| HOME | HELP | CONTACT US | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Radiation Oncology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA
Correspondence: Jay S. Loeffler, M.D., Massachusetts General Hospital, Department of Radiation Oncology, 100 Blossom Street, Boston, Massachusetts 02114, USA. Telephone: 617-726-8653; Fax: 617-726-3603; e-mail: jloeffler{at}partners.org
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionFuture Directions for...References |
|---|
Methods. We conducted a thorough review of the Anglo-American medical literature from 1960 to 2001 on the effects of radiotherapy on CVD, hypertension, diabetes mellitus, and IBD.
Results. Fifteen studies were identified that examined the effects of radiation therapy for cancer in patients with CVDs. Thirteen of 15 studies documented greater occurrences of acute and late toxicities (range 7%-100%). Higher rates of complications were noted especially for nonrheumatoid arthritis CVDs. Nine studies evaluated the effects of hypertension and diabetes on radiation tolerance. All nine studies documented higher rates of late toxicities than in a "control" group (range 34%-100%). When patients had both diabetes and hypertension, the risk of late toxicities was even higher. Six studies examined radiation tolerance of patients with IBD irradiated to the abdomen and pelvis. Five of these six studies showed greater occurrences of acute and late toxicities for patients with IBD, even with precautionary measures like reduced fraction size and volume and patient immobilization (13%-29%).
Conclusion. The majority of published studies documented lower radiation tolerance for patients who have CVD, diabetes mellitus, hypertension, and IBD. This may reflect a publication bias, as the majority of these studies are retrospective with small numbers of patients and use different scoring scales for complications. These factors may contribute to an overestimation of true radiation-induced morbidity. Although the paucity of data makes precise estimates difficult, a subset of patients, in particular, those with active CVD, IBD, or a combination of uncontrolled hypertension with type I diabetes, is likely to be at higher risk. Future prospective trials need to document these disease entities when reporting treatment-related complications and also must monitor toxicities associated with quiescent versus active IBD and CVD, type I versus type II diabetes, and levels of hypertension (controlled versus uncontrolled) matched for radiation-specific treatment sites, field size, fractionation, and total dose.
Key Words. Collagen vascular disease • Hypertension • Diabetes • Inflammatory bowel disease • Radiation therapy
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFuture Directions for...References |
|---|
CVDs constitute a heterogeneous consortium of ailments highlighted by alterations in immunoregulation. Although the precise mechanism and pathogenesis of immunological dysfunction remain an enigma, the resulting clinical entities, known as rheumatoid arthritis, systemic lupus erythematosus (SLE), polymyositis, systemic sclerosis, and mixed connective tissue disease, are characterized by a combination of periarticular, perimuscular, and perivascular inflammation [1, 2]. In some institutions, CVDs are considered relative contraindications for radiotherapy for fear of greater risks of acute and late complications [3–6].
Mechanistically, diabetes and hypertension have similar vascular pathologies as CVDs, albeit without the autoimmune etiology [1, 2, 7]. Von Lothar et al. [8] described marked medial hypertrophy with luminal narrowing in the arterioles of patients with hypertension. Similarly, chronic diabetes can cause microvascular occlusive changes leading to capillary hyalinization, arteriolar obliteration, and atherosclerosis with resultant tissue hypoxia [1]. Diabetics also have increased blood viscosity, which can lead to further tissue ischemia [1, 2].
IBD, comprised of ulcerative colitis and Crohn's disease, is a disease marked by inflammation of the bowel, a pattern of familial occurrence, and systemic manifestation [1, 2]. Much like the CVD publications, the literature on IBD treated with external beam radiation has been limited to case reports detailing one to five patients [9, 10]. In general, these reports emphasize the marked intolerance to radiation of patients with IBD.
Mechanistically, IBD shares similar characteristics with the previously described CVD. The microscopic hallmark of Crohn's disease is transmural nodular lymphoid aggregates, accompanied by proliferative changes of the muscularis mucosa, vessels, and nerves of the myenteric plexus [1, 2]. Endothelial edema in small blood vessels, transmural fibrosis, and excess collagen deposition are prominent. In ulcerative colitis, highly vascular granulation tissue develops in denuded mucosal areas. Although the level of fibrosis is usually less than that seen with Crohn's disease, chronic perivascular inflammatory infiltrate in the mucosa and submucosa is appreciable. Consequently, the potential for radiotherapy to add to this milieu of inflammatory vascular changes is certainly plausible.
