (43). However, the study comprised only 16 patients, further and larger studies are therefore clearly needed. The somatostatin analogue, octreotide, has previously been used in chronic radiation enteropathy due to its antisecretory, i.e. antidiarrhoic, action (44). Two animal studies (45,46), however, has demonstrated marked reduction in acute mucosal injury as well as chronic structural changes. Clinical studies are lacking.

Even in the acute setting the main therapeutic and prophylactic aim should be the avoidance or amelioration of chronic radiation enteropathy, which accounts for serious conditions in a substantial number of patients after RT. A permanent perturbation of biomolecular mechanisms is initiated during RT and is central in the pathogenesis of chronic radiation pathology. Therefore, prophylactic intervention should take place early during RT.

Future research on prophylaxis can now be based on the expanding knowledge on biomolecular phenomena in the intestine during RT. Possible modifications of these mechanisms have recently been extensively reviewed (47), some of which have undergone preclinical testing (46,48,49).

In summary, our findings indicate that sucralfate does not ameliorate the acute toxicity during pelvic RT. Sucralfate may even worsen diarrhoea in this setting. Our own results are in keeping with data obtained from a meta-analysis of the literature that does not disclose any overt benefit from sucralfate for the amelioration of the main radiation-induced symptom: diarrhoea. The prophylactic use of sucralfate can therefore not be recommended. 11

Legends to figures

Fig.l Development of the symptoms, abdominal pain, tenesmus, and bloating during pelvic RT. No stastically significant increase or difference between groups. Open columns: sucralfate; hatched columns: placebo.

Fig.2 Diarrhoea scores during and immediately after pelvic radiotherapy. Open columns: sucralfate; hatched columns: placebo. Significantly more diarrhoea is found in the sucralfate group at week 2 (p=0,049) and week 6 (p=0.033).

Fig. 3 A summary of published studies comparing sucralfate with placebo as prophylaxis against diarrhoea during pelvic radiotherapy, using the Comprehensive Meta-Analysis data package. 12

Sucralfate group Placebo group (N=20) (N=24) Tumour site Prostate 15 20 Urinary bladder 3 4 Uterine cervix 2 0

Age. Mean + SD 68 years ±6,8 64,1 years + 7,3

Radiation fields (mean dose, Gy, + SD) Pelvic fields 14 (64,70 + 0,67) 15 (65,00 + 0,30) Limited fields 6 (64,33 ± 0,33) 9 (65,25 ± 0,37)

Table 1. Patient characteristics in the two treatment groups.

Reference No. of patients Administration No. of patients Significance Diagnosis randomised S/P Dose (g/day) analysed S/P reported (p) Kneebone et al. N=338 Peroral N=335 NS *) Prostate cancer 166/172 suspension 164/171 6 g/day

O'Brien et al. N=86 Rectal N=86 NS Prostate cancer 41/42 3 g/day 41/42

Martenson et al. N=123 Peroral N=123 NS **) Pelvic cancer 62/61 capsules 62/61 2 g/day

Henriksson et N=70 Peroral N=66 p=0.003 Prostate and al. 35/35 granules 32/34 urinary bladder 6 g/day cancers

Henriksson et N=51 Peroral N=45 Gynecological al. 26/25 8 g/d 22/23 cancers

Hovdenak et al. N=51 Peroral tablets N=44 NS Pelvic cancers 24/27 6 g/d 20/24

Table 4. Studies included in the meta-analysis. N total evaluable patients = 699. *): Significantly more bleeding in the sucralfate group (p=0.001). **): Significantly more nausea (p=0.03), fecal incontinence (p=0.04), and total bowel dysfunction score (p=0.01) in the sucralfate group. Score o -*• w I I i 14

Diarrhoea

Week Citation Year Sucralfate Placebo Effect Lower Upper NTotal PValue 0.1 0.2 0.5 1 2 5 10 Henriksson R 1990 6/22 15/23 .200 .056 713 45 .011 9

