To the Editors: We have read with interest and some concern the recently published editorial, "We've Got a Treatment, but What's the Disease," by Rosenthal and Glatstein. This editorial enunciates these common anxieties (? "mid-life") about radiosurgery: A) that thedure as currently practiced worldwide, even in the United States, does not in all cases rely on the talents of radiation oncologists; B) that the technique disregards fundamental (? proven) principles about radiobiology, and C) that the authors of the editorial have chosen to ignore a tremendous body of historical and clinical literature relative to outcomes. In fact, long-term clinical data have been published in a wide variety of reports during the last ten years. Their reference list does not include a single article published beyond 1992. Now let's address the first issue. While it is true that the advantages obtained by closed skull focused stereotactic single session irradiation of a small but well-defined intracranial target volume (radiosurgery) were first espoused and practiced by neuosurgeons, the goal was not to impact upon the turf of the radiation oncologist. The goal was to provide a minimally invasive treatment for many problems deep within the brain for which traditional neurosurgical procedures were neither satisfactory nor effective. The tools that allowed neurosurgeons to accomplish this goal included focused particle beams or photon beams generated by the gamma knife or by linear accelerators. The technique also required highly precise (<1 mm) intracranial guiding systems (stereotactic technology). The initial evolution of this technology was cautious. It was based on more than 30 years of experimental and clinical work that preceded its introduction into the worldwide medical community beginning in the mid-1980s. In the United States, the vast majority of centers provide this technology based on the multidisciplinary input of neurosurgeons, radiation oncologists and medical physicists. The team provides both the necessary experience as well as the different perspectives that facilitate safe intervention and effective outcomes. The issue of responsibility in this multidisciplinary medical team should not be obfuscated by individual socio-economic concerns (who's in charge, who gets paid?). The recent purchase of an expensive deice for fractionated frameless radiotherapy by Southwest Medical Center may impact on Rosenthal and Glatstein's recent publication. In fact, stereotactic fractionated radiotherapy is the expensive treatment in search of a disease to pay for it Of greater concern is the authors' misconceptions (misunderstanding?) about the goals of radiosurgery (the second issue). Initially, radiosurgery was created to provide small volume destruction (in this case, true necrosis) of small target volumes withinthe basal ganglia, thalamus, or internal capsule for the treatment of intractable movement disorders, chronic pain, or medically refractory neuroses. With the redesign of the technology, deep-seated neoplasms and vascular malformations became more appealing targets with an entirely different radiobiologic goal. Instead, the goal became radiobiological inactivation of the ability of a tumor cell to divide and multiply (for tumors) or progressive luminal closure induced by endothelial hyperplasia (in the case of vascular malformations). Preservation of the surrounding normal brain (a feature brought about by the very sharp fall-off of the radiation dose delivered to small volumes with precise technology) reduced the risk of complications to normal brain, especially in contrast to surgical extirpation. Fractionated radiation therapy has rarely been an alternative to the usage of radiosurgery for these conditions. For malignant tumors, radiosurgery is most often used in conjunction with fractionated radiation therapy to take advantage of the single fraction destructive effects of radiosurgery followed or preceded by conventional fractionated radiation therapy. Such an approach enhances the likelihood of a satisfactory response based on the standard 4 Rs of curret radiobiological thinking. Stereotactic radiosurgery is a single "fraction" treatment; fractionated stereotactic radiosurgery is an absolute oxymoron. Certainly, renewed interest in the risk-benefit of fractionated radiation therapy is a logical outgrowth of the current tremendousave of enthusiasm for radiosurgery. In fact, the growth of radiosurgery has made radiation oncologists re-think their own practice of conventional radiation therapy. Similarly, it has had a profound impact on procedure selection by neurological surgeons. The third issue is addressed by the enormous volume of literature relative to outcomes in vascular malformations, malignant tumors, and benign tumors. The usage of radiosurgical technology should continue to stimulate thoughtful investigators to advance outcomes in these difficult conditions and reduce the risks of standard surgical techniques. It must be based on a collegial and multidisciplinary approach. The timing of Rosenthal and Glatstein's editorial was a mystery, appearing