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFuture Directions for...References |
|---|
|
RESULTS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFuture Directions for...References |
|---|
).
|
Another large series that implicated radiotherapy with significant morbidity in patients with collagen vascular disorder was published by Teo et al. [12]. When compared with other nasopharyngeal carcinoma patients diagnosed and treated in the same hospital, those with dermatomyositis had a higher rate of late toxicities. Two out of 10 patients developed chronic skin necrosis 7-9 months postradiation, and all 10 patients developed chronic indurated subcutaneous fibrosis in the treatment field. All 10 patients were treated to a median dose of 70 Gy at 1.8-2.0 Gy per fraction (fx).
A 1993 study by Ross et al. [13], had the strength of a matched-pair case-control study design. Sixty-one patients with CVD were compared with a matched control group of 61 patients without CVD. These patients were matched for age, stage, and treatment sites. The CVD population had very similar proportions of the different CVDs as the previous MGH study, with 64% rheumatoid arthritis, 21% lupus, 8% dermatomyositis, and 7% scleroderma. Sixty percent of these patients were treated with "curative" doses, exceeding 40 Gy. The remaining patients were treated with "palliative" doses of less than 40 Gy. With a median follow-up period of 18 months, Ross et al. reported 11% and 7% rates of acute toxicities (RTOG > Grade 3) for CVD and non-CVD patients, respectively and 10% and 7% late toxicity (RTOG > grade 3) for CVD and non-CVD patients, respectively. In both the acute and late toxicity comparisons, there was not a statistically significant difference between the CVD and non-CVD arms. In addition, this analysis could not correlate the risk of radiation effects with anatomical sites of irradiation. When the data were analyzed according to specific CVDs, patients with rheumatoid arthritis had a trend toward higher risk for late toxicity than patients in the control arm, at 24% and 5%, respectively (p = 0.125). Interestingly, patients with lupus had a higher risk for acute reactions compared with patients in the control arm, at 36% and 18%, respectively, although this also was statistically nonsignificant (p = 0.5). Overall, this study did not show greater risks for acute and late toxicities for patients with CVD when matched with a control group. Unlike the MGH study, nonrheumatoid patients in this trial did not fair worse than rheumatoid arthritis patients.
Two publications reported the effects of total lymphoid irradiation for refractory rheumatoid arthritis. Trentham et al. [14] reported on 10 patients who were treated with 30 Gy of radiation at 1.8-2.0 Gy/fx. Even with these modest doses, 20% of patients experienced acute diarrhea and 10% had chronic, persistent radiation enteritis. The second publication from Kotzin et al. [15] reported on Stanford University Medical Center's experience with total lymphoid irradiation for rheumatoid patients. Eleven patients were treated with 20 Gy of radiation at 1.5-2.2 Gy/fx. One patient developed pulmonary fibrosis, and two patients developed pericarditis requiring steroid therapy. Although the number of cases was small, there was a trend among the refractory rheumatoid patients toward a greater sensitivity to even the modest doses of radiation used in this treatment protocol.
Rakfal and Deutsch documented six patients with SLE treated with radiotherapy [16]. The doses utilized ranged from 25.4-60 Gy. Although the number of patients was small, this report represents the fourth largest publication of CVD patients treated with radiotherapy. None of the patients treated with radiotherapy developed unexpected acute or late complications. The authors consequently concluded that CVDs are not contraindications for radiation.
A study by Fleck et al. [17] revealed no complications among five patients whose CVD was diagnosed after radiation therapy, whereas three of four patients whose disease was diagnosed before radiation therapy had complications. In addition, Varga et al. [18] described one patient who received two courses of radiation therapy, one before the onset of scleroderma and the other 4 years later. The first course of radiation therapy was well tolerated, but the patient had severe complications after the second course delivered to a different anatomical site. Therefore, these studies suggest that the toxicities associated with radiation therapy in patients with CVD exist only if the CVD is present or "active" at the time of radiation delivery.