Henriksson R 1991 14/32 27/34 .202 .068 .597 66 .003 — O'Brien PC 1997 12/41 15/42 745 .296 1.874 83 .531 Martenson JA 2000 33/62 25/61 1.639 .803 3.345 123 .174 Kneebone J 2001 70/164 68/171 1.128 .730 1.743 335 .588 Hovdenak N 2003 11/24 7/27 2.418 .745 7.845 51 .138

Combined (6) 146/345 157/358 .939 .697 1.264 703 .676

Favors Sucralfate Favors Placebo 15

References

1. Tait DM, Nahum AE, Rigby L, Chow M, Mayles WP, Dearnaley DP, Horwich A. Conformal radiotherapy of the pelvis: assessment of acute toxicity. Radiother Oncol 1993; 29: 117-126.

2. Hovdenak N, Karlsdottir A, Sørbye H, Dahl O. Profiles and time course of acute radiation toxicity symptoms during conformal radiotherapy for cancer of the prostate. Acta Oncologica 2003; in press.

3. Babb RR. Radiation proctitis: a review. Am J Gastroenterol 1996; 91: 1309-1311.

4. Dahl O, Kardamakis D, Lind B, Rosenwald JC. Current status of conformal radiotherapy. Acta Oncol 1996; 35: 41-57.

5. Coleman CN. Radiation oncology - Linking tecnology and biology in the treatment of cancer. Acta Oncol 2002; 41: 6-13.

6. IMRT Collaborative Working Group. Intensity-modulated radiotherapy: current status and issues of interest. Int J Radiat Oncol Biol Phys 2003; 51: 880-914.

7. Muren LP, Mella O, Hafslund R, Dahl O. Norwegian oncologists' expectations of intensity-modulated radiotherapy. Acta Oncol 2003; 41: 562-565.

8. Lechner P, Cesnik H. Abdominopelvic omentopexy: preparation procedure for radiotherapy in rectal cancer. Dis Colon Rectum 1992; 35: 1157-1160.

9. Jensen SL, Funch JP. Role of sucralfate in peptic disease. Dig.Dis 1992; 10: 153-161.

10. Konturek SJ, Brzozowski T, Majka J, Szlachcic A, Bielanski W, Stachura J, Otto W. Fibroblast growth factor in gastroprotection and ulcer healing: interaction with sucralfate. Gut 1993; 34: 881-887.

11. Lucey MR, Park J, DelValle J, Wang LD, Yamada T. Sucrose octasulfate stimulates gastric somatostatin release. Am J Med 1991; 91: 52S-57S.

12. Hovdenak N, Fajardo LF, Hauer-Jensen M. Acute radiation proctitis: a sequential clinicopathologic study during pelvic radiotherapy. Int J Radiat Oncol Biol Phys 2000; 48: 1111-1117. 16

13. Hanauer SB, Robinson M, Pruitt R, Lazenby AJ, Persson T, Nilsson LG, Walton- Bowen K, Haskell LP, Levine JG. Budesonide enema for the treatment of active, distal ulcerative colitis and proctitis: a dose-ranging study. U.S. Budesonide enema study group. Gastroenterology 1998; 115: 525-532.

14. Henriksson R, Arevarn M, Franzen L, Persson H, Stendahl U. Beneficial effects of sucralphate in radiation induced diarrhea. An open randomized study in gynecological cancer patients. Eur.J.Gynaecol.Oncol. 1990; 11: 299-302.

15. Henriksson R, Franzen L, Littbrand B. Prevention of irradiation-induced bowel discomfort by sucralfate: a double-blind, placebo-controlled study when treating localized pelvic cancer. Am J Med 1991. 91: 151 S-l 57S.

