As the number of long-term cancer survivors increases, late complications of therapy will become an increasingly important concern to both patients and physicians. In administering radiation therapy, the physician typically has attempted to prevent complications primarily by limiting the irradiated dose and volume. Unfortunately, the relationship between radiation dose, volume of tissue irradiated, development of complications, and tumor control is complex and not precisely defined for most normal tissues and malignancies. Unlike medical oncology, there are few formal prospective studies in radiation oncology designed to establish the maximum-tolerated dose for a particular tumor or normal tissue. Commonly accepted normal tissue tolerance doses of radiation, for the most part, have been derived empirically, often with little supporting data. Although these estimates may be reasonably generalized, there remains considerable individual variation in radiation sensitivity. Recently, however, much progress has been made toward better elucidating the molecular mechanisms underlying radiation injury. Late radiation damage in most tissues is characterized by loss of parenchymal cells and excessive formation of fibrous tissue. For years, it was taught that radiation fibrosis is a permanent irreversible condition, but its underlying mechanism remained uncertain. It has been known for decades that the biologic effects of ionizing radiation begin almost instantaneously with the generation of reactive oxygen species. More recently, however, we are beginning to learn that these immediate biochemical events lead to clinically and histologically recognizable injury through a series of genetic and molecular responses. This process is dynamic and involves a number of proinflammatory cytokines, profibrotic cytokines, and chemokines produced by a variety of cell types, including macrophages, epithelial cells, and fibroblasts. Furthermore, these events seem to be sustained for months or even years after therapy is completed. For example, the presence of progressive hypoxia has been noted after radiation of the lung and spinal cord. Because hypoxia itself is known to generate reactive oxygen species, promote inflammation and vascular damage, activate profibrotic cytokines, and promote collagen formation, it has been suggested that postradiation hypoxia may be an important contributor to maintenance of the injured phenotype. These findings have also suggested that, because the process is dynamic, it may not be irreversible. In this issue, Delanian et al contribute significantly to the growing body of knowledge that indicates that, indeed, radiation-induced fibrosis is at least partially reversible. They also demonstrate for the first time that long-term antifibrotic therapy will be needed to sustain benefit. The authors report on a retrospective series of 44 patients with 55 distinct superficial progressive fibrotic lesions induced by radiation therapy for breast cancer. The patients were administered the combination of pentoxifylline and vitamin E twice daily for either 6 to 12 months or 24 to 48 months. Both drugs have antioxidant properties, and pentoxifylline may also have antifibrotic effects, although neither agent is clearly effective as a single agent in reversing radiation-induced fibrosis. The main end points were reduction in the size of the fibrotic region and reduction in the global score of late injury. Delanian et al demonstrated the following: both treatment regimens produced significant improvements in both outcome measures; the maximum regression to be expected is approximately 69%; improvement occurred sooner in younger lesions ( 6 years since radiotherapy); and stopping the drugs resulted in a rebound effect that was more severe in the patients treated for a short duration. Tolerance of the regimen was good, and no patient discontinued therapy as a result of toxicity. Although oncologists should be encouraged by the mounting evidence that radiation-induced fibrosis may be reversible, many questions remain unanswered. First, JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 23 NUMBER 34 DECEMBER 1 2005