Postradiotherapy Pelvic Fractures: Title and subTitle BreakCause for Concern or Opportunity for Future Research?

Radiotherapy has proven to be an indispensable tool in the curative management of malignant disease, especially for primary carcinomas of the pelvis including prostate, rectal, cervical, endometrial, and anal cancers. In this setting, radiotherapy is used either alone or in combination with systemic, hormonal, and/or surgical therapies. The ultimate goal of this radiation-based therapy is to provide long-term disease control or a “cure” with as little treatment-related morbidity as possible. Yet, the study of long-term toxicity associated with treatments that include radiation therapy has received little attention over the years.

Radiation toxicity can roughly be divided into early and late effects. Early or acute effects, such as nausea, skin reactions, diarrhea, and neutropenia, tend to be temporary and, for the most part, resolve shortly after the completion of therapy. Late effects, such as connective tissue fibrosis and secondary malignancies, can occur long after the completion of radiation therapy. The causes for late radiation injuries are not completely understood. The 2 main theories consist of damage to the cellular matrix and vascular injury.¹ However, the etiology for long-term sequelae is probably much more complex and likely involves a cascade of cellular, vascular, and cytokine changes induced by radiation.

In this issue of JAMA, Baxter and colleagues² provide compelling evidence for what many radiation oncologists have long believed, despite a paucity of literature about modern radiation delivery: that pelvic radiotherapy increases the risk of bone fractures. Of the 6428 women in the study, 556 were diagnosed with anal cancer, 1605 with cervical cancer, and 4267 with rectal cancer. Overall, 44.4% received radiation therapy and 55.6% did not. The cumulative incidence of pelvic fractures within the first 5 years of the study was increased for women in the irradiated group compared with women in the nonirradiated group: 14% vs 7.5% for women with anal cancer, 8.2% vs 5.9% for cervical cancer, and 11.2% vs 8.7% for rectal cancer. As Baxter et al show, an increase in the relative risk of pelvic fractures associated with the administration of pelvic radiotherapy is a significant public health issue that deserves attention.

However, there are several limitations to these findings. First, as noted by the authors, this work represents a retrospective cohort study that could lead to undetected bias. Despite controlling for age and race (as surrogates for the development of osteoporosis), there may be other potential confounding factors. Several medications including steroids, heparin, and thyroid hormone therapy have been known to contribute to the risk of osteoporosis^{3 - 5} and may have acted synergistically with radiation therapy. Additionally, women who had received pelvic irradiation for their cervical cancer may have undergone an earlier total abdominal hysterectomy and bilateral oophorectomy. Osteoporosis from estrogen deficiency may have already initiated bone resorption further contributing to the late effects of radiotherapy.

Second, the other significant limitation is the lack of discussion regarding the effects of combined radiotherapy and chemotherapy. There exist 2 different definitions regarding radiotherapy toxicity: (1) effects directly attributable to radiotherapy or (2) any adverse events after radiotherapy. When evaluating patients after radiotherapy for toxicity, clinicians often conclude that the effect is due to the radiotherapy alone, potentially leading to a mistaken diagnosis. In a recent article explaining 465 gastrointestinal complaints among 265 patients after pelvic radiotherapy who had been referred to a gastroenterologist, 155 complaints were unrelated to the previous pelvic radiotherapy.⁶

Other potential confounders may have contributed to the increase in pelvic fractures associated with radiation. Likewise, the use of concurrent chemotherapy may have increased the pelvic fracture risk. Combined modality therapy is associated with increased acute and late toxicity rates as compared with radiotherapy alone.⁷ All 3 cancers included in this study (anal, cervical, and rectal) are most often managed with combined therapy. The correct term for any toxicity associated with the combination of chemotherapy and radiotherapy is treatment-related toxicity. It seems counterintuitive that practicing oncologists discuss how the combination of chemotherapy and radiotherapy increases tumor control, but then commonly attribute any local toxicity associated with therapy to be radiation-related. An example is small bowel toxicity reported in patients receiving 5-fluorouracil and pelvic radiotherapy. Although 5-fluorouracil is known to cause gastrointestinal morbidity,⁸ it is common to describe chemoradiation-induced small bowel toxicity as radiation enteritis. The correct term should be treatment-related enteritis. The study by Baxter et al may have strengthened its conclusions if the patient population were limited to those who received only pelvic radiotherapy and not combined chemotherapy.

Despite these shortcomings, this study effectively demonstrates that patients with pelvic malignancies who receive therapy that includes radiation have an increased risk of bone fractures. Moreover, it is probably reasonable to conclude that the majority of the osseous effects are due to the radiation, although these effects certainly may have been potentiated by concurrent systemic therapy. The fact that no statistically significant difference was found in the rate of arm or spine fractures between the irradiated and nonirradiated groups supports this view.