16. Kneebone A, Mameghan H, Bolin T, Berry M, Turner S, Kearsley J, Graham P, Fisher R, Delaney G. The effect of oral sucralfate on the acute proctitis associated with prostate radiotherapy: a double-blind, randomized trial. Int J Radiat Oncol Biol Phys 2001;51: 628-635.

17. Martenson JA, Bollinger JW, Sloan JA, Novotny PJ, Unas RE, Michalak JC, Shanahan TG, Mailliard JA, Levitt R. Sucralfate in the prevention of treatment- induced diarrhea in patients receiving pelvic radiation therapy: A North Central Cancer Treatment Group phase III double-blind placebo-controlled trial. J Clin Oncol 2000; 18: 1239-1245.

18. O'Brien PC, Franklin CI, Dear KB, Hamilton CC, Poulsen M, Joseph DJ, Spry N, Denham JW. A phase III double-blind randomised study of rectal sucralfate suspension in the prevention of acute radiation proctitis. Radiother Oncol 1997; 45: 117-123.

19. Stellamans K, Lievens Y, Lambin P, Van den WD, Van den BW, Scalliet P, Hutsebaut L, Haustermans K. Does sucralfate reduce early side effects of pelvic radiation? A double- blind randomized trial. Radiother Oncol 2002; 65: 105-108.

20. Vails A, Pestchen I, Prats C, Pera J, Aragon G, Vidarte M, Algara M. [Multicenter double-blind clinical trial comparing sucralfate vs placebo in the prevention of diarrhea secondary to pelvic irradiation]. Med Clin (Bare.) 1999; 113: 681-684.

21. Hollander D, Tarnawski A. The protective and therapeutic mechanisms of sucralfate. Scand J Gastroenterol 1990; 25: 1-5. 17

22. Nakazawa S, Nagashima R, Saniloff IM. Selective binding of sucralfate to gastric ulcer in man. Dig Dis Sci 1981; 26: 297-300.

23. Romano M, Razandi M, Ivey KJ. Effect of sucralfate and its components on taurocholate-induced damage to rat gastric mucosal cells in tissue culture. Dig.Dis Sci 1990; 35: 467-476.

24. Szabo S, Vattay P, Scarbrough E, Folkman J. Role of vascular factors, including angiogenesis, in the mechanisms of action of sucralfate. Am J Med 1991 91: 158S- 160S.

25. Sandor Z, Nagata M, Kusstatscher S, Szabo S. Stimulation of mucosal glutathione and angiogenesis: new mechanisms of gastroprotection and ulcer healing by sucralfate. Scand J Gastroenterol Suppl 1995; 210: 19-21.

26. Northway MG, Scobey MW, Geisinger KR. Radiation proctitis in the rat. Sequential changes and effects of anti- inflammatory agents. Cancer 1988; 62: 1962-1969.

27. Taal BG, Vales Olmos RA, Boot H, Hoefnagel CA. Assessment of sucralfate coating by sequential scintigraphic imaging in radiation-induced esophageal lesions. Gastro int est. Endosc. 1995; 41: 109-114.

28. Henriksson R, Franzen L, Littbrand B. Prevention and therapy of radiation-induced bowel discomfort. Scand J Gastroenterol Suppl 1992; 191: 7-11

29. O'Brien PC, Franklin CI, Poulsen MG, Joseph DJ, Spry NS, Denham JW. Acute symptoms, not rectally administered sucralfate, predict for late radiation proctitis: longer term follow-up of a phase III trial—Trans- Tasman Radiation Oncology Group. Int J Radiat Oncol Biol Phys 2002; 54: 442-449.

30. Kochhar R, Sriram PV, Sharma SC, Goel RC, Patel F. Natural history of late radiation proctosigmoiditis treated with topical sucralfate suspension. Dig Dis Sci 1999; 44: 973-978.

31. Gul YA, Prasannan S, Jabar FM, Shaker AR, Moissinac K. Pharmacotherapy for chronic hemorrhagic radiation proctitis. World J Surg 2002; 26: 1499-1502.