Significant systemic exacerbation of the CVD process after localized radiation therapy has also been described in the literature. Robertson et al. [19] described two patients, one with rheumatoid arthritis and the other with scleroderma, who were treated with breast irradiation and subsequently developed diffuse activation of inflammation and fibrosis. The rheumatoid arthritis patient developed progressive joint inflammation outside the radiation port that required gold, nonsteroidal anti-inflammatory drugs, and intraarticular aspirations for symptomatic relief. The patient's rheumatoid arthritis had been quiescent prior to her irradiation, requiring no immunoregulatory drugs. The second patient, with scleroderma, developed severe fibrosis of the esophagus, which was in the radiation port, 27 months post radiation therapy.
Delanian et al. [20] attempted to ameliorate the effects of curative doses of radiotherapy in patients with CVD by significantly reducing the dose. In three patients with diagnosis of quiescent scleroderma, doses were dropped, from 65 Gy typically used, to 40-45 Gy at 1.8 Gy/fx. Two of the three patients developed significant late toxicities even with reduced doses. One patient developed fatal hemorrhagic alveolitis, and the other developed femoral artery thrombosis and skin necrosis. The authors concluded that both significant doses and volumes must be reduced for radiation to be safely utilized in patients with CVD.
Diabetes and Hypertension
There is a paucity of published literature documenting late effects of radiation therapy for patients with underlying diabetes mellitus or hypertension (Table 2). van Nagell et al. [21] reported on 271 patients treated with definitive radiotherapy (paracervical dose range 70-90 Gy) for locally advanced cervical cancer. Their study reported 11 cases of rectovaginal/vesicovaginal fistulas, with a median follow-up of 5 years. Five of the 11 fistulas were due to tumor progression, and six were free of local disease. When examining these six cases, all six fistulas occurred in patients with diabetes or hypertension. In fact, 3 out of 6 patients had diabetes, and the other 3 patients had chronic hypertension. Two of the 6 cases had both diseases.
|
Maruyama et al. [22] also reported on 270 cervical cancer patients treated with at least 40 Gy of radiation. With a median observation period of 60 months, this study documented 9 out of 270 cases of small bowel obstruction unrelated to tumor progression. When examined further for other medical comorbidities, the authors discovered that four of these nine patients had diabetes and another three had both hypertension and diabetes. They consequently concluded that patients with diabetes and hypertension, especially in combination, were at higher risk of late radiation-induced toxicities.
One published study looking exclusively at the effects of diabetes in irradiated patients comes from Herold et al. [23]. This study examined 944 prostate cancer patients treated with definitive radiation therapy, 121 of whom had a history of type I or II diabetes. An average dose of 72 Gy of radiation at 1.8 Gy/fx was delivered utilizing three-dimensional conformal therapy. With a median follow-up of 36 months, they reported significantly higher rates of late genitourinary/gastrointestinal (GU/GI) toxicities in the diabetic population than in the nondiabetic control arm, at 34% and 23%, respectively (p = 0.01). In addition, the diabetic group also developed their complications earlier than the control group, at 10 months and 24 months, respectively. The authors attributed the higher rate of late morbidities to altered microvasculature in these diabetic patients. With an already impaired ability to repair tissue damage due to microvascular occlusive changes, the additional radiation insult and its documented affect on vascular integrity seem to lessen their already precarious ability to heal and perfuse normal tissue.
A study by Goldstein et al. [24] linked radiation-induced vascular changes with impotence, especially in smokers. The authors tested the erectile function in 23 patients who received radiation therapy for prostate cancer. Vascular testing revealed abnormalities in 15 of 23 patients. In two of these patients, arteriography revealed bilateral occlusive disease in the internal pudendal and penile arteries. Cigarette smoking was found among the 15 men whose erectile capacity was decreased. The authors concluded that vasculogenic impotence was the primary cause for erectile abnormality after radiation therapy.
Eifel et al. [25] published on 3,495 patients treated with radiation therapy for Stage I or II carcinomas of the cervix. The authors observed that late GI complications of 1,172 smokers were more than twice that of 2,143 nonsmokers (11.6% versus 5.4% at 10 years; p = 0.0001). The authors also noted that the risk of GU complications was significantly higher for diabetics (5.5%) than for nondiabetics (3.8%) (p = 0.04). Microvascular change from smoking, diabetes, and radiation was postulated as the potential etiology for the observed greater morbidity.