How should clinicians use the knowledge of an increased risk of pelvic fractures associated with radiotherapy? This potential morbidity should be discussed with the patient at the time of radiation oncology consultation and factored into informed decision making and informed consent regarding the use of radiotherapy. More important, patients who have received prior pelvic radiation must receive long-term follow-up examinations, and must be carefully assessed when pelvic pain appears.

Pelvic insufficiency fractures may be a common postradiotherapy toxicity. These hairline fractures result from radiation-induced weakening of pelvic bone and may occur by virtue of the pelvis being unable to support body weight. In one study of postmenopausal women who received radiation therapy for advanced cervical cancer, the pelvic insufficiency fracture rate was 17%.⁹ Insufficiency fractures can be asymptomatic or associated with pain.¹⁰ Painful lesions tend to be associated with multiple fractures on magnetic resonance imaging. The majority of symptoms will improve with conservative measures. Primary care physicians and oncologists should be aware that pelvic insufficiency can occur after pelvic radiotherapy. Some patients who initially are diagnosed with metastatic disease related to pelvic pain and a positive bone scan ultimately may be diagnosed with an insufficiency fracture. This can lead to considerable patient anxiety and, in the extreme case, to unnecessary and potentially harmful application of anticancer therapy. Moreno and colleagues reviewed 8 symptomatic cases of pelvic insufficiency fractures and noted 5 of these patients to be originally diagnosed with metastasis.¹¹ Isolated pelvic bone metastasis in the absence of other metastatic disease is an uncommon event in cervical, rectal, and anal cancer. If patients develop pain and isolated pelvic bone scan findings after pelvic radiotherapy, conservative measures should be the first line of therapy.

In addition, there should be attempts to minimize the known risks for pelvic fractures. These include aggressive treatment of underlying osteopenia in concordance with current guidelines.¹² Efforts for reducing osseous effects should be pursued with secondary end points, such as bone density used to guide such study.

An important area of research involves the prevention of cancer treatment morbidity with improvements in patient quality of life. One solution is to minimize radiotherapy to only the at-risk regions—commonly referred to as the clinical target volume. In general, the clinical target volume is surrounded by a margin of normal tissue to account for patient set up variations and internal organ motion. This volume is commonly referred to as the planning target volume. The bony structures of the pelvis are rarely part of the clinical target volume for cancers of the rectum, cervix, and anus. Emerging evidence suggests that newer radiation treatment techniques have the ability to reduce the volume of normal tissue irradiated outside the clinical target volume.¹³ The most common approach is known as intensity modulated radiation therapy, and has been used in pelvic malignancies with reductions in toxicity.¹³ Although very promising, the use of intensity modulated radiation therapy for the curative management of cervical, rectal, and anal cancer requires further careful prospective data before being adopted as standard therapy and requires answering questions about the optimal clinical target volume, quality assurance, and costs. Despite these limitations, intensity modulated radiation therapy will be tested prospectively in a multi-institutional setting by the Radiation Therapy Oncology Group to administer postoperative radiation therapy in cervical and endometrial cancers, as well for the definitive therapy of anal cancers (W.S., unpublished data, 2005).

In addition to improved targeting of radiotherapy, consideration of agents that may reduce toxicity or improve the osseous environment after radiotherapy may be warranted. Amifostine (WR 2721) is a free radical scavenger that appears to be taken up in normal tissue preferentially over tumor cells, providing normal tissue radioprotection without protection of the cancer.¹⁴ This agent has been found to reduce xerostomia in patients who received head and neck radiotherapy¹⁵ and decreases toxicity in patients with ovarian cancer receiving chemotherapy.¹⁶ Preclinical data in a rat model suggest that amifostine reduces bone loss by decreasing osteoclastic bone resorption¹⁷ and improves bone mineral density after radiotherapyas compared with animal controls.¹⁸ Amifostine appears to improve other effects of pelvic radiotherapy¹⁹ such as mucosal toxicity and is under investigation in patients receiving radiotherapy for cervical cancer in a prospective multicenter Radiation Therapy Oncology Group trial.²⁰ Other attempts to improve the osseous environment after radiotherapy could include the prophylactic use of biphosphonates in nonosteopenic patients. The Radiation Therapy Oncology Group is currently developing a phase 3 trial (W.S. and L.K., unpublished data, 2005) to evaluate the potential benefit of bisphosphonate therapy in the prevention of osteoporosis-associated bone fractures in patients receiving hormonal therapy and pelvic radiation for locally advanced adenocarcinoma of the prostate.

In conclusion, Baxter et al have provided compelling evidence for a significant increase in pelvic fracture risk with the use of pelvic radiotherapy as a component of definitive cancer management. The morbidity associated with pelvic fractures and the widespread use of pelvic radiotherapy make research into reducing such osseous effects a high priority.