almost atavistic, especially considering the enormous growth of understanding and experience accumulated in the ten-year interval since both linac and gamma knife radiosurgical tecnologies became available in North America. AUTHORS' RESPONSE: In response to the Letter to the Editor by Lunsford, Flickinger and Larson, our main objectives in writing that article were twofold. The first was to review those principles of fractionation derived from a near century's experience in clinical radiobioloy. We have learned over and over again that, in general, hypofractionation leads to poorer tumor control, and more frequent and severe normal tissue complications. We believe that this point was, perhaps, not as fully appreciated during the development of radiosurgery because of a more surgical rather than radiotherapeutic influence. The second objective regards the safety issues of the even more widespread use of radiosurgery for brain tumors during the period when long-term follow-up data (ten years or more) are still emerging. Radiosurgery is in common use at our institution, the University of Pennsylvania Medical Center. We in no way wish to diminish the established safety and effectiveness of radiosurgery for arteriovenous malformations (AVMs). Additionally, we wholeheartedly encourage continued investigation for benign and malignant intracranial tumors. Our chief concern is the objective scientific validation of radiosurgery for these latter applications in prospective trials which have adequate long-term follow-up to establish safety. The central nervous system is the most unforgiving organ in terms of late radiation effects. Are all patients undergoing radiosurgery for benign tumors being accurately informed of the good results of modern fractionated radiotherapy, and those who undergo it for malignant tumors, that objective phase III validation and long-term safety data are NOT yet available? It frightens us even more that Lunsford et al. state, "In fact, the growth of radiosurgery has made radiation oncologists re-think their own practice of conventional raiation therapy." Just when do we evaluate the new clothes for the emperor Lunsford et al. tell us that radiosurgery technology has been "re-designed" with ".an entirely different radiobiologic goal. (the) inability of a tumor cell to divide and multiply." Radiation oncologists have long been taught as residents that raiologists accept the definition of "radiobiologic cell death" as the loss of continual clonogenicity. We all strive to this end in the treatment of tumors, but we are concerned about the extrapolation of the accepted application of radiosurgery for AVM tumors. More than 10,000 patients have had radiosurgery for brain tumors. Many of these have been benign, and more than 1,000 patients were treated with protons at the Harvard Cyclotron Unit, mostly for pituitary adenoma. Their experience has established safety, but the data for photon radiosurgery is not as large or mature, and one wonders how much photon radiosurgery adds to the excellent results achievable by conventional fractionated radiotherapy, especially for patients with pituitary tumors. With respect to malignant primary tumors or metastases, there have been fewer patients so treated. We recognize that longer term follow-up is not as important an issue for this unfortunate patient population whose survival period is generally short. Nonetheless, we reiterate that: A) hypofractionation has historically been shown to lead both to decreased control and increased complications, and B) that the higher the grade of a brain tumor, the more difficulty we have in localizing its extensions, especially when a treatment volume is <3 cc. There is absolutely no evidence that fractionated stereotactic treatment is an "oxymoron." Those data are only now beginning to emerge. It makes sense to encourage the investigation of radiosurgery as a boost followingonventional fractionated radiotherapy, or, for those who had the wherewithal to develop practical and cost-effective methods to treat with "fractionated radiosurgery" (read "stereotactic radiotherapy") to use those principles of clinical radiobiology twe have learned painstakingly over the last century to drive clinical investigation, and not rely solely on the impetus of new technology. Such investigation is ongoing at our institution, as we strive for the scientific evaluation of the comparative efficacy and long-term safety of radiosurgery for brain tumors. Had Coutard and Baclesse not pioneered fractionation, radiotherapy probably would have fallen into oblivion due to the morbidities of single shot treatment. Indeed, much of the first half of this century was spent learning that doses large enough to sterilize a mass of tumor cells (10 logs) cannot be predictably given safely. Instead, fractionation evolved which permitted us to exploit repopulation, redistribution, reoxygenation and repair. The use of these large single doses remains, at least in our minds, investigational in the treatment of especially malignant tumors. This is the way this subject is presented to patients here.