Harwood and Tierie [26] published a series from Holland documenting the effects of diabetes, hypertension, or both diabetes and hypertension on head and neck radiotherapy. A total of 204 localized glottic cancer patients were treated with definitive radiotherapy to a mean dose of 62 Gy in 30 fractions. The authors noted that diabetes and/or hypertension significantly contributed to subsequent risk for major complications. These major late complications included severe edema requiring tracheotomy, necrosis of the larynx, or laryngeal stenosis. When patients were both hypertensive (defined as having >100 mm Hg diastolic blood pressure or on antihypertensive medication) and diabetic (type I or II), the risk of major chronic complication was 67% versus 5% for nondiabetic, normotensive patients (p = 0.01). When diabetics were compared with nondiabetics, the risk for major complications was 30% versus 6% (p = 0.024). When hypertensive patients were compared with normotensive patients, there was a trend toward a higher major complication rate in the former group at 15% and 5.6%, respectively (p = 0.08). Although this study included only 39 patients with diabetes and/or hypertension, it further documents the potential greater toxicities associated with these comorbidities.
Interestingly, a series of articles relating central nervous system (CNS) toxicities with diabetes has been published. Smith et al. [27] reported on a 33-year-old male with a history of poorly controlled type I diabetes treated with definitive radiation therapy for localized nasopharyngeal carcinoma. The patient received 65 Gy in 30 fractions to the nasopharynx with 45 Gy to the spinal cord. The patient subsequently expired with the diagnosis of brain stem necrosis. On postmortem exam, foci of edema, demyelination, and infarction were noted along the brain stem. The small- and medium-sized vessels revealed hyaline deposition, endothelial edema, necrosis, and perivascular lymphocytic infiltration. Although these are typical findings associated with high-dose radiotherapy, the authors speculate that microvascular damage already induced by the patient's prior diabetes mellitus was responsible for his severe complications following standard-dose radiation therapy.
Correlating this pathological finding with clinical data, Debus et al. [28] published on brain stem tolerance to conformal radiotherapy for skull-based tumors. They examined 367 patients treated at MGH with combined photon/proton beam therapy. Seventeen of these patients developed brain stem toxicity. The mean dose to the surface of the brain stem was 64 cGy at 1.8 Gy/fx. Multivariate analysis identified three independent factors as important prognosticators, one of which was the prevalence of diabetes (p < 0.01).
In view of the published literature, diabetes and hypertension, especially in combination, appear to be associated with higher toxicity to radiotherapy. Potential watershed regions of perfusion, like the brain stem and the retina, may be at additional risk for injury [7, 26].
IBD
Apart from the anecdotal case reports, there are two large retrospective analyses evaluating the tolerance to radiation in patients with IBD (Table 3). The largest of these reported 28 patients with histories of IBD and pelvic/abdominal tumors treated with >40-Gy radiotherapy [29]. With a median follow-up of 32 months, severe acute and late effects were 21% and 29%, respectively. The risk for severe late effects was significantly higher for patients treated without precautionary measures (such as lateral decubitus position, proton beam therapy where available, smaller fields, and scheduled rest periods). The risk of late effects at 5 years was reduced by 50% by using some or all of the aforementioned precautionary measures (p = 0.01). The authors did not find a poorer prognosis for patients with Crohn's disease versus ulcerative colitis or active versus quiescent IBD.
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFuture Directions for...References |
|---|
Acutely, radiation therapy affects early responding tissues, like basal dermis and oral/gastric mucosa, by denuding and stunting their ability to proliferate. This acute effect of radiotherapy may also act in conjunction with the already targeted basal layer filled with immune complex and inflammatory cells. Such common targeting may be additive to the typical radiation-induced moist desquamation [13].
In the same manner, the additive injury induced by both radiation and the preexisting CVD process also may help to explain potentially exaggerated late effects seen in some patients following radiotherapy. In the connective tissues, perivascular infiltrates, resulting in immune complex vasculitis (typical of lupus, mixed connective tissue disorder, and diabetes mellitus), and vessel obliteration, via excessive fibroblast proliferation of the dermal connective tissue (typical of scleroderma and rheumatoid arthritis), are nearly universally observed [18]. In a similar manner, late radiation injury affects the same target structures as CVDs. For instance, cultured fibroblasts from animals treated with experimental radiation proliferate at an accelerated rate and produce excessive amounts of collagen in vitro compared with fibroblasts from matched nonirradiated animals [31, 32]. Furthermore, radiation-induced obliteration of capillaries and small vessels is well documented. This principle underlies the routine use of radiation to obliterate arteriovascular malformations in the CNS and capillary bleeding of tumors. Glomus jugulare tumors, which are typically benign outgrowths of jugular vasculature, are routinely controlled with relatively low-dose radiation therapy. In summary, the local effects of radiation therapy on fibroblasts and blood vessels may potentiate the natural progression of CVD.