32. Kochhar R, Sharma SC0BB, Mehta SK. Rectal sucralfate in radiation proctitis. Lancet 1988; 400. 18

33. Ladas SD, Raptis SA. Sucralfate enemas in the treatment of chronic postradiation proctitis. Am J Gastroenterol 1989; 84: 1587-1589.

34. Melko GP, Turco TF, Phelan TF, Sauers NM. Treatment of radiation-induced prodtitis with sucralfate enemas. Ann Pharmacother 1999; 33: 1274-1276.

35. Sasai T, Hiraishi H, Suzuki Y, Masuyama H, Ishida M, Terano A. Treatment of chronic post-radiation proctitis with oral administration of sucralfate. Am J Gastroenterol 1998; 93: 1593-1595.

36. Stockdale AD, Biswas A. Long-term control of radiation proctitis following treatment with sucralfate enemas. Br J Surg 1996; 84: 379.

37. Denton AS, Andreyev HJ, Forbes A, Maher EJ. Systematic review for non-surgical interventions for the management of late radiation proctitis. Br J Cancer 2002; 87: 134-143.

38. Grdina DJ, Murley JS, Kataoka Y. Radioprotectants: current status and new directions. Oncology 2002; 63 Suppl 2: 2-10.

39. Lindegaard JC, Grau C. Has the outlook improved for amifostine as a clinical radioprotector? Radiother Oncol 2000; 57: 113-118.

40. Lindegaard JC. Has the time come for routine use of amifostine in clinical radiotherapy practice? Ada Oncologica 2003; 42: 2-3.

41. Wasserman TH, Brizel DM. The role of amifostine as a radioprotector. Oncology (Huntingt) 2001; 15: 1349-1354.

42. Ben Josef E, Han S, Tobi M, Vargas BJ, Stamos B, Kelly L, Biggar S, Kaplan I. Intrarectal application of amifostine for the prevention of radiation- induced rectal injury. Semin.Radiat Oncol 2002; 12: 81-85.

43. Khan AM, Birk JW, Anderson JC, Georgsson M, Park TL, Smith CJ, Comer GM. A prospective randomized placebo-controlled double-blinded pilot study of misoprostol rectal suppositories in the prevention of acute and chronic radiation proctitis symptoms in prostate cancer patients. Am J Gastroenterol 2000; 95: 1961-1966.

44. Baillie-Johnson HR. Octreotide in the management of treatment-related diarrhoea. Anticancer Drugs 1996; 7 Suppl 1: 11-15. 19

45. Wang J, Zheng H, Sung CC, Hauer-Jensen M. The synthetic somatostatin analogue, octreotide, ameliorates acute and delayed intestinal radiation injury. Int J Radiat Oncol Biol Phys 1999; 45: 1289-1296.

46. Wang J, Zheng H, Hauer-Jensen M. Influence of Short-Term Octreotide Administration on Chronic Tissue Injury, Transforming Growth Factor beta (TGF- beta) Overexpression, and Collagen Accumulation in Irradiated Rat Intestine. J Pharmacol Exp Ther 2001; 297: 35-42.

47. Hauer-Jensen M, Wang J, Denham JW. Mechanisms and modification of the radiation response of gastrointestinal organs. In: Nieder C, Milas L, Ang KK, eds. Modification of radiation response. Berlin, Heidelberg, New York: Springer-Verlag, 2003: 49-72.

48. Ferguson MW. Skin wound healing: transforming growth factor beta antagonists decrease scarring and improve quality. J Interferon Res 1994; 14: 303-304.

49. Wang J, Albertson CM, Zheng H, Fink LM, Herbert JM, Hauer-Jensen M. Short-term inhibition of ADP-induced platelet aggregation by clopidogrel ameliorates radiation- induced toxicity in rat small intestine. Thromb.Haemost. 2002; 87: 122-128.