The intriguing issue of radiation therapy activating a quiescent state of CVD also remains. Several regulatory peptides have been shown to stimulate fibroblast growth, chemotaxis, and connective tissue macromolecule production. Transforming growth factor-ß (TGF-ß) is a cytokine produced by inflammatory cells known to stimulate collagen synthesis [33, 34]. Ionizing radiation may be capable of inducing the production or release of fibroblast-triggering signals from inflammatory cells, endothelial cells, and fibroblasts. In fact, TGF-ß was found to be markedly elevated in irradiated tissue from patients with GI cancer up to 40 weeks following radiation, suggesting that TGF-ß may be implicated in postirradiation fibrosis [33–35]. Furthermore, Rube and colleagues [36–38] correlated acute and long-lasting increases in the expression of TGF-ß in lung tissue with thoracic irradiation. Another potential mechanism for "systemic" activation is through radiation-induced basement membrane damage to capillary endothelial cells. This could potentially expose antigens to distant parts of the body for targeting by the preexisting autoimmune process. In this manner, the release of other potential cytokines from localized radiation therapy may induce a systemic flair in nonirradiated tissue by heightening an already hyperactive immunoregulated milieu. These potential mechanisms of radiation-induced acute and late effects are diagramed in Figure 1.
|
The effect of diabetes and chronic hypertension on radiation tolerance is better understood. Of paramount importance is to correct the underlying metabolic and autonomic disease. Clearly, not all patients with diabetes are at equal risk. The review of literature shows, specifically, that patients with uncontrolled hypertension in addition to type I diabetes may be at the highest risk for radiation-induced morbidity [21–23]. Once again, reduction in dose, dose per fraction, and treatment field size are practical options for patients with suspected vascular disease.
IBD patients have a higher incidence of late morbidities with radiation therapy. However, an encouraging study from Willett et al. [29] showed the value of specialized techniques and precautions in delivering the radiation to minimize bowel toxicity. Patients with longer courses of indolent IBD, quiescent disease at the time of radiation delivery, or on prophylactic anti-inflammatory medication are less likely to develop radiation complications than their counterparts.
In light of these findings, a limitation of this literature review needs to be highlighted. The data obtained for this review were largely gathered, via computer retrieval, from published retrospective data, which did not report associated comorbidities in detail. For instance, reports examining the effects of radiation therapy on diabetes did not distinguish between type I and type II. Published reports on CVDs did not routinely distinguish between active and quiescent disease. Consequently, future studies will need to document the comorbidities in more detail to better illuminate the effects of radiotherapy on nonmalignant diseases.
|
FUTURE DIRECTIONS FOR INVESTIGATIONS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFuture Directions for...References |
|---|
Radiation protectors and modifiers might also be helpful for these patients. The application of such therapy to these patients should also be actively researched [38, 39].
|
REFERENCES |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFuture Directions for...References |
|---|
This article has been cited by other articles:
 |
|
 |
|
  T. M. Pisansky External-beam radiotherapy for localized prostate cancer. N. Engl. J. Med., October 12, 2006; 355(15): 1583 - 1591. [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Whelan Use of Conventional Radiation Therapy As Part of Breast-Conserving Treatment J. Clin. Oncol., March 10, 2005; 23(8): 1718 - 1725. [Full Text] [PDF]
|
|
|
|
 |
|
 K. E. Rieger, W.-J. Hong, V. G. Tusher, J. Tang, R. Tibshirani, and G. Chu Toxicity from radiation therapy associated with abnormal transcriptional responses to DNA damage PNAS, April 27, 2004; 101(17): 6635 - 6640. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | CONTACT US | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| THE ONCOLOGIST | STEM CELLS | CME | ALPHAMED PRESS JOURNALS |
Grade 3) of radiation therapy and CVD
